tInitial revision - plan9port - [fork] Plan 9 from user space
HTML git clone git://src.adamsgaard.dk/plan9port
DIR Log
DIR Files
DIR Refs
DIR README
DIR LICENSE
---
DIR commit b2cfc4e2e71d0f0a5113ddfbd93c8285cc4d74e4
DIR parent 5f7d5e8d1899f41b0e5366c0251530ea1dc753d0
HTML Author: rsc <devnull@localhost>
Date: Tue, 30 Sep 2003 17:47:41 +0000
Initial revision
Diffstat:
A include/bio.h | 79 +++++++++++++++++++++++++++++++
A include/cursor.h | 7 +++++++
A include/draw.h | 520 +++++++++++++++++++++++++++++++
A include/event.h | 63 +++++++++++++++++++++++++++++++
A include/fmt.h | 100 +++++++++++++++++++++++++++++++
A include/frame.h | 85 +++++++++++++++++++++++++++++++
A include/keyboard.h | 36 +++++++++++++++++++++++++++++++
A include/lib9.h | 254 +++++++++++++++++++++++++++++++
A include/libc.h | 2 ++
A include/mouse.h | 44 +++++++++++++++++++++++++++++++
A include/regexp9.h | 71 +++++++++++++++++++++++++++++++
A include/thread.h | 132 +++++++++++++++++++++++++++++++
A include/u.h | 0
A include/utf.h | 51 +++++++++++++++++++++++++++++++
A man/man3/bio.3 | 336 +++++++++++++++++++++++++++++++
A man/man3/fmtinstall.3 | 346 +++++++++++++++++++++++++++++++
A man/man3/ioproc.3 | 179 +++++++++++++++++++++++++++++++
A man/man3/isalpharune.3 | 47 +++++++++++++++++++++++++++++++
A man/man3/print.3 | 469 +++++++++++++++++++++++++++++++
A man/man3/regexp9.3 | 227 +++++++++++++++++++++++++++++++
A man/man3/rune.3 | 187 +++++++++++++++++++++++++++++++
A man/man3/runestrcat.3 | 65 +++++++++++++++++++++++++++++++
A man/man3/thread.3 | 576 +++++++++++++++++++++++++++++++
A package | 19 +++++++++++++++++++
A src/lib9/LICENSE | 258 +++++++++++++++++++++++++++++++
A src/lib9/Make.Darwin-PowerMacintosh | 6 ++++++
A src/lib9/Make.FreeBSD-386 | 7 +++++++
A src/lib9/Make.HP-UX-9000 | 6 ++++++
A src/lib9/Make.Linux-386 | 7 +++++++
A src/lib9/Make.NetBSD-386 | 7 +++++++
A src/lib9/Make.OSF1-alpha | 6 ++++++
A src/lib9/Make.SunOS-sun4u | 2 ++
A src/lib9/Make.SunOS-sun4u-cc | 6 ++++++
A src/lib9/Make.SunOS-sun4u-gcc | 6 ++++++
A src/lib9/Makefile | 120 +++++++++++++++++++++++++++++++
A src/lib9/Makefile.MID | 49 +++++++++++++++++++++++++++++++
A src/lib9/_exits.c | 9 +++++++++
A src/lib9/argv0.c | 4 ++++
A src/lib9/await.c | 105 +++++++++++++++++++++++++++++++
A src/lib9/encodefmt.c | 69 ++++++++++++++++++++++++++++++
A src/lib9/errstr.c | 68 +++++++++++++++++++++++++++++++
A src/lib9/exits.c | 10 ++++++++++
A src/lib9/ffork-FreeBSD.c | 33 +++++++++++++++++++++++++++++++
A src/lib9/ffork-Linux.c | 39 +++++++++++++++++++++++++++++++
A src/lib9/getcallerpc-386.c | 7 +++++++
A src/lib9/getfields.c | 36 +++++++++++++++++++++++++++++++
A src/lib9/lib9.h | 246 +++++++++++++++++++++++++++++++
A src/lib9/lock.c | 54 +++++++++++++++++++++++++++++++
A src/lib9/malloctag.c | 15 +++++++++++++++
A src/lib9/mallocz.c | 14 ++++++++++++++
A src/lib9/mkfile | 2 ++
A src/lib9/nrand.c | 17 +++++++++++++++++
A src/lib9/qlock.c | 360 +++++++++++++++++++++++++++++++
A src/lib9/rand.c | 89 +++++++++++++++++++++++++++++++
A src/lib9/readn.c | 21 +++++++++++++++++++++
A src/lib9/rendez.c | 180 +++++++++++++++++++++++++++++++
A src/lib9/strecpy.c | 16 ++++++++++++++++
A src/lib9/sysfatal.c | 20 ++++++++++++++++++++
A src/lib9/tas-386.s | 6 ++++++
A src/lib9/test.c | 8 ++++++++
A src/lib9/testfork.c | 21 +++++++++++++++++++++
A src/lib9/tokenize.c | 106 ++++++++++++++++++++++++++++++
A src/lib9/u16.c | 52 +++++++++++++++++++++++++++++++
A src/lib9/u32.c | 109 +++++++++++++++++++++++++++++++
A src/lib9/u64.c | 126 +++++++++++++++++++++++++++++++
A src/lib9/wait.c | 30 ++++++++++++++++++++++++++++++
A src/lib9/werrstr.c | 13 +++++++++++++
A src/libbio/LICENSE | 258 +++++++++++++++++++++++++++++++
A src/libbio/Make.FreeBSD-386 | 7 +++++++
A src/libbio/Make.HP-UX-9000 | 6 ++++++
A src/libbio/Make.Linux-386 | 7 +++++++
A src/libbio/Make.NetBSD-386 | 7 +++++++
A src/libbio/Make.OSF1-alpha | 6 ++++++
A src/libbio/Make.SunOS-sun4u | 2 ++
A src/libbio/Make.SunOS-sun4u-cc | 6 ++++++
A src/libbio/Make.SunOS-sun4u-gcc | 6 ++++++
A src/libbio/Makefile | 108 +++++++++++++++++++++++++++++++
A src/libbio/Makefile.MID | 37 +++++++++++++++++++++++++++++++
A src/libbio/NOTICE | 27 +++++++++++++++++++++++++++
A src/libbio/README | 7 +++++++
A src/libbio/bbuffered.c | 20 ++++++++++++++++++++
A src/libbio/bcat.c | 42 +++++++++++++++++++++++++++++++
A src/libbio/bfildes.c | 9 +++++++++
A src/libbio/bflush.c | 33 +++++++++++++++++++++++++++++++
A src/libbio/bgetc.c | 52 +++++++++++++++++++++++++++++++
A src/libbio/bgetd.c | 36 +++++++++++++++++++++++++++++++
A src/libbio/bgetrune.c | 47 +++++++++++++++++++++++++++++++
A src/libbio/binit.c | 142 +++++++++++++++++++++++++++++++
A src/libbio/bio.3 | 336 +++++++++++++++++++++++++++++++
A src/libbio/bio.h | 79 +++++++++++++++++++++++++++++++
A src/libbio/boffset.c | 25 +++++++++++++++++++++++++
A src/libbio/bprint.c | 28 ++++++++++++++++++++++++++++
A src/libbio/bputc.c | 29 +++++++++++++++++++++++++++++
A src/libbio/bputrune.c | 23 +++++++++++++++++++++++
A src/libbio/brdline.c | 94 +++++++++++++++++++++++++++++++
A src/libbio/brdstr.c | 112 +++++++++++++++++++++++++++++++
A src/libbio/bread.c | 45 +++++++++++++++++++++++++++++++
A src/libbio/bseek.c | 56 +++++++++++++++++++++++++++++++
A src/libbio/bundle.ports | 45 +++++++++++++++++++++++++++++++
A src/libbio/bwrite.c | 38 +++++++++++++++++++++++++++++++
A src/libbio/lib9.h | 12 ++++++++++++
A src/libbio/mkfile | 1 +
A src/libbio/rpm.spec | 30 ++++++++++++++++++++++++++++++
A src/libfmt/LICENSE | 19 +++++++++++++++++++
A src/libfmt/Make.Darwin-PowerMacint… | 6 ++++++
A src/libfmt/Make.FreeBSD-386 | 7 +++++++
A src/libfmt/Make.HP-UX-9000 | 6 ++++++
A src/libfmt/Make.Linux-386 | 7 +++++++
A src/libfmt/Make.NetBSD-386 | 7 +++++++
A src/libfmt/Make.OSF1-alpha | 6 ++++++
A src/libfmt/Make.SunOS-sun4u | 2 ++
A src/libfmt/Make.SunOS-sun4u-cc | 6 ++++++
A src/libfmt/Make.SunOS-sun4u-gcc | 6 ++++++
A src/libfmt/Makefile | 134 +++++++++++++++++++++++++++++++
A src/libfmt/Makefile.MID | 63 +++++++++++++++++++++++++++++++
A src/libfmt/NOTICE | 19 +++++++++++++++++++
A src/libfmt/README | 19 +++++++++++++++++++
A src/libfmt/bundle.ports | 51 +++++++++++++++++++++++++++++++
A src/libfmt/charstod.c | 85 +++++++++++++++++++++++++++++++
A src/libfmt/dofmt.c | 558 +++++++++++++++++++++++++++++++
A src/libfmt/dorfmt.c | 61 +++++++++++++++++++++++++++++++
A src/libfmt/errfmt.c | 28 ++++++++++++++++++++++++++++
A src/libfmt/fltfmt.c | 610 +++++++++++++++++++++++++++++++
A src/libfmt/fmt.c | 221 +++++++++++++++++++++++++++++++
A src/libfmt/fmt.h | 100 +++++++++++++++++++++++++++++++
A src/libfmt/fmtdef.h | 121 +++++++++++++++++++++++++++++++
A src/libfmt/fmtfd.c | 46 +++++++++++++++++++++++++++++++
A src/libfmt/fmtfdflush.c | 33 +++++++++++++++++++++++++++++++
A src/libfmt/fmtinstall.3 | 346 +++++++++++++++++++++++++++++++
A src/libfmt/fmtlock.c | 28 ++++++++++++++++++++++++++++
A src/libfmt/fmtprint.c | 47 +++++++++++++++++++++++++++++++
A src/libfmt/fmtquote.c | 262 +++++++++++++++++++++++++++++++
A src/libfmt/fmtrune.c | 40 +++++++++++++++++++++++++++++++
A src/libfmt/fmtstr.c | 65 +++++++++++++++++++++++++++++++
A src/libfmt/fmtvprint.c | 46 +++++++++++++++++++++++++++++++
A src/libfmt/fprint.c | 28 ++++++++++++++++++++++++++++
A src/libfmt/mkfile | 1 +
A src/libfmt/nan.h | 4 ++++
A src/libfmt/nan64.c | 72 +++++++++++++++++++++++++++++++
A src/libfmt/pow10.c | 57 +++++++++++++++++++++++++++++++
A src/libfmt/print.3 | 469 +++++++++++++++++++++++++++++++
A src/libfmt/print.c | 28 ++++++++++++++++++++++++++++
A src/libfmt/rpm.spec | 34 +++++++++++++++++++++++++++++++
A src/libfmt/runefmtstr.c | 65 +++++++++++++++++++++++++++++++
A src/libfmt/runeseprint.c | 30 ++++++++++++++++++++++++++++++
A src/libfmt/runesmprint.c | 30 ++++++++++++++++++++++++++++++
A src/libfmt/runesnprint.c | 31 +++++++++++++++++++++++++++++++
A src/libfmt/runesprint.c | 30 ++++++++++++++++++++++++++++++
A src/libfmt/runevseprint.c | 39 +++++++++++++++++++++++++++++++
A src/libfmt/runevsmprint.c | 37 +++++++++++++++++++++++++++++++
A src/libfmt/runevsnprint.c | 38 +++++++++++++++++++++++++++++++
A src/libfmt/seprint.c | 27 +++++++++++++++++++++++++++
A src/libfmt/smprint.c | 27 +++++++++++++++++++++++++++
A src/libfmt/snprint.c | 28 ++++++++++++++++++++++++++++
A src/libfmt/sprint.c | 27 +++++++++++++++++++++++++++
A src/libfmt/strtod.c | 539 +++++++++++++++++++++++++++++++
A src/libfmt/strtod.h | 4 ++++
A src/libfmt/test.c | 39 +++++++++++++++++++++++++++++++
A src/libfmt/vfprint.c | 31 +++++++++++++++++++++++++++++++
A src/libfmt/vseprint.c | 37 +++++++++++++++++++++++++++++++
A src/libfmt/vsmprint.c | 36 +++++++++++++++++++++++++++++++
A src/libfmt/vsnprint.c | 37 +++++++++++++++++++++++++++++++
A src/libregexp/LICENSE | 258 +++++++++++++++++++++++++++++++
A src/libregexp/Make.Darwin-PowerMac… | 6 ++++++
A src/libregexp/Make.FreeBSD-386 | 7 +++++++
A src/libregexp/Make.HP-UX-9000 | 6 ++++++
A src/libregexp/Make.Linux-386 | 7 +++++++
A src/libregexp/Make.NetBSD-386 | 7 +++++++
A src/libregexp/Make.OSF1-alpha | 6 ++++++
A src/libregexp/Make.SunOS-sun4u | 2 ++
A src/libregexp/Make.SunOS-sun4u-cc | 6 ++++++
A src/libregexp/Make.SunOS-sun4u-gcc | 6 ++++++
A src/libregexp/Makefile | 106 ++++++++++++++++++++++++++++++
A src/libregexp/Makefile.MID | 34 +++++++++++++++++++++++++++++++
A src/libregexp/NOTICE | 25 +++++++++++++++++++++++++
A src/libregexp/README | 7 +++++++
A src/libregexp/bundle.ports | 51 +++++++++++++++++++++++++++++++
A src/libregexp/lib9.h | 6 ++++++
A src/libregexp/mkfile | 1 +
A src/libregexp/regaux.c | 76 +++++++++++++++++++++++++++++++
A src/libregexp/regcomp.c | 557 +++++++++++++++++++++++++++++++
A src/libregexp/regcomp.h | 74 +++++++++++++++++++++++++++++++
A src/libregexp/regerror.c | 14 ++++++++++++++
A src/libregexp/regexec.c | 219 +++++++++++++++++++++++++++++++
A src/libregexp/regexp9.3 | 227 +++++++++++++++++++++++++++++++
A src/libregexp/regexp9.7 | 150 +++++++++++++++++++++++++++++++
A src/libregexp/regexp9.h | 71 +++++++++++++++++++++++++++++++
A src/libregexp/regsub.c | 62 +++++++++++++++++++++++++++++++
A src/libregexp/rpm.spec | 34 +++++++++++++++++++++++++++++++
A src/libregexp/rregaux.c | 26 ++++++++++++++++++++++++++
A src/libregexp/rregexec.c | 213 +++++++++++++++++++++++++++++++
A src/libregexp/rregsub.c | 55 +++++++++++++++++++++++++++++++
A src/libregexp/test.c | 46 +++++++++++++++++++++++++++++++
A src/libregexp/test2.c | 20 ++++++++++++++++++++
A src/libutf/LICENSE | 13 +++++++++++++
A src/libutf/Make.Darwin-PowerMacint… | 6 ++++++
A src/libutf/Make.FreeBSD-386 | 7 +++++++
A src/libutf/Make.HP-UX-9000 | 6 ++++++
A src/libutf/Make.Linux-386 | 7 +++++++
A src/libutf/Make.NetBSD-386 | 7 +++++++
A src/libutf/Make.OSF1-alpha | 6 ++++++
A src/libutf/Make.SunOS-sun4u | 2 ++
A src/libutf/Make.SunOS-sun4u-cc | 6 ++++++
A src/libutf/Make.SunOS-sun4u-gcc | 6 ++++++
A src/libutf/Makefile | 112 +++++++++++++++++++++++++++++++
A src/libutf/Makefile.BOT | 47 +++++++++++++++++++++++++++++++
A src/libutf/Makefile.CMD | 5 +++++
A src/libutf/Makefile.LIB | 4 ++++
A src/libutf/Makefile.MID | 41 +++++++++++++++++++++++++++++++
A src/libutf/Makefile.TOP | 20 ++++++++++++++++++++
A src/libutf/NOTICE | 13 +++++++++++++
A src/libutf/README | 13 +++++++++++++
A src/libutf/bundle.ports | 43 ++++++++++++++++++++++++++++++
A src/libutf/isalpharune.3 | 47 +++++++++++++++++++++++++++++++
A src/libutf/lib9.h | 17 +++++++++++++++++
A src/libutf/mkfile | 9 +++++++++
A src/libutf/rpm.spec | 28 ++++++++++++++++++++++++++++
A src/libutf/rune.3 | 187 +++++++++++++++++++++++++++++++
A src/libutf/rune.c | 177 +++++++++++++++++++++++++++++++
A src/libutf/runestrcat.3 | 65 +++++++++++++++++++++++++++++++
A src/libutf/runestrcat.c | 25 +++++++++++++++++++++++++
A src/libutf/runestrchr.c | 35 +++++++++++++++++++++++++++++++
A src/libutf/runestrcmp.c | 35 +++++++++++++++++++++++++++++++
A src/libutf/runestrcpy.c | 28 ++++++++++++++++++++++++++++
A src/libutf/runestrdup.c | 30 ++++++++++++++++++++++++++++++
A src/libutf/runestrecpy.c | 32 +++++++++++++++++++++++++++++++
A src/libutf/runestrlen.c | 24 ++++++++++++++++++++++++
A src/libutf/runestrncat.c | 32 +++++++++++++++++++++++++++++++
A src/libutf/runestrncmp.c | 37 +++++++++++++++++++++++++++++++
A src/libutf/runestrncpy.c | 33 +++++++++++++++++++++++++++++++
A src/libutf/runestrrchr.c | 30 ++++++++++++++++++++++++++++++
A src/libutf/runestrstr.c | 44 +++++++++++++++++++++++++++++++
A src/libutf/runetype.c | 1152 +++++++++++++++++++++++++++++++
A src/libutf/utf.7 | 91 +++++++++++++++++++++++++++++++
A src/libutf/utf.h | 51 +++++++++++++++++++++++++++++++
A src/libutf/utfdef.h | 14 ++++++++++++++
A src/libutf/utfecpy.c | 36 +++++++++++++++++++++++++++++++
A src/libutf/utflen.c | 38 +++++++++++++++++++++++++++++++
A src/libutf/utfnlen.c | 41 +++++++++++++++++++++++++++++++
A src/libutf/utfrrune.c | 46 +++++++++++++++++++++++++++++++
A src/libutf/utfrune.c | 45 +++++++++++++++++++++++++++++++
A src/libutf/utfutf.c | 41 +++++++++++++++++++++++++++++++
242 files changed, 18177 insertions(+), 0 deletions(-)
---
DIR diff --git a/include/bio.h b/include/bio.h
t@@ -0,0 +1,79 @@
+#ifndef _BIOH_
+#define _BIOH_ 1
+
+#include <sys/types.h> /* for off_t */
+#include <fcntl.h> /* for O_RDONLY, O_WRONLY */
+
+typedef struct Biobuf Biobuf;
+
+enum
+{
+ Bsize = 8*1024,
+ Bungetsize = 4, /* space for ungetc */
+ Bmagic = 0x314159,
+ Beof = -1,
+ Bbad = -2,
+
+ Binactive = 0, /* states */
+ Bractive,
+ Bwactive,
+ Bracteof,
+
+ Bend
+};
+
+struct Biobuf
+{
+ int icount; /* neg num of bytes at eob */
+ int ocount; /* num of bytes at bob */
+ int rdline; /* num of bytes after rdline */
+ int runesize; /* num of bytes of last getrune */
+ int state; /* r/w/inactive */
+ int fid; /* open file */
+ int flag; /* magic if malloc'ed */
+ off_t offset; /* offset of buffer in file */
+ int bsize; /* size of buffer */
+ unsigned char* bbuf; /* pointer to beginning of buffer */
+ unsigned char* ebuf; /* pointer to end of buffer */
+ unsigned char* gbuf; /* pointer to good data in buf */
+ unsigned char b[Bungetsize+Bsize];
+};
+
+#define BGETC(bp)\
+ ((bp)->icount?(bp)->bbuf[(bp)->bsize+(bp)->icount++]:Bgetc((bp)))
+#define BPUTC(bp,c)\
+ ((bp)->ocount?(bp)->bbuf[(bp)->bsize+(bp)->ocount++]=(c),0:Bputc((bp),(c)))
+#define BOFFSET(bp)\
+ (((bp)->state==Bractive)?\
+ (bp)->offset + (bp)->icount:\
+ (((bp)->state==Bwactive)?\
+ (bp)->offset + ((bp)->bsize + (bp)->ocount):\
+ -1))
+#define BLINELEN(bp)\
+ (bp)->rdline
+#define BFILDES(bp)\
+ (bp)->fid
+
+int Bbuffered(Biobuf*);
+int Bfildes(Biobuf*);
+int Bflush(Biobuf*);
+int Bgetc(Biobuf*);
+int Bgetd(Biobuf*, double*);
+int Binit(Biobuf*, int, int);
+int Binits(Biobuf*, int, int, unsigned char*, int);
+int Blinelen(Biobuf*);
+off_t Boffset(Biobuf*);
+Biobuf* Bopen(char*, int);
+int Bprint(Biobuf*, char*, ...);
+int Bputc(Biobuf*, int);
+void* Brdline(Biobuf*, int);
+long Bread(Biobuf*, void*, long);
+off_t Bseek(Biobuf*, off_t, int);
+int Bterm(Biobuf*);
+int Bungetc(Biobuf*);
+long Bwrite(Biobuf*, void*, long);
+char* Brdstr(Biobuf*, int, int);
+long Bgetrune(Biobuf*);
+int Bputrune(Biobuf*, long);
+
+#endif
DIR diff --git a/include/cursor.h b/include/cursor.h
t@@ -0,0 +1,7 @@
+typedef struct Cursor Cursor;
+struct Cursor
+{
+ Point offset;
+ uchar clr[2*16];
+ uchar set[2*16];
+};
DIR diff --git a/include/draw.h b/include/draw.h
t@@ -0,0 +1,520 @@
+typedef struct Cachefont Cachefont;
+typedef struct Cacheinfo Cacheinfo;
+typedef struct Cachesubf Cachesubf;
+typedef struct Display Display;
+typedef struct Font Font;
+typedef struct Fontchar Fontchar;
+typedef struct Image Image;
+typedef struct Mouse Mouse;
+typedef struct Point Point;
+typedef struct Rectangle Rectangle;
+typedef struct RGB RGB;
+typedef struct Screen Screen;
+typedef struct Subfont Subfont;
+
+extern int Rfmt(Fmt*);
+extern int Pfmt(Fmt*);
+
+enum
+{
+ DOpaque = 0xFFFFFFFF,
+ DTransparent = 0x00000000, /* only useful for allocimage, memfillcolor */
+ DBlack = 0x000000FF,
+ DWhite = 0xFFFFFFFF,
+ DRed = 0xFF0000FF,
+ DGreen = 0x00FF00FF,
+ DBlue = 0x0000FFFF,
+ DCyan = 0x00FFFFFF,
+ DMagenta = 0xFF00FFFF,
+ DYellow = 0xFFFF00FF,
+ DPaleyellow = 0xFFFFAAFF,
+ DDarkyellow = 0xEEEE9EFF,
+ DDarkgreen = 0x448844FF,
+ DPalegreen = 0xAAFFAAFF,
+ DMedgreen = 0x88CC88FF,
+ DDarkblue = 0x000055FF,
+ DPalebluegreen= 0xAAFFFFFF,
+ DPaleblue = 0x0000BBFF,
+ DBluegreen = 0x008888FF,
+ DGreygreen = 0x55AAAAFF,
+ DPalegreygreen = 0x9EEEEEFF,
+ DYellowgreen = 0x99994CFF,
+ DMedblue = 0x000099FF,
+ DGreyblue = 0x005DBBFF,
+ DPalegreyblue = 0x4993DDFF,
+ DPurpleblue = 0x8888CCFF,
+
+ DNotacolor = 0xFFFFFF00,
+ DNofill = DNotacolor,
+
+};
+
+enum
+{
+ Displaybufsize = 8000,
+ ICOSSCALE = 1024,
+ Borderwidth = 4,
+};
+
+enum
+{
+ /* refresh methods */
+ Refbackup = 0,
+ Refnone = 1,
+ Refmesg = 2
+};
+#define NOREFRESH ((void*)-1)
+
+enum
+{
+ /* line ends */
+ Endsquare = 0,
+ Enddisc = 1,
+ Endarrow = 2,
+ Endmask = 0x1F
+};
+
+#define ARROW(a, b, c) (Endarrow|((a)<<5)|((b)<<14)|((c)<<23))
+
+typedef enum
+{
+ /* Porter-Duff compositing operators */
+ Clear = 0,
+
+ SinD = 8,
+ DinS = 4,
+ SoutD = 2,
+ DoutS = 1,
+
+ S = SinD|SoutD,
+ SoverD = SinD|SoutD|DoutS,
+ SatopD = SinD|DoutS,
+ SxorD = SoutD|DoutS,
+
+ D = DinS|DoutS,
+ DoverS = DinS|DoutS|SoutD,
+ DatopS = DinS|SoutD,
+ DxorS = DoutS|SoutD, /* == SxorD */
+
+ Ncomp = 12,
+} Drawop;
+
+/*
+ * image channel descriptors
+ */
+enum {
+ CRed = 0,
+ CGreen,
+ CBlue,
+ CGrey,
+ CAlpha,
+ CMap,
+ CIgnore,
+ NChan,
+};
+
+#define __DC(type, nbits) ((((type)&15)<<4)|((nbits)&15))
+#define CHAN1(a,b) __DC(a,b)
+#define CHAN2(a,b,c,d) (CHAN1((a),(b))<<8|__DC((c),(d)))
+#define CHAN3(a,b,c,d,e,f) (CHAN2((a),(b),(c),(d))<<8|__DC((e),(f)))
+#define CHAN4(a,b,c,d,e,f,g,h) (CHAN3((a),(b),(c),(d),(e),(f))<<8|__DC((g),(h)))
+
+#define NBITS(c) ((c)&15)
+#define TYPE(c) (((c)>>4)&15)
+
+enum {
+ GREY1 = CHAN1(CGrey, 1),
+ GREY2 = CHAN1(CGrey, 2),
+ GREY4 = CHAN1(CGrey, 4),
+ GREY8 = CHAN1(CGrey, 8),
+ CMAP8 = CHAN1(CMap, 8),
+ RGB15 = CHAN4(CIgnore, 1, CRed, 5, CGreen, 5, CBlue, 5),
+ RGB16 = CHAN3(CRed, 5, CGreen, 6, CBlue, 5),
+ RGB24 = CHAN3(CRed, 8, CGreen, 8, CBlue, 8),
+ BGR24 = CHAN3(CBlue, 8, CGreen, 8, CRed, 8),
+ RGBA32 = CHAN4(CRed, 8, CGreen, 8, CBlue, 8, CAlpha, 8),
+ ARGB32 = CHAN4(CAlpha, 8, CRed, 8, CGreen, 8, CBlue, 8), /* stupid VGAs */
+ XRGB32 = CHAN4(CIgnore, 8, CRed, 8, CGreen, 8, CBlue, 8),
+ XBGR32 = CHAN4(CIgnore, 8, CBlue, 8, CGreen, 8, CRed, 8),
+};
+
+extern char* chantostr(char*, u32int);
+extern u32int strtochan(char*);
+extern int chantodepth(u32int);
+
+struct Point
+{
+ int x;
+ int y;
+};
+
+struct Rectangle
+{
+ Point min;
+ Point max;
+};
+
+typedef void (*Reffn)(Image*, Rectangle, void*);
+
+struct Screen
+{
+ Display *display; /* display holding data */
+ int id; /* id of system-held Screen */
+ Image *image; /* unused; for reference only */
+ Image *fill; /* color to paint behind windows */
+};
+
+struct Display
+{
+ QLock qlock;
+ int locking; /*program is using lockdisplay */
+ int dirno;
+ int imageid;
+ int local;
+ void (*error)(Display*, char*);
+ char *devdir;
+ char *windir;
+ char oldlabel[64];
+ u32int dataqid;
+ Image *image;
+ Image *white;
+ Image *black;
+ Image *opaque;
+ Image *transparent;
+ uchar *buf;
+ int bufsize;
+ uchar *bufp;
+ uchar *obuf;
+ int obufsize;
+ uchar *obufp;
+ Font *defaultfont;
+ Subfont *defaultsubfont;
+ Image *windows;
+ Image *screenimage;
+ int _isnewdisplay;
+};
+
+struct Image
+{
+ Display *display; /* display holding data */
+ int id; /* id of system-held Image */
+ Rectangle r; /* rectangle in data area, local coords */
+ Rectangle clipr; /* clipping region */
+ int depth; /* number of bits per pixel */
+ u32int chan;
+ int repl; /* flag: data replicates to tile clipr */
+ Screen *screen; /* 0 if not a window */
+ Image *next; /* next in list of windows */
+};
+
+struct RGB
+{
+ u32int red;
+ u32int green;
+ u32int blue;
+};
+
+/*
+ * Subfonts
+ *
+ * given char c, Subfont *f, Fontchar *i, and Point p, one says
+ * i = f->info+c;
+ * draw(b, Rect(p.x+i->left, p.y+i->top,
+ * p.x+i->left+((i+1)->x-i->x), p.y+i->bottom),
+ * color, f->bits, Pt(i->x, i->top));
+ * p.x += i->width;
+ * to draw characters in the specified color (itself an Image) in Image b.
+ */
+
+struct Fontchar
+{
+ int x; /* left edge of bits */
+ uchar top; /* first non-zero scan-line */
+ uchar bottom; /* last non-zero scan-line + 1 */
+ char left; /* offset of baseline */
+ uchar width; /* width of baseline */
+};
+
+struct Subfont
+{
+ char *name;
+ short n; /* number of chars in font */
+ uchar height; /* height of image */
+ char ascent; /* top of image to baseline */
+ Fontchar *info; /* n+1 character descriptors */
+ Image *bits; /* of font */
+ int ref;
+};
+
+enum
+{
+ /* starting values */
+ LOG2NFCACHE = 6,
+ NFCACHE = (1<<LOG2NFCACHE), /* #chars cached */
+ NFLOOK = 5, /* #chars to scan in cache */
+ NFSUBF = 2, /* #subfonts to cache */
+ /* max value */
+ MAXFCACHE = 1024+NFLOOK, /* upper limit */
+ MAXSUBF = 50, /* generous upper limit */
+ /* deltas */
+ DSUBF = 4,
+ /* expiry ages */
+ SUBFAGE = 10000,
+ CACHEAGE = 10000
+};
+
+struct Cachefont
+{
+ Rune min; /* lowest rune value to be taken from subfont */
+ Rune max; /* highest rune value+1 to be taken from subfont */
+ int offset; /* position in subfont of character at min */
+ char *name; /* stored in font */
+ char *subfontname; /* to access subfont */
+};
+
+struct Cacheinfo
+{
+ ushort x; /* left edge of bits */
+ uchar width; /* width of baseline */
+ schar left; /* offset of baseline */
+ Rune value; /* value of character at this slot in cache */
+ ushort age;
+};
+
+struct Cachesubf
+{
+ u32int age; /* for replacement */
+ Cachefont *cf; /* font info that owns us */
+ Subfont *f; /* attached subfont */
+};
+
+struct Font
+{
+ char *name;
+ Display *display;
+ short height; /* max height of image, interline spacing */
+ short ascent; /* top of image to baseline */
+ short width; /* widest so far; used in caching only */
+ short nsub; /* number of subfonts */
+ u32int age; /* increasing counter; used for LRU */
+ int maxdepth; /* maximum depth of all loaded subfonts */
+ int ncache; /* size of cache */
+ int nsubf; /* size of subfont list */
+ Cacheinfo *cache;
+ Cachesubf *subf;
+ Cachefont **sub; /* as read from file */
+ Image *cacheimage;
+};
+
+#define Dx(r) ((r).max.x-(r).min.x)
+#define Dy(r) ((r).max.y-(r).min.y)
+
+/*
+ * Image management
+ */
+extern Image* _allocimage(Image*, Display*, Rectangle, u32int, int, u32int, int, int);
+extern Image* allocimage(Display*, Rectangle, u32int, int, u32int);
+extern uchar* bufimage(Display*, int);
+extern int bytesperline(Rectangle, int);
+extern void closedisplay(Display*);
+extern void drawerror(Display*, char*);
+extern int flushimage(Display*, int);
+extern int freeimage(Image*);
+extern int _freeimage1(Image*);
+extern int geninitdraw(char*, void(*)(Display*, char*), char*, char*, char*, int);
+extern int initdraw(void(*)(Display*, char*), char*, char*);
+extern int newwindow(char*);
+extern int loadimage(Image*, Rectangle, uchar*, int);
+extern int cloadimage(Image*, Rectangle, uchar*, int);
+extern int getwindow(Display*, int);
+extern int gengetwindow(Display*, char*, Image**, Screen**, int);
+extern Image* readimage(Display*, int, int);
+extern Image* creadimage(Display*, int, int);
+extern int unloadimage(Image*, Rectangle, uchar*, int);
+extern int wordsperline(Rectangle, int);
+extern int writeimage(int, Image*, int);
+extern Image* namedimage(Display*, char*);
+extern int nameimage(Image*, char*, int);
+extern Image* allocimagemix(Display*, u32int, u32int);
+
+/*
+ * Colors
+ */
+extern void readcolmap(Display*, RGB*);
+extern void writecolmap(Display*, RGB*);
+extern u32int setalpha(u32int, uchar);
+
+/*
+ * Windows
+ */
+extern Screen* allocscreen(Image*, Image*, int);
+extern Image* _allocwindow(Image*, Screen*, Rectangle, int, u32int);
+extern Image* allocwindow(Screen*, Rectangle, int, u32int);
+extern void bottomnwindows(Image**, int);
+extern void bottomwindow(Image*);
+extern int freescreen(Screen*);
+extern Screen* publicscreen(Display*, int, u32int);
+extern void topnwindows(Image**, int);
+extern void topwindow(Image*);
+extern int originwindow(Image*, Point, Point);
+
+/*
+ * Geometry
+ */
+extern Point Pt(int, int);
+extern Rectangle Rect(int, int, int, int);
+extern Rectangle Rpt(Point, Point);
+extern Point addpt(Point, Point);
+extern Point subpt(Point, Point);
+extern Point divpt(Point, int);
+extern Point mulpt(Point, int);
+extern int eqpt(Point, Point);
+extern int eqrect(Rectangle, Rectangle);
+extern Rectangle insetrect(Rectangle, int);
+extern Rectangle rectaddpt(Rectangle, Point);
+extern Rectangle rectsubpt(Rectangle, Point);
+extern Rectangle canonrect(Rectangle);
+extern int rectXrect(Rectangle, Rectangle);
+extern int rectinrect(Rectangle, Rectangle);
+extern void combinerect(Rectangle*, Rectangle);
+extern int rectclip(Rectangle*, Rectangle);
+extern int ptinrect(Point, Rectangle);
+extern void replclipr(Image*, int, Rectangle);
+extern int drawreplxy(int, int, int); /* used to be drawsetxy */
+extern Point drawrepl(Rectangle, Point);
+extern int rgb2cmap(int, int, int);
+extern int cmap2rgb(int);
+extern int cmap2rgba(int);
+extern void icossin(int, int*, int*);
+extern void icossin2(int, int, int*, int*);
+
+/*
+ * Graphics
+ */
+extern void draw(Image*, Rectangle, Image*, Image*, Point);
+extern void drawop(Image*, Rectangle, Image*, Image*, Point, Drawop);
+extern void gendraw(Image*, Rectangle, Image*, Point, Image*, Point);
+extern void gendrawop(Image*, Rectangle, Image*, Point, Image*, Point, Drawop);
+extern void line(Image*, Point, Point, int, int, int, Image*, Point);
+extern void lineop(Image*, Point, Point, int, int, int, Image*, Point, Drawop);
+extern void poly(Image*, Point*, int, int, int, int, Image*, Point);
+extern void polyop(Image*, Point*, int, int, int, int, Image*, Point, Drawop);
+extern void fillpoly(Image*, Point*, int, int, Image*, Point);
+extern void fillpolyop(Image*, Point*, int, int, Image*, Point, Drawop);
+extern Point string(Image*, Point, Image*, Point, Font*, char*);
+extern Point stringop(Image*, Point, Image*, Point, Font*, char*, Drawop);
+extern Point stringn(Image*, Point, Image*, Point, Font*, char*, int);
+extern Point stringnop(Image*, Point, Image*, Point, Font*, char*, int, Drawop);
+extern Point runestring(Image*, Point, Image*, Point, Font*, Rune*);
+extern Point runestringop(Image*, Point, Image*, Point, Font*, Rune*, Drawop);
+extern Point runestringn(Image*, Point, Image*, Point, Font*, Rune*, int);
+extern Point runestringnop(Image*, Point, Image*, Point, Font*, Rune*, int, Drawop);
+extern Point stringbg(Image*, Point, Image*, Point, Font*, char*, Image*, Point);
+extern Point stringbgop(Image*, Point, Image*, Point, Font*, char*, Image*, Point, Drawop);
+extern Point stringnbg(Image*, Point, Image*, Point, Font*, char*, int, Image*, Point);
+extern Point stringnbgop(Image*, Point, Image*, Point, Font*, char*, int, Image*, Point, Drawop);
+extern Point runestringbg(Image*, Point, Image*, Point, Font*, Rune*, Image*, Point);
+extern Point runestringbgop(Image*, Point, Image*, Point, Font*, Rune*, Image*, Point, Drawop);
+extern Point runestringnbg(Image*, Point, Image*, Point, Font*, Rune*, int, Image*, Point);
+extern Point runestringnbgop(Image*, Point, Image*, Point, Font*, Rune*, int, Image*, Point, Drawop);
+extern Point _string(Image*, Point, Image*, Point, Font*, char*, Rune*, int, Rectangle, Image*, Point, Drawop);
+extern Point stringsubfont(Image*, Point, Image*, Subfont*, char*);
+extern int bezier(Image*, Point, Point, Point, Point, int, int, int, Image*, Point);
+extern int bezierop(Image*, Point, Point, Point, Point, int, int, int, Image*, Point, Drawop);
+extern int bezspline(Image*, Point*, int, int, int, int, Image*, Point);
+extern int bezsplineop(Image*, Point*, int, int, int, int, Image*, Point, Drawop);
+extern int bezsplinepts(Point*, int, Point**);
+extern int fillbezier(Image*, Point, Point, Point, Point, int, Image*, Point);
+extern int fillbezierop(Image*, Point, Point, Point, Point, int, Image*, Point, Drawop);
+extern int fillbezspline(Image*, Point*, int, int, Image*, Point);
+extern int fillbezsplineop(Image*, Point*, int, int, Image*, Point, Drawop);
+extern void ellipse(Image*, Point, int, int, int, Image*, Point);
+extern void ellipseop(Image*, Point, int, int, int, Image*, Point, Drawop);
+extern void fillellipse(Image*, Point, int, int, Image*, Point);
+extern void fillellipseop(Image*, Point, int, int, Image*, Point, Drawop);
+extern void arc(Image*, Point, int, int, int, Image*, Point, int, int);
+extern void arcop(Image*, Point, int, int, int, Image*, Point, int, int, Drawop);
+extern void fillarc(Image*, Point, int, int, Image*, Point, int, int);
+extern void fillarcop(Image*, Point, int, int, Image*, Point, int, int, Drawop);
+extern void border(Image*, Rectangle, int, Image*, Point);
+extern void borderop(Image*, Rectangle, int, Image*, Point, Drawop);
+
+/*
+ * Font management
+ */
+extern Font* openfont(Display*, char*);
+extern Font* buildfont(Display*, char*, char*);
+extern void freefont(Font*);
+extern Font* mkfont(Subfont*, Rune);
+extern int cachechars(Font*, char**, Rune**, ushort*, int, int*, char**);
+extern void agefont(Font*);
+extern Subfont* allocsubfont(char*, int, int, int, Fontchar*, Image*);
+extern Subfont* lookupsubfont(Display*, char*);
+extern void installsubfont(char*, Subfont*);
+extern void uninstallsubfont(Subfont*);
+extern void freesubfont(Subfont*);
+extern Subfont* readsubfont(Display*, char*, int, int);
+extern Subfont* readsubfonti(Display*, char*, int, Image*, int);
+extern int writesubfont(int, Subfont*);
+extern void _unpackinfo(Fontchar*, uchar*, int);
+extern Point stringsize(Font*, char*);
+extern int stringwidth(Font*, char*);
+extern int stringnwidth(Font*, char*, int);
+extern Point runestringsize(Font*, Rune*);
+extern int runestringwidth(Font*, Rune*);
+extern int runestringnwidth(Font*, Rune*, int);
+extern Point strsubfontwidth(Subfont*, char*);
+extern int loadchar(Font*, Rune, Cacheinfo*, int, int, char**);
+extern char* subfontname(char*, char*, int);
+extern Subfont* _getsubfont(Display*, char*);
+extern Subfont* getdefont(Display*);
+extern void lockdisplay(Display*);
+extern void unlockdisplay(Display*);
+extern int drawlsetrefresh(u32int, int, void*, void*);
+
+/*
+ * Predefined
+ */
+extern uchar defontdata[];
+extern int sizeofdefont;
+extern Point ZP;
+extern Rectangle ZR;
+
+/*
+ * Set up by initdraw()
+ */
+extern Display *display;
+extern Font *font;
+extern Image *screen;
+extern Screen *_screen;
+extern int _cursorfd;
+extern int _drawdebug; /* set to 1 to see errors from flushimage */
+extern void _setdrawop(Display*, Drawop);
+extern Display *_initdisplay(void(*)(Display*,char*), char*);
+
+#define BGSHORT(p) (((p)[0]<<0) | ((p)[1]<<8))
+#define BGLONG(p) ((BGSHORT(p)<<0) | (BGSHORT(p+2)<<16))
+#define BPSHORT(p, v) ((p)[0]=(v), (p)[1]=((v)>>8))
+#define BPLONG(p, v) (BPSHORT(p, (v)), BPSHORT(p+2, (v)>>16))
+
+/*
+ * Compressed image file parameters and helper routines
+ */
+#define NMATCH 3 /* shortest match possible */
+#define NRUN (NMATCH+31) /* longest match possible */
+#define NMEM 1024 /* window size */
+#define NDUMP 128 /* maximum length of dump */
+#define NCBLOCK 6000 /* size of compressed blocks */
+extern void _twiddlecompressed(uchar*, int);
+extern int _compblocksize(Rectangle, int);
+
+/* XXX backwards helps; should go */
+extern int log2[];
+extern u32int drawld2chan[];
+extern void drawsetdebug(int);
+
+/*
+ * Port magic.
+ */
+int _drawmsgread(Display*, void*, int);
+int _drawmsgwrite(Display*, void*, int);
DIR diff --git a/include/event.h b/include/event.h
t@@ -0,0 +1,63 @@
+typedef struct Event Event;
+typedef struct Menu Menu;
+
+enum
+{
+ Emouse = 1,
+ Ekeyboard = 2,
+};
+
+enum
+{
+ MAXSLAVE = 32,
+ EMAXMSG = 128+8192, /* size of 9p header+data */
+};
+
+struct Mouse
+{
+ int buttons; /* bit array: LMR=124 */
+ Point xy;
+ ulong msec;
+};
+
+struct Event
+{
+ int kbdc;
+ Mouse mouse;
+ int n; /* number of characters in message */
+ void *v; /* data unpacked by general event-handling function */
+ uchar data[EMAXMSG]; /* message from an arbitrary file descriptor */
+};
+
+struct Menu
+{
+ char **item;
+ char *(*gen)(int);
+ int lasthit;
+};
+
+/*
+ * Events
+ */
+extern void einit(ulong);
+extern ulong estart(ulong, int, int);
+extern ulong estartfn(ulong, int, int, int (*fn)(int, Event*, uchar*, int));
+extern ulong etimer(ulong, int);
+extern ulong event(Event*);
+extern ulong eread(ulong, Event*);
+extern Mouse emouse(void);
+extern int ekbd(void);
+extern int ecanread(ulong);
+extern int ecanmouse(void);
+extern int ecankbd(void);
+extern void eresized(int); /* supplied by user */
+extern int emenuhit(int, Mouse*, Menu*);
+extern int eatomouse(Mouse*, char*, int);
+extern Rectangle getrect(int, Mouse*);
+struct Cursor;
+extern void esetcursor(struct Cursor*);
+extern void emoveto(Point);
+extern Rectangle egetrect(int, Mouse*);
+extern void edrawgetrect(Rectangle, int);
+extern int ereadmouse(Mouse*);
+extern int eatomouse(Mouse*, char*, int);
DIR diff --git a/include/fmt.h b/include/fmt.h
t@@ -0,0 +1,100 @@
+
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+
+#ifndef _FMTH_
+#define _FMTH_ 1
+
+#include <stdarg.h>
+
+#ifndef _UTFH_
+#include <utf.h>
+#endif
+
+typedef struct Fmt Fmt;
+struct Fmt{
+ unsigned char runes; /* output buffer is runes or chars? */
+ void *start; /* of buffer */
+ void *to; /* current place in the buffer */
+ void *stop; /* end of the buffer; overwritten if flush fails */
+ int (*flush)(Fmt *); /* called when to == stop */
+ void *farg; /* to make flush a closure */
+ int nfmt; /* num chars formatted so far */
+ va_list args; /* args passed to dofmt */
+ int r; /* % format Rune */
+ int width;
+ int prec;
+ unsigned long flags;
+};
+
+enum{
+ FmtWidth = 1,
+ FmtLeft = FmtWidth << 1,
+ FmtPrec = FmtLeft << 1,
+ FmtSharp = FmtPrec << 1,
+ FmtSpace = FmtSharp << 1,
+ FmtSign = FmtSpace << 1,
+ FmtZero = FmtSign << 1,
+ FmtUnsigned = FmtZero << 1,
+ FmtShort = FmtUnsigned << 1,
+ FmtLong = FmtShort << 1,
+ FmtVLong = FmtLong << 1,
+ FmtComma = FmtVLong << 1,
+ FmtByte = FmtComma << 1,
+ FmtLDouble = FmtByte << 1,
+
+ FmtFlag = FmtLDouble << 1
+};
+
+extern int print(char*, ...);
+extern char* seprint(char*, char*, char*, ...);
+extern char* vseprint(char*, char*, char*, va_list);
+extern int snprint(char*, int, char*, ...);
+extern int vsnprint(char*, int, char*, va_list);
+extern char* smprint(char*, ...);
+extern char* vsmprint(char*, va_list);
+extern int sprint(char*, char*, ...);
+extern int fprint(int, char*, ...);
+extern int vfprint(int, char*, va_list);
+
+extern int runesprint(Rune*, char*, ...);
+extern int runesnprint(Rune*, int, char*, ...);
+extern int runevsnprint(Rune*, int, char*, va_list);
+extern Rune* runeseprint(Rune*, Rune*, char*, ...);
+extern Rune* runevseprint(Rune*, Rune*, char*, va_list);
+extern Rune* runesmprint(char*, ...);
+extern Rune* runevsmprint(char*, va_list);
+
+extern int fmtfdinit(Fmt*, int, char*, int);
+extern int fmtfdflush(Fmt*);
+extern int fmtstrinit(Fmt*);
+extern char* fmtstrflush(Fmt*);
+extern int runefmtstrinit(Fmt*);
+
+extern int quotestrfmt(Fmt *f);
+extern void quotefmtinstall(void);
+extern int (*fmtdoquote)(int);
+
+
+extern int fmtinstall(int, int (*)(Fmt*));
+extern int dofmt(Fmt*, char*);
+extern int fmtprint(Fmt*, char*, ...);
+extern int fmtvprint(Fmt*, char*, va_list);
+extern int fmtrune(Fmt*, int);
+extern int fmtstrcpy(Fmt*, char*);
+
+extern double fmtstrtod(const char *, char **);
+extern double fmtcharstod(int(*)(void*), void*);
+
+#endif
DIR diff --git a/include/frame.h b/include/frame.h
t@@ -0,0 +1,85 @@
+typedef struct Frbox Frbox;
+typedef struct Frame Frame;
+
+enum{
+ BACK,
+ HIGH,
+ BORD,
+ TEXT,
+ HTEXT,
+ NCOL
+};
+
+#define FRTICKW 3
+
+struct Frbox
+{
+ long wid; /* in pixels */
+ long nrune; /* <0 ==> negate and treat as break char */
+ uchar *ptr;
+ short bc; /* break char */
+ short minwid;
+};
+
+struct Frame
+{
+ Font *font; /* of chars in the frame */
+ Display *display; /* on which frame appears */
+ Image *b; /* on which frame appears */
+ Image *cols[NCOL]; /* text and background colors */
+ Rectangle r; /* in which text appears */
+ Rectangle entire; /* of full frame */
+ void (*scroll)(Frame*, int); /* scroll function provided by application */
+ Frbox *box;
+ ulong p0, p1; /* selection */
+ ushort nbox, nalloc;
+ ushort maxtab; /* max size of tab, in pixels */
+ ushort nchars; /* # runes in frame */
+ ushort nlines; /* # lines with text */
+ ushort maxlines; /* total # lines in frame */
+ ushort lastlinefull; /* last line fills frame */
+ ushort modified; /* changed since frselect() */
+ Image *tick; /* typing tick */
+ Image *tickback; /* saved image under tick */
+ int ticked; /* flag: is tick onscreen? */
+};
+
+ulong frcharofpt(Frame*, Point);
+Point frptofchar(Frame*, ulong);
+int frdelete(Frame*, ulong, ulong);
+void frinsert(Frame*, Rune*, Rune*, ulong);
+void frselect(Frame*, Mousectl*);
+void frselectpaint(Frame*, Point, Point, Image*);
+void frdrawsel(Frame*, Point, ulong, ulong, int);
+void frdrawsel0(Frame*, Point, ulong, ulong, Image*, Image*);
+void frinit(Frame*, Rectangle, Font*, Image*, Image**);
+void frsetrects(Frame*, Rectangle, Image*);
+void frclear(Frame*, int);
+
+uchar *_frallocstr(Frame*, unsigned);
+void _frinsure(Frame*, int, unsigned);
+Point _frdraw(Frame*, Point);
+void _frgrowbox(Frame*, int);
+void _frfreebox(Frame*, int, int);
+void _frmergebox(Frame*, int);
+void _frdelbox(Frame*, int, int);
+void _frsplitbox(Frame*, int, int);
+int _frfindbox(Frame*, int, ulong, ulong);
+void _frclosebox(Frame*, int, int);
+int _frcanfit(Frame*, Point, Frbox*);
+void _frcklinewrap(Frame*, Point*, Frbox*);
+void _frcklinewrap0(Frame*, Point*, Frbox*);
+void _fradvance(Frame*, Point*, Frbox*);
+int _frnewwid(Frame*, Point, Frbox*);
+int _frnewwid0(Frame*, Point, Frbox*);
+void _frclean(Frame*, Point, int, int);
+void _frredraw(Frame*, Point);
+void _fraddbox(Frame*, int, int);
+Point _frptofcharptb(Frame*, ulong, Point, int);
+Point _frptofcharnb(Frame*, ulong, int);
+int _frstrlen(Frame*, int);
+void frtick(Frame*, Point, int);
+void frinittick(Frame*);
+
+#define NRUNE(b) ((b)->nrune<0? 1 : (b)->nrune)
+#define NBYTE(b) strlen((char*)(b)->ptr)
DIR diff --git a/include/keyboard.h b/include/keyboard.h
t@@ -0,0 +1,36 @@
+typedef struct Keyboardctl Keyboardctl;
+
+struct Keyboardctl
+{
+ struct Channel *c; /* chan(Rune)[20] */
+
+ char *file;
+ int consfd; /* to cons file */
+ int ctlfd; /* to ctl file */
+ int pid; /* of slave proc */
+};
+
+
+extern Keyboardctl* initkeyboard(char*);
+extern int ctlkeyboard(Keyboardctl*, char*);
+extern void closekeyboard(Keyboardctl*);
+
+enum {
+ KF= 0xF000, /* Rune: beginning of private Unicode space */
+ /* KF|1, KF|2, ..., KF|0xC is F1, F2, ..., F12 */
+ Khome= KF|0x0D,
+ Kup= KF|0x0E,
+ Kpgup= KF|0x0F,
+ Kprint= KF|0x10,
+ Kleft= KF|0x11,
+ Kright= KF|0x12,
+ Kdown= 0x80,
+ Kview= 0x80,
+ Kpgdown= KF|0x13,
+ Kins= KF|0x14,
+ Kend= '\r', /* [sic] */
+
+ Kalt= KF|0x15,
+ Kshift= KF|0x16,
+ Kctl= KF|0x17,
+};
DIR diff --git a/include/lib9.h b/include/lib9.h
t@@ -0,0 +1,254 @@
+/*
+ * Lib9 is miscellany from the Plan 9 C library that doesn't
+ * fit into libutf or into libfmt, but is still missing from traditional
+ * Unix C libraries.
+ */
+#ifndef _LIB9H_
+#define _LIB9H_ 1
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+
+#include <unistd.h>
+#include <string.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <fcntl.h>
+#include <assert.h>
+#include <setjmp.h>
+
+#ifndef _FMTH_
+# include <fmt.h>
+#endif
+
+#define nil ((void*)0)
+#define nelem(x) (sizeof(x)/sizeof((x)[0]))
+
+#define _NEEDUCHAR 1
+#define _NEEDUSHORT 1
+#define _NEEDUINT 1
+#define _NEEDULONG 1
+
+#if defined(__linux__)
+# include <sys/types.h>
+# if defined(__USE_MISC)
+# undef _NEEDUSHORT
+# undef _NEEDUINT
+# undef _NEEDULONG
+# endif
+#endif
+#if defined(__FreeBSD__)
+# include <sys/types.h>
+# if !defined(_POSIX_SOURCE)
+# undef _NEEDUSHORT
+# undef _NEEDUINT
+# endif
+#endif
+
+typedef signed char schar;
+typedef unsigned int u32int;
+#ifdef _NEEDUCHAR
+ typedef unsigned char uchar;
+#endif
+#ifdef _NEEDUSHORT
+ typedef unsigned short ushort;
+#endif
+#ifdef _NEEDUINT
+ typedef unsigned int uint;
+#endif
+#ifdef _NEEDULONG
+ typedef unsigned long ulong;
+#endif
+typedef unsigned long long uvlong;
+typedef long long vlong;
+
+/* rfork to create new process running fn(arg) */
+
+#if defined(__FreeBSD__)
+#undef RFFDG
+#undef RFNOTEG
+#undef RFPROC
+#undef RFMEM
+#undef RFNOWAIT
+#undef RFCFDG
+#endif
+
+enum
+{
+/* RFNAMEG = (1<<0), */
+/* RFENVG = (1<<1), */
+ RFFDG = (1<<2),
+ RFNOTEG = (1<<3),
+ RFPROC = (1<<4),
+ RFMEM = (1<<5),
+ RFNOWAIT = (1<<6),
+/* RFCNAMEG = (1<<10), */
+/* RFCENVG = (1<<11), */
+ RFCFDG = (1<<12),
+/* RFREND = (1<<13), */
+/* RFNOMNT = (1<<14) */
+};
+extern int ffork(int, void(*)(void*), void*);
+
+/* wait for processes */
+#define wait _p9wait
+typedef struct Waitmsg Waitmsg;
+struct Waitmsg
+{
+ int pid; /* of loved one */
+ ulong time[3]; /* of loved one & descendants */
+ char *msg;
+};
+extern int await(char*, int);
+extern Waitmsg* wait(void);
+
+/* synchronization */
+typedef struct Lock Lock;
+struct Lock
+{
+ int val;
+};
+
+extern int _tas(void*);
+extern void lock(Lock*);
+extern void unlock(Lock*);
+extern int canlock(Lock*);
+
+typedef struct QLp QLp;
+struct QLp
+{
+ int inuse;
+ QLp *next;
+ int state;
+};
+
+typedef struct QLock QLock;
+struct QLock
+{
+ Lock lock;
+ int locked;
+ QLp *head;
+ QLp *tail;
+};
+
+extern void qlock(QLock*);
+extern void qunlock(QLock*);
+extern int canqlock(QLock*);
+extern void _qlockinit(ulong (*)(ulong, ulong));
+
+typedef struct RWLock RWLock;
+struct RWLock
+{
+ Lock lock;
+ int readers;
+ int writer;
+ QLp *head;
+ QLp *tail;
+};
+
+extern void rlock(RWLock*);
+extern void runlock(RWLock*);
+extern int canrlock(RWLock*);
+extern void wlock(RWLock*);
+extern void wunlock(RWLock*);
+extern int canwlock(RWLock*);
+
+typedef struct Rendez Rendez;
+struct Rendez
+{
+ QLock *l;
+ QLp *head;
+ QLp *tail;
+};
+
+extern void rsleep(Rendez*);
+extern int rwakeup(Rendez*);
+extern int rwakeupall(Rendez*);
+
+extern ulong rendezvous(ulong, ulong);
+
+/* one of a kind */
+extern void sysfatal(char*, ...);
+extern int nrand(int);
+extern void setmalloctag(void*, ulong);
+extern void setrealloctag(void*, ulong);
+extern void *mallocz(ulong, int);
+extern long readn(int, void*, long);
+extern void exits(char*);
+extern void _exits(char*);
+extern ulong getcallerpc(void*);
+
+/* string routines */
+extern char* strecpy(char*, char*, char*);
+extern int tokenize(char*, char**, int);
+extern int cistrncmp(char*, char*, int);
+extern int cistrcmp(char*, char*);
+extern char* cistrstr(char*, char*);
+extern int getfields(char*, char**, int, int, char*);
+extern int gettokens(char *, char **, int, char *);
+
+/* formatting helpers */
+extern int dec64(uchar*, int, char*, int);
+extern int enc64(char*, int, uchar*, int);
+extern int dec32(uchar*, int, char*, int);
+extern int enc32(char*, int, uchar*, int);
+extern int dec16(uchar*, int, char*, int);
+extern int enc16(char*, int, uchar*, int);
+extern int encodefmt(Fmt*);
+
+/* error string */
+enum
+{
+ ERRMAX = 128
+};
+extern void rerrstr(char*, uint);
+extern void werrstr(char*, ...);
+extern int errstr(char*, uint);
+
+/* compiler directives on plan 9 */
+#define USED(x) if(x){}else{}
+#define SET(x) ((x)=0)
+
+/* command line */
+extern char *argv0;
+#define ARGBEGIN for((argv0||(argv0=*argv)),argv++,argc--;\
+ argv[0] && argv[0][0]=='-' && argv[0][1];\
+ argc--, argv++) {\
+ char *_args, *_argt;\
+ Rune _argc;\
+ _args = &argv[0][1];\
+ if(_args[0]=='-' && _args[1]==0){\
+ argc--; argv++; break;\
+ }\
+ _argc = 0;\
+ while(*_args && (_args += chartorune(&_argc, _args)))\
+ switch(_argc)
+#define ARGEND SET(_argt);USED(_argt);USED(_argc);USED(_args);}USED(argv);USED(argc);
+#define ARGF() (_argt=_args, _args="",\
+ (*_argt? _argt: argv[1]? (argc--, *++argv): 0))
+#define EARGF(x) (_argt=_args, _args="",\
+ (*_argt? _argt: argv[1]? (argc--, *++argv): ((x), abort(), (char*)0)))
+
+#define ARGC() _argc
+
+#define OREAD O_RDONLY
+#define OWRITE O_WRONLY
+#define AEXIST 0
+#define AREAD 4
+#define AWRITE 2
+#define AEXEC 1
+#define ORCLOSE 8
+#define OCEXEC 16
+
+#define dup dup2
+#define exec execv
+#define seek lseek
+#define getwd getcwd
+
+#if defined(__cplusplus)
+}
+#endif
+
+#endif /* _LIB9H_ */
DIR diff --git a/include/libc.h b/include/libc.h
t@@ -0,0 +1,2 @@
+#include <lib9.h>
+
DIR diff --git a/include/mouse.h b/include/mouse.h
t@@ -0,0 +1,44 @@
+typedef struct Menu Menu;
+typedef struct Mousectl Mousectl;
+
+struct Mouse
+{
+ int buttons; /* bit array: LMR=124 */
+ Point xy;
+ ulong msec;
+};
+
+struct Mousectl
+{
+ Mouse m;
+ struct Channel *c; /* chan(Mouse) */
+ struct Channel *resizec; /* chan(int)[2] */
+ /* buffered in case client is waiting for a mouse action before handling resize */
+
+ char *file;
+ int mfd; /* to mouse file */
+ int cfd; /* to cursor file */
+ int pid; /* of slave proc */
+ Display *display;
+ /*Image* image; / * of associated window/display */
+};
+
+struct Menu
+{
+ char **item;
+ char *(*gen)(int);
+ int lasthit;
+};
+
+/*
+ * Mouse
+ */
+extern Mousectl* initmouse(char*, Image*);
+extern void moveto(Mousectl*, Point);
+extern int readmouse(Mousectl*);
+extern void closemouse(Mousectl*);
+struct Cursor;
+extern void setcursor(Mousectl*, struct Cursor*);
+extern void drawgetrect(Rectangle, int);
+extern Rectangle getrect(int, Mousectl*);
+extern int menuhit(int, Mousectl*, Menu*, Screen*);
DIR diff --git a/include/regexp9.h b/include/regexp9.h
t@@ -0,0 +1,71 @@
+#ifndef _REGEXP9H_
+
+#define _REGEXP9H_ 1
+#include <utf.h>
+
+typedef struct Resub Resub;
+typedef struct Reclass Reclass;
+typedef struct Reinst Reinst;
+typedef struct Reprog Reprog;
+
+/*
+ * Sub expression matches
+ */
+struct Resub{
+ union
+ {
+ char *sp;
+ Rune *rsp;
+ }s;
+ union
+ {
+ char *ep;
+ Rune *rep;
+ }e;
+};
+
+/*
+ * character class, each pair of rune's defines a range
+ */
+struct Reclass{
+ Rune *end;
+ Rune spans[64];
+};
+
+/*
+ * Machine instructions
+ */
+struct Reinst{
+ int type;
+ union {
+ Reclass *cp; /* class pointer */
+ Rune r; /* character */
+ int subid; /* sub-expression id for RBRA and LBRA */
+ Reinst *right; /* right child of OR */
+ }u1;
+ union { /* regexp relies on these two being in the same union */
+ Reinst *left; /* left child of OR */
+ Reinst *next; /* next instruction for CAT & LBRA */
+ }u2;
+};
+
+/*
+ * Reprogram definition
+ */
+struct Reprog{
+ Reinst *startinst; /* start pc */
+ Reclass class[16]; /* .data */
+ Reinst firstinst[5]; /* .text */
+};
+
+extern Reprog *regcomp(char*);
+extern Reprog *regcomplit(char*);
+extern Reprog *regcompnl(char*);
+extern void regerror(char*);
+extern int regexec(Reprog*, char*, Resub*, int);
+extern void regsub(char*, char*, int, Resub*, int);
+
+extern int rregexec(Reprog*, Rune*, Resub*, int);
+extern void rregsub(Rune*, Rune*, Resub*, int);
+
+#endif
DIR diff --git a/include/thread.h b/include/thread.h
t@@ -0,0 +1,132 @@
+#ifndef _THREADH_
+#define _THREADH_ 1
+
+/* avoid conflicts with socket library */
+#undef send
+#define send _threadsend
+#undef recv
+#define recv _threadrecv
+
+typedef struct Alt Alt;
+typedef struct Channel Channel;
+typedef struct Ref Ref;
+
+/* Channel structure. S is the size of the buffer. For unbuffered channels
+ * s is zero. v is an array of s values. If s is zero, v is unused.
+ * f and n represent the state of the queue pointed to by v.
+ */
+
+enum {
+ Nqwds = 2,
+ Nqshift = 5, // 2log #of bits in long
+ Nqmask = - 1,
+ Nqbits = (1 << Nqshift) * 2,
+};
+
+struct Channel {
+ int s; // Size of the channel (may be zero)
+ unsigned int f; // Extraction point (insertion pt: (f + n) % s)
+ unsigned int n; // Number of values in the channel
+ int e; // Element size
+ int freed; // Set when channel is being deleted
+ volatile Alt **qentry; // Receivers/senders waiting (malloc)
+ volatile int nentry; // # of entries malloc-ed
+ unsigned char v[1]; // Array of s values in the channel
+};
+
+
+/* Channel operations for alt: */
+typedef enum {
+ CHANEND,
+ CHANSND,
+ CHANRCV,
+ CHANNOP,
+ CHANNOBLK,
+} ChanOp;
+
+struct Alt {
+ Channel *c; /* channel */
+ void *v; /* pointer to value */
+ ChanOp op; /* operation */
+
+ /* the next variables are used internally to alt
+ * they need not be initialized
+ */
+ Channel **tag; /* pointer to rendez-vous tag */
+ int entryno; /* entry number */
+};
+
+struct Ref {
+ long ref;
+};
+
+int alt(Alt alts[]);
+Channel* chancreate(int elemsize, int bufsize);
+int chaninit(Channel *c, int elemsize, int elemcnt);
+void chanfree(Channel *c);
+int chanprint(Channel *, char *, ...);
+long decref(Ref *r); /* returns 0 iff value is now zero */
+void incref(Ref *r);
+int nbrecv(Channel *c, void *v);
+void* nbrecvp(Channel *c);
+unsigned long nbrecvul(Channel *c);
+int nbsend(Channel *c, void *v);
+int nbsendp(Channel *c, void *v);
+int nbsendul(Channel *c, unsigned long v);
+int proccreate(void (*f)(void *arg), void *arg, unsigned int stacksize);
+int procrfork(void (*f)(void *arg), void *arg, unsigned int stacksize, int flag);
+void** procdata(void);
+void procexec(Channel *, char *, char *[]);
+void procexecl(Channel *, char *, ...);
+int recv(Channel *c, void *v);
+void* recvp(Channel *c);
+unsigned long recvul(Channel *c);
+int send(Channel *c, void *v);
+int sendp(Channel *c, void *v);
+int sendul(Channel *c, unsigned long v);
+int threadcreate(void (*f)(void *arg), void *arg, unsigned int stacksize);
+void** threaddata(void);
+void threadexits(char *);
+void threadexitsall(char *);
+int threadgetgrp(void); /* return thread group of current thread */
+char* threadgetname(void);
+void threadint(int); /* interrupt thread */
+void threadintgrp(int); /* interrupt threads in grp */
+void threadkill(int); /* kill thread */
+void threadkillgrp(int); /* kill threads in group */
+void threadmain(int argc, char *argv[]);
+void threadnonotes(void);
+int threadnotify(int (*f)(void*, char*), int in);
+int threadid(void);
+int threadpid(int);
+int threadsetgrp(int); /* set thread group, return old */
+void threadsetname(char *name);
+Channel* threadwaitchan(void);
+int tprivalloc(void);
+void tprivfree(int);
+void **tprivaddr(int);
+void yield(void);
+
+long threadstack(void);
+
+extern int mainstacksize;
+
+/* slave I/O processes */
+typedef struct Ioproc Ioproc;
+
+Ioproc* ioproc(void);
+void closeioproc(Ioproc*);
+void iointerrupt(Ioproc*);
+
+int ioclose(Ioproc*, int);
+int iodial(Ioproc*, char*, char*, char*, int*);
+int ioopen(Ioproc*, char*, int);
+long ioread(Ioproc*, int, void*, long);
+long ioreadn(Ioproc*, int, void*, long);
+long iowrite(Ioproc*, int, void*, long);
+int iosleep(Ioproc*, long);
+
+long iocall(Ioproc*, long (*)(va_list*), ...);
+void ioret(Ioproc*, int);
+
+#endif /* _THREADH_ */
DIR diff --git a/include/u.h b/include/u.h
DIR diff --git a/include/utf.h b/include/utf.h
t@@ -0,0 +1,51 @@
+#ifndef _UTFH_
+#define _UTFH_ 1
+
+typedef unsigned short Rune; /* 16 bits */
+
+enum
+{
+ UTFmax = 3, /* maximum bytes per rune */
+ Runesync = 0x80, /* cannot represent part of a UTF sequence (<) */
+ Runeself = 0x80, /* rune and UTF sequences are the same (<) */
+ Runeerror = 0x80, /* decoding error in UTF */
+};
+
+/*
+ * rune routines
+ */
+extern int runetochar(char*, Rune*);
+extern int chartorune(Rune*, char*);
+extern int runelen(long);
+extern int runenlen(Rune*, int);
+extern int fullrune(char*, int);
+extern int utflen(char*);
+extern int utfnlen(char*, long);
+extern char* utfrune(char*, long);
+extern char* utfrrune(char*, long);
+extern char* utfutf(char*, char*);
+extern char* utfecpy(char*, char*, char*);
+
+extern Rune* runestrcat(Rune*, Rune*);
+extern Rune* runestrchr(Rune*, Rune);
+extern int runestrcmp(Rune*, Rune*);
+extern Rune* runestrcpy(Rune*, Rune*);
+extern Rune* runestrncpy(Rune*, Rune*, long);
+extern Rune* runestrecpy(Rune*, Rune*, Rune*);
+extern Rune* runestrdup(Rune*);
+extern Rune* runestrncat(Rune*, Rune*, long);
+extern int runestrncmp(Rune*, Rune*, long);
+extern Rune* runestrrchr(Rune*, Rune);
+extern long runestrlen(Rune*);
+extern Rune* runestrstr(Rune*, Rune*);
+
+extern Rune tolowerrune(Rune);
+extern Rune totitlerune(Rune);
+extern Rune toupperrune(Rune);
+extern int isalpharune(Rune);
+extern int islowerrune(Rune);
+extern int isspacerune(Rune);
+extern int istitlerune(Rune);
+extern int isupperrune(Rune);
+
+#endif
DIR diff --git a/man/man3/bio.3 b/man/man3/bio.3
t@@ -0,0 +1,336 @@
+.TH BIO 3
+.SH NAME
+Bopen, Binit, Binits, Brdline, Brdstr, Bgetc, Bgetd, Bungetc, Bread, Bseek, Boffset, Bfildes, Blinelen, Bputc, Bprint, Bvprint, Bwrite, Bflush, Bterm, Bbuffered \- buffered input/output
+.SH SYNOPSIS
+.ta \w'Biobuf* 'u
+.B #include <fmt.h>
+.B #include <bio.h>
+.PP
+.B
+Biobuf* Bopen(char *file, int mode)
+.PP
+.B
+int Binit(Biobuf *bp, int fd, int mode)
+.PP
+.B
+int Bterm(Biobuf *bp)
+.PP
+.B
+int Bprint(Biobuf *bp, char *format, ...)
+.PP
+.B
+int Bvprint(Biobuf *bp, char *format, va_list arglist);
+.PP
+.B
+void* Brdline(Biobuf *bp, int delim)
+.PP
+.B
+char* Brdstr(Biobuf *bp, int delim, int nulldelim)
+.PP
+.B
+int Blinelen(Biobuf *bp)
+.PP
+.B
+off_t Boffset(Biobuf *bp)
+.PP
+.B
+int Bfildes(Biobuf *bp)
+.PP
+.B
+int Bgetc(Biobuf *bp)
+.PP
+.B
+long Bgetrune(Biobufhdr *bp)
+.PP
+.B
+int Bgetd(Biobuf *bp, double *d)
+.PP
+.B
+int Bungetc(Biobuf *bp)
+.PP
+.B
+int Bungetrune(Biobufhdr *bp)
+.PP
+.B
+off_t Bseek(Biobuf *bp, off_t n, int type)
+.PP
+.B
+int Bputc(Biobuf *bp, int c)
+.PP
+.B
+int Bputrune(Biobufhdr *bp, long c)
+.PP
+.B
+long Bread(Biobuf *bp, void *addr, long nbytes)
+.PP
+.B
+long Bwrite(Biobuf *bp, void *addr, long nbytes)
+.PP
+.B
+int Bflush(Biobuf *bp)
+.PP
+.B
+int Bbuffered(Biobuf *bp)
+.PP
+.SH DESCRIPTION
+These routines implement fast buffered I/O.
+I/O on different file descriptors is independent.
+.PP
+.I Bopen
+opens
+.I file
+for mode
+.B O_RDONLY
+or creates for mode
+.BR O_WRONLY .
+It calls
+.IR malloc (3)
+to allocate a buffer.
+.PP
+.I Binit
+initializes a buffer
+with the open file descriptor passed in
+by the user.
+.PP
+Arguments
+of types pointer to Biobuf and pointer to Biobuf
+can be used interchangeably in the following routines.
+.PP
+.IR Bopen ,
+.IR Binit ,
+or
+.I Binits
+should be called before any of the
+other routines on that buffer.
+.I Bfildes
+returns the integer file descriptor of the associated open file.
+.PP
+.I Bterm
+flushes the buffer for
+.IR bp .
+If the buffer was allocated by
+.IR Bopen ,
+the buffer is
+.I freed
+and the file is closed.
+.PP
+.I Brdline
+reads a string from the file associated with
+.I bp
+up to and including the first
+.I delim
+character.
+The delimiter character at the end of the line is
+not altered.
+.I Brdline
+returns a pointer to the start of the line or
+.L 0
+on end-of-file or read error.
+.I Blinelen
+returns the length (including the delimiter)
+of the most recent string returned by
+.IR Brdline .
+.PP
+.I Brdstr
+returns a
+.IR malloc (3)-allocated
+buffer containing the next line of input delimited by
+.IR delim ,
+terminated by a NUL (0) byte.
+Unlike
+.IR Brdline ,
+which returns when its buffer is full even if no delimiter has been found,
+.I Brdstr
+will return an arbitrarily long line in a single call.
+If
+.I nulldelim
+is set, the terminal delimiter will be overwritten with a NUL.
+After a successful call to
+.IR Brdstr ,
+the return value of
+.I Blinelen
+will be the length of the returned buffer, excluding the NUL.
+.PP
+.I Bgetc
+returns the next byte from
+.IR bp ,
+or a negative value
+at end of file.
+.I Bungetc
+may be called immediately after
+.I Bgetc
+to allow the same byte to be reread.
+.PP
+.I Bgetrune
+calls
+.I Bgetc
+to read the bytes of the next
+.SM UTF
+sequence in the input stream and returns the value of the rune
+represented by the sequence.
+It returns a negative value
+at end of file.
+.I Bungetrune
+may be called immediately after
+.I Bgetrune
+to allow the same
+.SM UTF
+sequence to be reread as either bytes or a rune.
+.I Bungetc
+and
+.I Bungetrune
+may back up a maximum of five bytes.
+.PP
+.I Bgetd
+uses
+.I fmtcharstod
+(undocumented)
+and
+.I Bgetc
+to read the formatted
+floating-point number in the input stream,
+skipping initial blanks and tabs.
+The value is stored in
+.BR *d.
+.PP
+.I Bread
+reads
+.I nbytes
+of data from
+.I bp
+into memory starting at
+.IR addr .
+The number of bytes read is returned on success
+and a negative value is returned if a read error occurred.
+.PP
+.I Bseek
+applies
+.IR lseek (2)
+to
+.IR bp .
+It returns the new file offset.
+.I Boffset
+returns the file offset of the next character to be processed.
+.PP
+.I Bputc
+outputs the low order 8 bits of
+.I c
+on
+.IR bp .
+If this causes a
+.IR write
+to occur and there is an error,
+a negative value is returned.
+Otherwise, a zero is returned.
+.PP
+.I Bputrune
+calls
+.I Bputc
+to output the low order
+16 bits of
+.I c
+as a rune
+in
+.SM UTF
+format
+on the output stream.
+.PP
+.I Bprint
+is a buffered interface to
+.IR print (2).
+If this causes a
+.IR write
+to occur and there is an error,
+a negative value
+.RB ( Beof )
+is returned.
+Otherwise, the number of bytes output is returned.
+.I Bvprint
+does the same except it takes as argument a
+.B va_list
+parameter, so it can be called within a variadic function.
+.PP
+.I Bwrite
+outputs
+.I nbytes
+of data starting at
+.I addr
+to
+.IR bp .
+If this causes a
+.IR write
+to occur and there is an error,
+a negative value is returned.
+Otherwise, the number of bytes written is returned.
+.PP
+.I Bflush
+causes any buffered output associated with
+.I bp
+to be written.
+The return is as for
+.IR Bputc .
+.I Bflush
+is called on
+exit for every buffer still open
+for writing.
+.PP
+.I Bbuffered
+returns the number of bytes in the buffer.
+When reading, this is the number of bytes still available from the last
+read on the file; when writing, it is the number of bytes ready to be
+written.
+.PP
+This library uses
+.IR fmt (3)
+for diagnostic messages about internal errors,
+as well as for the implementation of
+.I Bprint
+and
+.IR Bvprint .
+It uses
+.IR utf (3)
+for the implementation of
+.I Bgetrune
+and
+.IR Bputrune .
+.SH SEE ALSO
+.IR atexit (3).
+.IR open (2),
+.IR print (3),
+.IR utf (7)
+.SH DIAGNOSTICS
+.I Bio
+routines that return integers yield
+.B Beof
+if
+.I bp
+is not the descriptor of an open file.
+.I Bopen
+returns zero if the file cannot be opened in the given mode.
+.SH HISTORY
+The
+.IR bio (3)
+library originally appeared in Plan 9.
+This is a port of the Plan 9 bio library.
+.SH BUGS
+.I Brdline
+returns an error on strings longer than the buffer associated
+with the file
+and also if the end-of-file is encountered
+before a delimiter.
+.I Blinelen
+will tell how many characters are available
+in these cases.
+In the case of a true end-of-file,
+.I Blinelen
+will return zero.
+At the cost of allocating a buffer,
+.I Brdstr
+sidesteps these issues.
+.PP
+The data returned by
+.I Brdline
+may be overwritten by calls to any other
+.I bio
+routine on the same
+.IR bp.
DIR diff --git a/man/man3/fmtinstall.3 b/man/man3/fmtinstall.3
t@@ -0,0 +1,346 @@
+.TH FMTINSTALL 3
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.SH NAME
+fmtinstall, dofmt, fmtprint, fmtvprint, fmtstrcpy, fmtfdinit, fmtfdflush, fmtstrinit, fmtstrflush \- support for user-defined print formats and output routines
+.SH SYNOPSIS
+.B #include <fmt.h>
+.PP
+.ft L
+.nf
+.ta \w' 'u +\w' 'u +\w' 'u +\w' 'u +\w' 'u
+typedef struct Fmt Fmt;
+struct Fmt{
+ void *start; /* of buffer */
+ void *to; /* current place in the buffer */
+ void *stop; /* end of the buffer; overwritten if flush fails */
+ int (*flush)(Fmt*); /* called when to == stop */
+ void *farg; /* to make flush a closure */
+ int nfmt; /* num chars formatted so far */
+ va_list args; /* args passed to dofmt */
+ int r; /* % format character */
+ int width;
+ int prec;
+ unsigned long flags;
+};
+
+enum{
+ FmtWidth = 1,
+ FmtLeft = FmtWidth << 1,
+ FmtPrec = FmtLeft << 1,
+ FmtSharp = FmtPrec << 1,
+ FmtSpace = FmtSharp << 1,
+ FmtSign = FmtSpace << 1,
+ FmtZero = FmtSign << 1,
+ FmtUnsigned = FmtZero << 1,
+ FmtShort = FmtUnsigned << 1,
+ FmtLong = FmtShort << 1,
+ FmtVLong = FmtLong << 1,
+ FmtComma = FmtVLong << 1,
+ FmtByte = FmtComma << 1,
+ FmtLDouble = FmtByte << 1,
+
+ FmtFlag = FmtLDouble << 1
+};
+.fi
+.PP
+.B
+.ta \w'\fLchar* 'u
+
+.PP
+.B
+int fmtfdinit(Fmt *f, int fd, char *buf, int nbuf);
+.PP
+.B
+int fmtfdflush(Fmt *f);
+.PP
+.B
+int fmtstrinit(Fmt *f);
+.PP
+.B
+char* fmtstrflush(Fmt *f);
+.PP
+.B
+int fmtinstall(int c, int (*fn)(Fmt*));
+.PP
+.B
+int dofmt(Fmt *f, char *fmt);
+.PP
+.B
+int fmtprint(Fmt *f, char *fmt, ...);
+.PP
+.B
+int fmtvprint(Fmt *f, char *fmt, va_list v);
+.PP
+.B
+int fmtrune(Fmt *f, int r);
+.PP
+.B
+int fmtstrcpy(Fmt *f, char *s);
+.SH DESCRIPTION
+The interface described here allows the construction of custom
+.IR print (3)
+verbs and output routines.
+In essence, they provide access to the workings of the formatted print code.
+.PP
+The
+.IR print (3)
+suite maintains its state with a data structure called
+.BR Fmt .
+A typical call to
+.IR print (3)
+or its relatives initializes a
+.B Fmt
+structure, passes it to subsidiary routines to process the output,
+and finishes by emitting any saved state recorded in the
+.BR Fmt .
+The details of the
+.B Fmt
+are unimportant to outside users, except insofar as the general
+design influences the interface.
+The
+.B Fmt
+records
+the verb being processed, its precision and width,
+and buffering parameters.
+Most important, it also records a
+.I flush
+routine that the library will call if a buffer overflows.
+When printing to a file descriptor, the flush routine will
+emit saved characters and reset the buffer; when printing
+to an allocated string, it will resize the string to receive more output.
+The flush routine is nil when printing to fixed-size buffers.
+User code need never provide a flush routine; this is done internally
+by the library.
+.SS Custom output routines
+To write a custom output routine, such as an error handler that
+formats and prints custom error messages, the output sequence can be run
+from outside the library using the routines described here.
+There are two main cases: output to an open file descriptor
+and output to a string.
+.PP
+To write to a file descriptor, call
+.I fmtfdinit
+to initialize the local
+.B Fmt
+structure
+.IR f ,
+giving the file descriptor
+.IR fd ,
+the buffer
+.IR buf ,
+and its size
+.IR nbuf .
+Then call
+.IR fmtprint
+or
+.IR fmtvprint
+to generate the output.
+These behave just like
+.B fprint
+(see
+.IR print (3))
+or
+.B vfprint
+except that the characters are buffered until
+.I fmtfdflush
+is called.
+A typical example of this sequence appears in the Examples section.
+.PP
+The same basic sequence applies when outputting to an allocated string:
+call
+.I fmtstrinit
+to initialize the
+.BR Fmt ,
+then call
+.I fmtprint
+and
+.I fmtvprint
+to generate the output.
+Finally,
+.I fmtstrflush
+will return the allocated string, which should be freed after use.
+Regardless of the output style or type,
+.I fmtprint
+or
+.I fmtvprint
+generates the characters.
+.SS Custom format verbs
+.I Fmtinstall
+is used to install custom verbs and flags labeled by character
+.IR c ,
+which may be any non-zero Unicode character.
+.I Fn
+should be declared as
+.IP
+.EX
+int fn(Fmt*)
+.EE
+.PP
+.IB Fp ->r
+is the flag or verb character to cause
+.I fn
+to be called.
+In
+.IR fn ,
+.IB fp ->width ,
+.IB fp ->prec
+are the width and precision, and
+.IB fp ->flags
+the decoded flags for the verb (see
+.IR print (3)
+for a description of these items).
+The standard flag values are:
+.B FmtSign
+.RB ( + ),
+.B FmtLeft
+.RB ( - ),
+.B FmtSpace
+.RB ( '\ ' ),
+.B FmtSharp
+.RB ( # ),
+.B FmtComma
+.RB ( , ),
+.B FmtLong
+.RB ( l ),
+.B FmtShort
+.RB ( h ),
+.B FmtByte
+.RB ( hh ),
+.B FmtUnsigned
+.RB ( u ),
+.B FmtLDouble
+.RB ( L ),
+and
+.B FmtVLong
+.RB ( ll ).
+The flag bits
+.B FmtWidth
+and
+.B FmtPrec
+identify whether a width and precision were specified.
+.PP
+.I Fn
+is passed a pointer to the
+.B Fmt
+structure recording the state of the output.
+If
+.IB fp ->r
+is a verb (rather than a flag),
+.I fn
+should use
+.B Fmt->args
+to fetch its argument from the list,
+then format it, and return zero.
+If
+.IB fp ->r
+is a flag,
+.I fn
+should return a negative value:
+the negation of one of the above flag values, or some otherwise unused power of two.
+All interpretation of
+.IB fp ->width\f1,
+.IB fp ->prec\f1,
+and
+.IB fp-> flags
+is left up to the conversion routine.
+.I Fmtinstall
+returns 0 if the installation succeeds, \-1 if it fails.
+.PP
+.IR Fmtprint
+and
+.IR fmtvprint
+may be called to
+help prepare output in custom conversion routines.
+However, these functions clear the width, precision, and flags.
+The function
+.I dofmt
+is the underlying formatter; it
+uses the existing contents of
+.B Fmt
+and should be called only by sophisticated conversion routines.
+All these routines return the number of characters
+produced.
+.PP
+Some internal functions may be useful to format primitive types.
+They honor the width, precision and flags as described in
+.IR print (3).
+.I Fmtrune
+formats a single character
+.BR r .
+.I Fmtstrcpy
+formats a string
+.BR s .
+All these routines return zero for successful execution.
+.SH EXAMPLES
+This function prints an error message with a variable
+number of arguments and then quits.
+Compared to the corresponding example in
+.IR print (3),
+this version uses a smaller buffer, will never truncate
+the output message, but might generate multiple
+.B write
+system calls to produce its output.
+.IP
+.EX
+.ta 6n +6n +6n +6n +6n +6n +6n +6n +6n
+
+void fatal(char *fmt, ...)
+{
+ Fmt f;
+ char buf[64];
+ va_list arg;
+
+ fmtfdinit(&f, 1, buf, sizeof buf);
+ fmtprint(&f, "fatal: ");
+ va_start(arg, fmt);
+ fmtvprint(&f, fmt, arg);
+ va_end(arg);
+ fmtprint(&f, "\en");
+ fmtfdflush(&f);
+ exits("fatal error");
+}
+.EE
+.PP
+This example adds a verb to print complex numbers.
+.IP
+.EX
+typedef
+struct {
+ double r, i;
+} Complex;
+
+int
+Xfmt(Fmt *f)
+{
+ Complex c;
+
+ c = va_arg(f->args, Complex);
+ return fmtprint(f, "(%g,%g)", c.r, c.i);
+}
+
+main(...)
+{
+ Complex x;
+
+ x.r = 1.5;
+ x.i = -2.3;
+
+ fmtinstall('X', Xfmt);
+ print("x = %X\en", x);
+}
+.EE
+.SH SEE ALSO
+.IR print (3)
+.SH HISTORY
+This formatted print library originally
+appeared as part of the Plan 9 C library.
+.SH BUGS
+The Plan 9 version supports Unicode strings and produces UTF output.
+This version assumes that characters are always represented by 1-byte values.
DIR diff --git a/man/man3/ioproc.3 b/man/man3/ioproc.3
t@@ -0,0 +1,179 @@
+.TH IOPROC 2
+.SH NAME
+closeioproc,
+iocall,
+ioclose,
+iointerrupt,
+iodial,
+ioopen,
+ioproc,
+ioread,
+ioreadn,
+iowrite \- slave I/O processes for threaded programs
+.SH SYNOPSIS
+.PP
+.de XX
+.ift .sp 0.5
+.ifn .sp
+..
+.EX
+.ta \w'Ioproc* 'u
+#include <u.h>
+#include <libc.h>
+#include <thread.h>
+.sp
+typedef struct Ioproc Ioproc;
+.sp
+Ioproc* ioproc(void);
+.XX
+int ioopen(Ioproc *io, char *file, int omode);
+int ioclose(Ioproc *io, int fd);
+long ioread(Ioproc *io, int fd, void *a, long n);
+long ioreadn(Ioproc *io, int fd, void *a, long n);
+long iowrite(Ioproc *io, int fd, void *a, long n);
+int iodial(Ioproc *io, char *addr, char *local, char *dir, char *cdfp);
+.XX
+void iointerrupt(Ioproc *io);
+void closeioproc(Ioproc *io);
+.XX
+long iocall(Ioproc *io, long (*op)(va_list *arg), ...);
+.EE
+.SH DESCRIPTION
+.PP
+These routines provide access to I/O in slave procs.
+Since the I/O itself is done in a slave proc, other threads
+in the calling proc can run while the calling thread
+waits for the I/O to complete.
+.PP
+.I Ioproc
+forks a new slave proc and returns a pointer to the
+.B Ioproc
+associated with it.
+.I Ioproc
+uses
+.I mallocz
+and
+.IR proccreate ;
+if either fails, it calls
+.I sysfatal
+rather than return an error.
+.PP
+.IR Ioopen ,
+.IR ioclose ,
+.IR ioread ,
+.IR ioreadn ,
+.IR iowrite ,
+and
+.IR iodial
+are execute the
+similarly named library or system calls
+(see
+.IR open (2),
+.IR read (2),
+and
+.IR dial (2))
+in the slave process associated with
+.IR io .
+It is an error to execute more than one call
+at a time in an I/O proc.
+.PP
+.I Iointerrupt
+interrupts the call currently executing in the I/O proc.
+If no call is executing,
+.IR iointerrupt
+is a no-op.
+.PP
+.I Closeioproc
+terminates the I/O proc and frees the associated
+.B Ioproc .
+.PP
+.I Iocall
+is a primitive that may be used to implement
+more slave I/O routines.
+.I Iocall
+arranges for
+.I op
+to be called in
+.IR io 's
+proc, with
+.I arg
+set to the variable parameter list,
+returning the value that
+.I op
+returns.
+.SH EXAMPLE
+Relay messages between two file descriptors,
+counting the total number of bytes seen:
+.IP
+.EX
+.ta +\w'xxxx'u +\w'xxxx'u +\w'xxxx'u
+int tot;
+
+void
+relaythread(void *v)
+{
+ int *fd, n;
+ char buf[1024];
+ Ioproc *io;
+
+ fd = v;
+ io = ioproc();
+ while((n = ioread(io, fd[0], buf, sizeof buf)) > 0){
+ if(iowrite(io, fd[1], buf, n) != n)
+ sysfatal("iowrite: %r");
+ tot += n;
+ }
+ closeioproc(io);
+}
+
+void
+relay(int fd0, int fd1)
+{
+ int fd[4];
+
+ fd[0] = fd[3] = fd0;
+ fd[1] = fd[2] = fd1;
+ threadcreate(relaythread, fd, 8192);
+ threadcreate(relaythread, fd+2, 8192);
+}
+.EE
+.LP
+If the two
+.I relaythread
+instances were running in different procs, the
+common access to
+.I tot
+would be unsafe.
+.EE
+.PP
+Implement
+.IR ioread :
+.IP
+.EX
+static long
+_ioread(va_list *arg)
+{
+ int fd;
+ void *a;
+ long n;
+
+ fd = va_arg(*arg, int);
+ a = va_arg(*arg, void*);
+ n = va_arg(*arg, long);
+ return read(fd, a, n);
+}
+
+long
+ioread(Ioproc *io, int fd, void *a, long n)
+{
+ return iocall(io, _ioread, fd, a, n);
+}
+.EE
+.SH SOURCE
+.B /sys/src/libthread/io*.c
+.SH SEE ALSO
+.IR dial (2),
+.IR open (2),
+.IR read (2),
+.IR thread (2)
+
DIR diff --git a/man/man3/isalpharune.3 b/man/man3/isalpharune.3
t@@ -0,0 +1,47 @@
+.TH ISALPHARUNE 3
+.SH NAME
+isalpharune, islowerrune, isspacerune, istitlerune, isupperrune, tolowerrune, totitlerune, toupperrune \- Unicode character classes and cases
+.SH SYNOPSIS
+.B #include <utf.h>
+.PP
+.B
+int isalpharune(Rune c)
+.PP
+.B
+int islowerrune(Rune c)
+.PP
+.B
+int isspacerune(Rune c)
+.PP
+.B
+int istitlerune(Rune c)
+.PP
+.B
+int isupperrune(Rune c)
+.PP
+.B
+Rune tolowerrune(Rune c)
+.PP
+.B
+Rune totitlerune(Rune c)
+.PP
+.B
+Rune toupperrune(Rune c)
+.SH DESCRIPTION
+These routines examine and operate on Unicode characters,
+in particular a subset of their properties as defined in the Unicode standard.
+Unicode defines some characters as alphabetic and specifies three cases:
+upper, lower, and title.
+Analogously to
+.IR ctype (3)
+for
+.SM ASCII\c
+,
+these routines
+test types and modify cases for Unicode characters.
+The names are self-explanatory.
+.PP
+The case-conversion routines return the character unchanged if it has no case.
+.SH "SEE ALSO
+.IR ctype (3) ,
+.IR "The Unicode Standard" .
DIR diff --git a/man/man3/print.3 b/man/man3/print.3
t@@ -0,0 +1,469 @@
+.TH PRINT 3
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.SH NAME
+print, fprint, sprint, snprint, seprint, smprint, vfprint, vsnprint, vseprint, vsmprint \- print formatted output
+.SH SYNOPSIS
+.B #include <utf.h>
+.PP
+.B #include <fmt.h>
+.PP
+.ta \w'\fLchar* 'u
+.B
+int print(char *format, ...)
+.PP
+.B
+int fprint(int fd, char *format, ...)
+.PP
+.B
+int sprint(char *s, char *format, ...)
+.PP
+.B
+int snprint(char *s, int len, char *format, ...)
+.PP
+.B
+char* seprint(char *s, char *e, char *format, ...)
+.PP
+.B
+char* smprint(char *format, ...)
+.PP
+.B
+int runesprint(Rune *s, char *format, ...)
+.PP
+.B
+int runesnprint(Rune *s, int len, char *format, ...)
+.PP
+.B
+Rune* runeseprint(Rune *s, Rune *e, char *format, ...)
+.PP
+.B
+Rune* runesmprint(char *format, ...)
+.PP
+.B
+int vfprint(int fd, char *format, va_list v)
+.PP
+.B
+int vsnprint(char *s, int len, char *format, va_list v)
+.PP
+.B
+char* vseprint(char *s, char *e, char *format, va_list v)
+.PP
+.B
+char* vsmprint(char *format, va_list v)
+.PP
+.B
+int runevsnprint(Rune *s, int len, char *format, va_list v)
+.PP
+.B
+Rune* runevseprint(Rune *s, Rune *e, char *format, va_list v)
+.PP
+.B
+Rune* runevsmprint(Rune *format, va_list v)
+.PP
+.B
+.SH DESCRIPTION
+.I Print
+writes text to the standard output.
+.I Fprint
+writes to the named output
+file descriptor.
+.I Sprint
+places text
+followed by the NUL character
+.RB ( \e0 )
+in consecutive bytes starting at
+.IR s ;
+it is the user's responsibility to ensure that
+enough storage is available.
+Each function returns the number of bytes
+transmitted (not including the NUL
+in the case of
+.IR sprint ),
+or
+a negative value if an output error was encountered.
+.PP
+.I Snprint
+is like
+.IR sprint ,
+but will not place more than
+.I len
+bytes in
+.IR s .
+Its result is always NUL-terminated and holds the maximal
+number of characters that can fit.
+.I Seprint
+is like
+.IR snprint ,
+except that the end is indicated by a pointer
+.I e
+rather than a count and the return value points to the terminating NUL of the
+resulting string.
+.I Smprint
+is like
+.IR sprint ,
+except that it prints into and returns a string of the required length, which is
+allocated by
+.IR malloc (3).
+.PP
+The routines
+.IR runesprint ,
+.IR runesnprint ,
+.IR runeseprint ,
+and
+.I runesmprint
+are the same as
+.IR sprint ,
+.IR snprint ,
+.IR seprint
+and
+.I smprint
+except that their output is rune strings instead of byte strings.
+.PP
+Finally, the routines
+.IR vfprint ,
+.IR vsnprint ,
+.IR vseprint ,
+.IR vsmprint ,
+.IR runevsnprint ,
+.IR runevseprint ,
+and
+.I runevsmprint
+are like their
+.BR v-less
+relatives except they take as arguments a
+.B va_list
+parameter, so they can be called within a variadic function.
+The Example section shows a representative usage.
+.PP
+Each of these functions
+converts, formats, and prints its
+trailing arguments
+under control of a
+.IR format
+string.
+The
+format
+contains two types of objects:
+plain characters, which are simply copied to the
+output stream,
+and conversion specifications,
+each of which results in fetching of
+zero or more
+arguments.
+The results are undefined if there are arguments of the
+wrong type or too few
+arguments for the format.
+If the format is exhausted while
+arguments remain, the excess
+is ignored.
+.PP
+Each conversion specification has the following format:
+.IP
+.B "% [flags] verb
+.PP
+The verb is a single character and each flag is a single character or a
+(decimal) numeric string.
+Up to two numeric strings may be used;
+the first is called
+.IR width ,
+the second
+.IR precision .
+A period can be used to separate them, and if the period is
+present then
+.I width
+and
+.I precision
+are taken to be zero if missing, otherwise they are `omitted'.
+Either or both of the numbers may be replaced with the character
+.BR * ,
+meaning that the actual number will be obtained from the argument list
+as an integer.
+The flags and numbers are arguments to
+the
+.I verb
+described below.
+.PP
+The numeric verbs
+.BR d ,
+.BR i ,
+.BR u ,
+.BR o ,
+.BR b ,
+.BR x ,
+and
+.B X
+format their arguments in decimal, decimal,
+unsigned decimal, octal, binary, hexadecimal, and upper case hexadecimal.
+Each interprets the flags
+.BR 0 ,
+.BR h ,
+.BR hh ,
+.BR l ,
+.BR + ,
+.BR - ,
+.BR , ,
+and
+.B #
+to mean pad with zeros,
+short, byte, long, always print a sign, left justified, commas every three digits,
+and alternate format.
+Also, a space character in the flag
+position is like
+.BR + ,
+but prints a space instead of a plus sign for non-negative values.
+If neither
+short nor long is specified,
+then the argument is an
+.BR int .
+If an unsigned verb is specified,
+then the argument is interpreted as a
+positive number and no sign is output;
+space and
+.B +
+flags are ignored for unsigned verbs.
+If two
+.B l
+flags are given,
+then the argument is interpreted as a
+.B vlong
+(usually an 8-byte, sometimes a 4-byte integer).
+If
+.I precision
+is not omitted, the number is padded on the left with zeros
+until at least
+.I precision
+digits appear.
+If
+.I precision
+is explicitly 0, and the number is 0,
+no digits are generated, and alternate formatting
+does not apply.
+Then, if alternate format is specified,
+for
+.B o
+conversion, the number is preceded by a
+.B 0
+if it doesn't already begin with one.
+For non-zero numbers and
+.B x
+conversion, the number is preceded by
+.BR 0x ;
+for
+.B X
+conversion, the number is preceded by
+.BR 0X .
+Finally, if
+.I width
+is not omitted, the number is padded on the left (or right, if
+left justification is specified) with enough blanks to
+make the field at least
+.I width
+characters long.
+.PP
+The floating point verbs
+.BR f ,
+.BR e ,
+.BR E ,
+.BR g ,
+and
+.B G
+take a
+.B double
+argument.
+Each interprets the flags
+.BR 0 ,
+.BR L
+.BR + ,
+.BR - ,
+and
+.B #
+to mean pad with zeros,
+long double argument,
+always print a sign,
+left justified,
+and
+alternate format.
+.I Width
+is the minimum field width and,
+if the converted value takes up less than
+.I width
+characters, it is padded on the left (or right, if `left justified')
+with spaces.
+.I Precision
+is the number of digits that are converted after the decimal place for
+.BR e ,
+.BR E ,
+and
+.B f
+conversions,
+and
+.I precision
+is the maximum number of significant digits for
+.B g
+and
+.B G
+conversions.
+The
+.B f
+verb produces output of the form
+.RB [ - ] digits [ .digits\fR].
+.B E
+conversion appends an exponent
+.BR E [ - ] digits ,
+and
+.B e
+conversion appends an exponent
+.BR e [ - ] digits .
+The
+.B g
+verb will output the argument in either
+.B e
+or
+.B f
+with the goal of producing the smallest output.
+Also, trailing zeros are omitted from the fraction part of
+the output, and a trailing decimal point appears only if it is followed
+by a digit.
+The
+.B G
+verb is similar, but uses
+.B E
+format instead of
+.BR e .
+When alternate format is specified, the result will always contain a decimal point,
+and for
+.B g
+and
+.B G
+conversions, trailing zeros are not removed.
+.PP
+The
+.B s
+verb copies a string
+(pointer to
+.BR char )
+to the output.
+The number of characters copied
+.RI ( n )
+is the minimum
+of the size of the string and
+.IR precision .
+These
+.I n
+characters are justified within a field of
+.I width
+characters as described above.
+If a
+.I precision
+is given, it is safe for the string not to be nul-terminated
+as long as it is at least
+.I precision
+characters (not bytes!) long.
+The
+.B S
+verb is similar, but it interprets its pointer as an array
+of runes (see
+.IR utf (7));
+the runes are converted to
+.SM UTF
+before output.
+.PP
+The
+.B c
+verb copies a single
+.B char
+(promoted to
+.BR int )
+justified within a field of
+.I width
+characters as described above.
+The
+.B C
+verb is similar, but works on runes.
+.PP
+The
+.B p
+verb formats a pointer value.
+At the moment, it is a synonym for
+.BR x ,
+but that will change if pointers and integers are different sizes.
+.PP
+The
+.B r
+verb takes no arguments; it copies the error string returned by a call to
+.IR strerror (3)
+with an argument of
+.IR errno.
+.PP
+Custom verbs may be installed using
+.IR fmtinstall (3).
+.SH EXAMPLE
+This function prints an error message with a variable
+number of arguments and then quits.
+.IP
+.EX
+.ta 6n +6n +6n
+void fatal(char *msg, ...)
+{
+ char buf[1024], *out;
+ va_list arg;
+
+ out = vseprint(buf, buf+sizeof buf, "Fatal error: ");
+ va_start(arg, msg);
+ out = vseprint(out, buf+sizeof buf, msg, arg);
+ va_end(arg);
+ write(2, buf, out-buf);
+ exit(1);
+}
+.EE
+.SH SEE ALSO
+.IR fmtinstall (3),
+.IR fprintf (3),
+.IR utf (7)
+.SH DIAGNOSTICS
+Routines that write to a file descriptor or call
+.IR malloc
+set
+.IR errstr .
+.SH BUGS
+The formatting is close to that specified for ANSI
+.IR fprintf (3);
+the main difference is that
+.B b
+and
+.B r
+are not in ANSI and some
+.B C9X
+verbs are missing.
+Also, and distinctly not a bug,
+.I print
+and friends generate
+.SM UTF
+rather than
+.SM ASCII.
+.PP
+There is no
+.BR runeprint ,
+.BR runefprint ,
+etc. because runes are byte-order dependent and should not be written directly to a file; use the
+UTF output of
+.I print
+or
+.I fprint
+instead.
+Also,
+.I sprint
+is deprecated for safety reasons; use
+.IR snprint ,
+.IR seprint ,
+or
+.I smprint
+instead.
+Safety also precludes the existence of
+.IR runesprint .
DIR diff --git a/man/man3/regexp9.3 b/man/man3/regexp9.3
t@@ -0,0 +1,227 @@
+.TH REGEXP9 3
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.de LR
+.if t .BR \\$1 \\$2
+.if n .RB ` \\$1 '\\$2
+..
+.de L
+.nh
+.if t .B \\$1
+.if n .RB ` \\$1 '
+..
+.SH NAME
+regcomp, regcomplit, regcompnl, regexec, regsub, regerror \- Plan 9 regular expression library
+.SH SYNOPSIS
+.B #include <regexp9.h>
+.PP
+.ta \w'\fLRegprog 'u
+.B
+Reprog *regcomp(char *exp)
+.PP
+.B
+Reprog *regcomplit(char *exp)
+.PP
+.B
+Reprog *regcompnl(char *exp)
+.PP
+.nf
+.B
+int regexec(Reprog *prog, char *string, Resub *match, int msize)
+.PP
+.nf
+.B
+void regsub(char *source, char *dest, int dlen, Resub *match, int msize)
+.PP
+.nf
+.B
+int rregexec(Reprog *prog, Rune *string, Resub *match, int msize)
+.PP
+.nf
+.B
+void rregsub(Rune *source, Rune *dest, int dlen, Resub *match, int msize)
+.PP
+.B
+void regerror(char *msg)
+.SH DESCRIPTION
+.I Regcomp
+compiles a
+regular expression and returns
+a pointer to the generated description.
+The space is allocated by
+.IR malloc (3)
+and may be released by
+.IR free .
+Regular expressions are exactly as in
+.IR regexp9 (7).
+.PP
+.I Regcomplit
+is like
+.I regcomp
+except that all characters are treated literally.
+.I Regcompnl
+is like
+.I regcomp
+except that the
+.B .
+metacharacter matches all characters, including newlines.
+.PP
+.I Regexec
+matches a null-terminated
+.I string
+against the compiled regular expression in
+.IR prog .
+If it matches,
+.I regexec
+returns
+.B 1
+and fills in the array
+.I match
+with character pointers to the substrings of
+.I string
+that correspond to the
+parenthesized subexpressions of
+.IR exp :
+.BI match[ i ].sp
+points to the beginning and
+.BI match[ i ].ep
+points just beyond
+the end of the
+.IR i th
+substring.
+(Subexpression
+.I i
+begins at the
+.IR i th
+left parenthesis, counting from 1.)
+Pointers in
+.B match[0]
+pick out the substring that corresponds to
+the whole regular expression.
+Unused elements of
+.I match
+are filled with zeros.
+Matches involving
+.LR * ,
+.LR + ,
+and
+.L ?
+are extended as far as possible.
+The number of array elements in
+.I match
+is given by
+.IR msize .
+The structure of elements of
+.I match
+is:
+.IP
+.EX
+typedef struct {
+ union {
+ char *sp;
+ Rune *rsp;
+ } s;
+ union {
+ char *ep;
+ Rune *rep;
+ } e;
+} Resub;
+.EE
+.LP
+If
+.B match[0].s.sp
+is nonzero on entry,
+.I regexec
+starts matching at that point within
+.IR string .
+If
+.B match[0].e.ep
+is nonzero on entry,
+the last character matched is the one
+preceding that point.
+.PP
+.I Regsub
+places in
+.I dest
+a substitution instance of
+.I source
+in the context of the last
+.I regexec
+performed using
+.IR match .
+Each instance of
+.BI \e n\f1,
+where
+.I n
+is a digit, is replaced by the
+string delimited by
+.BI match[ n ].s.sp
+and
+.BI match[ n ].e.ep\f1.
+Each instance of
+.L &
+is replaced by the string delimited by
+.B match[0].s.sp
+and
+.BR match[0].e.ep .
+The substitution will always be null terminated and
+trimmed to fit into dlen bytes.
+.PP
+.IR Regerror ,
+called whenever an error is detected in
+.IR regcomp ,
+writes the string
+.I msg
+on the standard error file and exits.
+.I Regerror
+can be replaced to perform
+special error processing.
+If the user supplied
+.I regerror
+returns rather than exits,
+.I regcomp
+will return 0.
+.PP
+.I Rregexec
+and
+.I rregsub
+are variants of
+.I regexec
+and
+.I regsub
+that use strings of
+.B Runes
+instead of strings of
+.BR chars .
+With these routines, the
+.I rsp
+and
+.I rep
+fields of the
+.I match
+array elements should be used.
+.SH "SEE ALSO"
+.IR grep (1),
+.IR regexp9 (7)
+.SH DIAGNOSTICS
+.I Regcomp
+returns
+.B 0
+for an illegal expression
+or other failure.
+.I Regexec
+returns 0
+if
+.I string
+is not matched.
+.SH HISTORY
+This particular regular expression was first written by Rob Pike for Plan 9.
+It has also appeared as part of the Inferno operating system.
+.SH BUGS
+There is no way to specify or match a NUL character; NULs terminate patterns and strings.
DIR diff --git a/man/man3/rune.3 b/man/man3/rune.3
t@@ -0,0 +1,187 @@
+.TH RUNE 3
+.SH NAME
+runetochar, chartorune, runelen, runenlen, fullrune, utfecpy, utflen, utfnlen, utfrune, utfrrune, utfutf \- rune/UTF conversion
+.SH SYNOPSIS
+.ta \w'\fLchar*xx'u
+.B #include <utf.h>
+.PP
+.B
+int runetochar(char *s, Rune *r)
+.PP
+.B
+int chartorune(Rune *r, char *s)
+.PP
+.B
+int runelen(long r)
+.PP
+.B
+int runenlen(Rune *r, int n)
+.PP
+.B
+int fullrune(char *s, int n)
+.PP
+.B
+char* utfecpy(char *s1, char *es1, char *s2)
+.PP
+.B
+int utflen(char *s)
+.PP
+.B
+int utfnlen(char *s, long n)
+.PP
+.B
+char* utfrune(char *s, long c)
+.PP
+.B
+char* utfrrune(char *s, long c)
+.PP
+.B
+char* utfutf(char *s1, char *s2)
+.SH DESCRIPTION
+These routines convert to and from a
+.SM UTF
+byte stream and runes.
+.PP
+.I Runetochar
+copies one rune at
+.I r
+to at most
+.B UTFmax
+bytes starting at
+.I s
+and returns the number of bytes copied.
+.BR UTFmax ,
+defined as
+.B 3
+in
+.BR <libc.h> ,
+is the maximum number of bytes required to represent a rune.
+.PP
+.I Chartorune
+copies at most
+.B UTFmax
+bytes starting at
+.I s
+to one rune at
+.I r
+and returns the number of bytes copied.
+If the input is not exactly in
+.SM UTF
+format,
+.I chartorune
+will convert to 0x80 and return 1.
+.PP
+.I Runelen
+returns the number of bytes
+required to convert
+.I r
+into
+.SM UTF.
+.PP
+.I Runenlen
+returns the number of bytes
+required to convert the
+.I n
+runes pointed to by
+.I r
+into
+.SM UTF.
+.PP
+.I Fullrune
+returns 1 if the string
+.I s
+of length
+.I n
+is long enough to be decoded by
+.I chartorune
+and 0 otherwise.
+This does not guarantee that the string
+contains a legal
+.SM UTF
+encoding.
+This routine is used by programs that
+obtain input a byte at
+a time and need to know when a full rune
+has arrived.
+.PP
+The following routines are analogous to the
+corresponding string routines with
+.B utf
+substituted for
+.B str
+and
+.B rune
+substituted for
+.BR chr .
+.PP
+.I Utfecpy
+copies UTF sequences until a null sequence has been copied, but writes no
+sequences beyond
+.IR es1 .
+If any sequences are copied,
+.I s1
+is terminated by a null sequence, and a pointer to that sequence is returned.
+Otherwise, the original
+.I s1
+is returned.
+.PP
+.I Utflen
+returns the number of runes that
+are represented by the
+.SM UTF
+string
+.IR s .
+.PP
+.I Utfnlen
+returns the number of complete runes that
+are represented by the first
+.I n
+bytes of
+.SM UTF
+string
+.IR s .
+If the last few bytes of the string contain an incompletely coded rune,
+.I utfnlen
+will not count them; in this way, it differs from
+.IR utflen ,
+which includes every byte of the string.
+.PP
+.I Utfrune
+.RI ( utfrrune )
+returns a pointer to the first (last)
+occurrence of rune
+.I c
+in the
+.SM UTF
+string
+.IR s ,
+or 0 if
+.I c
+does not occur in the string.
+The NUL byte terminating a string is considered to
+be part of the string
+.IR s .
+.PP
+.I Utfutf
+returns a pointer to the first occurrence of
+the
+.SM UTF
+string
+.I s2
+as a
+.SM UTF
+substring of
+.IR s1 ,
+or 0 if there is none.
+If
+.I s2
+is the null string,
+.I utfutf
+returns
+.IR s1 .
+.SH HISTORY
+These routines were written by Rob Pike and Ken Thompson
+and first appeared in Plan 9.
+.SH SEE ALSO
+.IR utf (7),
+.IR tcs (1)
DIR diff --git a/man/man3/runestrcat.3 b/man/man3/runestrcat.3
t@@ -0,0 +1,65 @@
+.TH RUNESTRCAT 3
+.SH NAME
+runestrcat,
+runestrncat,
+runestrcmp,
+runestrncmp,
+runestrcpy,
+runestrncpy,
+runestrecpy,
+runestrlen,
+runestrchr,
+runestrrchr,
+runestrdup,
+runestrstr \- rune string operations
+.SH SYNOPSIS
+.B #include <utf.h>
+.PP
+.ta \w'\fLRune* \fP'u
+.B
+Rune* runestrcat(Rune *s1, Rune *s2)
+.PP
+.B
+Rune* runestrncat(Rune *s1, Rune *s2, long n)
+.PP
+.B
+int runestrcmp(Rune *s1, Rune *s2)
+.PP
+.B
+int runestrncmp(Rune *s1, Rune *s2, long n)
+.PP
+.B
+Rune* runestrcpy(Rune *s1, Rune *s2)
+.PP
+.B
+Rune* runestrncpy(Rune *s1, Rune *s2, long n)
+.PP
+.B
+Rune* runestrecpy(Rune *s1, Rune *es1, Rune *s2)
+.PP
+.B
+long runestrlen(Rune *s)
+.PP
+.B
+Rune* runestrchr(Rune *s, Rune c)
+.PP
+.B
+Rune* runestrrchr(Rune *s, Rune c)
+.PP
+.B
+Rune* runestrdup(Rune *s)
+.PP
+.B
+Rune* runestrstr(Rune *s1, Rune *s2)
+.SH DESCRIPTION
+These functions are rune string analogues of
+the corresponding functions in
+.IR strcat (3).
+.SH HISTORY
+These routines first appeared in Plan 9.
+.SH SEE ALSO
+.IR memmove (3),
+.IR rune (3),
+.IR strcat (2)
+.SH BUGS
+The outcome of overlapping moves varies among implementations.
DIR diff --git a/man/man3/thread.3 b/man/man3/thread.3
t@@ -0,0 +1,576 @@
+.TH THREAD 2
+.SH NAME
+alt,
+chancreate,
+chanfree,
+chaninit,
+chanprint,
+mainstacksize,
+proccreate,
+procdata,
+procexec,
+procexecl,
+procrfork,
+recv,
+recvp,
+recvul,
+send,
+sendp,
+sendul,
+nbrecv,
+nbrecvp,
+nbrecvul,
+nbsend,
+nbsendp,
+nbsendul,
+threadcreate,
+threaddata,
+threadexits,
+threadexitsall,
+threadgetgrp,
+threadgetname,
+threadint,
+threadintgrp,
+threadkill,
+threadkillgrp,
+threadmain,
+threadnotify,
+threadid,
+threadpid,
+threadsetgrp,
+threadsetname,
+threadwaitchan,
+yield \- thread and proc management
+.SH SYNOPSIS
+.PP
+.EX
+.ta 4n +4n +4n +4n +4n +4n +4n
+#include <u.h>
+#include <libc.h>
+#include <thread.h>
+.sp
+#define CHANEND 0
+#define CHANSND 1
+#define CHANRCV 2
+#define CHANNOP 3
+#define CHANNOBLK 4
+.sp
+.ta \w' 'u +\w'Channel 'u
+typedef struct Alt Alt;
+struct Alt {
+ Channel *c;
+ void *v;
+ int op;
+ Channel **tag;
+ int entryno;
+};
+.fi
+.de XX
+.ift .sp 0.5
+.ifn .sp
+..
+.PP
+.nf
+.ft L
+.ta \w'\fLChannel* 'u +4n +4n +4n +4n
+void threadmain(int argc, char *argv[])
+int mainstacksize
+int proccreate(void (*fn)(void*), void *arg, uint stacksize)
+int procrfork(void (*fn)(void*), void *arg, uint stacksize,
+ int rforkflag)
+int threadcreate(void (*fn)(void*), void *arg, uint stacksize)
+void threadexits(char *status)
+void threadexitsall(char *status)
+void yield(void)
+.XX
+int threadid(void)
+int threadgrp(void)
+int threadsetgrp(int group)
+int threadpid(int id)
+.XX
+int threadint(int id)
+int threadintgrp(int group)
+int threadkill(int id)
+int threadkillgrp(int group)
+.XX
+void threadsetname(char *name)
+char* threadgetname(void)
+.XX
+void** threaddata(void)
+void** procdata(void)
+.XX
+int chaninit(Channel *c, int elsize, int nel)
+Channel* chancreate(int elsize, int nel)
+void chanfree(Channel *c)
+.XX
+int alt(Alt *alts)
+int recv(Channel *c, void *v)
+void* recvp(Channel *c)
+ulong recvul(Channel *c)
+int nbrecv(Channel *c, void *v)
+void* nbrecvp(Channel *c)
+ulong nbrecvul(Channel *c)
+int send(Channel *c, void *v)
+int sendp(Channel *c, void *v)
+int sendul(Channel *c, ulong v)
+int nbsend(Channel *c, void *v)
+int nbsendp(Channel *c, void *v)
+int nbsendul(Channel *c, ulong v)
+int chanprint(Channel *c, char *fmt, ...)
+.XX
+int procexecl(Channel *cpid, char *file, ...)
+int procexec(Channel *cpid, char *file, char *args[])
+Channel* threadwaitchan(void)
+.XX
+int threadnotify(int (*f)(void*, char*), int in)
+.EE
+.SH DESCRIPTION
+.PP
+The thread library provides parallel programming support similar to that
+of the languages
+Alef and Newsqueak.
+Threads
+and
+procs
+occupy a shared address space,
+communicating and synchronizing through
+.I channels
+and shared variables.
+.PP
+A
+.I proc
+is a Plan 9 process that contains one or more cooperatively scheduled
+.IR threads .
+Programs using threads must replace
+.I main
+by
+.IR threadmain .
+The thread library provides a
+.I main
+function that sets up a proc with a single thread executing
+.I threadmain
+on a stack of size
+.I mainstacksize
+(default eight kilobytes).
+To set
+.IR mainstacksize ,
+declare a global variable
+initialized to the desired value
+.RI ( e.g. ,
+.B int
+.B mainstacksize
+.B =
+.BR 1024 ).
+.PP
+.I Threadcreate
+creates a new thread in the calling proc, returning a unique integer
+identifying the thread; the thread
+executes
+.I fn(arg)
+on a stack of size
+.IR stacksize .
+Thread stacks are allocated in shared memory, making it valid to pass
+pointers to stack variables between threads and procs.
+.I Procrfork
+creates a new proc, and inside that proc creates
+a single thread as
+.I threadcreate
+would,
+returning the id of the created thread.
+.I Procrfork
+creates the new proc by calling
+.B rfork
+(see
+.IR fork (2))
+with flags
+.BR RFPROC|RFMEM|RFNOWAIT| \fIrforkflag\fR.
+(The thread library depends on all its procs
+running in the same rendezvous group.
+Do not include
+.B RFREND
+in
+.IR rforkflag .)
+.I Proccreate
+is identical to
+.I procrfork
+with
+.I rforkflag
+set to zero.
+Be aware that the calling thread may continue
+execution before
+the newly created proc and thread
+are scheduled.
+Because of this,
+.I arg
+should not point to data on the stack of a function that could
+return before the new process is scheduled.
+.PP
+.I Threadexits
+terminates the calling thread.
+If the thread is the last in its proc,
+.I threadexits
+also terminates the proc, using
+.I status
+as the exit status.
+.I Threadexitsall
+terminates all procs in the program,
+using
+.I status
+as the exit status.
+.PP
+The threads in a proc are coroutines, scheduled nonpreemptively
+in a round-robin fashion.
+A thread must explicitly relinquish control of the processor
+before another thread in the same proc is run.
+Calls that do this are
+.IR yield ,
+.IR proccreate ,
+.IR procexec ,
+.IR procexecl ,
+.IR threadexits ,
+.IR alt ,
+.IR send ,
+and
+.I recv
+(and the calls related to
+.I send
+and
+.IR recv \(emsee
+their descriptions further on).
+Procs are scheduled by the operating system.
+Therefore, threads in different procs can preempt one another
+in arbitrary ways and should synchronize their
+actions using
+.B qlocks
+(see
+.IR lock (2))
+or channel communication.
+System calls such as
+.IR read (2)
+block the entire proc;
+all threads in a proc block until the system call finishes.
+.PP
+As mentioned above, each thread has a unique integer thread id.
+Thread ids are not reused; they are unique across the life of the program.
+.I Threadid
+returns the id for the current thread.
+Each thread also has a thread group id.
+The initial thread has a group id of zero.
+Each new thread inherits the group id of
+the thread that created it.
+.I Threadgrp
+returns the group id for the current thread;
+.I threadsetgrp
+sets it.
+.I Threadpid
+returns the pid of the Plan 9 process containing
+the thread identified by
+.IR id ,
+or \-1
+if no such thread is found.
+.PP
+.I Threadint
+interrupts a thread that is blocked in a channel operation
+or system call.
+.I Threadintgrp
+interrupts all threads with the given group id.
+.I Threadkill
+marks a thread to die when it next relinquishes the processor
+(via one of the calls listed above).
+If the thread is blocked in a channel operation or system call,
+it is also interrupted.
+.I Threadkillgrp
+kills all threads with the given group id.
+Note that
+.I threadkill
+and
+.I threadkillgrp
+will not terminate a thread that never relinquishes
+the processor.
+.PP
+Primarily for debugging,
+threads can have string names associated with them.
+.I Threadgetname
+returns the current thread's name;
+.I threadsetname
+sets it.
+The pointer returned by
+.I threadgetname
+is only valid until the next call to
+.IR threadsetname .
+.PP
+.I Threaddata
+returns a pointer to a per-thread pointer
+that may be modified by threaded programs for
+per-thread storage.
+Similarly,
+.I procdata
+returns a pointer to a per-proc pointer.
+.PP
+.I Procexecl
+and
+.I procexec
+are threaded analogues of
+.I exec
+and
+.I execl
+(see
+.IR exec (2));
+on success,
+they replace the calling thread (which must be the only thread in its proc)
+and invoke the external program, never returning.
+On error, they return \-1.
+If
+.I cpid
+is not null, the pid of the invoked program
+will be sent along
+.I cpid
+once the program has been started, or \-1 will be sent if an
+error occurs.
+.I Procexec
+and
+.I procexecl
+will not access their arguments after sending a result
+along
+.IR cpid .
+Thus, programs that malloc the
+.I argv
+passed to
+.I procexec
+can safely free it once they have
+received the
+.I cpid
+response.
+.I Threadwaitchan
+returns a channel of pointers to
+.B Waitmsg
+structures (see
+.IR wait (2)).
+When an exec'ed process exits, a pointer to a
+.B Waitmsg
+is sent to this channel.
+These
+.B Waitmsg
+structures have been allocated with
+.IR malloc (2)
+and should be freed after use.
+.PP
+A
+.B Channel
+is a buffered or unbuffered queue for fixed-size messages.
+Procs and threads
+.I send
+messages into the channel and
+.I recv
+messages from the channel. If the channel is unbuffered, a
+.I send
+operation blocks until the corresponding
+.I recv
+operation occurs and
+.IR "vice versa" .
+.I Chaninit
+initializes a
+.B Channel
+for messages of size
+.I elsize
+and with a buffer holding
+.I nel
+messages.
+If
+.I nel
+is zero, the channel is unbuffered.
+.IR Chancreate
+allocates a new channel and initializes it.
+.I Chanfree
+frees a channel that is no longer used.
+.I Chanfree
+can be called by either sender or receiver after the last item has been
+sent or received. Freeing the channel will be delayed if there is a thread
+blocked on it until that thread unblocks (but
+.I chanfree
+returns immediately).
+.PP
+.I Send
+sends the element pointed at by
+.I v
+to the channel
+.IR c .
+If
+.I v
+is null, zeros are sent.
+.I Recv
+receives an element from
+.I c
+and stores it in
+.IR v .
+If
+.I v
+is null,
+the received value is discarded.
+.I Send
+and
+.I recv
+return 1 on success, \-1 if interrupted.
+.I Nbsend
+and
+.I nbrecv
+behave similarly, but return 0 rather than blocking.
+.PP
+.IR Sendp ,
+.IR nbsendp ,
+.IR sendul ,
+and
+.I nbsendul
+send a pointer or an unsigned long; the channel must
+have been initialized with the appropriate
+.IR elsize .
+.IR Recvp ,
+.IR nbrecvp ,
+.IR recvul ,
+and
+.I nbrecvul
+receive a pointer or an unsigned long;
+they return zero when a zero is received,
+when interrupted, or
+(for
+.I nbrecvp
+and
+.IR nbrecvul )
+when the operation would have blocked.
+To distinguish between these three cases,
+use
+.I recv
+or
+.IR nbrecv .
+.PP
+.I Alt
+can be used to recv from or send to one of a number of channels,
+as directed by an array of
+.B Alt
+structures,
+each of which describes a potential send or receive operation.
+In an
+.B Alt
+structure,
+.B c
+is the channel;
+.B v
+the value pointer (which may be null); and
+.B op
+the operation:
+.B CHANSND
+for a send operation,
+.B CHANRECV
+for a recv operation;
+.B CHANNOP
+for no operation
+(useful
+when
+.I alt
+is called with a varying set of operations).
+The array of
+.B Alt
+structures is terminated by an entry with
+.I op
+.B CHANEND
+or
+.BR CHANNOBLK .
+If at least one
+.B Alt
+structure can proceed, one of them is
+chosen at random to be executed.
+.I Alt
+returns the index of the chosen structure.
+If no operations can proceed and the list is terminated with
+.BR CHANNOBLK ,
+.I alt
+returns the index of the terminating
+.B CHANNOBLK
+structure.
+Otherwise,
+.I alt
+blocks until one of the operations can proceed,
+eventually returning the index of the structure executes.
+.I Alt
+returns \-1 when interrupted.
+The
+.B tag
+and
+.B entryno
+fields in the
+.B Alt
+structure are used internally by
+.I alt
+and need not be initialized.
+They are not used between
+.I alt
+calls.
+.PP
+.I Chanprint
+formats its arguments in the manner of
+.IR print (2)
+and sends the result to the channel
+.IR c.
+The string delivered by
+.I chanprint
+is allocated with
+.IR malloc (2)
+and should be freed upon receipt.
+.PP
+Thread library functions do not return on failure;
+if errors occur, the entire program is aborted.
+.PP
+Threaded programs should use
+.I threadnotify
+in place of
+.I atnotify
+(see
+.IR notify (2)).
+.PP
+It is safe to use
+.B sysfatal
+(see
+.IR perror (2))
+in threaded programs.
+.I Sysfatal
+will print the error string and call
+.IR threadexitsall .
+.PP
+It is safe to use
+.IR rfork
+(see
+.IR fork (2))
+to manage the namespace, file descriptors, note group, and environment of a
+single process.
+That is, it is safe to call
+.I rfork
+with the flags
+.BR RFNAMEG ,
+.BR RFFDG ,
+.BR RFCFDG ,
+.BR RFNOTEG ,
+.BR RFENVG ,
+and
+.BR RFCENVG.
+(To create new processes, use
+.I proccreate
+and
+.IR procrfork .)
+As mentioned above,
+the thread library depends on all procs being in the
+same rendezvous group; do not change the rendezvous
+group with
+.IR rfork .
+.SH FILES
+.B /sys/lib/acid/thread
+contains useful
+.IR acid (1)
+functions for debugging threaded programs.
+.PP
+.B /sys/src/libthread/example.c
+contains a full example program.
+.SH SOURCE
+.B /sys/src/libthread
+.SH SEE ALSO
+.IR intro (2),
+.IR ioproc (2)
DIR diff --git a/package b/package
t@@ -0,0 +1,19 @@
+#!/bin/sh
+
+rm -rf ../9tmp
+mkdir ../9tmp
+mkdir ../9pkg
+cp -R * ../9tmp
+cd ../9tmp
+rm -rf `find . -name CVS`
+rm lib/*
+rm */*.o
+rm */*.a
+rm mk/mk
+rm sam/sam
+rm man/*/*
+for i in libfmt libbio lib9 libthread libutf libregexp sam mk
+do
+ tar cf - $i |gzip >../9pkg/$i.tar.gz
+done
+tar cf - . | gzip > ../9pkg/all.tar.gz
DIR diff --git a/src/lib9/LICENSE b/src/lib9/LICENSE
t@@ -0,0 +1,258 @@
+The Plan 9 software is provided under the terms of the
+Lucent Public License, Version 1.02, reproduced below,
+with the following exceptions:
+
+1. No right is granted to create derivative works of or
+ to redistribute (other than with the Plan 9 Operating System)
+ the screen imprinter fonts identified in subdirectory
+ /lib/font/bit/lucida and printer fonts (Lucida Sans Unicode, Lucida
+ Sans Italic, Lucida Sans Demibold, Lucida Typewriter, Lucida Sans
+ Typewriter83), identified in subdirectory /sys/lib/postscript/font.
+ These directories contain material copyrights by B&H Inc. and Y&Y Inc.
+
+2. The printer fonts identified in subdirectory /sys/lib/ghostscript/font
+ are subject to the GNU GPL, reproduced in the file /LICENSE.gpl.
+
+3. The ghostscript program in the subdirectory /sys/src/cmd/gs is
+ covered by the Aladdin Free Public License, reproduced in the file
+ /LICENSE.afpl.
+
+===================================================================
+
+Lucent Public License Version 1.02
+
+THE ACCOMPANYING PROGRAM IS PROVIDED UNDER THE TERMS OF THIS PUBLIC
+LICENSE ("AGREEMENT"). ANY USE, REPRODUCTION OR DISTRIBUTION OF THE
+PROGRAM CONSTITUTES RECIPIENT'S ACCEPTANCE OF THIS AGREEMENT.
+
+1. DEFINITIONS
+
+"Contribution" means:
+
+ a. in the case of Lucent Technologies Inc. ("LUCENT"), the Original
+ Program, and
+ b. in the case of each Contributor,
+
+ i. changes to the Program, and
+ ii. additions to the Program;
+
+ where such changes and/or additions to the Program were added to the
+ Program by such Contributor itself or anyone acting on such
+ Contributor's behalf, and the Contributor explicitly consents, in
+ accordance with Section 3C, to characterization of the changes and/or
+ additions as Contributions.
+
+"Contributor" means LUCENT and any other entity that has Contributed a
+Contribution to the Program.
+
+"Distributor" means a Recipient that distributes the Program,
+modifications to the Program, or any part thereof.
+
+"Licensed Patents" mean patent claims licensable by a Contributor
+which are necessarily infringed by the use or sale of its Contribution
+alone or when combined with the Program.
+
+"Original Program" means the original version of the software
+accompanying this Agreement as released by LUCENT, including source
+code, object code and documentation, if any.
+
+"Program" means the Original Program and Contributions or any part
+thereof
+
+"Recipient" means anyone who receives the Program under this
+Agreement, including all Contributors.
+
+2. GRANT OF RIGHTS
+
+ a. Subject to the terms of this Agreement, each Contributor hereby
+ grants Recipient a non-exclusive, worldwide, royalty-free copyright
+ license to reproduce, prepare derivative works of, publicly display,
+ publicly perform, distribute and sublicense the Contribution of such
+ Contributor, if any, and such derivative works, in source code and
+ object code form.
+
+ b. Subject to the terms of this Agreement, each Contributor hereby
+ grants Recipient a non-exclusive, worldwide, royalty-free patent
+ license under Licensed Patents to make, use, sell, offer to sell,
+ import and otherwise transfer the Contribution of such Contributor, if
+ any, in source code and object code form. The patent license granted
+ by a Contributor shall also apply to the combination of the
+ Contribution of that Contributor and the Program if, at the time the
+ Contribution is added by the Contributor, such addition of the
+ Contribution causes such combination to be covered by the Licensed
+ Patents. The patent license granted by a Contributor shall not apply
+ to (i) any other combinations which include the Contribution, nor to
+ (ii) Contributions of other Contributors. No hardware per se is
+ licensed hereunder.
+
+ c. Recipient understands that although each Contributor grants the
+ licenses to its Contributions set forth herein, no assurances are
+ provided by any Contributor that the Program does not infringe the
+ patent or other intellectual property rights of any other entity. Each
+ Contributor disclaims any liability to Recipient for claims brought by
+ any other entity based on infringement of intellectual property rights
+ or otherwise. As a condition to exercising the rights and licenses
+ granted hereunder, each Recipient hereby assumes sole responsibility
+ to secure any other intellectual property rights needed, if any. For
+ example, if a third party patent license is required to allow
+ Recipient to distribute the Program, it is Recipient's responsibility
+ to acquire that license before distributing the Program.
+
+ d. Each Contributor represents that to its knowledge it has sufficient
+ copyright rights in its Contribution, if any, to grant the copyright
+ license set forth in this Agreement.
+
+3. REQUIREMENTS
+
+A. Distributor may choose to distribute the Program in any form under
+this Agreement or under its own license agreement, provided that:
+
+ a. it complies with the terms and conditions of this Agreement;
+
+ b. if the Program is distributed in source code or other tangible
+ form, a copy of this Agreement or Distributor's own license agreement
+ is included with each copy of the Program; and
+
+ c. if distributed under Distributor's own license agreement, such
+ license agreement:
+
+ i. effectively disclaims on behalf of all Contributors all warranties
+ and conditions, express and implied, including warranties or
+ conditions of title and non-infringement, and implied warranties or
+ conditions of merchantability and fitness for a particular purpose;
+ ii. effectively excludes on behalf of all Contributors all liability
+ for damages, including direct, indirect, special, incidental and
+ consequential damages, such as lost profits; and
+ iii. states that any provisions which differ from this Agreement are
+ offered by that Contributor alone and not by any other party.
+
+B. Each Distributor must include the following in a conspicuous
+ location in the Program:
+
+ Copyright (C) 2003, Lucent Technologies Inc. and others. All Rights
+ Reserved.
+
+C. In addition, each Contributor must identify itself as the
+originator of its Contribution in a manner that reasonably allows
+subsequent Recipients to identify the originator of the Contribution.
+Also, each Contributor must agree that the additions and/or changes
+are intended to be a Contribution. Once a Contribution is contributed,
+it may not thereafter be revoked.
+
+4. COMMERCIAL DISTRIBUTION
+
+Commercial distributors of software may accept certain
+responsibilities with respect to end users, business partners and the
+like. While this license is intended to facilitate the commercial use
+of the Program, the Distributor who includes the Program in a
+commercial product offering should do so in a manner which does not
+create potential liability for Contributors. Therefore, if a
+Distributor includes the Program in a commercial product offering,
+such Distributor ("Commercial Distributor") hereby agrees to defend
+and indemnify every Contributor ("Indemnified Contributor") against
+any losses, damages and costs (collectively"Losses") arising from
+claims, lawsuits and other legal actions brought by a third party
+against the Indemnified Contributor to the extent caused by the acts
+or omissions of such Commercial Distributor in connection with its
+distribution of the Program in a commercial product offering. The
+obligations in this section do not apply to any claims or Losses
+relating to any actual or alleged intellectual property infringement.
+In order to qualify, an Indemnified Contributor must: a) promptly
+notify the Commercial Distributor in writing of such claim, and b)
+allow the Commercial Distributor to control, and cooperate with the
+Commercial Distributor in, the defense and any related settlement
+negotiations. The Indemnified Contributor may participate in any such
+claim at its own expense.
+
+For example, a Distributor might include the Program in a commercial
+product offering, Product X. That Distributor is then a Commercial
+Distributor. If that Commercial Distributor then makes performance
+claims, or offers warranties related to Product X, those performance
+claims and warranties are such Commercial Distributor's responsibility
+alone. Under this section, the Commercial Distributor would have to
+defend claims against the Contributors related to those performance
+claims and warranties, and if a court requires any Contributor to pay
+any damages as a result, the Commercial Distributor must pay those
+damages.
+
+5. NO WARRANTY
+
+EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, THE PROGRAM IS
+PROVIDED ON AN"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT LIMITATION, ANY
+WARRANTIES OR CONDITIONS OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY
+OR FITNESS FOR A PARTICULAR PURPOSE. Each Recipient is solely
+responsible for determining the appropriateness of using and
+distributing the Program and assumes all risks associated with its
+exercise of rights under this Agreement, including but not limited to
+the risks and costs of program errors, compliance with applicable
+laws, damage to or loss of data, programs or equipment, and
+unavailability or interruption of operations.
+
+6. DISCLAIMER OF LIABILITY
+
+EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, NEITHER RECIPIENT NOR
+ANY CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING
+WITHOUT LIMITATION LOST PROFITS), HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OR
+DISTRIBUTION OF THE PROGRAM OR THE EXERCISE OF ANY RIGHTS GRANTED
+HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+7. EXPORT CONTROL
+
+Recipient agrees that Recipient alone is responsible for compliance
+with the United States export administration regulations (and the
+export control laws and regulation of any other countries).
+
+8. GENERAL
+
+If any provision of this Agreement is invalid or unenforceable under
+applicable law, it shall not affect the validity or enforceability of
+the remainder of the terms of this Agreement, and without further
+action by the parties hereto, such provision shall be reformed to the
+minimum extent necessary to make such provision valid and enforceable.
+
+If Recipient institutes patent litigation against a Contributor with
+respect to a patent applicable to software (including a cross-claim or
+counterclaim in a lawsuit), then any patent licenses granted by that
+Contributor to such Recipient under this Agreement shall terminate as
+of the date such litigation is filed. In addition, if Recipient
+institutes patent litigation against any entity (including a
+cross-claim or counterclaim in a lawsuit) alleging that the Program
+itself (excluding combinations of the Program with other software or
+hardware) infringes such Recipient's patent(s), then such Recipient's
+rights granted under Section 2(b) shall terminate as of the date such
+litigation is filed.
+
+All Recipient's rights under this Agreement shall terminate if it
+fails to comply with any of the material terms or conditions of this
+Agreement and does not cure such failure in a reasonable period of
+time after becoming aware of such noncompliance. If all Recipient's
+rights under this Agreement terminate, Recipient agrees to cease use
+and distribution of the Program as soon as reasonably practicable.
+However, Recipient's obligations under this Agreement and any licenses
+granted by Recipient relating to the Program shall continue and
+survive.
+
+LUCENT may publish new versions (including revisions) of this
+Agreement from time to time. Each new version of the Agreement will be
+given a distinguishing version number. The Program (including
+Contributions) may always be distributed subject to the version of the
+Agreement under which it was received. In addition, after a new
+version of the Agreement is published, Contributor may elect to
+distribute the Program (including its Contributions) under the new
+version. No one other than LUCENT has the right to modify this
+Agreement. Except as expressly stated in Sections 2(a) and 2(b) above,
+Recipient receives no rights or licenses to the intellectual property
+of any Contributor under this Agreement, whether expressly, by
+implication, estoppel or otherwise. All rights in the Program not
+expressly granted under this Agreement are reserved.
+
+This Agreement is governed by the laws of the State of New York and
+the intellectual property laws of the United States of America. No
+party to this Agreement will bring a legal action under this Agreement
+more than one year after the cause of action arose. Each party waives
+its rights to a jury trial in any resulting litigation.
+
DIR diff --git a/src/lib9/Make.Darwin-PowerMacintosh b/src/lib9/Make.Darwin-PowerMacintosh
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I${PREFIX}/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.FreeBSD-386 b/src/lib9/Make.FreeBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.HP-UX-9000 b/src/lib9/Make.HP-UX-9000
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS=-O -c -Ae -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.Linux-386 b/src/lib9/Make.Linux-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.NetBSD-386 b/src/lib9/Make.NetBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.OSF1-alpha b/src/lib9/Make.OSF1-alpha
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.SunOS-sun4u b/src/lib9/Make.SunOS-sun4u
t@@ -0,0 +1,2 @@
+include Make.SunOS-sun4u-$(CC)
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.SunOS-sun4u-cc b/src/lib9/Make.SunOS-sun4u-cc
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I. -O
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/lib9/Make.SunOS-sun4u-gcc b/src/lib9/Make.SunOS-sun4u-gcc
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/lib9/Makefile b/src/lib9/Makefile
t@@ -0,0 +1,120 @@
+
+# this works in gnu make
+SYSNAME:=${shell uname}
+OBJTYPE:=${shell uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'}
+
+# this works in bsd make
+SYSNAME!=uname
+OBJTYPE!=uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'
+
+# the gnu rules will mess up bsd but not vice versa,
+# hence the gnu rules come first.
+
+include Make.$(SYSNAME)-$(OBJTYPE)
+
+PREFIX=/usr/local
+
+NUKEFILES=
+
+TGZFILES=
+
+LIB=lib9.a
+VERSION=2.0
+PORTPLACE=devel/lib9
+NAME=lib9
+
+OFILES=\
+ _exits.$O\
+ argv0.$O\
+ await.$O\
+ encodefmt.$O\
+ errstr.$O\
+ exits.$O\
+ ffork-$(SYSNAME).$O\
+ getcallerpc-$(OBJTYPE).$O\
+ getfields.$O\
+ lock.$O\
+ malloctag.$O\
+ mallocz.$O\
+ nrand.$O\
+ qlock.$O\
+ readn.$O\
+ rendez.$O\
+ strecpy.$O\
+ sysfatal.$O\
+ tas-$(OBJTYPE).$O\
+ tokenize.$O\
+ u16.$O\
+ u32.$O\
+ u64.$O\
+ wait.$O\
+ werrstr.$O\
+
+HFILES=\
+ lib9.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ # install -m 0644 lib9.3 $(PREFIX)/man/man3/lib9.3
+ install -m 0644 lib9.h $(PREFIX)/include/lib9.h
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+test: $(LIB) test.$O
+ $(CC) -o test test.$O $(LIB) -L$(PREFIX)/lib -lfmt -lutf
+
+testfork: $(LIB) testfork.$O
+ $(CC) -o testfork testfork.$O $(LIB) -L$(PREFIX)/lib -lfmt -lutf
+
+$(LIB): $(OFILES)
+ $(AR) $(ARFLAGS) $(LIB) $(OFILES)
+
+NUKEFILES+=$(LIB)
+.c.$O:
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+%.$O: %.c
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+
+$(OFILES): $(HFILES)
+
+tgz:
+ rm -rf $(NAME)-$(VERSION)
+ mkdir $(NAME)-$(VERSION)
+ cp Makefile Make.* README LICENSE NOTICE *.[ch137] rpm.spec bundle.ports $(TGZFILES) $(NAME)-$(VERSION)
+ tar cf - $(NAME)-$(VERSION) | gzip >$(NAME)-$(VERSION).tgz
+ rm -rf $(NAME)-$(VERSION)
+
+clean:
+ rm -f $(OFILES) $(LIB)
+
+nuke:
+ rm -f $(OFILES) *.tgz *.rpm $(NUKEFILES)
+
+rpm:
+ make tgz
+ cp $(NAME)-$(VERSION).tgz /usr/src/RPM/SOURCES
+ rpm -ba rpm.spec
+ cp /usr/src/RPM/SRPMS/$(NAME)-$(VERSION)-1.src.rpm .
+ cp /usr/src/RPM/RPMS/i586/$(NAME)-$(VERSION)-1.i586.rpm .
+ scp *.rpm rsc@amsterdam.lcs.mit.edu:public_html/software
+
+PORTDIR=/usr/ports/$(PORTPLACE)
+
+ports:
+ make tgz
+ rm -rf $(PORTDIR)
+ mkdir $(PORTDIR)
+ cp $(NAME)-$(VERSION).tgz /usr/ports/distfiles
+ cat bundle.ports | (cd $(PORTDIR) && awk '$$1=="---" && $$3=="---" { ofile=$$2; next} {if(ofile) print >ofile}')
+ (cd $(PORTDIR); make makesum)
+ (cd $(PORTDIR); make)
+ (cd $(PORTDIR); /usr/local/bin/portlint)
+ rm -rf $(PORTDIR)/work
+ shar `find $(PORTDIR)` > ports.shar
+ (cd $(PORTDIR); tar cf - *) | gzip >$(NAME)-$(VERSION)-ports.tgz
+ scp *.tgz rsc@amsterdam.lcs.mit.edu:public_html/software
+
+.phony: all clean nuke install tgz rpm ports
DIR diff --git a/src/lib9/Makefile.MID b/src/lib9/Makefile.MID
t@@ -0,0 +1,49 @@
+LIB=lib9.a
+VERSION=2.0
+PORTPLACE=devel/lib9
+NAME=lib9
+
+OFILES=\
+ _exits.$O\
+ argv0.$O\
+ await.$O\
+ encodefmt.$O\
+ errstr.$O\
+ exits.$O\
+ ffork-$(SYSNAME).$O\
+ getcallerpc-$(OBJTYPE).$O\
+ getfields.$O\
+ lock.$O\
+ malloctag.$O\
+ mallocz.$O\
+ nrand.$O\
+ qlock.$O\
+ readn.$O\
+ rendez.$O\
+ strecpy.$O\
+ sysfatal.$O\
+ tas-$(OBJTYPE).$O\
+ tokenize.$O\
+ u16.$O\
+ u32.$O\
+ u64.$O\
+ wait.$O\
+ werrstr.$O\
+
+HFILES=\
+ lib9.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ # install -m 0644 lib9.3 $(PREFIX)/man/man3/lib9.3
+ install -m 0644 lib9.h $(PREFIX)/include/lib9.h
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+test: $(LIB) test.$O
+ $(CC) -o test test.$O $(LIB) -L$(PREFIX)/lib -lfmt -lutf
+
+testfork: $(LIB) testfork.$O
+ $(CC) -o testfork testfork.$O $(LIB) -L$(PREFIX)/lib -lfmt -lutf
+
DIR diff --git a/src/lib9/_exits.c b/src/lib9/_exits.c
t@@ -0,0 +1,9 @@
+#include <lib9.h>
+
+void
+_exits(char *s)
+{
+ if(s && *s)
+ _exit(1);
+ _exit(0);
+}
DIR diff --git a/src/lib9/argv0.c b/src/lib9/argv0.c
t@@ -0,0 +1,4 @@
+#include <lib9.h>
+
+char *argv0;
+
DIR diff --git a/src/lib9/await.c b/src/lib9/await.c
t@@ -0,0 +1,105 @@
+#include <signal.h>
+#include <sys/types.h>
+#include <sys/resource.h>
+#include <sys/wait.h>
+#include <sys/time.h>
+#include <string.h>
+#include <errno.h>
+#include <lib9.h>
+
+static struct {
+ int sig;
+ char *str;
+} tab[] = {
+ SIGHUP, "hangup",
+ SIGINT, "interrupt",
+ SIGQUIT, "quit",
+ SIGILL, "sys: trap: illegal instruction",
+ SIGTRAP, "sys: trace trap",
+ SIGABRT, "sys: abort",
+#ifdef SIGEMT
+ SIGEMT, "sys: emulate instruction executed",
+#endif
+ SIGFPE, "sys: fp: trap",
+ SIGKILL, "sys: kill",
+ SIGBUS, "sys: bus error",
+ SIGSEGV, "sys: segmentation violation",
+ SIGALRM, "alarm",
+ SIGTERM, "kill",
+ SIGURG, "sys: urgent condition on socket",
+ SIGSTOP, "sys: stop",
+ SIGTSTP, "sys: tstp",
+ SIGCONT, "sys: cont",
+ SIGCHLD, "sys: child",
+ SIGTTIN, "sys: ttin",
+ SIGTTOU, "sys: ttou",
+ SIGIO, "sys: i/o possible on fd",
+ SIGXCPU, "sys: cpu time limit exceeded",
+ SIGXFSZ, "sys: file size limit exceeded",
+ SIGVTALRM, "sys: virtual time alarm",
+ SIGPROF, "sys: profiling timer alarm",
+ SIGWINCH, "sys: window size change",
+#ifdef SIGINFO
+ SIGINFO, "sys: status request",
+#endif
+ SIGUSR1, "sys: usr1",
+ SIGUSR2, "sys: usr2",
+};
+
+static char*
+_p9sigstr(int sig, char *tmp)
+{
+ int i;
+
+ for(i=0; i<nelem(tab); i++)
+ if(tab[i].sig == sig)
+ return tab[i].str;
+ sprint(tmp, "sys: signal %d", sig);
+ return tmp;
+}
+
+/*
+static int
+_p9strsig(char *s)
+{
+ int i;
+
+ for(i=0; i<nelem(tab); i++)
+ if(strcmp(s, tab[i].str) == 0)
+ return tab[i].sig;
+ return 0;
+}
+*/
+
+int
+await(char *str, int n)
+{
+ int pid, status, cd;
+ struct rusage ru;
+ char buf[128], tmp[64];
+ ulong u, s;
+
+ for(;;){
+ pid = wait3(&status, 0, &ru);
+ if(pid < 0)
+ return -1;
+ u = ru.ru_utime.tv_sec*1000+((ru.ru_utime.tv_usec+500)/1000);
+ s = ru.ru_stime.tv_sec*1000+((ru.ru_stime.tv_usec+500)/1000);
+ if(WIFEXITED(status)){
+ status = WEXITSTATUS(status);
+ if(status)
+ snprint(buf, sizeof buf, "%d %lu %lu %lu %d", pid, u, s, u+s, status);
+ else
+ snprint(buf, sizeof buf, "%d %lu %lu %lu ''", pid, u, s, u+s);
+ strecpy(str, str+n, buf);
+ return strlen(str);
+ }
+ if(WIFSIGNALED(status)){
+ cd = WCOREDUMP(status);
+ USED(cd);
+ snprint(buf, sizeof buf, "%d %lu %lu %lu '%s'", pid, u, s, u+s, _p9sigstr(WTERMSIG(status), tmp));
+ strecpy(str, str+n, buf);
+ return strlen(str);
+ }
+ }
+}
DIR diff --git a/src/lib9/encodefmt.c b/src/lib9/encodefmt.c
t@@ -0,0 +1,69 @@
+#include <lib9.h>
+
+int
+encodefmt(Fmt *f)
+{
+ char *out;
+ char *buf;
+ int len;
+ int ilen;
+ int rv;
+ uchar *b;
+ char obuf[64]; // rsc optimization
+
+ if(!(f->flags&FmtPrec) || f->prec < 1)
+ goto error;
+
+ b = va_arg(f->args, uchar*);
+
+ ilen = f->prec;
+ f->prec = 0;
+ f->flags &= ~FmtPrec;
+ switch(f->r){
+ case '<':
+ len = (8*ilen+4)/5 + 3;
+ break;
+ case '[':
+ len = (8*ilen+5)/6 + 4;
+ break;
+ case 'H':
+ len = 2*ilen + 1;
+ break;
+ default:
+ goto error;
+ }
+
+ if(len > sizeof(obuf)){
+ buf = malloc(len);
+ if(buf == nil)
+ goto error;
+ } else
+ buf = obuf;
+
+ // convert
+ out = buf;
+ switch(f->r){
+ case '<':
+ rv = enc32(out, len, b, ilen);
+ break;
+ case '[':
+ rv = enc64(out, len, b, ilen);
+ break;
+ case 'H':
+ rv = enc16(out, len, b, ilen);
+ break;
+ default:
+ rv = -1;
+ break;
+ }
+ if(rv < 0)
+ goto error;
+
+ fmtstrcpy(f, buf);
+ if(buf != obuf)
+ free(buf);
+ return 0;
+
+error:
+ return fmtstrcpy(f, "<encodefmt>");
+}
DIR diff --git a/src/lib9/errstr.c b/src/lib9/errstr.c
t@@ -0,0 +1,68 @@
+/*
+ * We assume there's only one error buffer for the whole system.
+ * If you use ffork, you need to provide a _syserrstr. Since most
+ * people will use libthread (which provides a _syserrstr), this is
+ * okay.
+ */
+
+#include <errno.h>
+#include <string.h>
+#include <lib9.h>
+
+enum
+{
+ EPLAN9 = 0x19283745,
+};
+
+char *(*_syserrstr)(void);
+static char xsyserr[ERRMAX];
+static char*
+getsyserr(void)
+{
+ char *s;
+
+ s = nil;
+ if(_syserrstr)
+ s = (*_syserrstr)();
+ if(s == nil)
+ s = xsyserr;
+ return s;
+}
+
+int
+errstr(char *err, uint n)
+{
+ char tmp[ERRMAX];
+ char *syserr;
+
+ syserr = getsyserr();
+ if(errno != EPLAN9)
+ strcpy(syserr, strerror(errno));
+
+ strecpy(tmp, tmp+ERRMAX, syserr);
+ strecpy(syserr, syserr+ERRMAX, err);
+ strecpy(err, err+n, tmp);
+ errno = EPLAN9;
+ return 0;
+}
+
+void
+rerrstr(char *err, uint n)
+{
+ char *syserr;
+
+ syserr = getsyserr();
+ if(errno != EPLAN9)
+ strcpy(syserr, strerror(errno));
+ strecpy(err, err+n, syserr);
+}
+
+/* replaces __errfmt in libfmt */
+
+int
+__errfmt(Fmt *f)
+{
+ if(errno == EPLAN9)
+ return fmtstrcpy(f, getsyserr());
+ return fmtstrcpy(f, strerror(errno));
+}
DIR diff --git a/src/lib9/exits.c b/src/lib9/exits.c
t@@ -0,0 +1,10 @@
+#include <lib9.h>
+
+void
+exits(char *s)
+{
+ if(s && *s)
+ exit(1);
+ exit(0);
+}
+
DIR diff --git a/src/lib9/ffork-FreeBSD.c b/src/lib9/ffork-FreeBSD.c
t@@ -0,0 +1,33 @@
+#include <lib9.h>
+
+extern int __isthreaded;
+int
+ffork(int flags, void(*fn)(void*), void *arg)
+{
+ void *p;
+
+ __isthreaded = 1;
+ p = malloc(16384);
+ if(p == nil)
+ return -1;
+ memset(p, 0xFE, 16384);
+ return rfork_thread(RFPROC|flags, (char*)p+16000, (int(*)(void*))fn, arg);
+}
+
+/*
+ * For FreeBSD libc.
+ */
+
+typedef struct {
+ volatile long access_lock;
+ volatile long lock_owner;
+ volatile char *fname;
+ volatile int lineno;
+} spinlock_t;
+
+void
+_spinlock(spinlock_t *lk)
+{
+ lock((Lock*)&lk->access_lock);
+}
+
DIR diff --git a/src/lib9/ffork-Linux.c b/src/lib9/ffork-Linux.c
t@@ -0,0 +1,39 @@
+#include <sched.h>
+#include <signal.h>
+#include <lib9.h>
+
+int fforkstacksize = 16384;
+
+int
+ffork(int flags, void (*fn)(void*), void *arg)
+{
+ char *p;
+ int cloneflag, pid;
+
+ p = malloc(fforkstacksize);
+ if(p == nil)
+ return -1;
+ cloneflag = 0;
+ flags &= ~RFPROC;
+ if(flags&RFMEM){
+ cloneflag |= CLONE_VM;
+ flags &= ~RFMEM;
+ }
+ if(!(flags&RFFDG))
+ cloneflag |= CLONE_FILES;
+ else
+ flags &= ~RFFDG;
+ if(!(flags&RFNOWAIT))
+ cloneflag |= SIGCHLD;
+ else
+ flags &= ~RFNOWAIT;
+ if(flags){
+ fprint(2, "unknown rfork flags %x\n", flags);
+ return -1;
+ }
+ pid = clone((int(*)(void*))fn, p+fforkstacksize-16, cloneflag, arg);
+ if(pid < 0)
+ free(p);
+ return pid;
+}
+
DIR diff --git a/src/lib9/getcallerpc-386.c b/src/lib9/getcallerpc-386.c
t@@ -0,0 +1,7 @@
+#include <lib9.h>
+
+ulong
+getcallerpc(void *x)
+{
+ return (((ulong*)(x))[-1]);
+}
DIR diff --git a/src/lib9/getfields.c b/src/lib9/getfields.c
t@@ -0,0 +1,36 @@
+#include <lib9.h>
+
+int
+getfields(char *str, char **args, int max, int mflag, char *set)
+{
+ Rune r;
+ int nr, intok, narg;
+
+ if(max <= 0)
+ return 0;
+
+ narg = 0;
+ args[narg] = str;
+ if(!mflag)
+ narg++;
+ intok = 0;
+ for(;; str += nr) {
+ nr = chartorune(&r, str);
+ if(r == 0)
+ break;
+ if(utfrune(set, r)) {
+ if(narg >= max)
+ break;
+ *str = 0;
+ intok = 0;
+ args[narg] = str + nr;
+ if(!mflag)
+ narg++;
+ } else {
+ if(!intok && mflag)
+ narg++;
+ intok = 1;
+ }
+ }
+ return narg;
+}
DIR diff --git a/src/lib9/lib9.h b/src/lib9/lib9.h
t@@ -0,0 +1,246 @@
+/*
+ * Lib9 is miscellany from the Plan 9 C library that doesn't
+ * fit into libutf or into libfmt, but is still missing from traditional
+ * Unix C libraries.
+ */
+#ifndef _LIB9H_
+#define _LIB9H_ 1
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+
+#include <unistd.h>
+#include <string.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <fcntl.h>
+#include <assert.h>
+
+#ifndef _FMTH_
+# include <fmt.h>
+#endif
+
+#define nil ((void*)0)
+#define nelem(x) (sizeof(x)/sizeof((x)[0]))
+
+#define _NEEDUCHAR 1
+#define _NEEDUSHORT 1
+#define _NEEDUINT 1
+#define _NEEDULONG 1
+
+#if defined(__linux__)
+# include <sys/types.h>
+# if defined(__USE_MISC)
+# undef _NEEDUSHORT
+# undef _NEEDUINT
+# undef _NEEDULONG
+# endif
+#endif
+#if defined(__FreeBSD__)
+# include <sys/types.h>
+# if !defined(_POSIX_SOURCE)
+# undef _NEEDUSHORT
+# undef _NEEDUINT
+# endif
+#endif
+
+typedef signed char schar;
+typedef unsigned int u32int;
+#ifdef _NEEDUCHAR
+ typedef unsigned char uchar;
+#endif
+#ifdef _NEEDUSHORT
+ typedef unsigned short ushort;
+#endif
+#ifdef _NEEDUINT
+ typedef unsigned int uint;
+#endif
+#ifdef _NEEDULONG
+ typedef unsigned long ulong;
+#endif
+typedef unsigned long long uvlong;
+typedef long long vlong;
+
+/* rfork to create new process running fn(arg) */
+
+#if defined(__FreeBSD__)
+#undef RFFDG
+#undef RFNOTEG
+#undef RFPROC
+#undef RFMEM
+#undef RFNOWAIT
+#undef RFCFDG
+#endif
+
+enum
+{
+/* RFNAMEG = (1<<0), */
+/* RFENVG = (1<<1), */
+ RFFDG = (1<<2),
+ RFNOTEG = (1<<3),
+ RFPROC = (1<<4),
+ RFMEM = (1<<5),
+ RFNOWAIT = (1<<6),
+/* RFCNAMEG = (1<<10), */
+/* RFCENVG = (1<<11), */
+ RFCFDG = (1<<12),
+/* RFREND = (1<<13), */
+/* RFNOMNT = (1<<14) */
+};
+extern int ffork(int, void(*)(void*), void*);
+
+/* wait for processes */
+#define wait _p9wait
+typedef struct Waitmsg Waitmsg;
+struct Waitmsg
+{
+ int pid; /* of loved one */
+ ulong time[3]; /* of loved one & descendants */
+ char *msg;
+};
+extern int await(char*, int);
+extern Waitmsg* wait(void);
+
+/* synchronization */
+typedef struct Lock Lock;
+struct Lock
+{
+ int val;
+};
+
+extern int _tas(void*);
+extern void lock(Lock*);
+extern void unlock(Lock*);
+extern int canlock(Lock*);
+
+typedef struct QLp QLp;
+struct QLp
+{
+ int inuse;
+ QLp *next;
+ int state;
+};
+
+typedef struct QLock QLock;
+struct QLock
+{
+ Lock lock;
+ int locked;
+ QLp *head;
+ QLp *tail;
+};
+
+extern void qlock(QLock*);
+extern void qunlock(QLock*);
+extern int canqlock(QLock*);
+extern void _qlockinit(ulong (*)(ulong, ulong));
+
+typedef struct RWLock RWLock;
+struct RWLock
+{
+ Lock lock;
+ int readers;
+ int writer;
+ QLp *head;
+ QLp *tail;
+};
+
+extern void rlock(RWLock*);
+extern void runlock(RWLock*);
+extern int canrlock(RWLock*);
+extern void wlock(RWLock*);
+extern void wunlock(RWLock*);
+extern int canwlock(RWLock*);
+
+typedef struct Rendez Rendez;
+struct Rendez
+{
+ QLock *l;
+ QLp *head;
+ QLp *tail;
+};
+
+extern void rsleep(Rendez*);
+extern int rwakeup(Rendez*);
+extern int rwakeupall(Rendez*);
+
+extern ulong rendezvous(ulong, ulong);
+
+/* one of a kind */
+extern void sysfatal(char*, ...);
+extern int nrand(int);
+extern void setmalloctag(void*, ulong);
+extern void setrealloctag(void*, ulong);
+extern void *mallocz(ulong, int);
+extern long readn(int, void*, long);
+extern void exits(char*);
+extern void _exits(char*);
+extern ulong getcallerpc(void*);
+
+/* string routines */
+extern char* strecpy(char*, char*, char*);
+extern int tokenize(char*, char**, int);
+extern int cistrncmp(char*, char*, int);
+extern int cistrcmp(char*, char*);
+extern char* cistrstr(char*, char*);
+extern int getfields(char*, char**, int, int, char*);
+extern int gettokens(char *, char **, int, char *);
+
+/* formatting helpers */
+extern int dec64(uchar*, int, char*, int);
+extern int enc64(char*, int, uchar*, int);
+extern int dec32(uchar*, int, char*, int);
+extern int enc32(char*, int, uchar*, int);
+extern int dec16(uchar*, int, char*, int);
+extern int enc16(char*, int, uchar*, int);
+extern int encodefmt(Fmt*);
+
+/* error string */
+enum
+{
+ ERRMAX = 128
+};
+extern void rerrstr(char*, uint);
+extern void werrstr(char*, ...);
+extern int errstr(char*, uint);
+
+/* compiler directives on plan 9 */
+#define USED(x) if(x){}else{}
+#define SET(x) ((x)=0)
+
+/* command line */
+extern char *argv0;
+#define ARGBEGIN for((argv0||(argv0=*argv)),argv++,argc--;\
+ argv[0] && argv[0][0]=='-' && argv[0][1];\
+ argc--, argv++) {\
+ char *_args, *_argt;\
+ Rune _argc;\
+ _args = &argv[0][1];\
+ if(_args[0]=='-' && _args[1]==0){\
+ argc--; argv++; break;\
+ }\
+ _argc = 0;\
+ while(*_args && (_args += chartorune(&_argc, _args)))\
+ switch(_argc)
+#define ARGEND SET(_argt);USED(_argt);USED(_argc);USED(_args);}USED(argv);USED(argc);
+#define ARGF() (_argt=_args, _args="",\
+ (*_argt? _argt: argv[1]? (argc--, *++argv): 0))
+#define EARGF(x) (_argt=_args, _args="",\
+ (*_argt? _argt: argv[1]? (argc--, *++argv): ((x), abort(), (char*)0)))
+
+#define ARGC() _argc
+
+#define OREAD O_RDONLY
+#define OWRITE O_WRONLY
+#define AEXIST 0
+#define AREAD 4
+#define AWRITE 2
+#define AEXEC 1
+
+#if defined(__cplusplus)
+}
+#endif
+
+#endif /* _LIB9H_ */
DIR diff --git a/src/lib9/lock.c b/src/lib9/lock.c
t@@ -0,0 +1,54 @@
+#include <unistd.h>
+#include <sched.h>
+#include <lib9.h>
+
+int _ntas;
+static int
+_xtas(void *v)
+{
+ int x;
+
+_ntas++;
+ x = _tas(v);
+ if(x == 0 || x == 0xCAFEBABE)
+ return x;
+ fprint(2, "%d: tas %p got %ux\n", getpid(), v, x);
+ abort();
+}
+
+int
+canlock(Lock *l)
+{
+ return !_xtas(&l->val);
+}
+
+void
+unlock(Lock *l)
+{
+ l->val = 0;
+}
+
+void
+lock(Lock *lk)
+{
+ int i;
+
+ /* once fast */
+ if(!_xtas(&lk->val))
+ return;
+ /* a thousand times pretty fast */
+ for(i=0; i<1000; i++){
+ if(!_xtas(&lk->val))
+ return;
+ sched_yield();
+ }
+ /* now nice and slow */
+ for(i=0; i<1000; i++){
+ if(!_xtas(&lk->val))
+ return;
+ usleep(100*1000);
+ }
+ /* take your time */
+ while(_xtas(&lk->val))
+ usleep(1000*1000);
+}
DIR diff --git a/src/lib9/malloctag.c b/src/lib9/malloctag.c
t@@ -0,0 +1,15 @@
+#include <lib9.h>
+
+void
+setmalloctag(void *v, ulong t)
+{
+ USED(v);
+ USED(t);
+}
+
+void
+setrealloctag(void *v, ulong t)
+{
+ USED(v);
+ USED(t);
+}
DIR diff --git a/src/lib9/mallocz.c b/src/lib9/mallocz.c
t@@ -0,0 +1,14 @@
+#include <unistd.h>
+#include <string.h>
+#include <lib9.h>
+
+void*
+mallocz(unsigned long n, int clr)
+{
+ void *v;
+
+ v = malloc(n);
+ if(clr && v)
+ memset(v, 0, n);
+ return v;
+}
DIR diff --git a/src/lib9/mkfile b/src/lib9/mkfile
t@@ -0,0 +1,2 @@
+<../libutf/mkfile
+
DIR diff --git a/src/lib9/nrand.c b/src/lib9/nrand.c
t@@ -0,0 +1,17 @@
+#include <lib9.h>
+
+#define MASK 0x7fffffffL
+
+int
+nrand(int n)
+{
+ long slop, v;
+
+ if(n < 0)
+ return n;
+ slop = MASK % n;
+ do
+ v = lrand();
+ while(v <= slop);
+ return v % n;
+}
DIR diff --git a/src/lib9/qlock.c b/src/lib9/qlock.c
t@@ -0,0 +1,360 @@
+#include <lib9.h>
+
+static struct {
+ QLp *p;
+ QLp x[1024];
+} ql = {
+ ql.x
+};
+
+enum
+{
+ Queuing,
+ QueuingR,
+ QueuingW,
+ Sleeping,
+};
+
+static ulong (*_rendezvousp)(ulong, ulong) = rendezvous;
+
+/* this gets called by the thread library ONLY to get us to use its rendezvous */
+void
+_qlockinit(ulong (*r)(ulong, ulong))
+{
+ _rendezvousp = r;
+}
+
+/* find a free shared memory location to queue ourselves in */
+static QLp*
+getqlp(void)
+{
+ QLp *p, *op;
+
+ op = ql.p;
+ for(p = op+1; ; p++){
+ if(p == &ql.x[nelem(ql.x)])
+ p = ql.x;
+ if(p == op)
+ abort();
+ if(_tas(&(p->inuse)) == 0){
+ ql.p = p;
+ p->next = nil;
+ break;
+ }
+ }
+ return p;
+}
+
+void
+qlock(QLock *q)
+{
+ QLp *p, *mp;
+
+ lock(&q->lock);
+ if(!q->locked){
+ q->locked = 1;
+ unlock(&q->lock);
+ return;
+ }
+
+
+ /* chain into waiting list */
+ mp = getqlp();
+ p = q->tail;
+ if(p == nil)
+ q->head = mp;
+ else
+ p->next = mp;
+ q->tail = mp;
+ mp->state = Queuing;
+ unlock(&q->lock);
+
+ /* wait */
+ while((*_rendezvousp)((ulong)mp, 1) == ~0)
+ ;
+ mp->inuse = 0;
+}
+
+void
+qunlock(QLock *q)
+{
+ QLp *p;
+
+ lock(&q->lock);
+ p = q->head;
+ if(p != nil){
+ /* wakeup head waiting process */
+ q->head = p->next;
+ if(q->head == nil)
+ q->tail = nil;
+ unlock(&q->lock);
+ while((*_rendezvousp)((ulong)p, 0x12345) == ~0)
+ ;
+ return;
+ }
+ q->locked = 0;
+ unlock(&q->lock);
+}
+
+int
+canqlock(QLock *q)
+{
+ if(!canlock(&q->lock))
+ return 0;
+ if(!q->locked){
+ q->locked = 1;
+ unlock(&q->lock);
+ return 1;
+ }
+ unlock(&q->lock);
+ return 0;
+}
+
+void
+rlock(RWLock *q)
+{
+ QLp *p, *mp;
+
+ lock(&q->lock);
+ if(q->writer == 0 && q->head == nil){
+ /* no writer, go for it */
+ q->readers++;
+ unlock(&q->lock);
+ return;
+ }
+
+ mp = getqlp();
+ p = q->tail;
+ if(p == 0)
+ q->head = mp;
+ else
+ p->next = mp;
+ q->tail = mp;
+ mp->next = nil;
+ mp->state = QueuingR;
+ unlock(&q->lock);
+
+ /* wait in kernel */
+ while((*_rendezvousp)((ulong)mp, 1) == ~0)
+ ;
+ mp->inuse = 0;
+}
+
+int
+canrlock(RWLock *q)
+{
+ lock(&q->lock);
+ if (q->writer == 0 && q->head == nil) {
+ /* no writer; go for it */
+ q->readers++;
+ unlock(&q->lock);
+ return 1;
+ }
+ unlock(&q->lock);
+ return 0;
+}
+
+void
+runlock(RWLock *q)
+{
+ QLp *p;
+
+ lock(&q->lock);
+ if(q->readers <= 0)
+ abort();
+ p = q->head;
+ if(--(q->readers) > 0 || p == nil){
+ unlock(&q->lock);
+ return;
+ }
+
+ /* start waiting writer */
+ if(p->state != QueuingW)
+ abort();
+ q->head = p->next;
+ if(q->head == 0)
+ q->tail = 0;
+ q->writer = 1;
+ unlock(&q->lock);
+
+ /* wakeup waiter */
+ while((*_rendezvousp)((ulong)p, 0) == ~0)
+ ;
+}
+
+void
+wlock(RWLock *q)
+{
+ QLp *p, *mp;
+
+ lock(&q->lock);
+ if(q->readers == 0 && q->writer == 0){
+ /* noone waiting, go for it */
+ q->writer = 1;
+ unlock(&q->lock);
+ return;
+ }
+
+ /* wait */
+ p = q->tail;
+ mp = getqlp();
+ if(p == nil)
+ q->head = mp;
+ else
+ p->next = mp;
+ q->tail = mp;
+ mp->next = nil;
+ mp->state = QueuingW;
+ unlock(&q->lock);
+
+ /* wait in kernel */
+ while((*_rendezvousp)((ulong)mp, 1) == ~0)
+ ;
+ mp->inuse = 0;
+}
+
+int
+canwlock(RWLock *q)
+{
+ lock(&q->lock);
+ if (q->readers == 0 && q->writer == 0) {
+ /* no one waiting; go for it */
+ q->writer = 1;
+ unlock(&q->lock);
+ return 1;
+ }
+ unlock(&q->lock);
+ return 0;
+}
+
+void
+wunlock(RWLock *q)
+{
+ QLp *p;
+
+ lock(&q->lock);
+ if(q->writer == 0)
+ abort();
+ p = q->head;
+ if(p == nil){
+ q->writer = 0;
+ unlock(&q->lock);
+ return;
+ }
+ if(p->state == QueuingW){
+ /* start waiting writer */
+ q->head = p->next;
+ if(q->head == nil)
+ q->tail = nil;
+ unlock(&q->lock);
+ while((*_rendezvousp)((ulong)p, 0) == ~0)
+ ;
+ return;
+ }
+
+ if(p->state != QueuingR)
+ abort();
+
+ /* wake waiting readers */
+ while(q->head != nil && q->head->state == QueuingR){
+ p = q->head;
+ q->head = p->next;
+ q->readers++;
+ while((*_rendezvousp)((ulong)p, 0) == ~0)
+ ;
+ }
+ if(q->head == nil)
+ q->tail = nil;
+ q->writer = 0;
+ unlock(&q->lock);
+}
+
+void
+rsleep(Rendez *r)
+{
+ QLp *t, *me;
+
+ if(!r->l)
+ abort();
+ lock(&r->l->lock);
+ /* we should hold the qlock */
+ if(!r->l->locked)
+ abort();
+
+ /* add ourselves to the wait list */
+ me = getqlp();
+ me->state = Sleeping;
+ if(r->head == nil)
+ r->head = me;
+ else
+ r->tail->next = me;
+ me->next = nil;
+ r->tail = me;
+
+ /* pass the qlock to the next guy */
+ t = r->l->head;
+ if(t){
+ r->l->head = t->next;
+ if(r->l->head == nil)
+ r->l->tail = nil;
+ unlock(&r->l->lock);
+ while((*_rendezvousp)((ulong)t, 0x12345) == ~0)
+ ;
+ }else{
+ r->l->locked = 0;
+ unlock(&r->l->lock);
+ }
+
+ /* wait for a wakeup */
+ while((*_rendezvousp)((ulong)me, 0x23456) == ~0)
+ ;
+ me->inuse = 0;
+ if(!r->l->locked)
+ abort();
+}
+
+int
+rwakeup(Rendez *r)
+{
+ QLp *t;
+
+ /*
+ * take off wait and put on front of queue
+ * put on front so guys that have been waiting will not get starved
+ */
+
+ if(!r->l)
+ abort();
+ lock(&r->l->lock);
+ if(!r->l->locked)
+ abort();
+
+ t = r->head;
+ if(t == nil){
+ unlock(&r->l->lock);
+ return 0;
+ }
+
+ r->head = t->next;
+ if(r->head == nil)
+ r->tail = nil;
+
+ t->next = r->l->head;
+ r->l->head = t;
+ if(r->l->tail == nil)
+ r->l->tail = t;
+
+ t->state = Queuing;
+ unlock(&r->l->lock);
+ return 1;
+}
+
+int
+rwakeupall(Rendez *r)
+{
+ int i;
+
+ for(i=0; rwakeup(r); i++)
+ ;
+ return i;
+}
DIR diff --git a/src/lib9/rand.c b/src/lib9/rand.c
t@@ -0,0 +1,89 @@
+#include <lib9.h>
+
+/*
+ * algorithm by
+ * D. P. Mitchell & J. A. Reeds
+ */
+
+#define LEN 607
+#define TAP 273
+#define MASK 0x7fffffffL
+#define A 48271
+#define M 2147483647
+#define Q 44488
+#define R 3399
+#define NORM (1.0/(1.0+MASK))
+
+static ulong rng_vec[LEN];
+static ulong* rng_tap = rng_vec;
+static ulong* rng_feed = 0;
+static Lock lk;
+
+static void
+isrand(long seed)
+{
+ long lo, hi, x;
+ int i;
+
+ rng_tap = rng_vec;
+ rng_feed = rng_vec+LEN-TAP;
+ seed = seed%M;
+ if(seed < 0)
+ seed += M;
+ if(seed == 0)
+ seed = 89482311;
+ x = seed;
+ /*
+ * Initialize by x[n+1] = 48271 * x[n] mod (2**31 - 1)
+ */
+ for(i = -20; i < LEN; i++) {
+ hi = x / Q;
+ lo = x % Q;
+ x = A*lo - R*hi;
+ if(x < 0)
+ x += M;
+ if(i >= 0)
+ rng_vec[i] = x;
+ }
+}
+
+void
+srand(long seed)
+{
+ lock(&lk);
+ isrand(seed);
+ unlock(&lk);
+}
+
+long
+lrand(void)
+{
+ ulong x;
+
+ lock(&lk);
+
+ rng_tap--;
+ if(rng_tap < rng_vec) {
+ if(rng_feed == 0) {
+ isrand(1);
+ rng_tap--;
+ }
+ rng_tap += LEN;
+ }
+ rng_feed--;
+ if(rng_feed < rng_vec)
+ rng_feed += LEN;
+ x = (*rng_feed + *rng_tap) & MASK;
+ *rng_feed = x;
+
+ unlock(&lk);
+
+ return x;
+}
+
+int
+rand(void)
+{
+ return lrand() & 0x7fff;
+}
+
DIR diff --git a/src/lib9/readn.c b/src/lib9/readn.c
t@@ -0,0 +1,21 @@
+#include <lib9.h>
+
+long
+readn(int f, void *av, long n)
+{
+ char *a;
+ long m, t;
+
+ a = av;
+ t = 0;
+ while(t < n){
+ m = read(f, a+t, n-t);
+ if(m <= 0){
+ if(t == 0)
+ return m;
+ break;
+ }
+ t += m;
+ }
+ return t;
+}
DIR diff --git a/src/lib9/rendez.c b/src/lib9/rendez.c
t@@ -0,0 +1,180 @@
+/*
+ NAME
+ rendezvous - user level process synchronization
+
+ SYNOPSIS
+ ulong rendezvous(ulong tag, ulong value)
+
+ DESCRIPTION
+ The rendezvous system call allows two processes to synchro-
+ nize and exchange a value. In conjunction with the shared
+ memory system calls (see segattach(2) and fork(2)), it
+ enables parallel programs to control their scheduling.
+
+ Two processes wishing to synchronize call rendezvous with a
+ common tag, typically an address in memory they share. One
+ process will arrive at the rendezvous first; it suspends
+ execution until a second arrives. When a second process
+ meets the rendezvous the value arguments are exchanged
+ between the processes and returned as the result of the
+ respective rendezvous system calls. Both processes are
+ awakened when the rendezvous succeeds.
+
+ The set of tag values which two processes may use to
+ rendezvous-their tag space-is inherited when a process
+ forks, unless RFREND is set in the argument to rfork; see
+ fork(2).
+
+ If a rendezvous is interrupted the return value is ~0, so
+ that value should not be used in normal communication.
+
+ * This simulates rendezvous with shared memory, pause, and SIGUSR1.
+ */
+
+#include <signal.h>
+#include <lib9.h>
+
+enum
+{
+ VOUSHASH = 257,
+};
+
+typedef struct Vous Vous;
+struct Vous
+{
+ Vous *link;
+ Lock lk;
+ int pid;
+ ulong val;
+ ulong tag;
+};
+
+static void
+ign(int x)
+{
+ USED(x);
+}
+
+void /*__attribute__((constructor))*/
+ignusr1(void)
+{
+ signal(SIGUSR1, ign);
+}
+
+static Vous vouspool[2048];
+static int nvousused;
+static Vous *vousfree;
+static Vous *voushash[VOUSHASH];
+static Lock vouslock;
+
+static Vous*
+getvous(void)
+{
+ Vous *v;
+
+ if(vousfree){
+ v = vousfree;
+ vousfree = v->link;
+ }else if(nvousused < nelem(vouspool))
+ v = &vouspool[nvousused++];
+ else
+ abort();
+ return v;
+}
+
+static void
+putvous(Vous *v)
+{
+ lock(&vouslock);
+ v->link = vousfree;
+ vousfree = v;
+ unlock(&vouslock);
+}
+
+static Vous*
+findvous(ulong tag, ulong val, int pid)
+{
+ int h;
+ Vous *v, **l;
+
+ lock(&vouslock);
+ h = tag%VOUSHASH;
+ for(l=&voushash[h], v=*l; v; l=&(*l)->link, v=*l){
+ if(v->tag == tag){
+ *l = v->link;
+ unlock(&vouslock);
+ return v;
+ }
+ }
+ v = getvous();
+ v->pid = pid;
+ v->link = voushash[h];
+ v->val = val;
+ v->tag = tag;
+ lock(&v->lk);
+ voushash[h] = v;
+ unlock(&vouslock);
+ return v;
+}
+
+#define DBG 0
+ulong
+rendezvous(ulong tag, ulong val)
+{
+ int me, vpid;
+ ulong rval;
+ Vous *v;
+ sigset_t mask;
+
+ me = getpid();
+ v = findvous(tag, val, me);
+ if(v->pid == me){
+ if(DBG)fprint(2, "pid is %d tag %lux, sleeping\n", me, tag);
+ /*
+ * No rendezvous partner was found; the next guy
+ * through will find v and wake us, so we must go
+ * to sleep.
+ *
+ * To go to sleep:
+ * 1. disable USR1 signals.
+ * 2. unlock v->lk (tells waker okay to signal us).
+ * 3. atomically suspend and enable USR1 signals.
+ *
+ * The call to ignusr1() could be done once at
+ * process creation instead of every time through rendezvous.
+ */
+ v->val = val;
+ ignusr1();
+ sigprocmask(SIG_SETMASK, NULL, &mask);
+ sigaddset(&mask, SIGUSR1);
+ sigprocmask(SIG_SETMASK, &mask, NULL);
+ sigdelset(&mask, SIGUSR1);
+ unlock(&v->lk);
+ sigsuspend(&mask);
+ rval = v->val;
+ if(DBG)fprint(2, "pid is %d, awake\n", me);
+ putvous(v);
+ }else{
+ /*
+ * Found someone to meet. Wake him:
+ *
+ * A. lock v->lk (waits for him to get to his step 2)
+ * B. send a USR1
+ *
+ * He won't get the USR1 until he suspends, which
+ * means it must wake him up (it can't get delivered
+ * before he sleeps).
+ */
+ vpid = v->pid;
+ lock(&v->lk);
+ rval = v->val;
+ v->val = val;
+ unlock(&v->lk);
+ if(kill(vpid, SIGUSR1) < 0){
+ if(DBG)fprint(2, "pid is %d, kill %d failed: %r\n", me, vpid);
+ abort();
+ }
+ }
+ return rval;
+}
+
DIR diff --git a/src/lib9/strecpy.c b/src/lib9/strecpy.c
t@@ -0,0 +1,16 @@
+#include <lib9.h>
+
+char*
+strecpy(char *to, char *e, char *from)
+{
+ if(to >= e)
+ return to;
+ to = memccpy(to, from, '\0', e - to);
+ if(to == nil){
+ to = e - 1;
+ *to = '\0';
+ }else{
+ to--;
+ }
+ return to;
+}
DIR diff --git a/src/lib9/sysfatal.c b/src/lib9/sysfatal.c
t@@ -0,0 +1,20 @@
+#include <lib9.h>
+
+void (*_sysfatal)(char*, ...);
+
+void
+sysfatal(char *fmt, ...)
+{
+ char buf[256];
+ va_list arg;
+
+ va_start(arg, fmt);
+ if(_sysfatal)
+ (*_sysfatal)(fmt, arg);
+ vseprint(buf, buf+sizeof buf, fmt, arg);
+ va_end(arg);
+
+ fprint(2, "%s; %s\n", argv0 ? argv0 : "<prog>", buf);
+ exits("fatal");
+}
+
DIR diff --git a/src/lib9/tas-386.s b/src/lib9/tas-386.s
t@@ -0,0 +1,6 @@
+.globl _tas
+_tas:
+ movl $0xCAFEBABE, %eax
+ movl 4(%esp), %ecx
+ xchgl %eax, 0(%ecx)
+ ret
DIR diff --git a/src/lib9/test.c b/src/lib9/test.c
t@@ -0,0 +1,8 @@
+#include <lib9.h>
+
+int
+main(int argc, char **argv)
+{
+ werrstr("hello world");
+ print("%r\n");
+}
DIR diff --git a/src/lib9/testfork.c b/src/lib9/testfork.c
t@@ -0,0 +1,21 @@
+#include <lib9.h>
+
+void
+sayhi(void *v)
+{
+ USED(v);
+
+ print("hello from subproc\n");
+ print("rendez got %lu from main\n", rendezvous(0x1234, 1234));
+ exits(0);
+}
+
+int
+main(int argc, char **argv)
+{
+ print("hello from main\n");
+ ffork(RFMEM|RFPROC, sayhi, nil);
+
+ print("rendez got %lu from subproc\n", rendezvous(0x1234, 0));
+ exits(0);
+}
DIR diff --git a/src/lib9/tokenize.c b/src/lib9/tokenize.c
t@@ -0,0 +1,106 @@
+#include <lib9.h>
+
+static char qsep[] = " \t\r\n";
+
+static char*
+qtoken(char *s, char *sep)
+{
+ int quoting;
+ char *t;
+
+ quoting = 0;
+ t = s; /* s is output string, t is input string */
+ while(*t!='\0' && (quoting || utfrune(sep, *t)==nil)){
+ if(*t != '\''){
+ *s++ = *t++;
+ continue;
+ }
+ /* *t is a quote */
+ if(!quoting){
+ quoting = 1;
+ t++;
+ continue;
+ }
+ /* quoting and we're on a quote */
+ if(t[1] != '\''){
+ /* end of quoted section; absorb closing quote */
+ t++;
+ quoting = 0;
+ continue;
+ }
+ /* doubled quote; fold one quote into two */
+ t++;
+ *s++ = *t++;
+ }
+ if(*s != '\0'){
+ *s = '\0';
+ if(t == s)
+ t++;
+ }
+ return t;
+}
+
+static char*
+etoken(char *t, char *sep)
+{
+ int quoting;
+
+ /* move to end of next token */
+ quoting = 0;
+ while(*t!='\0' && (quoting || utfrune(sep, *t)==nil)){
+ if(*t != '\''){
+ t++;
+ continue;
+ }
+ /* *t is a quote */
+ if(!quoting){
+ quoting = 1;
+ t++;
+ continue;
+ }
+ /* quoting and we're on a quote */
+ if(t[1] != '\''){
+ /* end of quoted section; absorb closing quote */
+ t++;
+ quoting = 0;
+ continue;
+ }
+ /* doubled quote; fold one quote into two */
+ t += 2;
+ }
+ return t;
+}
+
+int
+gettokens(char *s, char **args, int maxargs, char *sep)
+{
+ int nargs;
+
+ for(nargs=0; nargs<maxargs; nargs++){
+ while(*s!='\0' && utfrune(sep, *s)!=nil)
+ *s++ = '\0';
+ if(*s == '\0')
+ break;
+ args[nargs] = s;
+ s = etoken(s, sep);
+ }
+
+ return nargs;
+}
+
+int
+tokenize(char *s, char **args, int maxargs)
+{
+ int nargs;
+
+ for(nargs=0; nargs<maxargs; nargs++){
+ while(*s!='\0' && utfrune(qsep, *s)!=nil)
+ s++;
+ if(*s == '\0')
+ break;
+ args[nargs] = s;
+ s = qtoken(s, qsep);
+ }
+
+ return nargs;
+}
DIR diff --git a/src/lib9/u16.c b/src/lib9/u16.c
t@@ -0,0 +1,52 @@
+#include <lib9.h>
+static char t16e[] = "0123456789ABCDEF";
+
+int
+dec16(uchar *out, int lim, char *in, int n)
+{
+ int c, w = 0, i = 0;
+ uchar *start = out;
+ uchar *eout = out + lim;
+
+ while(n-- > 0){
+ c = *in++;
+ if('0' <= c && c <= '9')
+ c = c - '0';
+ else if('a' <= c && c <= 'z')
+ c = c - 'a' + 10;
+ else if('A' <= c && c <= 'Z')
+ c = c - 'A' + 10;
+ else
+ continue;
+ w = (w<<4) + c;
+ i++;
+ if(i == 2){
+ if(out + 1 > eout)
+ goto exhausted;
+ *out++ = w;
+ w = 0;
+ i = 0;
+ }
+ }
+exhausted:
+ return out - start;
+}
+
+int
+enc16(char *out, int lim, uchar *in, int n)
+{
+ uint c;
+ char *eout = out + lim;
+ char *start = out;
+
+ while(n-- > 0){
+ c = *in++;
+ if(out + 2 >= eout)
+ goto exhausted;
+ *out++ = t16e[c>>4];
+ *out++ = t16e[c&0xf];
+ }
+exhausted:
+ *out = 0;
+ return out - start;
+}
DIR diff --git a/src/lib9/u32.c b/src/lib9/u32.c
t@@ -0,0 +1,109 @@
+#include <lib9.h>
+
+int
+dec32(uchar *dest, int ndest, char *src, int nsrc)
+{
+ char *s, *tab;
+ uchar *start;
+ int i, u[8];
+
+ if(ndest+1 < (5*nsrc+7)/8)
+ return -1;
+ start = dest;
+ tab = "23456789abcdefghijkmnpqrstuvwxyz";
+ while(nsrc>=8){
+ for(i=0; i<8; i++){
+ s = strchr(tab,(int)src[i]);
+ u[i] = s ? s-tab : 0;
+ }
+ *dest++ = (u[0]<<3) | (0x7 & (u[1]>>2));
+ *dest++ = ((0x3 & u[1])<<6) | (u[2]<<1) | (0x1 & (u[3]>>4));
+ *dest++ = ((0xf & u[3])<<4) | (0xf & (u[4]>>1));
+ *dest++ = ((0x1 & u[4])<<7) | (u[5]<<2) | (0x3 & (u[6]>>3));
+ *dest++ = ((0x7 & u[6])<<5) | u[7];
+ src += 8;
+ nsrc -= 8;
+ }
+ if(nsrc > 0){
+ if(nsrc == 1 || nsrc == 3 || nsrc == 6)
+ return -1;
+ for(i=0; i<nsrc; i++){
+ s = strchr(tab,(int)src[i]);
+ u[i] = s ? s-tab : 0;
+ }
+ *dest++ = (u[0]<<3) | (0x7 & (u[1]>>2));
+ if(nsrc == 2)
+ goto out;
+ *dest++ = ((0x3 & u[1])<<6) | (u[2]<<1) | (0x1 & (u[3]>>4));
+ if(nsrc == 4)
+ goto out;
+ *dest++ = ((0xf & u[3])<<4) | (0xf & (u[4]>>1));
+ if(nsrc == 5)
+ goto out;
+ *dest++ = ((0x1 & u[4])<<7) | (u[5]<<2) | (0x3 & (u[6]>>3));
+ }
+out:
+ return dest-start;
+}
+
+int
+enc32(char *dest, int ndest, uchar *src, int nsrc)
+{
+ char *tab, *start;
+ int j;
+
+ if(ndest <= (8*nsrc+4)/5 )
+ return -1;
+ start = dest;
+ tab = "23456789abcdefghijkmnpqrstuvwxyz";
+ while(nsrc>=5){
+ j = (0x1f & (src[0]>>3));
+ *dest++ = tab[j];
+ j = (0x1c & (src[0]<<2)) | (0x03 & (src[1]>>6));
+ *dest++ = tab[j];
+ j = (0x1f & (src[1]>>1));
+ *dest++ = tab[j];
+ j = (0x10 & (src[1]<<4)) | (0x0f & (src[2]>>4));
+ *dest++ = tab[j];
+ j = (0x1e & (src[2]<<1)) | (0x01 & (src[3]>>7));
+ *dest++ = tab[j];
+ j = (0x1f & (src[3]>>2));
+ *dest++ = tab[j];
+ j = (0x18 & (src[3]<<3)) | (0x07 & (src[4]>>5));
+ *dest++ = tab[j];
+ j = (0x1f & (src[4]));
+ *dest++ = tab[j];
+ src += 5;
+ nsrc -= 5;
+ }
+ if(nsrc){
+ j = (0x1f & (src[0]>>3));
+ *dest++ = tab[j];
+ j = (0x1c & (src[0]<<2));
+ if(nsrc == 1)
+ goto out;
+ j |= (0x03 & (src[1]>>6));
+ *dest++ = tab[j];
+ j = (0x1f & (src[1]>>1));
+ if(nsrc == 2)
+ goto out;
+ *dest++ = tab[j];
+ j = (0x10 & (src[1]<<4));
+ if(nsrc == 3)
+ goto out;
+ j |= (0x0f & (src[2]>>4));
+ *dest++ = tab[j];
+ j = (0x1e & (src[2]<<1));
+ if(nsrc == 4)
+ goto out;
+ j |= (0x01 & (src[3]>>7));
+ *dest++ = tab[j];
+ j = (0x1f & (src[3]>>2));
+ *dest++ = tab[j];
+ j = (0x18 & (src[3]<<3));
+out:
+ *dest++ = tab[j];
+ }
+ *dest = 0;
+ return dest-start;
+}
DIR diff --git a/src/lib9/u64.c b/src/lib9/u64.c
t@@ -0,0 +1,126 @@
+#include <lib9.h>
+
+enum {
+ INVAL= 255
+};
+
+static uchar t64d[256] = {
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL, 62,INVAL,INVAL,INVAL, 63,
+ 52, 53, 54, 55, 56, 57, 58, 59, 60, 61,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
+ 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
+ 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,
+ INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL,INVAL
+};
+static char t64e[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+
+int
+dec64(uchar *out, int lim, char *in, int n)
+{
+ ulong b24;
+ uchar *start = out;
+ uchar *e = out + lim;
+ int i, c;
+
+ b24 = 0;
+ i = 0;
+ while(n-- > 0){
+
+ c = t64d[*(uchar*)in++];
+ if(c == INVAL)
+ continue;
+ switch(i){
+ case 0:
+ b24 = c<<18;
+ break;
+ case 1:
+ b24 |= c<<12;
+ break;
+ case 2:
+ b24 |= c<<6;
+ break;
+ case 3:
+ if(out + 3 > e)
+ goto exhausted;
+
+ b24 |= c;
+ *out++ = b24>>16;
+ *out++ = b24>>8;
+ *out++ = b24;
+ i = -1;
+ break;
+ }
+ i++;
+ }
+ switch(i){
+ case 2:
+ if(out + 1 > e)
+ goto exhausted;
+ *out++ = b24>>16;
+ break;
+ case 3:
+ if(out + 2 > e)
+ goto exhausted;
+ *out++ = b24>>16;
+ *out++ = b24>>8;
+ break;
+ }
+exhausted:
+ return out - start;
+}
+
+int
+enc64(char *out, int lim, uchar *in, int n)
+{
+ int i;
+ ulong b24;
+ char *start = out;
+ char *e = out + lim;
+
+ for(i = n/3; i > 0; i--){
+ b24 = (*in++)<<16;
+ b24 |= (*in++)<<8;
+ b24 |= *in++;
+ if(out + 4 >= e)
+ goto exhausted;
+ *out++ = t64e[(b24>>18)];
+ *out++ = t64e[(b24>>12)&0x3f];
+ *out++ = t64e[(b24>>6)&0x3f];
+ *out++ = t64e[(b24)&0x3f];
+ }
+
+ switch(n%3){
+ case 2:
+ b24 = (*in++)<<16;
+ b24 |= (*in)<<8;
+ if(out + 4 >= e)
+ goto exhausted;
+ *out++ = t64e[(b24>>18)];
+ *out++ = t64e[(b24>>12)&0x3f];
+ *out++ = t64e[(b24>>6)&0x3f];
+ *out++ = '=';
+ break;
+ case 1:
+ b24 = (*in)<<16;
+ if(out + 4 >= e)
+ goto exhausted;
+ *out++ = t64e[(b24>>18)];
+ *out++ = t64e[(b24>>12)&0x3f];
+ *out++ = '=';
+ *out++ = '=';
+ break;
+ }
+exhausted:
+ *out = 0;
+ return out - start;
+}
DIR diff --git a/src/lib9/wait.c b/src/lib9/wait.c
t@@ -0,0 +1,30 @@
+#include <lib9.h>
+
+Waitmsg*
+wait(void)
+{
+ int n, l;
+ char buf[512], *fld[5];
+ Waitmsg *w;
+
+ n = await(buf, sizeof buf-1);
+ if(n < 0)
+ return nil;
+ buf[n] = '\0';
+ if(tokenize(buf, fld, nelem(fld)) != nelem(fld)){
+ werrstr("couldn't parse wait message");
+ return nil;
+ }
+ l = strlen(fld[4])+1;
+ w = malloc(sizeof(Waitmsg)+l);
+ if(w == nil)
+ return nil;
+ w->pid = atoi(fld[0]);
+ w->time[0] = atoi(fld[1]);
+ w->time[1] = atoi(fld[2]);
+ w->time[2] = atoi(fld[3]);
+ w->msg = (char*)&w[1];
+ memmove(w->msg, fld[4], l);
+ return w;
+}
+
DIR diff --git a/src/lib9/werrstr.c b/src/lib9/werrstr.c
t@@ -0,0 +1,13 @@
+#include <lib9.h>
+
+void
+werrstr(char *fmt, ...)
+{
+ va_list arg;
+ char buf[ERRMAX];
+
+ va_start(arg, fmt);
+ vseprint(buf, buf+ERRMAX, fmt, arg);
+ va_end(arg);
+ errstr(buf, ERRMAX);
+}
DIR diff --git a/src/libbio/LICENSE b/src/libbio/LICENSE
t@@ -0,0 +1,258 @@
+The Plan 9 software is provided under the terms of the
+Lucent Public License, Version 1.02, reproduced below,
+with the following exceptions:
+
+1. No right is granted to create derivative works of or
+ to redistribute (other than with the Plan 9 Operating System)
+ the screen imprinter fonts identified in subdirectory
+ /lib/font/bit/lucida and printer fonts (Lucida Sans Unicode, Lucida
+ Sans Italic, Lucida Sans Demibold, Lucida Typewriter, Lucida Sans
+ Typewriter83), identified in subdirectory /sys/lib/postscript/font.
+ These directories contain material copyrights by B&H Inc. and Y&Y Inc.
+
+2. The printer fonts identified in subdirectory /sys/lib/ghostscript/font
+ are subject to the GNU GPL, reproduced in the file /LICENSE.gpl.
+
+3. The ghostscript program in the subdirectory /sys/src/cmd/gs is
+ covered by the Aladdin Free Public License, reproduced in the file
+ /LICENSE.afpl.
+
+===================================================================
+
+Lucent Public License Version 1.02
+
+THE ACCOMPANYING PROGRAM IS PROVIDED UNDER THE TERMS OF THIS PUBLIC
+LICENSE ("AGREEMENT"). ANY USE, REPRODUCTION OR DISTRIBUTION OF THE
+PROGRAM CONSTITUTES RECIPIENT'S ACCEPTANCE OF THIS AGREEMENT.
+
+1. DEFINITIONS
+
+"Contribution" means:
+
+ a. in the case of Lucent Technologies Inc. ("LUCENT"), the Original
+ Program, and
+ b. in the case of each Contributor,
+
+ i. changes to the Program, and
+ ii. additions to the Program;
+
+ where such changes and/or additions to the Program were added to the
+ Program by such Contributor itself or anyone acting on such
+ Contributor's behalf, and the Contributor explicitly consents, in
+ accordance with Section 3C, to characterization of the changes and/or
+ additions as Contributions.
+
+"Contributor" means LUCENT and any other entity that has Contributed a
+Contribution to the Program.
+
+"Distributor" means a Recipient that distributes the Program,
+modifications to the Program, or any part thereof.
+
+"Licensed Patents" mean patent claims licensable by a Contributor
+which are necessarily infringed by the use or sale of its Contribution
+alone or when combined with the Program.
+
+"Original Program" means the original version of the software
+accompanying this Agreement as released by LUCENT, including source
+code, object code and documentation, if any.
+
+"Program" means the Original Program and Contributions or any part
+thereof
+
+"Recipient" means anyone who receives the Program under this
+Agreement, including all Contributors.
+
+2. GRANT OF RIGHTS
+
+ a. Subject to the terms of this Agreement, each Contributor hereby
+ grants Recipient a non-exclusive, worldwide, royalty-free copyright
+ license to reproduce, prepare derivative works of, publicly display,
+ publicly perform, distribute and sublicense the Contribution of such
+ Contributor, if any, and such derivative works, in source code and
+ object code form.
+
+ b. Subject to the terms of this Agreement, each Contributor hereby
+ grants Recipient a non-exclusive, worldwide, royalty-free patent
+ license under Licensed Patents to make, use, sell, offer to sell,
+ import and otherwise transfer the Contribution of such Contributor, if
+ any, in source code and object code form. The patent license granted
+ by a Contributor shall also apply to the combination of the
+ Contribution of that Contributor and the Program if, at the time the
+ Contribution is added by the Contributor, such addition of the
+ Contribution causes such combination to be covered by the Licensed
+ Patents. The patent license granted by a Contributor shall not apply
+ to (i) any other combinations which include the Contribution, nor to
+ (ii) Contributions of other Contributors. No hardware per se is
+ licensed hereunder.
+
+ c. Recipient understands that although each Contributor grants the
+ licenses to its Contributions set forth herein, no assurances are
+ provided by any Contributor that the Program does not infringe the
+ patent or other intellectual property rights of any other entity. Each
+ Contributor disclaims any liability to Recipient for claims brought by
+ any other entity based on infringement of intellectual property rights
+ or otherwise. As a condition to exercising the rights and licenses
+ granted hereunder, each Recipient hereby assumes sole responsibility
+ to secure any other intellectual property rights needed, if any. For
+ example, if a third party patent license is required to allow
+ Recipient to distribute the Program, it is Recipient's responsibility
+ to acquire that license before distributing the Program.
+
+ d. Each Contributor represents that to its knowledge it has sufficient
+ copyright rights in its Contribution, if any, to grant the copyright
+ license set forth in this Agreement.
+
+3. REQUIREMENTS
+
+A. Distributor may choose to distribute the Program in any form under
+this Agreement or under its own license agreement, provided that:
+
+ a. it complies with the terms and conditions of this Agreement;
+
+ b. if the Program is distributed in source code or other tangible
+ form, a copy of this Agreement or Distributor's own license agreement
+ is included with each copy of the Program; and
+
+ c. if distributed under Distributor's own license agreement, such
+ license agreement:
+
+ i. effectively disclaims on behalf of all Contributors all warranties
+ and conditions, express and implied, including warranties or
+ conditions of title and non-infringement, and implied warranties or
+ conditions of merchantability and fitness for a particular purpose;
+ ii. effectively excludes on behalf of all Contributors all liability
+ for damages, including direct, indirect, special, incidental and
+ consequential damages, such as lost profits; and
+ iii. states that any provisions which differ from this Agreement are
+ offered by that Contributor alone and not by any other party.
+
+B. Each Distributor must include the following in a conspicuous
+ location in the Program:
+
+ Copyright (C) 2003, Lucent Technologies Inc. and others. All Rights
+ Reserved.
+
+C. In addition, each Contributor must identify itself as the
+originator of its Contribution in a manner that reasonably allows
+subsequent Recipients to identify the originator of the Contribution.
+Also, each Contributor must agree that the additions and/or changes
+are intended to be a Contribution. Once a Contribution is contributed,
+it may not thereafter be revoked.
+
+4. COMMERCIAL DISTRIBUTION
+
+Commercial distributors of software may accept certain
+responsibilities with respect to end users, business partners and the
+like. While this license is intended to facilitate the commercial use
+of the Program, the Distributor who includes the Program in a
+commercial product offering should do so in a manner which does not
+create potential liability for Contributors. Therefore, if a
+Distributor includes the Program in a commercial product offering,
+such Distributor ("Commercial Distributor") hereby agrees to defend
+and indemnify every Contributor ("Indemnified Contributor") against
+any losses, damages and costs (collectively"Losses") arising from
+claims, lawsuits and other legal actions brought by a third party
+against the Indemnified Contributor to the extent caused by the acts
+or omissions of such Commercial Distributor in connection with its
+distribution of the Program in a commercial product offering. The
+obligations in this section do not apply to any claims or Losses
+relating to any actual or alleged intellectual property infringement.
+In order to qualify, an Indemnified Contributor must: a) promptly
+notify the Commercial Distributor in writing of such claim, and b)
+allow the Commercial Distributor to control, and cooperate with the
+Commercial Distributor in, the defense and any related settlement
+negotiations. The Indemnified Contributor may participate in any such
+claim at its own expense.
+
+For example, a Distributor might include the Program in a commercial
+product offering, Product X. That Distributor is then a Commercial
+Distributor. If that Commercial Distributor then makes performance
+claims, or offers warranties related to Product X, those performance
+claims and warranties are such Commercial Distributor's responsibility
+alone. Under this section, the Commercial Distributor would have to
+defend claims against the Contributors related to those performance
+claims and warranties, and if a court requires any Contributor to pay
+any damages as a result, the Commercial Distributor must pay those
+damages.
+
+5. NO WARRANTY
+
+EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, THE PROGRAM IS
+PROVIDED ON AN"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT LIMITATION, ANY
+WARRANTIES OR CONDITIONS OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY
+OR FITNESS FOR A PARTICULAR PURPOSE. Each Recipient is solely
+responsible for determining the appropriateness of using and
+distributing the Program and assumes all risks associated with its
+exercise of rights under this Agreement, including but not limited to
+the risks and costs of program errors, compliance with applicable
+laws, damage to or loss of data, programs or equipment, and
+unavailability or interruption of operations.
+
+6. DISCLAIMER OF LIABILITY
+
+EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, NEITHER RECIPIENT NOR
+ANY CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING
+WITHOUT LIMITATION LOST PROFITS), HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OR
+DISTRIBUTION OF THE PROGRAM OR THE EXERCISE OF ANY RIGHTS GRANTED
+HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+7. EXPORT CONTROL
+
+Recipient agrees that Recipient alone is responsible for compliance
+with the United States export administration regulations (and the
+export control laws and regulation of any other countries).
+
+8. GENERAL
+
+If any provision of this Agreement is invalid or unenforceable under
+applicable law, it shall not affect the validity or enforceability of
+the remainder of the terms of this Agreement, and without further
+action by the parties hereto, such provision shall be reformed to the
+minimum extent necessary to make such provision valid and enforceable.
+
+If Recipient institutes patent litigation against a Contributor with
+respect to a patent applicable to software (including a cross-claim or
+counterclaim in a lawsuit), then any patent licenses granted by that
+Contributor to such Recipient under this Agreement shall terminate as
+of the date such litigation is filed. In addition, if Recipient
+institutes patent litigation against any entity (including a
+cross-claim or counterclaim in a lawsuit) alleging that the Program
+itself (excluding combinations of the Program with other software or
+hardware) infringes such Recipient's patent(s), then such Recipient's
+rights granted under Section 2(b) shall terminate as of the date such
+litigation is filed.
+
+All Recipient's rights under this Agreement shall terminate if it
+fails to comply with any of the material terms or conditions of this
+Agreement and does not cure such failure in a reasonable period of
+time after becoming aware of such noncompliance. If all Recipient's
+rights under this Agreement terminate, Recipient agrees to cease use
+and distribution of the Program as soon as reasonably practicable.
+However, Recipient's obligations under this Agreement and any licenses
+granted by Recipient relating to the Program shall continue and
+survive.
+
+LUCENT may publish new versions (including revisions) of this
+Agreement from time to time. Each new version of the Agreement will be
+given a distinguishing version number. The Program (including
+Contributions) may always be distributed subject to the version of the
+Agreement under which it was received. In addition, after a new
+version of the Agreement is published, Contributor may elect to
+distribute the Program (including its Contributions) under the new
+version. No one other than LUCENT has the right to modify this
+Agreement. Except as expressly stated in Sections 2(a) and 2(b) above,
+Recipient receives no rights or licenses to the intellectual property
+of any Contributor under this Agreement, whether expressly, by
+implication, estoppel or otherwise. All rights in the Program not
+expressly granted under this Agreement are reserved.
+
+This Agreement is governed by the laws of the State of New York and
+the intellectual property laws of the United States of America. No
+party to this Agreement will bring a legal action under this Agreement
+more than one year after the cause of action arose. Each party waives
+its rights to a jury trial in any resulting litigation.
+
DIR diff --git a/src/libbio/Make.FreeBSD-386 b/src/libbio/Make.FreeBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.HP-UX-9000 b/src/libbio/Make.HP-UX-9000
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS=-O -c -Ae -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.Linux-386 b/src/libbio/Make.Linux-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.NetBSD-386 b/src/libbio/Make.NetBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.OSF1-alpha b/src/libbio/Make.OSF1-alpha
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.SunOS-sun4u b/src/libbio/Make.SunOS-sun4u
t@@ -0,0 +1,2 @@
+include Make.SunOS-sun4u-$(CC)
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.SunOS-sun4u-cc b/src/libbio/Make.SunOS-sun4u-cc
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I. -O
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libbio/Make.SunOS-sun4u-gcc b/src/libbio/Make.SunOS-sun4u-gcc
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libbio/Makefile b/src/libbio/Makefile
t@@ -0,0 +1,108 @@
+
+# this works in gnu make
+SYSNAME:=${shell uname}
+OBJTYPE:=${shell uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'}
+
+# this works in bsd make
+SYSNAME!=uname
+OBJTYPE!=uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'
+
+# the gnu rules will mess up bsd but not vice versa,
+# hence the gnu rules come first.
+
+include Make.$(SYSNAME)-$(OBJTYPE)
+
+PREFIX=/usr/local
+
+NUKEFILES=
+
+TGZFILES=
+
+LIB=libbio.a
+VERSION=2.0
+PORTPLACE=devel/libbio
+NAME=libbio
+
+OFILES=\
+ bbuffered.$O\
+ bfildes.$O\
+ bflush.$O\
+ bgetc.$O\
+ bgetd.$O\
+ binit.$O\
+ boffset.$O\
+ bprint.$O\
+ bputc.$O\
+ brdline.$O\
+ brdstr.$O\
+ bread.$O\
+ bseek.$O\
+ bwrite.$O\
+ bgetrune.$O\
+ bputrune.$O\
+
+HFILES=\
+ bio.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ install -m 0644 bio.3 $(PREFIX)/man/man3/bio.3
+ install -m 0644 bio.h $(PREFIX)/include
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+bcat: bcat.$O $(LIB)
+ $(CC) -o bcat bcat.$O $(LIB) -L/usr/local/lib -lfmt -lutf
+
+$(LIB): $(OFILES)
+ $(AR) $(ARFLAGS) $(LIB) $(OFILES)
+
+NUKEFILES+=$(LIB)
+.c.$O:
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+%.$O: %.c
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+
+$(OFILES): $(HFILES)
+
+tgz:
+ rm -rf $(NAME)-$(VERSION)
+ mkdir $(NAME)-$(VERSION)
+ cp Makefile Make.* README LICENSE NOTICE *.[ch137] rpm.spec bundle.ports $(TGZFILES) $(NAME)-$(VERSION)
+ tar cf - $(NAME)-$(VERSION) | gzip >$(NAME)-$(VERSION).tgz
+ rm -rf $(NAME)-$(VERSION)
+
+clean:
+ rm -f $(OFILES) $(LIB)
+
+nuke:
+ rm -f $(OFILES) *.tgz *.rpm $(NUKEFILES)
+
+rpm:
+ make tgz
+ cp $(NAME)-$(VERSION).tgz /usr/src/RPM/SOURCES
+ rpm -ba rpm.spec
+ cp /usr/src/RPM/SRPMS/$(NAME)-$(VERSION)-1.src.rpm .
+ cp /usr/src/RPM/RPMS/i586/$(NAME)-$(VERSION)-1.i586.rpm .
+ scp *.rpm rsc@amsterdam.lcs.mit.edu:public_html/software
+
+PORTDIR=/usr/ports/$(PORTPLACE)
+
+ports:
+ make tgz
+ rm -rf $(PORTDIR)
+ mkdir $(PORTDIR)
+ cp $(NAME)-$(VERSION).tgz /usr/ports/distfiles
+ cat bundle.ports | (cd $(PORTDIR) && awk '$$1=="---" && $$3=="---" { ofile=$$2; next} {if(ofile) print >ofile}')
+ (cd $(PORTDIR); make makesum)
+ (cd $(PORTDIR); make)
+ (cd $(PORTDIR); /usr/local/bin/portlint)
+ rm -rf $(PORTDIR)/work
+ shar `find $(PORTDIR)` > ports.shar
+ (cd $(PORTDIR); tar cf - *) | gzip >$(NAME)-$(VERSION)-ports.tgz
+ scp *.tgz rsc@amsterdam.lcs.mit.edu:public_html/software
+
+.phony: all clean nuke install tgz rpm ports
DIR diff --git a/src/libbio/Makefile.MID b/src/libbio/Makefile.MID
t@@ -0,0 +1,37 @@
+LIB=libbio.a
+VERSION=2.0
+PORTPLACE=devel/libbio
+NAME=libbio
+
+OFILES=\
+ bbuffered.$O\
+ bfildes.$O\
+ bflush.$O\
+ bgetc.$O\
+ bgetd.$O\
+ binit.$O\
+ boffset.$O\
+ bprint.$O\
+ bputc.$O\
+ brdline.$O\
+ brdstr.$O\
+ bread.$O\
+ bseek.$O\
+ bwrite.$O\
+ bgetrune.$O\
+ bputrune.$O\
+
+HFILES=\
+ bio.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ install -m 0644 bio.3 $(PREFIX)/man/man3/bio.3
+ install -m 0644 bio.h $(PREFIX)/include
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+bcat: bcat.$O $(LIB)
+ $(CC) -o bcat bcat.$O $(LIB) -L/usr/local/lib -lfmt -lutf
+
DIR diff --git a/src/libbio/NOTICE b/src/libbio/NOTICE
t@@ -0,0 +1,27 @@
+Copyright © 1994-1999 Lucent Technologies Inc. All rights reserved.
+Portions Copyright © 1995-1997 C H Forsyth (forsyth@caldo.demon.co.uk). All rights reserved.
+Portions Copyright © 1997-1999 Vita Nuova Limited. All rights reserved.
+Portions Copyright © 2000-2002 Vita Nuova Holdings Limited (www.vitanuova.com). All rights reserved.
+
+Under a licence agreement with Lucent Technologies Inc. effective 1st March 2000,
+Vita Nuova Holdings Limited has the right to determine (within a specified scope)
+the form and content of sublicences for this software.
+
+Vita Nuova Holdings Limited now makes this software available as Free
+Software under the terms of the `GNU General Public LIcense, Version 2'
+(see the file LICENCE or http://www.fsf.org/copyleft/gpl.html for
+the full terms and conditions). One of the conditions of that licence
+is that you must keep intact all notices that refer to that licence and to the absence of
+of any warranty: for this software, note that includes this NOTICE file in particular.
+
+This suite of programs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+`GNU General Public License' for more details.
+
+This copyright NOTICE applies to all files in this directory and
+subdirectories, unless another copyright notice appears in a given
+file or subdirectory. If you take code from this software to use in
+other programs, you must somehow include with it an appropriate
+copyright notice that includes the copyright notice and the other
+notices above.
DIR diff --git a/src/libbio/README b/src/libbio/README
t@@ -0,0 +1,7 @@
+This is a Unix port of the Plan 9 buffered I/O library,
+originally done for the Inferno operating system.
+
+Russ Cox repackaged this to build as a standalone
+Unix library. Send comments about packaging to
+Russ Cox <rsc@post.harvard.edu>
+
DIR diff --git a/src/libbio/bbuffered.c b/src/libbio/bbuffered.c
t@@ -0,0 +1,20 @@
+#include "lib9.h"
+#include <bio.h>
+
+int
+Bbuffered(Biobuf *bp)
+{
+ switch(bp->state) {
+ case Bracteof:
+ case Bractive:
+ return -bp->icount;
+
+ case Bwactive:
+ return bp->bsize + bp->ocount;
+
+ case Binactive:
+ return 0;
+ }
+ fprint(2, "Bbuffered: unknown state %d\n", bp->state);
+ return 0;
+}
DIR diff --git a/src/libbio/bcat.c b/src/libbio/bcat.c
t@@ -0,0 +1,42 @@
+#include <fmt.h>
+#include "bio.h"
+
+Biobuf bout;
+
+void
+bcat(Biobuf *b, char *name)
+{
+ char buf[1000];
+ int n;
+
+ while((n = Bread(b, buf, sizeof buf)) > 0){
+ if(Bwrite(&bout, buf, n) < 0)
+ fprint(2, "writing during %s: %r\n", name);
+ }
+ if(n < 0)
+ fprint(2, "reading %s: %r\n", name);
+}
+
+int
+main(int argc, char **argv)
+{
+ int i;
+ Biobuf b, *bp;
+
+ Binit(&bout, 1, O_WRONLY);
+
+ if(argc == 1){
+ Binit(&b, 0, O_RDONLY);
+ bcat(&b, "<stdin>");
+ }else{
+ for(i=1; i<argc; i++){
+ if((bp = Bopen(argv[i], O_RDONLY)) == 0){
+ fprint(2, "Bopen %s: %r\n", argv[i]);
+ continue;
+ }
+ bcat(bp, argv[i]);
+ Bterm(bp);
+ }
+ }
+ exit(0);
+}
DIR diff --git a/src/libbio/bfildes.c b/src/libbio/bfildes.c
t@@ -0,0 +1,9 @@
+#include "lib9.h"
+#include <bio.h>
+
+int
+Bfildes(Biobuf *bp)
+{
+
+ return bp->fid;
+}
DIR diff --git a/src/libbio/bflush.c b/src/libbio/bflush.c
t@@ -0,0 +1,33 @@
+#include "lib9.h"
+#include <bio.h>
+
+int
+Bflush(Biobuf *bp)
+{
+ int n, c;
+
+ switch(bp->state) {
+ case Bwactive:
+ n = bp->bsize+bp->ocount;
+ if(n == 0)
+ return 0;
+ c = write(bp->fid, bp->bbuf, n);
+ if(n == c) {
+ bp->offset += n;
+ bp->ocount = -bp->bsize;
+ return 0;
+ }
+ bp->state = Binactive;
+ bp->ocount = 0;
+ break;
+
+ case Bracteof:
+ bp->state = Bractive;
+
+ case Bractive:
+ bp->icount = 0;
+ bp->gbuf = bp->ebuf;
+ return 0;
+ }
+ return Beof;
+}
DIR diff --git a/src/libbio/bgetc.c b/src/libbio/bgetc.c
t@@ -0,0 +1,52 @@
+#include "lib9.h"
+#include <bio.h>
+
+int
+Bgetc(Biobuf *bp)
+{
+ int i;
+
+loop:
+ i = bp->icount;
+ if(i != 0) {
+ bp->icount = i+1;
+ return bp->ebuf[i];
+ }
+ if(bp->state != Bractive) {
+ if(bp->state == Bracteof)
+ bp->state = Bractive;
+ return Beof;
+ }
+ /*
+ * get next buffer, try to keep Bungetsize
+ * characters pre-catenated from the previous
+ * buffer to allow that many ungets.
+ */
+ memmove(bp->bbuf-Bungetsize, bp->ebuf-Bungetsize, Bungetsize);
+ i = read(bp->fid, bp->bbuf, bp->bsize);
+ bp->gbuf = bp->bbuf;
+ if(i <= 0) {
+ if(i < 0)
+ bp->state = Binactive;
+ return Beof;
+ }
+ if(i < bp->bsize) {
+ memmove(bp->ebuf-i-Bungetsize, bp->bbuf-Bungetsize, i+Bungetsize);
+ bp->gbuf = bp->ebuf-i;
+ }
+ bp->icount = -i;
+ bp->offset += i;
+ goto loop;
+}
+
+int
+Bungetc(Biobuf *bp)
+{
+
+ if(bp->state == Bracteof)
+ bp->state = Bractive;
+ if(bp->state != Bractive)
+ return Beof;
+ bp->icount--;
+ return 1;
+}
DIR diff --git a/src/libbio/bgetd.c b/src/libbio/bgetd.c
t@@ -0,0 +1,36 @@
+#include "lib9.h"
+#include <bio.h>
+
+struct bgetd
+{
+ Biobuf* b;
+ int eof;
+};
+
+static int
+Bgetdf(void *vp)
+{
+ int c;
+ struct bgetd *bg = vp;
+
+ c = Bgetc(bg->b);
+ if(c == Beof)
+ bg->eof = 1;
+ return c;
+}
+
+int
+Bgetd(Biobuf *bp, double *dp)
+{
+ double d;
+ struct bgetd b;
+
+ b.b = bp;
+ b.eof = 0;
+ d = fmtcharstod(Bgetdf, &b);
+ if(b.eof)
+ return -1;
+ Bungetc(bp);
+ *dp = d;
+ return 1;
+}
DIR diff --git a/src/libbio/bgetrune.c b/src/libbio/bgetrune.c
t@@ -0,0 +1,47 @@
+#include "lib9.h"
+#include <bio.h>
+#include <utf.h>
+
+long
+Bgetrune(Biobuf *bp)
+{
+ int c, i;
+ Rune rune;
+ char str[4];
+
+ c = Bgetc(bp);
+ if(c < Runeself) { /* one char */
+ bp->runesize = 1;
+ return c;
+ }
+ str[0] = c;
+
+ for(i=1;;) {
+ c = Bgetc(bp);
+ if(c < 0)
+ return c;
+ str[i++] = c;
+
+ if(fullrune(str, i)) {
+ bp->runesize = chartorune(&rune, str);
+ while(i > bp->runesize) {
+ Bungetc(bp);
+ i--;
+ }
+ return rune;
+ }
+ }
+}
+
+int
+Bungetrune(Biobuf *bp)
+{
+
+ if(bp->state == Bracteof)
+ bp->state = Bractive;
+ if(bp->state != Bractive)
+ return Beof;
+ bp->icount -= bp->runesize;
+ bp->runesize = 0;
+ return 1;
+}
DIR diff --git a/src/libbio/binit.c b/src/libbio/binit.c
t@@ -0,0 +1,142 @@
+#include "lib9.h"
+#include <bio.h>
+
+enum
+{
+ MAXBUFS = 20
+};
+
+static Biobuf* wbufs[MAXBUFS];
+static int atexitflag;
+
+static
+void
+batexit(void)
+{
+ Biobuf *bp;
+ int i;
+
+ for(i=0; i<MAXBUFS; i++) {
+ bp = wbufs[i];
+ if(bp != 0) {
+ wbufs[i] = 0;
+ Bflush(bp);
+ }
+ }
+}
+
+static
+void
+deinstall(Biobuf *bp)
+{
+ int i;
+
+ for(i=0; i<MAXBUFS; i++)
+ if(wbufs[i] == bp)
+ wbufs[i] = 0;
+}
+
+static
+void
+install(Biobuf *bp)
+{
+ int i;
+
+ deinstall(bp);
+ for(i=0; i<MAXBUFS; i++)
+ if(wbufs[i] == 0) {
+ wbufs[i] = bp;
+ break;
+ }
+ if(atexitflag == 0) {
+ atexitflag = 1;
+ atexit(batexit);
+ }
+}
+
+int
+Binits(Biobuf *bp, int f, int mode, unsigned char *p, int size)
+{
+
+ p += Bungetsize; /* make room for Bungets */
+ size -= Bungetsize;
+
+ switch(mode) {
+ default:
+ fprint(2, "Bopen: unknown mode %d\n", mode);
+ return Beof;
+
+ case OREAD:
+ bp->state = Bractive;
+ bp->ocount = 0;
+ break;
+
+ case OWRITE:
+ install(bp);
+ bp->state = Bwactive;
+ bp->ocount = -size;
+ break;
+ }
+ bp->bbuf = p;
+ bp->ebuf = p+size;
+ bp->bsize = size;
+ bp->icount = 0;
+ bp->gbuf = bp->ebuf;
+ bp->fid = f;
+ bp->flag = 0;
+ bp->rdline = 0;
+ bp->offset = 0;
+/* bp->runesize = 0; */
+ return 0;
+}
+
+
+int
+Binit(Biobuf *bp, int f, int mode)
+{
+ return Binits(bp, f, mode, bp->b, sizeof(bp->b));
+}
+
+Biobuf*
+Bopen(char *name, int mode)
+{
+ Biobuf *bp;
+ int f;
+
+ switch(mode) {
+ default:
+ fprint(2, "Bopen: unknown mode %d\n", mode);
+ return 0;
+
+ case OREAD:
+ f = open(name, OREAD);
+ if(f < 0)
+ return 0;
+ break;
+
+ case OWRITE:
+ f = creat(name, 0666);
+ if(f < 0)
+ return 0;
+ }
+ bp = malloc(sizeof(Biobuf));
+ if(bp == 0)
+ return 0;
+ Binits(bp, f, mode, bp->b, sizeof(bp->b));
+ bp->flag = Bmagic;
+ return bp;
+}
+
+int
+Bterm(Biobuf *bp)
+{
+
+ deinstall(bp);
+ Bflush(bp);
+ if(bp->flag == Bmagic) {
+ bp->flag = 0;
+ close(bp->fid);
+ free(bp);
+ }
+ return 0;
+}
DIR diff --git a/src/libbio/bio.3 b/src/libbio/bio.3
t@@ -0,0 +1,336 @@
+.TH BIO 3
+.SH NAME
+Bopen, Binit, Binits, Brdline, Brdstr, Bgetc, Bgetd, Bungetc, Bread, Bseek, Boffset, Bfildes, Blinelen, Bputc, Bprint, Bvprint, Bwrite, Bflush, Bterm, Bbuffered \- buffered input/output
+.SH SYNOPSIS
+.ta \w'Biobuf* 'u
+.B #include <fmt.h>
+.B #include <bio.h>
+.PP
+.B
+Biobuf* Bopen(char *file, int mode)
+.PP
+.B
+int Binit(Biobuf *bp, int fd, int mode)
+.PP
+.B
+int Bterm(Biobuf *bp)
+.PP
+.B
+int Bprint(Biobuf *bp, char *format, ...)
+.PP
+.B
+int Bvprint(Biobuf *bp, char *format, va_list arglist);
+.PP
+.B
+void* Brdline(Biobuf *bp, int delim)
+.PP
+.B
+char* Brdstr(Biobuf *bp, int delim, int nulldelim)
+.PP
+.B
+int Blinelen(Biobuf *bp)
+.PP
+.B
+off_t Boffset(Biobuf *bp)
+.PP
+.B
+int Bfildes(Biobuf *bp)
+.PP
+.B
+int Bgetc(Biobuf *bp)
+.PP
+.B
+long Bgetrune(Biobufhdr *bp)
+.PP
+.B
+int Bgetd(Biobuf *bp, double *d)
+.PP
+.B
+int Bungetc(Biobuf *bp)
+.PP
+.B
+int Bungetrune(Biobufhdr *bp)
+.PP
+.B
+off_t Bseek(Biobuf *bp, off_t n, int type)
+.PP
+.B
+int Bputc(Biobuf *bp, int c)
+.PP
+.B
+int Bputrune(Biobufhdr *bp, long c)
+.PP
+.B
+long Bread(Biobuf *bp, void *addr, long nbytes)
+.PP
+.B
+long Bwrite(Biobuf *bp, void *addr, long nbytes)
+.PP
+.B
+int Bflush(Biobuf *bp)
+.PP
+.B
+int Bbuffered(Biobuf *bp)
+.PP
+.SH DESCRIPTION
+These routines implement fast buffered I/O.
+I/O on different file descriptors is independent.
+.PP
+.I Bopen
+opens
+.I file
+for mode
+.B O_RDONLY
+or creates for mode
+.BR O_WRONLY .
+It calls
+.IR malloc (3)
+to allocate a buffer.
+.PP
+.I Binit
+initializes a buffer
+with the open file descriptor passed in
+by the user.
+.PP
+Arguments
+of types pointer to Biobuf and pointer to Biobuf
+can be used interchangeably in the following routines.
+.PP
+.IR Bopen ,
+.IR Binit ,
+or
+.I Binits
+should be called before any of the
+other routines on that buffer.
+.I Bfildes
+returns the integer file descriptor of the associated open file.
+.PP
+.I Bterm
+flushes the buffer for
+.IR bp .
+If the buffer was allocated by
+.IR Bopen ,
+the buffer is
+.I freed
+and the file is closed.
+.PP
+.I Brdline
+reads a string from the file associated with
+.I bp
+up to and including the first
+.I delim
+character.
+The delimiter character at the end of the line is
+not altered.
+.I Brdline
+returns a pointer to the start of the line or
+.L 0
+on end-of-file or read error.
+.I Blinelen
+returns the length (including the delimiter)
+of the most recent string returned by
+.IR Brdline .
+.PP
+.I Brdstr
+returns a
+.IR malloc (3)-allocated
+buffer containing the next line of input delimited by
+.IR delim ,
+terminated by a NUL (0) byte.
+Unlike
+.IR Brdline ,
+which returns when its buffer is full even if no delimiter has been found,
+.I Brdstr
+will return an arbitrarily long line in a single call.
+If
+.I nulldelim
+is set, the terminal delimiter will be overwritten with a NUL.
+After a successful call to
+.IR Brdstr ,
+the return value of
+.I Blinelen
+will be the length of the returned buffer, excluding the NUL.
+.PP
+.I Bgetc
+returns the next byte from
+.IR bp ,
+or a negative value
+at end of file.
+.I Bungetc
+may be called immediately after
+.I Bgetc
+to allow the same byte to be reread.
+.PP
+.I Bgetrune
+calls
+.I Bgetc
+to read the bytes of the next
+.SM UTF
+sequence in the input stream and returns the value of the rune
+represented by the sequence.
+It returns a negative value
+at end of file.
+.I Bungetrune
+may be called immediately after
+.I Bgetrune
+to allow the same
+.SM UTF
+sequence to be reread as either bytes or a rune.
+.I Bungetc
+and
+.I Bungetrune
+may back up a maximum of five bytes.
+.PP
+.I Bgetd
+uses
+.I fmtcharstod
+(undocumented)
+and
+.I Bgetc
+to read the formatted
+floating-point number in the input stream,
+skipping initial blanks and tabs.
+The value is stored in
+.BR *d.
+.PP
+.I Bread
+reads
+.I nbytes
+of data from
+.I bp
+into memory starting at
+.IR addr .
+The number of bytes read is returned on success
+and a negative value is returned if a read error occurred.
+.PP
+.I Bseek
+applies
+.IR lseek (2)
+to
+.IR bp .
+It returns the new file offset.
+.I Boffset
+returns the file offset of the next character to be processed.
+.PP
+.I Bputc
+outputs the low order 8 bits of
+.I c
+on
+.IR bp .
+If this causes a
+.IR write
+to occur and there is an error,
+a negative value is returned.
+Otherwise, a zero is returned.
+.PP
+.I Bputrune
+calls
+.I Bputc
+to output the low order
+16 bits of
+.I c
+as a rune
+in
+.SM UTF
+format
+on the output stream.
+.PP
+.I Bprint
+is a buffered interface to
+.IR print (2).
+If this causes a
+.IR write
+to occur and there is an error,
+a negative value
+.RB ( Beof )
+is returned.
+Otherwise, the number of bytes output is returned.
+.I Bvprint
+does the same except it takes as argument a
+.B va_list
+parameter, so it can be called within a variadic function.
+.PP
+.I Bwrite
+outputs
+.I nbytes
+of data starting at
+.I addr
+to
+.IR bp .
+If this causes a
+.IR write
+to occur and there is an error,
+a negative value is returned.
+Otherwise, the number of bytes written is returned.
+.PP
+.I Bflush
+causes any buffered output associated with
+.I bp
+to be written.
+The return is as for
+.IR Bputc .
+.I Bflush
+is called on
+exit for every buffer still open
+for writing.
+.PP
+.I Bbuffered
+returns the number of bytes in the buffer.
+When reading, this is the number of bytes still available from the last
+read on the file; when writing, it is the number of bytes ready to be
+written.
+.PP
+This library uses
+.IR fmt (3)
+for diagnostic messages about internal errors,
+as well as for the implementation of
+.I Bprint
+and
+.IR Bvprint .
+It uses
+.IR utf (3)
+for the implementation of
+.I Bgetrune
+and
+.IR Bputrune .
+.SH SEE ALSO
+.IR atexit (3).
+.IR open (2),
+.IR print (3),
+.IR utf (7)
+.SH DIAGNOSTICS
+.I Bio
+routines that return integers yield
+.B Beof
+if
+.I bp
+is not the descriptor of an open file.
+.I Bopen
+returns zero if the file cannot be opened in the given mode.
+.SH HISTORY
+The
+.IR bio (3)
+library originally appeared in Plan 9.
+This is a port of the Plan 9 bio library.
+.SH BUGS
+.I Brdline
+returns an error on strings longer than the buffer associated
+with the file
+and also if the end-of-file is encountered
+before a delimiter.
+.I Blinelen
+will tell how many characters are available
+in these cases.
+In the case of a true end-of-file,
+.I Blinelen
+will return zero.
+At the cost of allocating a buffer,
+.I Brdstr
+sidesteps these issues.
+.PP
+The data returned by
+.I Brdline
+may be overwritten by calls to any other
+.I bio
+routine on the same
+.IR bp.
DIR diff --git a/src/libbio/bio.h b/src/libbio/bio.h
t@@ -0,0 +1,79 @@
+#ifndef _BIOH_
+#define _BIOH_ 1
+
+#include <sys/types.h> /* for off_t */
+#include <fcntl.h> /* for O_RDONLY, O_WRONLY */
+
+typedef struct Biobuf Biobuf;
+
+enum
+{
+ Bsize = 8*1024,
+ Bungetsize = 4, /* space for ungetc */
+ Bmagic = 0x314159,
+ Beof = -1,
+ Bbad = -2,
+
+ Binactive = 0, /* states */
+ Bractive,
+ Bwactive,
+ Bracteof,
+
+ Bend
+};
+
+struct Biobuf
+{
+ int icount; /* neg num of bytes at eob */
+ int ocount; /* num of bytes at bob */
+ int rdline; /* num of bytes after rdline */
+ int runesize; /* num of bytes of last getrune */
+ int state; /* r/w/inactive */
+ int fid; /* open file */
+ int flag; /* magic if malloc'ed */
+ off_t offset; /* offset of buffer in file */
+ int bsize; /* size of buffer */
+ unsigned char* bbuf; /* pointer to beginning of buffer */
+ unsigned char* ebuf; /* pointer to end of buffer */
+ unsigned char* gbuf; /* pointer to good data in buf */
+ unsigned char b[Bungetsize+Bsize];
+};
+
+#define BGETC(bp)\
+ ((bp)->icount?(bp)->bbuf[(bp)->bsize+(bp)->icount++]:Bgetc((bp)))
+#define BPUTC(bp,c)\
+ ((bp)->ocount?(bp)->bbuf[(bp)->bsize+(bp)->ocount++]=(c),0:Bputc((bp),(c)))
+#define BOFFSET(bp)\
+ (((bp)->state==Bractive)?\
+ (bp)->offset + (bp)->icount:\
+ (((bp)->state==Bwactive)?\
+ (bp)->offset + ((bp)->bsize + (bp)->ocount):\
+ -1))
+#define BLINELEN(bp)\
+ (bp)->rdline
+#define BFILDES(bp)\
+ (bp)->fid
+
+int Bbuffered(Biobuf*);
+int Bfildes(Biobuf*);
+int Bflush(Biobuf*);
+int Bgetc(Biobuf*);
+int Bgetd(Biobuf*, double*);
+int Binit(Biobuf*, int, int);
+int Binits(Biobuf*, int, int, unsigned char*, int);
+int Blinelen(Biobuf*);
+off_t Boffset(Biobuf*);
+Biobuf* Bopen(char*, int);
+int Bprint(Biobuf*, char*, ...);
+int Bputc(Biobuf*, int);
+void* Brdline(Biobuf*, int);
+long Bread(Biobuf*, void*, long);
+off_t Bseek(Biobuf*, off_t, int);
+int Bterm(Biobuf*);
+int Bungetc(Biobuf*);
+long Bwrite(Biobuf*, void*, long);
+char* Brdstr(Biobuf*, int, int);
+long Bgetrune(Biobuf*);
+int Bputrune(Biobuf*, long);
+
+#endif
DIR diff --git a/src/libbio/boffset.c b/src/libbio/boffset.c
t@@ -0,0 +1,25 @@
+#include "lib9.h"
+#include <bio.h>
+
+off_t
+Boffset(Biobuf *bp)
+{
+ off_t n;
+
+ switch(bp->state) {
+ default:
+ fprint(2, "Boffset: unknown state %d\n", bp->state);
+ n = Beof;
+ break;
+
+ case Bracteof:
+ case Bractive:
+ n = bp->offset + bp->icount;
+ break;
+
+ case Bwactive:
+ n = bp->offset + (bp->bsize + bp->ocount);
+ break;
+ }
+ return n;
+}
DIR diff --git a/src/libbio/bprint.c b/src/libbio/bprint.c
t@@ -0,0 +1,28 @@
+#include "lib9.h"
+#include <bio.h>
+
+int
+Bprint(Biobuf *bp, char *fmt, ...)
+{
+ va_list ap;
+ char *ip, *ep, *out;
+ int n;
+
+ ep = (char*)bp->ebuf;
+ ip = ep + bp->ocount;
+ va_start(ap, fmt);
+ out = vseprint(ip, ep, fmt, ap);
+ va_end(ap);
+ if(out == 0 || out >= ep-5) {
+ Bflush(bp);
+ ip = ep + bp->ocount;
+ va_start(ap, fmt);
+ out = vseprint(ip, ep, fmt, ap);
+ va_end(ap);
+ if(out >= ep-5)
+ return Beof;
+ }
+ n = out-ip;
+ bp->ocount += n;
+ return n;
+}
DIR diff --git a/src/libbio/bputc.c b/src/libbio/bputc.c
t@@ -0,0 +1,29 @@
+#include "lib9.h"
+#include <bio.h>
+
+int
+Bputc(Biobuf *bp, int c)
+{
+ int i, j;
+
+loop:
+ i = bp->ocount;
+ j = i+1;
+ if(i != 0) {
+ bp->ocount = j;
+ bp->ebuf[i] = c;
+ return 0;
+ }
+ if(bp->state != Bwactive)
+ return Beof;
+ j = write(bp->fid, bp->bbuf, bp->bsize);
+ if(j == bp->bsize) {
+ bp->ocount = -bp->bsize;
+ bp->offset += j;
+ goto loop;
+ }
+ fprint(2, "Bputc: write error\n");
+ bp->state = Binactive;
+ bp->ocount = 0;
+ return Beof;
+}
DIR diff --git a/src/libbio/bputrune.c b/src/libbio/bputrune.c
t@@ -0,0 +1,23 @@
+#include "lib9.h"
+#include <bio.h>
+#include <utf.h>
+
+int
+Bputrune(Biobuf *bp, long c)
+{
+ Rune rune;
+ char str[4];
+ int n;
+
+ rune = c;
+ if(rune < Runeself) {
+ Bputc(bp, rune);
+ return 1;
+ }
+ n = runetochar(str, &rune);
+ if(n == 0)
+ return Bbad;
+ if(Bwrite(bp, str, n) != n)
+ return Beof;
+ return n;
+}
DIR diff --git a/src/libbio/brdline.c b/src/libbio/brdline.c
t@@ -0,0 +1,94 @@
+#include "lib9.h"
+#include <bio.h>
+
+void*
+Brdline(Biobuf *bp, int delim)
+{
+ char *ip, *ep;
+ int i, j;
+
+ i = -bp->icount;
+ if(i == 0) {
+ /*
+ * eof or other error
+ */
+ if(bp->state != Bractive) {
+ if(bp->state == Bracteof)
+ bp->state = Bractive;
+ bp->rdline = 0;
+ bp->gbuf = bp->ebuf;
+ return 0;
+ }
+ }
+
+ /*
+ * first try in remainder of buffer (gbuf doesn't change)
+ */
+ ip = (char*)bp->ebuf - i;
+ ep = memchr(ip, delim, i);
+ if(ep) {
+ j = (ep - ip) + 1;
+ bp->rdline = j;
+ bp->icount += j;
+ return ip;
+ }
+
+ /*
+ * copy data to beginning of buffer
+ */
+ if(i < bp->bsize)
+ memmove(bp->bbuf, ip, i);
+ bp->gbuf = bp->bbuf;
+
+ /*
+ * append to buffer looking for the delim
+ */
+ ip = (char*)bp->bbuf + i;
+ while(i < bp->bsize) {
+ j = read(bp->fid, ip, bp->bsize-i);
+ if(j <= 0) {
+ /*
+ * end of file with no delim
+ */
+ memmove(bp->ebuf-i, bp->bbuf, i);
+ bp->rdline = i;
+ bp->icount = -i;
+ bp->gbuf = bp->ebuf-i;
+ return 0;
+ }
+ bp->offset += j;
+ i += j;
+ ep = memchr(ip, delim, j);
+ if(ep) {
+ /*
+ * found in new piece
+ * copy back up and reset everything
+ */
+ ip = (char*)bp->ebuf - i;
+ if(i < bp->bsize){
+ memmove(ip, bp->bbuf, i);
+ bp->gbuf = (unsigned char*)ip;
+ }
+ j = (ep - (char*)bp->bbuf) + 1;
+ bp->rdline = j;
+ bp->icount = j - i;
+ return ip;
+ }
+ ip += j;
+ }
+
+ /*
+ * full buffer without finding
+ */
+ bp->rdline = bp->bsize;
+ bp->icount = -bp->bsize;
+ bp->gbuf = bp->bbuf;
+ return 0;
+}
+
+int
+Blinelen(Biobuf *bp)
+{
+
+ return bp->rdline;
+}
DIR diff --git a/src/libbio/brdstr.c b/src/libbio/brdstr.c
t@@ -0,0 +1,112 @@
+#include "lib9.h"
+#include <bio.h>
+
+static char*
+badd(char *p, int *np, char *data, int ndata, int delim, int nulldelim)
+{
+ int n;
+
+ n = *np;
+ p = realloc(p, n+ndata+1);
+ if(p){
+ memmove(p+n, data, ndata);
+ n += ndata;
+ if(n>0 && nulldelim && p[n-1]==delim)
+ p[--n] = '\0';
+ else
+ p[n] = '\0';
+ *np = n;
+ }
+ return p;
+}
+
+char*
+Brdstr(Biobuf *bp, int delim, int nulldelim)
+{
+ char *ip, *ep, *p;
+ int i, j;
+
+ i = -bp->icount;
+ bp->rdline = 0;
+ if(i == 0) {
+ /*
+ * eof or other error
+ */
+ if(bp->state != Bractive) {
+ if(bp->state == Bracteof)
+ bp->state = Bractive;
+ bp->gbuf = bp->ebuf;
+ return nil;
+ }
+ }
+
+ /*
+ * first try in remainder of buffer (gbuf doesn't change)
+ */
+ ip = (char*)bp->ebuf - i;
+ ep = memchr(ip, delim, i);
+ if(ep) {
+ j = (ep - ip) + 1;
+ bp->icount += j;
+ return badd(nil, &bp->rdline, ip, j, delim, nulldelim);
+ }
+
+ /*
+ * copy data to beginning of buffer
+ */
+ if(i < bp->bsize)
+ memmove(bp->bbuf, ip, i);
+ bp->gbuf = bp->bbuf;
+
+ /*
+ * append to buffer looking for the delim
+ */
+ p = nil;
+ for(;;){
+ ip = (char*)bp->bbuf + i;
+ while(i < bp->bsize) {
+ j = read(bp->fid, ip, bp->bsize-i);
+ if(j <= 0 && i == 0)
+ return p;
+ if(j <= 0 && i > 0){
+ /*
+ * end of file but no delim. pretend we got a delim
+ * by making the delim \0 and smashing it with nulldelim.
+ */
+ j = 1;
+ ep = ip;
+ delim = '\0';
+ nulldelim = 1;
+ *ep = delim; /* there will be room for this */
+ }else{
+ bp->offset += j;
+ ep = memchr(ip, delim, j);
+ }
+ i += j;
+ if(ep) {
+ /*
+ * found in new piece
+ * copy back up and reset everything
+ */
+ ip = (char*)bp->ebuf - i;
+ if(i < bp->bsize){
+ memmove(ip, bp->bbuf, i);
+ bp->gbuf = (unsigned char*)ip;
+ }
+ j = (ep - (char*)bp->bbuf) + 1;
+ bp->icount = j - i;
+ return badd(p, &bp->rdline, ip, j, delim, nulldelim);
+ }
+ ip += j;
+ }
+
+ /*
+ * full buffer without finding; add to user string and continue
+ */
+ p = badd(p, &bp->rdline, (char*)bp->bbuf, bp->bsize, 0, 0);
+ i = 0;
+ bp->icount = 0;
+ bp->gbuf = bp->ebuf;
+ }
+ return 0; /* never happens */
+}
DIR diff --git a/src/libbio/bread.c b/src/libbio/bread.c
t@@ -0,0 +1,45 @@
+#include "lib9.h"
+#include <bio.h>
+
+long
+Bread(Biobuf *bp, void *ap, long count)
+{
+ long c;
+ unsigned char *p;
+ int i, n, ic;
+
+ p = ap;
+ c = count;
+ ic = bp->icount;
+
+ while(c > 0) {
+ n = -ic;
+ if(n > c)
+ n = c;
+ if(n == 0) {
+ if(bp->state != Bractive)
+ break;
+ i = read(bp->fid, bp->bbuf, bp->bsize);
+ if(i <= 0) {
+ bp->state = Bracteof;
+ if(i < 0)
+ bp->state = Binactive;
+ break;
+ }
+ bp->gbuf = bp->bbuf;
+ bp->offset += i;
+ if(i < bp->bsize) {
+ memmove(bp->ebuf-i, bp->bbuf, i);
+ bp->gbuf = bp->ebuf-i;
+ }
+ ic = -i;
+ continue;
+ }
+ memmove(p, bp->ebuf+ic, n);
+ c -= n;
+ ic += n;
+ p += n;
+ }
+ bp->icount = ic;
+ return count-c;
+}
DIR diff --git a/src/libbio/bseek.c b/src/libbio/bseek.c
t@@ -0,0 +1,56 @@
+#include "lib9.h"
+#include <bio.h>
+
+off_t
+Bseek(Biobuf *bp, off_t offset, int base)
+{
+ off_t n, d;
+
+ switch(bp->state) {
+ default:
+ fprint(2, "Bseek: unknown state %d\n", bp->state);
+ return Beof;
+
+ case Bracteof:
+ bp->state = Bractive;
+ bp->icount = 0;
+ bp->gbuf = bp->ebuf;
+
+ case Bractive:
+ n = offset;
+ if(base == 1) {
+ n += Boffset(bp);
+ base = 0;
+ }
+
+ /*
+ * try to seek within buffer
+ */
+ if(base == 0) {
+ d = n - Boffset(bp);
+ bp->icount += d;
+ if(d >= 0) {
+ if(bp->icount <= 0)
+ return n;
+ } else {
+ if(bp->ebuf - bp->gbuf >= -bp->icount)
+ return n;
+ }
+ }
+
+ /*
+ * reset the buffer
+ */
+ n = lseek(bp->fid, n, base);
+ bp->icount = 0;
+ bp->gbuf = bp->ebuf;
+ break;
+
+ case Bwactive:
+ Bflush(bp);
+ n = lseek(bp->fid, offset, base);
+ break;
+ }
+ bp->offset = n;
+ return n;
+}
DIR diff --git a/src/libbio/bundle.ports b/src/libbio/bundle.ports
t@@ -0,0 +1,45 @@
+--- Makefile ---
+# New ports collection makefile for: libbio
+# Date Created: 11 Feb 2003
+# Whom: rsc
+#
+# THIS LINE NEEDS REPLACING. IT'S HERE TO GET BY PORTLINT
+# $FreeBSD: ports/devel/libbio/Makefile,v 1.1 2003/02/12 00:51:22 rsc Exp $
+
+PORTNAME= libbio
+PORTVERSION= 2.0
+CATEGORIES= devel
+MASTER_SITES= http://pdos.lcs.mit.edu/~rsc/software/
+EXTRACT_SUFX= .tgz
+
+MAINTAINER= rsc@post.harvard.edu
+
+DEPENDS= ${PORTSDIR}/devel/libfmt ${PORTSDIR}/devel/libutf
+
+MAN3= bio.3
+USE_REINPLACE= yes
+
+.include <bsd.port.pre.mk>
+
+post-patch:
+ ${REINPLACE_CMD} -e 's,$$(PREFIX),${PREFIX},g' ${WRKSRC}/Makefile
+
+.include <bsd.port.post.mk>
+--- pkg-comment ---
+Simple buffered I/O library from Plan 9
+--- pkg-descr ---
+Libbio is a port of Plan 9's formatted I/O library.
+It provides most of the same functionality as stdio or sfio,
+but with a simpler interface and smaller footprint.
+
+WWW: http://pdos.lcs.mit.edu/~rsc/software/#libbio
+http://plan9.bell-labs.com/magic/man2html/2/bio
+
+Russ Cox
+rsc@post.harvard.edu
+--- pkg-plist ---
+lib/libbio.a
+include/bio.h
+--- /dev/null ---
+This is just a way to make sure blank lines don't
+creep into pkg-plist.
DIR diff --git a/src/libbio/bwrite.c b/src/libbio/bwrite.c
t@@ -0,0 +1,38 @@
+#include "lib9.h"
+#include <bio.h>
+
+long
+Bwrite(Biobuf *bp, void *ap, long count)
+{
+ long c;
+ unsigned char *p;
+ int i, n, oc;
+
+ p = ap;
+ c = count;
+ oc = bp->ocount;
+
+ while(c > 0) {
+ n = -oc;
+ if(n > c)
+ n = c;
+ if(n == 0) {
+ if(bp->state != Bwactive)
+ return Beof;
+ i = write(bp->fid, bp->bbuf, bp->bsize);
+ if(i != bp->bsize) {
+ bp->state = Binactive;
+ return Beof;
+ }
+ bp->offset += i;
+ oc = -bp->bsize;
+ continue;
+ }
+ memmove(bp->ebuf+oc, p, n);
+ oc += n;
+ c -= n;
+ p += n;
+ }
+ bp->ocount = oc;
+ return count-c;
+}
DIR diff --git a/src/libbio/lib9.h b/src/libbio/lib9.h
t@@ -0,0 +1,12 @@
+#include <fmt.h>
+#include <fcntl.h>
+#include <string.h>
+#include <unistd.h>
+#include <stdlib.h>
+
+#define OREAD O_RDONLY
+#define OWRITE O_WRONLY
+
+#include <utf.h>
+
+#define nil ((void*)0)
DIR diff --git a/src/libbio/mkfile b/src/libbio/mkfile
t@@ -0,0 +1 @@
+<../libutf/mkfile
DIR diff --git a/src/libbio/rpm.spec b/src/libbio/rpm.spec
t@@ -0,0 +1,30 @@
+Summary: Simple buffered I/O library from Plan 9
+Name: libbio
+Version: 2.0
+Release: 1
+Group: Development/C
+Copyright: LGPL
+Packager: Russ Cox <rsc@post.harvard.edu>
+Source: http://pdos.lcs.mit.edu/~rsc/software/libbio-2.0.tgz
+URL: http://pdos.lcs.mit.edu/~rsc/software/#libbio
+Requires: libfmt libutf
+
+%description
+Libbio is a port of Plan 9's formatted I/O library.
+It provides most of the same functionality as stdio or sfio,
+but with a simpler interface and smaller footprint.
+
+http://plan9.bell-labs.com/magic/man2html/2/bio
+%prep
+%setup
+
+%build
+make
+
+%install
+make install
+
+%files
+/usr/local/include/bio.h
+/usr/local/lib/libbio.a
+/usr/local/man/man3/bio.3
DIR diff --git a/src/libfmt/LICENSE b/src/libfmt/LICENSE
t@@ -0,0 +1,19 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+*/
+
+This is a Unix port of the Plan 9 formatted I/O package.
+
+Please send comments about the packaging
+to Russ Cox <rsc@post.harvard.edu>.
+
DIR diff --git a/src/libfmt/Make.Darwin-PowerMacintosh b/src/libfmt/Make.Darwin-PowerMacintosh
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I${PREFIX}/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.FreeBSD-386 b/src/libfmt/Make.FreeBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include -pg
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.HP-UX-9000 b/src/libfmt/Make.HP-UX-9000
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS=-O -c -Ae -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.Linux-386 b/src/libfmt/Make.Linux-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -DNEEDLL
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.NetBSD-386 b/src/libfmt/Make.NetBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.OSF1-alpha b/src/libfmt/Make.OSF1-alpha
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.SunOS-sun4u b/src/libfmt/Make.SunOS-sun4u
t@@ -0,0 +1,2 @@
+include Make.SunOS-sun4u-$(CC)
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.SunOS-sun4u-cc b/src/libfmt/Make.SunOS-sun4u-cc
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I. -O
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Make.SunOS-sun4u-gcc b/src/libfmt/Make.SunOS-sun4u-gcc
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libfmt/Makefile b/src/libfmt/Makefile
t@@ -0,0 +1,134 @@
+
+# this works in gnu make
+SYSNAME:=${shell uname}
+OBJTYPE:=${shell uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'}
+
+# this works in bsd make
+SYSNAME!=uname
+OBJTYPE!=uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'
+
+# the gnu rules will mess up bsd but not vice versa,
+# hence the gnu rules come first.
+
+include Make.$(SYSNAME)-$(OBJTYPE)
+
+PREFIX=/usr/local
+
+NUKEFILES=
+
+TGZFILES=
+
+LIB=libfmt.a
+VERSION=2.0
+PORTPLACE=devel/libfmt
+NAME=libfmt
+
+NUM=\
+ charstod.$O\
+ pow10.$O\
+
+OFILES=\
+ dofmt.$O\
+ errfmt.$O\
+ fltfmt.$O\
+ fmt.$O\
+ fmtfd.$O\
+ fmtfdflush.$O\
+ fmtlock.$O\
+ fmtprint.$O\
+ fmtquote.$O\
+ fmtrune.$O\
+ fmtstr.$O\
+ fmtvprint.$O\
+ fprint.$O\
+ print.$O\
+ runefmtstr.$O\
+ runeseprint.$O\
+ runesmprint.$O\
+ runesnprint.$O\
+ runesprint.$O\
+ runevseprint.$O\
+ runevsmprint.$O\
+ runevsnprint.$O\
+ seprint.$O\
+ smprint.$O\
+ snprint.$O\
+ sprint.$O\
+ strtod.$O\
+ vfprint.$O\
+ vseprint.$O\
+ vsmprint.$O\
+ vsnprint.$O\
+ $(NUM)\
+ $(NAN)\
+
+HFILES=\
+ fmtdef.h\
+ fmt.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ install -m 0644 print.3 $(PREFIX)/man/man3/print.3
+ install -m 0644 fmtinstall.3 $(PREFIX)/man/man3/fmtinstall.3
+ install -m 0644 fmt.h $(PREFIX)/include/fmt.h
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+$(NAN).$O: nan.h
+strtod.$O: nan.h
+
+test: $(LIB) test.$O
+ $(CC) -o test test.$O $(LIB) -L$(PREFIX)/lib -lutf
+
+$(LIB): $(OFILES)
+ $(AR) $(ARFLAGS) $(LIB) $(OFILES)
+
+NUKEFILES+=$(LIB)
+.c.$O:
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+%.$O: %.c
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+
+$(OFILES): $(HFILES)
+
+tgz:
+ rm -rf $(NAME)-$(VERSION)
+ mkdir $(NAME)-$(VERSION)
+ cp Makefile Make.* README LICENSE NOTICE *.[ch137] rpm.spec bundle.ports $(TGZFILES) $(NAME)-$(VERSION)
+ tar cf - $(NAME)-$(VERSION) | gzip >$(NAME)-$(VERSION).tgz
+ rm -rf $(NAME)-$(VERSION)
+
+clean:
+ rm -f $(OFILES) $(LIB)
+
+nuke:
+ rm -f $(OFILES) *.tgz *.rpm $(NUKEFILES)
+
+rpm:
+ make tgz
+ cp $(NAME)-$(VERSION).tgz /usr/src/RPM/SOURCES
+ rpm -ba rpm.spec
+ cp /usr/src/RPM/SRPMS/$(NAME)-$(VERSION)-1.src.rpm .
+ cp /usr/src/RPM/RPMS/i586/$(NAME)-$(VERSION)-1.i586.rpm .
+ scp *.rpm rsc@amsterdam.lcs.mit.edu:public_html/software
+
+PORTDIR=/usr/ports/$(PORTPLACE)
+
+ports:
+ make tgz
+ rm -rf $(PORTDIR)
+ mkdir $(PORTDIR)
+ cp $(NAME)-$(VERSION).tgz /usr/ports/distfiles
+ cat bundle.ports | (cd $(PORTDIR) && awk '$$1=="---" && $$3=="---" { ofile=$$2; next} {if(ofile) print >ofile}')
+ (cd $(PORTDIR); make makesum)
+ (cd $(PORTDIR); make)
+ (cd $(PORTDIR); /usr/local/bin/portlint)
+ rm -rf $(PORTDIR)/work
+ shar `find $(PORTDIR)` > ports.shar
+ (cd $(PORTDIR); tar cf - *) | gzip >$(NAME)-$(VERSION)-ports.tgz
+ scp *.tgz rsc@amsterdam.lcs.mit.edu:public_html/software
+
+.phony: all clean nuke install tgz rpm ports
DIR diff --git a/src/libfmt/Makefile.MID b/src/libfmt/Makefile.MID
t@@ -0,0 +1,63 @@
+LIB=libfmt.a
+VERSION=2.0
+PORTPLACE=devel/libfmt
+NAME=libfmt
+
+NUM=\
+ charstod.$O\
+ pow10.$O\
+
+OFILES=\
+ dofmt.$O\
+ errfmt.$O\
+ fltfmt.$O\
+ fmt.$O\
+ fmtfd.$O\
+ fmtfdflush.$O\
+ fmtlock.$O\
+ fmtprint.$O\
+ fmtquote.$O\
+ fmtrune.$O\
+ fmtstr.$O\
+ fmtvprint.$O\
+ fprint.$O\
+ print.$O\
+ runefmtstr.$O\
+ runeseprint.$O\
+ runesmprint.$O\
+ runesnprint.$O\
+ runesprint.$O\
+ runevseprint.$O\
+ runevsmprint.$O\
+ runevsnprint.$O\
+ seprint.$O\
+ smprint.$O\
+ snprint.$O\
+ sprint.$O\
+ strtod.$O\
+ vfprint.$O\
+ vseprint.$O\
+ vsmprint.$O\
+ vsnprint.$O\
+ $(NUM)\
+ $(NAN)\
+
+HFILES=\
+ fmtdef.h\
+ fmt.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ install -m 0644 print.3 $(PREFIX)/man/man3/print.3
+ install -m 0644 fmtinstall.3 $(PREFIX)/man/man3/fmtinstall.3
+ install -m 0644 fmt.h $(PREFIX)/include/fmt.h
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+$(NAN).$O: nan.h
+strtod.$O: nan.h
+
+test: $(LIB) test.$O
+ $(CC) -o test test.$O $(LIB) -L$(PREFIX)/lib -lutf
+
DIR diff --git a/src/libfmt/NOTICE b/src/libfmt/NOTICE
t@@ -0,0 +1,19 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+*/
+
+This is a Unix port of the Plan 9 formatted I/O package.
+
+Please send comments about the packaging
+to Russ Cox <rsc@post.harvard.edu>.
+
DIR diff --git a/src/libfmt/README b/src/libfmt/README
t@@ -0,0 +1,19 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+*/
+
+This is a Unix port of the Plan 9 formatted I/O package.
+
+Please send comments about the packaging
+to Russ Cox <rsc@post.harvard.edu>.
+
DIR diff --git a/src/libfmt/bundle.ports b/src/libfmt/bundle.ports
t@@ -0,0 +1,51 @@
+--- Makefile ---
+# New ports collection makefile for: libfmt
+# Date Created: 11 Feb 2003
+# Whom: rsc
+#
+# THIS LINE NEEDS REPLACING. IT'S HERE TO GET BY PORTLINT
+# $FreeBSD: ports/devel/libfmt/Makefile,v 1.1 2003/02/12 00:51:22 rsc Exp $
+
+PORTNAME= libfmt
+PORTVERSION= 2.0
+CATEGORIES= devel
+MASTER_SITES= http://pdos.lcs.mit.edu/~rsc/software/
+EXTRACT_SUFX= .tgz
+
+MAINTAINER= rsc@post.harvard.edu
+
+DEPENDS= ${PORTSDIR}/devel/libutf
+
+MAN3= print.3 fmtinstall.3
+
+USE_REINPLACE=yes
+
+.include <bsd.port.pre.mk>
+
+post-patch:
+ ${REINPLACE_CMD} -e 's,$$(PREFIX),${PREFIX},g' ${WRKSRC}/Makefile
+
+.include <bsd.port.post.mk>
+--- pkg-comment ---
+Extensible formatted print C library (printf with user-defined verbs)
+--- pkg-descr ---
+Libfmt is a port of Plan 9's formatted print library.
+As a base it provides all the syntax of ANSI printf
+but adds the ability for client programs to install
+new print verbs. One such print verb (installed by
+default) is %r, which prints the system error string.
+Instead of perror("foo"), you can write fprint(2, "foo: %r\n").
+This is especially nice when you write verbs to format
+the data structures used by your particular program.
+
+WWW: http://pdos.lcs.mit.edu/~rsc/software/#libfmt
+http://plan9.bell-labs.com/magic/man2html/2/print
+
+Russ Cox
+rsc@post.harvard.edu
+--- pkg-plist ---
+lib/libfmt.a
+include/fmt.h
+--- /dev/null ---
+This is just a way to make sure blank lines don't
+creep into pkg-plist.
DIR diff --git a/src/libfmt/charstod.c b/src/libfmt/charstod.c
t@@ -0,0 +1,85 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * Reads a floating-point number by interpreting successive characters
+ * returned by (*f)(vp). The last call it makes to f terminates the
+ * scan, so is not a character in the number. It may therefore be
+ * necessary to back up the input stream up one byte after calling charstod.
+ */
+
+double
+fmtcharstod(int(*f)(void*), void *vp)
+{
+ double num, dem;
+ int neg, eneg, dig, exp, c;
+
+ num = 0;
+ neg = 0;
+ dig = 0;
+ exp = 0;
+ eneg = 0;
+
+ c = (*f)(vp);
+ while(c == ' ' || c == '\t')
+ c = (*f)(vp);
+ if(c == '-' || c == '+'){
+ if(c == '-')
+ neg = 1;
+ c = (*f)(vp);
+ }
+ while(c >= '0' && c <= '9'){
+ num = num*10 + c-'0';
+ c = (*f)(vp);
+ }
+ if(c == '.')
+ c = (*f)(vp);
+ while(c >= '0' && c <= '9'){
+ num = num*10 + c-'0';
+ dig++;
+ c = (*f)(vp);
+ }
+ if(c == 'e' || c == 'E'){
+ c = (*f)(vp);
+ if(c == '-' || c == '+'){
+ if(c == '-'){
+ dig = -dig;
+ eneg = 1;
+ }
+ c = (*f)(vp);
+ }
+ while(c >= '0' && c <= '9'){
+ exp = exp*10 + c-'0';
+ c = (*f)(vp);
+ }
+ }
+ exp -= dig;
+ if(exp < 0){
+ exp = -exp;
+ eneg = !eneg;
+ }
+ dem = __fmtpow10(exp);
+ if(eneg)
+ num /= dem;
+ else
+ num *= dem;
+ if(neg)
+ return -num;
+ return num;
+}
DIR diff --git a/src/libfmt/dofmt.c b/src/libfmt/dofmt.c
t@@ -0,0 +1,558 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/* format the output into f->to and return the number of characters fmted */
+int
+dofmt(Fmt *f, char *fmt)
+{
+ Rune rune, *rt, *rs;
+ int r;
+ char *t, *s;
+ int n, nfmt;
+
+ nfmt = f->nfmt;
+ for(;;){
+ if(f->runes){
+ rt = (Rune*)f->to;
+ rs = (Rune*)f->stop;
+ while((r = *(uchar*)fmt) && r != '%'){
+ if(r < Runeself)
+ fmt++;
+ else{
+ fmt += chartorune(&rune, fmt);
+ r = rune;
+ }
+ FMTRCHAR(f, rt, rs, r);
+ }
+ fmt++;
+ f->nfmt += rt - (Rune *)f->to;
+ f->to = rt;
+ if(!r)
+ return f->nfmt - nfmt;
+ f->stop = rs;
+ }else{
+ t = (char*)f->to;
+ s = (char*)f->stop;
+ while((r = *(uchar*)fmt) && r != '%'){
+ if(r < Runeself){
+ FMTCHAR(f, t, s, r);
+ fmt++;
+ }else{
+ n = chartorune(&rune, fmt);
+ if(t + n > s){
+ t = (char*)__fmtflush(f, t, n);
+ if(t != nil)
+ s = (char*)f->stop;
+ else
+ return -1;
+ }
+ while(n--)
+ *t++ = *fmt++;
+ }
+ }
+ fmt++;
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ if(!r)
+ return f->nfmt - nfmt;
+ f->stop = s;
+ }
+
+ fmt = (char*)__fmtdispatch(f, fmt, 0);
+ if(fmt == nil)
+ return -1;
+ }
+ return 0; /* not reached */
+}
+
+void *
+__fmtflush(Fmt *f, void *t, int len)
+{
+ if(f->runes)
+ f->nfmt += (Rune*)t - (Rune*)f->to;
+ else
+ f->nfmt += (char*)t - (char *)f->to;
+ f->to = t;
+ if(f->flush == 0 || (*f->flush)(f) == 0 || (char*)f->to + len > (char*)f->stop){
+ f->stop = f->to;
+ return nil;
+ }
+ return f->to;
+}
+
+/*
+ * put a formatted block of memory sz bytes long of n runes into the output buffer,
+ * left/right justified in a field of at least f->width charactes
+ */
+int
+__fmtpad(Fmt *f, int n)
+{
+ char *t, *s;
+ int i;
+
+ t = (char*)f->to;
+ s = (char*)f->stop;
+ for(i = 0; i < n; i++)
+ FMTCHAR(f, t, s, ' ');
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ return 0;
+}
+
+int
+__rfmtpad(Fmt *f, int n)
+{
+ Rune *t, *s;
+ int i;
+
+ t = (Rune*)f->to;
+ s = (Rune*)f->stop;
+ for(i = 0; i < n; i++)
+ FMTRCHAR(f, t, s, ' ');
+ f->nfmt += t - (Rune *)f->to;
+ f->to = t;
+ return 0;
+}
+
+int
+__fmtcpy(Fmt *f, const void *vm, int n, int sz)
+{
+ Rune *rt, *rs, r;
+ char *t, *s, *m, *me;
+ ulong fl;
+ int nc, w;
+
+ m = (char*)vm;
+ me = m + sz;
+ w = f->width;
+ fl = f->flags;
+ if((fl & FmtPrec) && n > f->prec)
+ n = f->prec;
+ if(f->runes){
+ if(!(fl & FmtLeft) && __rfmtpad(f, w - n) < 0)
+ return -1;
+ rt = (Rune*)f->to;
+ rs = (Rune*)f->stop;
+ for(nc = n; nc > 0; nc--){
+ r = *(uchar*)m;
+ if(r < Runeself)
+ m++;
+ else if((me - m) >= UTFmax || fullrune(m, me-m))
+ m += chartorune(&r, m);
+ else
+ break;
+ FMTRCHAR(f, rt, rs, r);
+ }
+ f->nfmt += rt - (Rune *)f->to;
+ f->to = rt;
+ if(m < me)
+ return -1;
+ if(fl & FmtLeft && __rfmtpad(f, w - n) < 0)
+ return -1;
+ }else{
+ if(!(fl & FmtLeft) && __fmtpad(f, w - n) < 0)
+ return -1;
+ t = (char*)f->to;
+ s = (char*)f->stop;
+ for(nc = n; nc > 0; nc--){
+ r = *(uchar*)m;
+ if(r < Runeself)
+ m++;
+ else if((me - m) >= UTFmax || fullrune(m, me-m))
+ m += chartorune(&r, m);
+ else
+ break;
+ FMTRUNE(f, t, s, r);
+ }
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ if(fl & FmtLeft && __fmtpad(f, w - n) < 0)
+ return -1;
+ }
+ return 0;
+}
+
+int
+__fmtrcpy(Fmt *f, const void *vm, int n)
+{
+ Rune r, *m, *me, *rt, *rs;
+ char *t, *s;
+ ulong fl;
+ int w;
+
+ m = (Rune*)vm;
+ w = f->width;
+ fl = f->flags;
+ if((fl & FmtPrec) && n > f->prec)
+ n = f->prec;
+ if(f->runes){
+ if(!(fl & FmtLeft) && __rfmtpad(f, w - n) < 0)
+ return -1;
+ rt = (Rune*)f->to;
+ rs = (Rune*)f->stop;
+ for(me = m + n; m < me; m++)
+ FMTRCHAR(f, rt, rs, *m);
+ f->nfmt += rt - (Rune *)f->to;
+ f->to = rt;
+ if(fl & FmtLeft && __rfmtpad(f, w - n) < 0)
+ return -1;
+ }else{
+ if(!(fl & FmtLeft) && __fmtpad(f, w - n) < 0)
+ return -1;
+ t = (char*)f->to;
+ s = (char*)f->stop;
+ for(me = m + n; m < me; m++){
+ r = *m;
+ FMTRUNE(f, t, s, r);
+ }
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ if(fl & FmtLeft && __fmtpad(f, w - n) < 0)
+ return -1;
+ }
+ return 0;
+}
+
+/* fmt out one character */
+int
+__charfmt(Fmt *f)
+{
+ char x[1];
+
+ x[0] = va_arg(f->args, int);
+ f->prec = 1;
+ return __fmtcpy(f, (const char*)x, 1, 1);
+}
+
+/* fmt out one rune */
+int
+__runefmt(Fmt *f)
+{
+ Rune x[1];
+
+ x[0] = va_arg(f->args, int);
+ return __fmtrcpy(f, (const void*)x, 1);
+}
+
+/* public helper routine: fmt out a null terminated string already in hand */
+int
+fmtstrcpy(Fmt *f, char *s)
+{
+ int p, i;
+ if(!s)
+ return __fmtcpy(f, "<nil>", 5, 5);
+ /* if precision is specified, make sure we don't wander off the end */
+ if(f->flags & FmtPrec){
+ p = f->prec;
+ for(i = 0; i < p; i++)
+ if(s[i] == 0)
+ break;
+ return __fmtcpy(f, s, utfnlen(s, i), i); /* BUG?: won't print a partial rune at end */
+ }
+
+ return __fmtcpy(f, s, utflen(s), strlen(s));
+}
+
+/* fmt out a null terminated utf string */
+int
+__strfmt(Fmt *f)
+{
+ char *s;
+
+ s = va_arg(f->args, char *);
+ return fmtstrcpy(f, s);
+}
+
+/* public helper routine: fmt out a null terminated rune string already in hand */
+int
+fmtrunestrcpy(Fmt *f, Rune *s)
+{
+ Rune *e;
+ int n, p;
+
+ if(!s)
+ return __fmtcpy(f, "<nil>", 5, 5);
+ /* if precision is specified, make sure we don't wander off the end */
+ if(f->flags & FmtPrec){
+ p = f->prec;
+ for(n = 0; n < p; n++)
+ if(s[n] == 0)
+ break;
+ }else{
+ for(e = s; *e; e++)
+ ;
+ n = e - s;
+ }
+ return __fmtrcpy(f, s, n);
+}
+
+/* fmt out a null terminated rune string */
+int
+__runesfmt(Fmt *f)
+{
+ Rune *s;
+
+ s = va_arg(f->args, Rune *);
+ return fmtrunestrcpy(f, s);
+}
+
+/* fmt a % */
+int
+__percentfmt(Fmt *f)
+{
+ Rune x[1];
+
+ x[0] = f->r;
+ f->prec = 1;
+ return __fmtrcpy(f, (const void*)x, 1);
+}
+
+/* fmt an integer */
+int
+__ifmt(Fmt *f)
+{
+ char buf[70], *p, *conv;
+ uvlong vu;
+ ulong u;
+ int neg, base, i, n, fl, w, isv;
+
+ neg = 0;
+ fl = f->flags;
+ isv = 0;
+ vu = 0;
+ u = 0;
+ /*
+ * Unsigned verbs
+ */
+ switch(f->r){
+ case 'o':
+ case 'u':
+ case 'x':
+ case 'X':
+ fl |= FmtUnsigned;
+ break;
+ }
+ if(f->r == 'p'){
+ u = (ulong)va_arg(f->args, void*);
+ f->r = 'x';
+ fl |= FmtUnsigned;
+ }else if(fl & FmtVLong){
+ isv = 1;
+ if(fl & FmtUnsigned)
+ vu = va_arg(f->args, uvlong);
+ else
+ vu = va_arg(f->args, vlong);
+ }else if(fl & FmtLong){
+ if(fl & FmtUnsigned)
+ u = va_arg(f->args, ulong);
+ else
+ u = va_arg(f->args, long);
+ }else if(fl & FmtByte){
+ if(fl & FmtUnsigned)
+ u = (uchar)va_arg(f->args, int);
+ else
+ u = (char)va_arg(f->args, int);
+ }else if(fl & FmtShort){
+ if(fl & FmtUnsigned)
+ u = (ushort)va_arg(f->args, int);
+ else
+ u = (short)va_arg(f->args, int);
+ }else{
+ if(fl & FmtUnsigned)
+ u = va_arg(f->args, uint);
+ else
+ u = va_arg(f->args, int);
+ }
+ conv = "0123456789abcdef";
+ switch(f->r){
+ case 'd':
+ case 'i':
+ base = 10;
+ break;
+ case 'u':
+ base = 10;
+ break;
+ case 'x':
+ base = 16;
+ break;
+ case 'X':
+ base = 16;
+ conv = "0123456789ABCDEF";
+ break;
+ case 'b':
+ base = 2;
+ break;
+ case 'o':
+ base = 8;
+ break;
+ default:
+ return -1;
+ }
+ if(!(fl & FmtUnsigned)){
+ if(isv && (vlong)vu < 0){
+ vu = -(vlong)vu;
+ neg = 1;
+ }else if(!isv && (long)u < 0){
+ u = -(long)u;
+ neg = 1;
+ }
+ }else{
+ fl &= ~(FmtSign|FmtSpace); /* no + for unsigned conversions */
+ }
+ p = buf + sizeof buf - 1;
+ n = 0;
+ if(isv){
+ while(vu){
+ i = vu % base;
+ vu /= base;
+ if((fl & FmtComma) && n % 4 == 3){
+ *p-- = ',';
+ n++;
+ }
+ *p-- = conv[i];
+ n++;
+ }
+ }else{
+ while(u){
+ i = u % base;
+ u /= base;
+ if((fl & FmtComma) && n % 4 == 3){
+ *p-- = ',';
+ n++;
+ }
+ *p-- = conv[i];
+ n++;
+ }
+ }
+ if(n == 0){
+ if(!(fl & FmtPrec) || f->prec != 0){
+ *p-- = '0';
+ n = 1;
+ }
+ fl &= ~FmtSharp;
+ }
+ for(w = f->prec; n < w && p > buf+3; n++)
+ *p-- = '0';
+ if(neg || (fl & (FmtSign|FmtSpace)))
+ n++;
+ if(fl & FmtSharp){
+ if(base == 16)
+ n += 2;
+ else if(base == 8){
+ if(p[1] == '0')
+ fl &= ~FmtSharp;
+ else
+ n++;
+ }
+ }
+ if((fl & FmtZero) && !(fl & (FmtLeft|FmtPrec))){
+ for(w = f->width; n < w && p > buf+3; n++)
+ *p-- = '0';
+ f->width = 0;
+ }
+ if(fl & FmtSharp){
+ if(base == 16)
+ *p-- = f->r;
+ if(base == 16 || base == 8)
+ *p-- = '0';
+ }
+ if(neg)
+ *p-- = '-';
+ else if(fl & FmtSign)
+ *p-- = '+';
+ else if(fl & FmtSpace)
+ *p-- = ' ';
+ f->flags &= ~FmtPrec;
+ return __fmtcpy(f, p + 1, n, n);
+}
+
+int
+__countfmt(Fmt *f)
+{
+ void *p;
+ ulong fl;
+
+ fl = f->flags;
+ p = va_arg(f->args, void*);
+ if(fl & FmtVLong){
+ *(vlong*)p = f->nfmt;
+ }else if(fl & FmtLong){
+ *(long*)p = f->nfmt;
+ }else if(fl & FmtByte){
+ *(char*)p = f->nfmt;
+ }else if(fl & FmtShort){
+ *(short*)p = f->nfmt;
+ }else{
+ *(int*)p = f->nfmt;
+ }
+ return 0;
+}
+
+int
+__flagfmt(Fmt *f)
+{
+ switch(f->r){
+ case ',':
+ f->flags |= FmtComma;
+ break;
+ case '-':
+ f->flags |= FmtLeft;
+ break;
+ case '+':
+ f->flags |= FmtSign;
+ break;
+ case '#':
+ f->flags |= FmtSharp;
+ break;
+ case ' ':
+ f->flags |= FmtSpace;
+ break;
+ case 'u':
+ f->flags |= FmtUnsigned;
+ break;
+ case 'h':
+ if(f->flags & FmtShort)
+ f->flags |= FmtByte;
+ f->flags |= FmtShort;
+ break;
+ case 'L':
+ f->flags |= FmtLDouble;
+ break;
+ case 'l':
+ if(f->flags & FmtLong)
+ f->flags |= FmtVLong;
+ f->flags |= FmtLong;
+ break;
+ }
+ return 1;
+}
+
+/* default error format */
+int
+__badfmt(Fmt *f)
+{
+ char x[3];
+
+ x[0] = '%';
+ x[1] = f->r;
+ x[2] = '%';
+ f->prec = 3;
+ __fmtcpy(f, (const void*)x, 3, 3);
+ return 0;
+}
DIR diff --git a/src/libfmt/dorfmt.c b/src/libfmt/dorfmt.c
t@@ -0,0 +1,61 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/* format the output into f->to and return the number of characters fmted */
+
+int
+dorfmt(Fmt *f, const Rune *fmt)
+{
+ Rune *rt, *rs;
+ int r;
+ char *t, *s;
+ int nfmt;
+
+ nfmt = f->nfmt;
+ for(;;){
+ if(f->runes){
+ rt = f->to;
+ rs = f->stop;
+ while((r = *fmt++) && r != '%'){
+ FMTRCHAR(f, rt, rs, r);
+ }
+ f->nfmt += rt - (Rune *)f->to;
+ f->to = rt;
+ if(!r)
+ return f->nfmt - nfmt;
+ f->stop = rs;
+ }else{
+ t = f->to;
+ s = f->stop;
+ while((r = *fmt++) && r != '%'){
+ FMTRUNE(f, t, f->stop, r);
+ }
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ if(!r)
+ return f->nfmt - nfmt;
+ f->stop = s;
+ }
+
+ fmt = __fmtdispatch(f, fmt, 1);
+ if(fmt == nil)
+ return -1;
+ }
+ return 0; /* not reached */
+}
DIR diff --git a/src/libfmt/errfmt.c b/src/libfmt/errfmt.c
t@@ -0,0 +1,28 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <errno.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+__errfmt(Fmt *f)
+{
+ char *s;
+
+ s = strerror(errno);
+ return fmtstrcpy(f, s);
+}
DIR diff --git a/src/libfmt/fltfmt.c b/src/libfmt/fltfmt.c
t@@ -0,0 +1,610 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdio.h>
+#include <math.h>
+#include <float.h>
+#include <string.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <stdarg.h>
+#include "fmt.h"
+#include "fmtdef.h"
+#include "nan.h"
+
+enum
+{
+ FDEFLT = 6,
+ NSIGNIF = 17
+};
+
+/*
+ * first few powers of 10, enough for about 1/2 of the
+ * total space for doubles.
+ */
+static double pows10[] =
+{
+ 1e0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7, 1e8, 1e9,
+ 1e10, 1e11, 1e12, 1e13, 1e14, 1e15, 1e16, 1e17, 1e18, 1e19,
+ 1e20, 1e21, 1e22, 1e23, 1e24, 1e25, 1e26, 1e27, 1e28, 1e29,
+ 1e30, 1e31, 1e32, 1e33, 1e34, 1e35, 1e36, 1e37, 1e38, 1e39,
+ 1e40, 1e41, 1e42, 1e43, 1e44, 1e45, 1e46, 1e47, 1e48, 1e49,
+ 1e50, 1e51, 1e52, 1e53, 1e54, 1e55, 1e56, 1e57, 1e58, 1e59,
+ 1e60, 1e61, 1e62, 1e63, 1e64, 1e65, 1e66, 1e67, 1e68, 1e69,
+ 1e70, 1e71, 1e72, 1e73, 1e74, 1e75, 1e76, 1e77, 1e78, 1e79,
+ 1e80, 1e81, 1e82, 1e83, 1e84, 1e85, 1e86, 1e87, 1e88, 1e89,
+ 1e90, 1e91, 1e92, 1e93, 1e94, 1e95, 1e96, 1e97, 1e98, 1e99,
+ 1e100, 1e101, 1e102, 1e103, 1e104, 1e105, 1e106, 1e107, 1e108, 1e109,
+ 1e110, 1e111, 1e112, 1e113, 1e114, 1e115, 1e116, 1e117, 1e118, 1e119,
+ 1e120, 1e121, 1e122, 1e123, 1e124, 1e125, 1e126, 1e127, 1e128, 1e129,
+ 1e130, 1e131, 1e132, 1e133, 1e134, 1e135, 1e136, 1e137, 1e138, 1e139,
+ 1e140, 1e141, 1e142, 1e143, 1e144, 1e145, 1e146, 1e147, 1e148, 1e149,
+ 1e150, 1e151, 1e152, 1e153, 1e154, 1e155, 1e156, 1e157, 1e158, 1e159,
+};
+
+static double
+pow10(int n)
+{
+ double d;
+ int neg;
+
+ neg = 0;
+ if(n < 0){
+ if(n < DBL_MIN_10_EXP){
+ return 0.;
+ }
+ neg = 1;
+ n = -n;
+ }else if(n > DBL_MAX_10_EXP){
+ return HUGE_VAL;
+ }
+ if(n < (int)(sizeof(pows10)/sizeof(pows10[0])))
+ d = pows10[n];
+ else{
+ d = pows10[sizeof(pows10)/sizeof(pows10[0]) - 1];
+ for(;;){
+ n -= sizeof(pows10)/sizeof(pows10[0]) - 1;
+ if(n < (int)(sizeof(pows10)/sizeof(pows10[0]))){
+ d *= pows10[n];
+ break;
+ }
+ d *= pows10[sizeof(pows10)/sizeof(pows10[0]) - 1];
+ }
+ }
+ if(neg){
+ return 1./d;
+ }
+ return d;
+}
+
+static int
+xadd(char *a, int n, int v)
+{
+ char *b;
+ int c;
+
+ if(n < 0 || n >= NSIGNIF)
+ return 0;
+ for(b = a+n; b >= a; b--) {
+ c = *b + v;
+ if(c <= '9') {
+ *b = c;
+ return 0;
+ }
+ *b = '0';
+ v = 1;
+ }
+ *a = '1'; /* overflow adding */
+ return 1;
+}
+
+static int
+xsub(char *a, int n, int v)
+{
+ char *b;
+ int c;
+
+ for(b = a+n; b >= a; b--) {
+ c = *b - v;
+ if(c >= '0') {
+ *b = c;
+ return 0;
+ }
+ *b = '9';
+ v = 1;
+ }
+ *a = '9'; /* underflow subtracting */
+ return 1;
+}
+
+static void
+xaddexp(char *p, int e)
+{
+ char se[9];
+ int i;
+
+ *p++ = 'e';
+ if(e < 0) {
+ *p++ = '-';
+ e = -e;
+ }
+ i = 0;
+ while(e) {
+ se[i++] = e % 10 + '0';
+ e /= 10;
+ }
+ if(i == 0) {
+ *p++ = '0';
+ } else {
+ while(i > 0)
+ *p++ = se[--i];
+ }
+ *p++ = '\0';
+}
+
+static char*
+xdodtoa(char *s1, double f, int chr, int prec, int *decpt, int *rsign)
+{
+ char s2[NSIGNIF+10];
+ double g, h;
+ int e, d, i;
+ int c2, sign, oerr;
+
+ if(chr == 'F')
+ chr = 'f';
+ if(prec > NSIGNIF)
+ prec = NSIGNIF;
+ if(prec < 0)
+ prec = 0;
+ if(__isNaN(f)) {
+ *decpt = 9999;
+ *rsign = 0;
+ strcpy(s1, "nan");
+ return &s1[3];
+ }
+ sign = 0;
+ if(f < 0) {
+ f = -f;
+ sign++;
+ }
+ *rsign = sign;
+ if(__isInf(f, 1) || __isInf(f, -1)) {
+ *decpt = 9999;
+ strcpy(s1, "inf");
+ return &s1[3];
+ }
+
+ e = 0;
+ g = f;
+ if(g != 0) {
+ frexp(f, &e);
+ e = (int)(e * .301029995664);
+ if(e >= -150 && e <= +150) {
+ d = 0;
+ h = f;
+ } else {
+ d = e/2;
+ h = f * pow10(-d);
+ }
+ g = h * pow10(d-e);
+ while(g < 1) {
+ e--;
+ g = h * pow10(d-e);
+ }
+ while(g >= 10) {
+ e++;
+ g = h * pow10(d-e);
+ }
+ }
+
+ /*
+ * convert NSIGNIF digits and convert
+ * back to get accuracy.
+ */
+ for(i=0; i<NSIGNIF; i++) {
+ d = (int)g;
+ s1[i] = d + '0';
+ g = (g - d) * 10;
+ }
+ s1[i] = 0;
+
+ /*
+ * try decimal rounding to eliminate 9s
+ */
+ c2 = prec + 1;
+ if(chr == 'f')
+ c2 += e;
+ oerr = errno;
+ if(c2 >= NSIGNIF-2) {
+ strcpy(s2, s1);
+ d = e;
+ s1[NSIGNIF-2] = '0';
+ s1[NSIGNIF-1] = '0';
+ xaddexp(s1+NSIGNIF, e-NSIGNIF+1);
+ g = fmtstrtod(s1, nil);
+ if(g == f)
+ goto found;
+ if(xadd(s1, NSIGNIF-3, 1)) {
+ e++;
+ xaddexp(s1+NSIGNIF, e-NSIGNIF+1);
+ }
+ g = fmtstrtod(s1, nil);
+ if(g == f)
+ goto found;
+ strcpy(s1, s2);
+ e = d;
+ }
+
+ /*
+ * convert back so s1 gets exact answer
+ */
+ for(d = 0; d < 10; d++) {
+ xaddexp(s1+NSIGNIF, e-NSIGNIF+1);
+ g = fmtstrtod(s1, nil);
+ if(f > g) {
+ if(xadd(s1, NSIGNIF-1, 1))
+ e--;
+ continue;
+ }
+ if(f < g) {
+ if(xsub(s1, NSIGNIF-1, 1))
+ e++;
+ continue;
+ }
+ break;
+ }
+
+found:
+ errno = oerr;
+
+ /*
+ * sign
+ */
+ d = 0;
+ i = 0;
+
+ /*
+ * round & adjust 'f' digits
+ */
+ c2 = prec + 1;
+ if(chr == 'f'){
+ if(xadd(s1, c2+e, 5))
+ e++;
+ c2 += e;
+ if(c2 < 0){
+ c2 = 0;
+ e = -prec - 1;
+ }
+ }else{
+ if(xadd(s1, c2, 5))
+ e++;
+ }
+ if(c2 > NSIGNIF){
+ c2 = NSIGNIF;
+ }
+
+ *decpt = e + 1;
+
+ /*
+ * terminate the converted digits
+ */
+ s1[c2] = '\0';
+ return &s1[c2];
+}
+
+/*
+ * this function works like the standard dtoa, if you want it.
+ */
+#if 0
+static char*
+__dtoa(double f, int mode, int ndigits, int *decpt, int *rsign, char **rve)
+{
+ static char s2[NSIGNIF + 10];
+ char *es;
+ int chr, prec;
+
+ switch(mode) {
+ /* like 'e' */
+ case 2:
+ case 4:
+ case 6:
+ case 8:
+ chr = 'e';
+ break;
+ /* like 'g' */
+ case 0:
+ case 1:
+ default:
+ chr = 'g';
+ break;
+ /* like 'f' */
+ case 3:
+ case 5:
+ case 7:
+ case 9:
+ chr = 'f';
+ break;
+ }
+
+ if(chr != 'f' && ndigits){
+ ndigits--;
+ }
+ prec = ndigits;
+ if(prec > NSIGNIF)
+ prec = NSIGNIF;
+ if(ndigits == 0)
+ prec = NSIGNIF;
+ es = xdodtoa(s2, f, chr, prec, decpt, rsign);
+
+ /*
+ * strip trailing 0
+ */
+ for(; es > s2 + 1; es--){
+ if(es[-1] != '0'){
+ break;
+ }
+ }
+ *es = '\0';
+ if(rve != NULL)
+ *rve = es;
+ return s2;
+}
+#endif
+
+static int
+fmtzdotpad(Fmt *f, int n, int pt)
+{
+ char *t, *s;
+ int i;
+ Rune *rt, *rs;
+
+ if(f->runes){
+ rt = (Rune*)f->to;
+ rs = (Rune*)f->stop;
+ for(i = 0; i < n; i++){
+ if(i == pt){
+ FMTRCHAR(f, rt, rs, '.');
+ }
+ FMTRCHAR(f, rt, rs, '0');
+ }
+ f->nfmt += rt - (Rune*)f->to;
+ f->to = rt;
+ }else{
+ t = (char*)f->to;
+ s = (char*)f->stop;
+ for(i = 0; i < n; i++){
+ if(i == pt){
+ FMTCHAR(f, t, s, '.');
+ }
+ FMTCHAR(f, t, s, '0');
+ }
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ }
+ return 0;
+}
+
+int
+__efgfmt(Fmt *fmt)
+{
+ double f;
+ char s1[NSIGNIF+10];
+ int e, d, n;
+ int c1, c2, c3, c4, ucase, sign, chr, prec, fl;
+
+ f = va_arg(fmt->args, double);
+ prec = FDEFLT;
+ fl = fmt->flags;
+ fmt->flags = 0;
+ if(fl & FmtPrec)
+ prec = fmt->prec;
+ chr = fmt->r;
+ ucase = 0;
+ if(chr == 'E'){
+ chr = 'e';
+ ucase = 1;
+ }else if(chr == 'F'){
+ chr = 'f';
+ ucase = 1;
+ }else if(chr == 'G'){
+ chr = 'g';
+ ucase = 1;
+ }
+ if(prec > 0 && chr == 'g')
+ prec--;
+ if(prec < 0)
+ prec = 0;
+
+ xdodtoa(s1, f, chr, prec, &e, &sign);
+ e--;
+ if(*s1 == 'i' || *s1 == 'n'){
+ if(ucase){
+ if(*s1 == 'i'){
+ strcpy(s1, "INF");
+ }else{
+ strcpy(s1, "NAN");
+ }
+ }
+ fmt->flags = fl & (FmtWidth|FmtLeft);
+ return __fmtcpy(fmt, (const void*)s1, 3, 3);
+ }
+
+ /*
+ * copy into final place
+ * c1 digits of leading '0'
+ * c2 digits from conversion
+ * c3 digits of trailing '0'
+ * c4 digits after '.'
+ */
+ c1 = 0;
+ c2 = prec + 1;
+ c3 = 0;
+ c4 = prec;
+ switch(chr) {
+ default:
+ chr = 'e';
+ break;
+ case 'g':
+ /*
+ * decide on 'e' of 'f' style convers
+ */
+ if(e >= -4 && e <= prec) {
+ c1 = -e;
+ c4 = prec - e;
+ chr = 'h'; /* flag for 'f' style */
+ }
+ break;
+ case 'f':
+ c1 = -e;
+ if(c1 > prec)
+ c1 = prec + 1;
+ c2 += e;
+ break;
+ }
+
+ /*
+ * clean up c1 c2 and c3
+ */
+ if(c1 < 0)
+ c1 = 0;
+ if(c2 < 0)
+ c2 = 0;
+ if(c2 > NSIGNIF) {
+ c3 = c2-NSIGNIF;
+ c2 = NSIGNIF;
+ }
+
+ /*
+ * trim trailing zeros for %g
+ */
+ if(!(fl & FmtSharp)
+ && (chr == 'g' || chr == 'h')){
+ if(c4 >= c3){
+ c4 -= c3;
+ c3 = 0;
+ }else{
+ c3 -= c4;
+ c4 = 0;
+ }
+ while(c4 && c2 > 1 && s1[c2 - 1] == '0'){
+ c4--;
+ c2--;
+ }
+ }
+
+ /*
+ * calculate the total length
+ */
+ n = c1 + c2 + c3;
+ if(sign || (fl & (FmtSign|FmtSpace)))
+ n++;
+ if(c4 || (fl & FmtSharp)){
+ n++;
+ }
+ if(chr == 'e' || chr == 'g'){
+ n += 4;
+ if(e >= 100)
+ n++;
+ }
+
+ /*
+ * pad to width if right justified
+ */
+ if((fl & (FmtWidth|FmtLeft)) == FmtWidth && n < fmt->width){
+ if(fl & FmtZero){
+ c1 += fmt->width - n;
+ }else{
+ if(__fmtpad(fmt, fmt->width - n) < 0){
+ return -1;
+ }
+ }
+ }
+
+ /*
+ * sign
+ */
+ d = 0;
+ if(sign)
+ d = '-';
+ else if(fl & FmtSign)
+ d = '+';
+ else if(fl & FmtSpace)
+ d = ' ';
+ if(d && fmtrune(fmt, d) < 0){
+ return -1;
+ }
+
+ /*
+ * copy digits
+ */
+ c4 = c1 + c2 + c3 - c4;
+ if(c1 > 0){
+ if(fmtzdotpad(fmt, c1, c4) < 0){
+ return -1;
+ }
+ c4 -= c1;
+ }
+ d = 0;
+ if(c4 >= 0 && c4 < c2){
+ if(__fmtcpy(fmt, s1, c4, c4) < 0 || fmtrune(fmt, '.') < 0)
+ return -1;
+ d = c4;
+ c2 -= c4;
+ c4 = -1;
+ }
+ if(__fmtcpy(fmt, (const void*)(s1 + d), c2, c2) < 0){
+ return -1;
+ }
+ c4 -= c2;
+ if(c3 > 0){
+ if(fmtzdotpad(fmt, c3, c4) < 0){
+ return -1;
+ }
+ c4 -= c3;
+ }
+
+ /*
+ * strip trailing '0' on g conv
+ */
+ if((fl & FmtSharp) && c4 == 0 && fmtrune(fmt, '.') < 0){
+ return -1;
+ }
+ if(chr == 'e' || chr == 'g') {
+ d = 0;
+ if(ucase)
+ s1[d++] = 'E';
+ else
+ s1[d++] = 'e';
+ c1 = e;
+ if(c1 < 0) {
+ s1[d++] = '-';
+ c1 = -c1;
+ } else
+ s1[d++] = '+';
+ if(c1 >= 100) {
+ s1[d++] = c1/100 + '0';
+ c1 = c1%100;
+ }
+ s1[d++] = c1/10 + '0';
+ s1[d++] = c1%10 + '0';
+ if(__fmtcpy(fmt, s1, d, d) < 0){
+ return -1;
+ }
+ }
+ if((fl & (FmtWidth|FmtLeft)) == (FmtWidth|FmtLeft) && n < fmt->width){
+ if(__fmtpad(fmt, fmt->width - n) < 0){
+ return -1;
+ }
+ }
+ return 0;
+}
DIR diff --git a/src/libfmt/fmt.c b/src/libfmt/fmt.c
t@@ -0,0 +1,221 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+enum
+{
+ Maxfmt = 64
+};
+
+typedef struct Convfmt Convfmt;
+struct Convfmt
+{
+ int c;
+ volatile Fmts fmt; /* for spin lock in fmtfmt; avoids race due to write order */
+};
+
+struct
+{
+ /* lock by calling __fmtlock, __fmtunlock */
+ int nfmt;
+ Convfmt fmt[Maxfmt];
+} fmtalloc;
+
+static Convfmt knownfmt[] = {
+ ' ', __flagfmt,
+ '#', __flagfmt,
+ '%', __percentfmt,
+ '+', __flagfmt,
+ ',', __flagfmt,
+ '-', __flagfmt,
+ 'C', __runefmt, /* Plan 9 addition */
+ 'E', __efgfmt,
+ 'F', __efgfmt, /* ANSI only */
+ 'G', __efgfmt,
+ 'L', __flagfmt, /* ANSI only */
+ 'S', __runesfmt, /* Plan 9 addition */
+ 'X', __ifmt,
+ 'b', __ifmt, /* Plan 9 addition */
+ 'c', __charfmt,
+ 'd', __ifmt,
+ 'e', __efgfmt,
+ 'f', __efgfmt,
+ 'g', __efgfmt,
+ 'h', __flagfmt,
+ 'i', __ifmt, /* ANSI only */
+ 'l', __flagfmt,
+ 'n', __countfmt,
+ 'o', __ifmt,
+ 'p', __ifmt,
+ 'r', __errfmt,
+ 's', __strfmt,
+ 'u', __ifmt, /* in Plan 9, __flagfmt */
+ 'x', __ifmt,
+ 0, nil,
+};
+
+
+int (*fmtdoquote)(int);
+
+/*
+ * __fmtlock() must be set
+ */
+static int
+__fmtinstall(int c, Fmts f)
+{
+ Convfmt *p, *ep;
+
+ if(c<=0 || c>=65536)
+ return -1;
+ if(!f)
+ f = __badfmt;
+
+ ep = &fmtalloc.fmt[fmtalloc.nfmt];
+ for(p=fmtalloc.fmt; p<ep; p++)
+ if(p->c == c)
+ break;
+
+ if(p == &fmtalloc.fmt[Maxfmt])
+ return -1;
+
+ p->fmt = f;
+ if(p == ep){ /* installing a new format character */
+ fmtalloc.nfmt++;
+ p->c = c;
+ }
+
+ return 0;
+}
+
+int
+fmtinstall(int c, Fmts f)
+{
+ int ret;
+
+ __fmtlock();
+ ret = __fmtinstall(c, f);
+ __fmtunlock();
+ return ret;
+}
+
+static Fmts
+fmtfmt(int c)
+{
+ Convfmt *p, *ep;
+
+ ep = &fmtalloc.fmt[fmtalloc.nfmt];
+ for(p=fmtalloc.fmt; p<ep; p++)
+ if(p->c == c){
+ while(p->fmt == nil) /* loop until value is updated */
+ ;
+ return p->fmt;
+ }
+
+ /* is this a predefined format char? */
+ __fmtlock();
+ for(p=knownfmt; p->c; p++)
+ if(p->c == c){
+ __fmtinstall(p->c, p->fmt);
+ __fmtunlock();
+ return p->fmt;
+ }
+ __fmtunlock();
+
+ return __badfmt;
+}
+
+void*
+__fmtdispatch(Fmt *f, void *fmt, int isrunes)
+{
+ Rune rune, r;
+ int i, n;
+
+ f->flags = 0;
+ f->width = f->prec = 0;
+
+ for(;;){
+ if(isrunes){
+ r = *(Rune*)fmt;
+ fmt = (Rune*)fmt + 1;
+ }else{
+ fmt = (char*)fmt + chartorune(&rune, (char*)fmt);
+ r = rune;
+ }
+ f->r = r;
+ switch(r){
+ case '\0':
+ return nil;
+ case '.':
+ f->flags |= FmtWidth|FmtPrec;
+ continue;
+ case '0':
+ if(!(f->flags & FmtWidth)){
+ f->flags |= FmtZero;
+ continue;
+ }
+ /* fall through */
+ case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9':
+ i = 0;
+ while(r >= '0' && r <= '9'){
+ i = i * 10 + r - '0';
+ if(isrunes){
+ r = *(Rune*)fmt;
+ fmt = (Rune*)fmt + 1;
+ }else{
+ r = *(char*)fmt;
+ fmt = (char*)fmt + 1;
+ }
+ }
+ if(isrunes)
+ fmt = (Rune*)fmt - 1;
+ else
+ fmt = (char*)fmt - 1;
+ numflag:
+ if(f->flags & FmtWidth){
+ f->flags |= FmtPrec;
+ f->prec = i;
+ }else{
+ f->flags |= FmtWidth;
+ f->width = i;
+ }
+ continue;
+ case '*':
+ i = va_arg(f->args, int);
+ if(i < 0){
+ /*
+ * negative precision =>
+ * ignore the precision.
+ */
+ if(f->flags & FmtPrec){
+ f->flags &= ~FmtPrec;
+ f->prec = 0;
+ continue;
+ }
+ i = -i;
+ f->flags |= FmtLeft;
+ }
+ goto numflag;
+ }
+ n = (*fmtfmt(r))(f);
+ if(n < 0)
+ return nil;
+ if(n == 0)
+ return fmt;
+ }
+}
DIR diff --git a/src/libfmt/fmt.h b/src/libfmt/fmt.h
t@@ -0,0 +1,100 @@
+
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+
+#ifndef _FMTH_
+#define _FMTH_ 1
+
+#include <stdarg.h>
+
+#ifndef _UTFH_
+#include <utf.h>
+#endif
+
+typedef struct Fmt Fmt;
+struct Fmt{
+ unsigned char runes; /* output buffer is runes or chars? */
+ void *start; /* of buffer */
+ void *to; /* current place in the buffer */
+ void *stop; /* end of the buffer; overwritten if flush fails */
+ int (*flush)(Fmt *); /* called when to == stop */
+ void *farg; /* to make flush a closure */
+ int nfmt; /* num chars formatted so far */
+ va_list args; /* args passed to dofmt */
+ int r; /* % format Rune */
+ int width;
+ int prec;
+ unsigned long flags;
+};
+
+enum{
+ FmtWidth = 1,
+ FmtLeft = FmtWidth << 1,
+ FmtPrec = FmtLeft << 1,
+ FmtSharp = FmtPrec << 1,
+ FmtSpace = FmtSharp << 1,
+ FmtSign = FmtSpace << 1,
+ FmtZero = FmtSign << 1,
+ FmtUnsigned = FmtZero << 1,
+ FmtShort = FmtUnsigned << 1,
+ FmtLong = FmtShort << 1,
+ FmtVLong = FmtLong << 1,
+ FmtComma = FmtVLong << 1,
+ FmtByte = FmtComma << 1,
+ FmtLDouble = FmtByte << 1,
+
+ FmtFlag = FmtLDouble << 1
+};
+
+extern int print(char*, ...);
+extern char* seprint(char*, char*, char*, ...);
+extern char* vseprint(char*, char*, char*, va_list);
+extern int snprint(char*, int, char*, ...);
+extern int vsnprint(char*, int, char*, va_list);
+extern char* smprint(char*, ...);
+extern char* vsmprint(char*, va_list);
+extern int sprint(char*, char*, ...);
+extern int fprint(int, char*, ...);
+extern int vfprint(int, char*, va_list);
+
+extern int runesprint(Rune*, char*, ...);
+extern int runesnprint(Rune*, int, char*, ...);
+extern int runevsnprint(Rune*, int, char*, va_list);
+extern Rune* runeseprint(Rune*, Rune*, char*, ...);
+extern Rune* runevseprint(Rune*, Rune*, char*, va_list);
+extern Rune* runesmprint(char*, ...);
+extern Rune* runevsmprint(char*, va_list);
+
+extern int fmtfdinit(Fmt*, int, char*, int);
+extern int fmtfdflush(Fmt*);
+extern int fmtstrinit(Fmt*);
+extern char* fmtstrflush(Fmt*);
+extern int runefmtstrinit(Fmt*);
+
+extern int quotestrfmt(Fmt *f);
+extern void quotefmtinstall(void);
+extern int (*fmtdoquote)(int);
+
+
+extern int fmtinstall(int, int (*)(Fmt*));
+extern int dofmt(Fmt*, char*);
+extern int fmtprint(Fmt*, char*, ...);
+extern int fmtvprint(Fmt*, char*, va_list);
+extern int fmtrune(Fmt*, int);
+extern int fmtstrcpy(Fmt*, char*);
+
+extern double fmtstrtod(const char *, char **);
+extern double fmtcharstod(int(*)(void*), void*);
+
+#endif
DIR diff --git a/src/libfmt/fmtdef.h b/src/libfmt/fmtdef.h
t@@ -0,0 +1,121 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+/*
+ * dofmt -- format to a buffer
+ * the number of characters formatted is returned,
+ * or -1 if there was an error.
+ * if the buffer is ever filled, flush is called.
+ * it should reset the buffer and return whether formatting should continue.
+ */
+#define uchar _fmtuchar
+#define ushort _fmtushort
+#define uint _fmtuint
+#define ulong _fmtulong
+#define vlong _fmtvlong
+#define uvlong _fmtuvlong
+
+#define USED(x) if(x);else
+
+typedef unsigned char uchar;
+typedef unsigned short ushort;
+typedef unsigned int uint;
+typedef unsigned long ulong;
+
+#ifndef NOVLONGS
+typedef unsigned long long uvlong;
+typedef long long vlong;
+#endif
+
+#define nil 0 /* cannot be ((void*)0) because used for function pointers */
+
+typedef int (*Fmts)(Fmt*);
+
+typedef struct Quoteinfo Quoteinfo;
+struct Quoteinfo
+{
+ int quoted; /* if set, string must be quoted */
+ int nrunesin; /* number of input runes that can be accepted */
+ int nbytesin; /* number of input bytes that can be accepted */
+ int nrunesout; /* number of runes that will be generated */
+ int nbytesout; /* number of bytes that will be generated */
+};
+
+void *__fmtflush(Fmt*, void*, int);
+void *__fmtdispatch(Fmt*, void*, int);
+int __floatfmt(Fmt*, double);
+int __fmtpad(Fmt*, int);
+int __rfmtpad(Fmt*, int);
+int __fmtFdFlush(Fmt*);
+
+int __efgfmt(Fmt*);
+int __charfmt(Fmt*);
+int __runefmt(Fmt*);
+int __runesfmt(Fmt*);
+int __countfmt(Fmt*);
+int __flagfmt(Fmt*);
+int __percentfmt(Fmt*);
+int __ifmt(Fmt*);
+int __strfmt(Fmt*);
+int __badfmt(Fmt*);
+int __fmtcpy(Fmt*, const void*, int, int);
+int __fmtrcpy(Fmt*, const void*, int n);
+int __errfmt(Fmt *f);
+
+double __fmtpow10(int);
+
+void __fmtlock(void);
+void __fmtunlock(void);
+
+#define FMTCHAR(f, t, s, c)\
+ do{\
+ if(t + 1 > (char*)s){\
+ t = __fmtflush(f, t, 1);\
+ if(t != nil)\
+ s = f->stop;\
+ else\
+ return -1;\
+ }\
+ *t++ = c;\
+ }while(0)
+
+#define FMTRCHAR(f, t, s, c)\
+ do{\
+ if(t + 1 > (Rune*)s){\
+ t = __fmtflush(f, t, sizeof(Rune));\
+ if(t != nil)\
+ s = f->stop;\
+ else\
+ return -1;\
+ }\
+ *t++ = c;\
+ }while(0)
+
+#define FMTRUNE(f, t, s, r)\
+ do{\
+ Rune _rune;\
+ int _runelen;\
+ if(t + UTFmax > (char*)s && t + (_runelen = runelen(r)) > (char*)s){\
+ t = __fmtflush(f, t, _runelen);\
+ if(t != nil)\
+ s = f->stop;\
+ else\
+ return -1;\
+ }\
+ if(r < Runeself)\
+ *t++ = r;\
+ else{\
+ _rune = r;\
+ t += runetochar(t, &_rune);\
+ }\
+ }while(0)
DIR diff --git a/src/libfmt/fmtfd.c b/src/libfmt/fmtfd.c
t@@ -0,0 +1,46 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * public routine for final flush of a formatting buffer
+ * to a file descriptor; returns total char count.
+ */
+int
+fmtfdflush(Fmt *f)
+{
+ if(__fmtFdFlush(f) <= 0)
+ return -1;
+ return f->nfmt;
+}
+
+/*
+ * initialize an output buffer for buffered printing
+ */
+int
+fmtfdinit(Fmt *f, int fd, char *buf, int size)
+{
+ f->runes = 0;
+ f->start = buf;
+ f->to = buf;
+ f->stop = buf + size;
+ f->flush = __fmtFdFlush;
+ f->farg = (void*)fd;
+ f->nfmt = 0;
+ return 0;
+}
DIR diff --git a/src/libfmt/fmtfdflush.c b/src/libfmt/fmtfdflush.c
t@@ -0,0 +1,33 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <unistd.h>
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * generic routine for flushing a formatting buffer
+ * to a file descriptor
+ */
+int
+__fmtFdFlush(Fmt *f)
+{
+ int n;
+
+ n = (char*)f->to - (char*)f->start;
+ if(n && write((int)f->farg, f->start, n) != n)
+ return 0;
+ f->to = f->start;
+ return 1;
+}
DIR diff --git a/src/libfmt/fmtinstall.3 b/src/libfmt/fmtinstall.3
t@@ -0,0 +1,346 @@
+.TH FMTINSTALL 3
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.SH NAME
+fmtinstall, dofmt, fmtprint, fmtvprint, fmtstrcpy, fmtfdinit, fmtfdflush, fmtstrinit, fmtstrflush \- support for user-defined print formats and output routines
+.SH SYNOPSIS
+.B #include <fmt.h>
+.PP
+.ft L
+.nf
+.ta \w' 'u +\w' 'u +\w' 'u +\w' 'u +\w' 'u
+typedef struct Fmt Fmt;
+struct Fmt{
+ void *start; /* of buffer */
+ void *to; /* current place in the buffer */
+ void *stop; /* end of the buffer; overwritten if flush fails */
+ int (*flush)(Fmt*); /* called when to == stop */
+ void *farg; /* to make flush a closure */
+ int nfmt; /* num chars formatted so far */
+ va_list args; /* args passed to dofmt */
+ int r; /* % format character */
+ int width;
+ int prec;
+ unsigned long flags;
+};
+
+enum{
+ FmtWidth = 1,
+ FmtLeft = FmtWidth << 1,
+ FmtPrec = FmtLeft << 1,
+ FmtSharp = FmtPrec << 1,
+ FmtSpace = FmtSharp << 1,
+ FmtSign = FmtSpace << 1,
+ FmtZero = FmtSign << 1,
+ FmtUnsigned = FmtZero << 1,
+ FmtShort = FmtUnsigned << 1,
+ FmtLong = FmtShort << 1,
+ FmtVLong = FmtLong << 1,
+ FmtComma = FmtVLong << 1,
+ FmtByte = FmtComma << 1,
+ FmtLDouble = FmtByte << 1,
+
+ FmtFlag = FmtLDouble << 1
+};
+.fi
+.PP
+.B
+.ta \w'\fLchar* 'u
+
+.PP
+.B
+int fmtfdinit(Fmt *f, int fd, char *buf, int nbuf);
+.PP
+.B
+int fmtfdflush(Fmt *f);
+.PP
+.B
+int fmtstrinit(Fmt *f);
+.PP
+.B
+char* fmtstrflush(Fmt *f);
+.PP
+.B
+int fmtinstall(int c, int (*fn)(Fmt*));
+.PP
+.B
+int dofmt(Fmt *f, char *fmt);
+.PP
+.B
+int fmtprint(Fmt *f, char *fmt, ...);
+.PP
+.B
+int fmtvprint(Fmt *f, char *fmt, va_list v);
+.PP
+.B
+int fmtrune(Fmt *f, int r);
+.PP
+.B
+int fmtstrcpy(Fmt *f, char *s);
+.SH DESCRIPTION
+The interface described here allows the construction of custom
+.IR print (3)
+verbs and output routines.
+In essence, they provide access to the workings of the formatted print code.
+.PP
+The
+.IR print (3)
+suite maintains its state with a data structure called
+.BR Fmt .
+A typical call to
+.IR print (3)
+or its relatives initializes a
+.B Fmt
+structure, passes it to subsidiary routines to process the output,
+and finishes by emitting any saved state recorded in the
+.BR Fmt .
+The details of the
+.B Fmt
+are unimportant to outside users, except insofar as the general
+design influences the interface.
+The
+.B Fmt
+records
+the verb being processed, its precision and width,
+and buffering parameters.
+Most important, it also records a
+.I flush
+routine that the library will call if a buffer overflows.
+When printing to a file descriptor, the flush routine will
+emit saved characters and reset the buffer; when printing
+to an allocated string, it will resize the string to receive more output.
+The flush routine is nil when printing to fixed-size buffers.
+User code need never provide a flush routine; this is done internally
+by the library.
+.SS Custom output routines
+To write a custom output routine, such as an error handler that
+formats and prints custom error messages, the output sequence can be run
+from outside the library using the routines described here.
+There are two main cases: output to an open file descriptor
+and output to a string.
+.PP
+To write to a file descriptor, call
+.I fmtfdinit
+to initialize the local
+.B Fmt
+structure
+.IR f ,
+giving the file descriptor
+.IR fd ,
+the buffer
+.IR buf ,
+and its size
+.IR nbuf .
+Then call
+.IR fmtprint
+or
+.IR fmtvprint
+to generate the output.
+These behave just like
+.B fprint
+(see
+.IR print (3))
+or
+.B vfprint
+except that the characters are buffered until
+.I fmtfdflush
+is called.
+A typical example of this sequence appears in the Examples section.
+.PP
+The same basic sequence applies when outputting to an allocated string:
+call
+.I fmtstrinit
+to initialize the
+.BR Fmt ,
+then call
+.I fmtprint
+and
+.I fmtvprint
+to generate the output.
+Finally,
+.I fmtstrflush
+will return the allocated string, which should be freed after use.
+Regardless of the output style or type,
+.I fmtprint
+or
+.I fmtvprint
+generates the characters.
+.SS Custom format verbs
+.I Fmtinstall
+is used to install custom verbs and flags labeled by character
+.IR c ,
+which may be any non-zero Unicode character.
+.I Fn
+should be declared as
+.IP
+.EX
+int fn(Fmt*)
+.EE
+.PP
+.IB Fp ->r
+is the flag or verb character to cause
+.I fn
+to be called.
+In
+.IR fn ,
+.IB fp ->width ,
+.IB fp ->prec
+are the width and precision, and
+.IB fp ->flags
+the decoded flags for the verb (see
+.IR print (3)
+for a description of these items).
+The standard flag values are:
+.B FmtSign
+.RB ( + ),
+.B FmtLeft
+.RB ( - ),
+.B FmtSpace
+.RB ( '\ ' ),
+.B FmtSharp
+.RB ( # ),
+.B FmtComma
+.RB ( , ),
+.B FmtLong
+.RB ( l ),
+.B FmtShort
+.RB ( h ),
+.B FmtByte
+.RB ( hh ),
+.B FmtUnsigned
+.RB ( u ),
+.B FmtLDouble
+.RB ( L ),
+and
+.B FmtVLong
+.RB ( ll ).
+The flag bits
+.B FmtWidth
+and
+.B FmtPrec
+identify whether a width and precision were specified.
+.PP
+.I Fn
+is passed a pointer to the
+.B Fmt
+structure recording the state of the output.
+If
+.IB fp ->r
+is a verb (rather than a flag),
+.I fn
+should use
+.B Fmt->args
+to fetch its argument from the list,
+then format it, and return zero.
+If
+.IB fp ->r
+is a flag,
+.I fn
+should return a negative value:
+the negation of one of the above flag values, or some otherwise unused power of two.
+All interpretation of
+.IB fp ->width\f1,
+.IB fp ->prec\f1,
+and
+.IB fp-> flags
+is left up to the conversion routine.
+.I Fmtinstall
+returns 0 if the installation succeeds, \-1 if it fails.
+.PP
+.IR Fmtprint
+and
+.IR fmtvprint
+may be called to
+help prepare output in custom conversion routines.
+However, these functions clear the width, precision, and flags.
+The function
+.I dofmt
+is the underlying formatter; it
+uses the existing contents of
+.B Fmt
+and should be called only by sophisticated conversion routines.
+All these routines return the number of characters
+produced.
+.PP
+Some internal functions may be useful to format primitive types.
+They honor the width, precision and flags as described in
+.IR print (3).
+.I Fmtrune
+formats a single character
+.BR r .
+.I Fmtstrcpy
+formats a string
+.BR s .
+All these routines return zero for successful execution.
+.SH EXAMPLES
+This function prints an error message with a variable
+number of arguments and then quits.
+Compared to the corresponding example in
+.IR print (3),
+this version uses a smaller buffer, will never truncate
+the output message, but might generate multiple
+.B write
+system calls to produce its output.
+.IP
+.EX
+.ta 6n +6n +6n +6n +6n +6n +6n +6n +6n
+
+void fatal(char *fmt, ...)
+{
+ Fmt f;
+ char buf[64];
+ va_list arg;
+
+ fmtfdinit(&f, 1, buf, sizeof buf);
+ fmtprint(&f, "fatal: ");
+ va_start(arg, fmt);
+ fmtvprint(&f, fmt, arg);
+ va_end(arg);
+ fmtprint(&f, "\en");
+ fmtfdflush(&f);
+ exits("fatal error");
+}
+.EE
+.PP
+This example adds a verb to print complex numbers.
+.IP
+.EX
+typedef
+struct {
+ double r, i;
+} Complex;
+
+int
+Xfmt(Fmt *f)
+{
+ Complex c;
+
+ c = va_arg(f->args, Complex);
+ return fmtprint(f, "(%g,%g)", c.r, c.i);
+}
+
+main(...)
+{
+ Complex x;
+
+ x.r = 1.5;
+ x.i = -2.3;
+
+ fmtinstall('X', Xfmt);
+ print("x = %X\en", x);
+}
+.EE
+.SH SEE ALSO
+.IR print (3)
+.SH HISTORY
+This formatted print library originally
+appeared as part of the Plan 9 C library.
+.SH BUGS
+The Plan 9 version supports Unicode strings and produces UTF output.
+This version assumes that characters are always represented by 1-byte values.
DIR diff --git a/src/libfmt/fmtlock.c b/src/libfmt/fmtlock.c
t@@ -0,0 +1,28 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+#include "fmtdef.h"
+
+void
+__fmtlock(void)
+{
+ ;
+}
+
+void
+__fmtunlock(void)
+{
+ ;
+}
DIR diff --git a/src/libfmt/fmtprint.c b/src/libfmt/fmtprint.c
t@@ -0,0 +1,47 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+
+/*
+ * format a string into the output buffer
+ * designed for formats which themselves call fmt,
+ * but ignore any width flags
+ */
+int
+fmtprint(Fmt *f, char *fmt, ...)
+{
+ va_list va;
+ int n;
+
+ f->flags = 0;
+ f->width = 0;
+ f->prec = 0;
+ va = f->args;
+ va_start(f->args, fmt);
+ n = dofmt(f, fmt);
+ va_end(f->args);
+ f->flags = 0;
+ f->width = 0;
+ f->prec = 0;
+ f->args = va;
+ if(n >= 0)
+ return 0;
+ return n;
+}
+
DIR diff --git a/src/libfmt/fmtquote.c b/src/libfmt/fmtquote.c
t@@ -0,0 +1,262 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * How many bytes of output UTF will be produced by quoting (if necessary) this string?
+ * How many runes? How much of the input will be consumed?
+ * The parameter q is filled in by __quotesetup.
+ * The string may be UTF or Runes (s or r).
+ * Return count does not include NUL.
+ * Terminate the scan at the first of:
+ * NUL in input
+ * count exceeded in input
+ * count exceeded on output
+ * *ninp is set to number of input bytes accepted.
+ * nin may be <0 initially, to avoid checking input by count.
+ */
+void
+__quotesetup(char *s, Rune *r, int nin, int nout, Quoteinfo *q, int sharp, int runesout)
+{
+ int w;
+ Rune c;
+
+ q->quoted = 0;
+ q->nbytesout = 0;
+ q->nrunesout = 0;
+ q->nbytesin = 0;
+ q->nrunesin = 0;
+ if(sharp || nin==0 || (s && *s=='\0') || (r && *r=='\0')){
+ if(nout < 2)
+ return;
+ q->quoted = 1;
+ q->nbytesout = 2;
+ q->nrunesout = 2;
+ }
+ for(; nin!=0; nin-=w){
+ if(s)
+ w = chartorune(&c, s);
+ else{
+ c = *r;
+ w = runelen(c);
+ }
+
+ if(c == '\0')
+ break;
+ if(runesout){
+ if(q->nrunesout+1 > nout)
+ break;
+ }else{
+ if(q->nbytesout+w > nout)
+ break;
+ }
+
+ if((c <= L' ') || (c == L'\'') || (fmtdoquote!=nil && fmtdoquote(c))){
+ if(!q->quoted){
+ if(runesout){
+ if(1+q->nrunesout+1+1 > nout) /* no room for quotes */
+ break;
+ }else{
+ if(1+q->nbytesout+w+1 > nout) /* no room for quotes */
+ break;
+ }
+ q->nrunesout += 2; /* include quotes */
+ q->nbytesout += 2; /* include quotes */
+ q->quoted = 1;
+ }
+ if(c == '\'') {
+ if(runesout){
+ if(1+q->nrunesout+1 > nout) /* no room for quotes */
+ break;
+ }else{
+ if(1+q->nbytesout+w > nout) /* no room for quotes */
+ break;
+ }
+ q->nbytesout++;
+ q->nrunesout++; /* quotes reproduce as two characters */
+ }
+ }
+
+ /* advance input */
+ if(s)
+ s += w;
+ else
+ r++;
+ q->nbytesin += w;
+ q->nrunesin++;
+
+ /* advance output */
+ q->nbytesout += w;
+ q->nrunesout++;
+ }
+}
+
+static int
+qstrfmt(char *sin, Rune *rin, Quoteinfo *q, Fmt *f)
+{
+ Rune r, *rm, *rme;
+ char *t, *s, *m, *me;
+ Rune *rt, *rs;
+ ulong fl;
+ int nc, w;
+
+ m = sin;
+ me = m + q->nbytesin;
+ rm = rin;
+ rme = rm + q->nrunesin;
+
+ w = f->width;
+ fl = f->flags;
+ if(f->runes){
+ if(!(fl & FmtLeft) && __rfmtpad(f, w - q->nrunesout) < 0)
+ return -1;
+ }else{
+ if(!(fl & FmtLeft) && __fmtpad(f, w - q->nbytesout) < 0)
+ return -1;
+ }
+ t = (char*)f->to;
+ s = (char*)f->stop;
+ rt = (Rune*)f->to;
+ rs = (Rune*)f->stop;
+ if(f->runes)
+ FMTRCHAR(f, rt, rs, '\'');
+ else
+ FMTRUNE(f, t, s, '\'');
+ for(nc = q->nrunesin; nc > 0; nc--){
+ if(sin){
+ r = *(uchar*)m;
+ if(r < Runeself)
+ m++;
+ else if((me - m) >= UTFmax || fullrune(m, me-m))
+ m += chartorune(&r, m);
+ else
+ break;
+ }else{
+ if(rm >= rme)
+ break;
+ r = *(uchar*)rm++;
+ }
+ if(f->runes){
+ FMTRCHAR(f, rt, rs, r);
+ if(r == '\'')
+ FMTRCHAR(f, rt, rs, r);
+ }else{
+ FMTRUNE(f, t, s, r);
+ if(r == '\'')
+ FMTRUNE(f, t, s, r);
+ }
+ }
+
+ if(f->runes){
+ FMTRCHAR(f, rt, rs, '\'');
+ USED(rs);
+ f->nfmt += rt - (Rune *)f->to;
+ f->to = rt;
+ if(fl & FmtLeft && __rfmtpad(f, w - q->nrunesout) < 0)
+ return -1;
+ }else{
+ FMTRUNE(f, t, s, '\'');
+ USED(s);
+ f->nfmt += t - (char *)f->to;
+ f->to = t;
+ if(fl & FmtLeft && __fmtpad(f, w - q->nbytesout) < 0)
+ return -1;
+ }
+ return 0;
+}
+
+int
+__quotestrfmt(int runesin, Fmt *f)
+{
+ int outlen;
+ Rune *r;
+ char *s;
+ Quoteinfo q;
+
+ f->flags &= ~FmtPrec; /* ignored for %q %Q, so disable for %s %S in easy case */
+ if(runesin){
+ r = va_arg(f->args, Rune *);
+ s = nil;
+ }else{
+ s = va_arg(f->args, char *);
+ r = nil;
+ }
+ if(!s && !r)
+ return __fmtcpy(f, (void*)"<nil>", 5, 5);
+
+ if(f->flush)
+ outlen = 0x7FFFFFFF; /* if we can flush, no output limit */
+ else if(f->runes)
+ outlen = (Rune*)f->stop - (Rune*)f->to;
+ else
+ outlen = (char*)f->stop - (char*)f->to;
+
+ __quotesetup(s, r, -1, outlen, &q, f->flags&FmtSharp, f->runes);
+//print("bytes in %d bytes out %d runes in %d runesout %d\n", q.nbytesin, q.nbytesout, q.nrunesin, q.nrunesout);
+
+ if(runesin){
+ if(!q.quoted)
+ return __fmtrcpy(f, r, q.nrunesin);
+ return qstrfmt(nil, r, &q, f);
+ }
+
+ if(!q.quoted)
+ return __fmtcpy(f, s, q.nrunesin, q.nbytesin);
+ return qstrfmt(s, nil, &q, f);
+}
+
+int
+quotestrfmt(Fmt *f)
+{
+ return __quotestrfmt(0, f);
+}
+
+int
+quoterunestrfmt(Fmt *f)
+{
+ return __quotestrfmt(1, f);
+}
+
+void
+quotefmtinstall(void)
+{
+ fmtinstall('q', quotestrfmt);
+ fmtinstall('Q', quoterunestrfmt);
+}
+
+int
+__needsquotes(char *s, int *quotelenp)
+{
+ Quoteinfo q;
+
+ __quotesetup(s, nil, -1, 0x7FFFFFFF, &q, 0, 0);
+ *quotelenp = q.nbytesout;
+
+ return q.quoted;
+}
+
+int
+__runeneedsquotes(Rune *r, int *quotelenp)
+{
+ Quoteinfo q;
+
+ __quotesetup(nil, r, -1, 0x7FFFFFFF, &q, 0, 0);
+ *quotelenp = q.nrunesout;
+
+ return q.quoted;
+}
DIR diff --git a/src/libfmt/fmtrune.c b/src/libfmt/fmtrune.c
t@@ -0,0 +1,40 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+fmtrune(Fmt *f, int r)
+{
+ Rune *rt;
+ char *t;
+ int n;
+
+ if(f->runes){
+ rt = (Rune*)f->to;
+ FMTRCHAR(f, rt, f->stop, r);
+ f->to = rt;
+ n = 1;
+ }else{
+ t = (char*)f->to;
+ FMTRUNE(f, t, f->stop, r);
+ n = t - (char*)f->to;
+ f->to = t;
+ }
+ f->nfmt += n;
+ return 0;
+}
DIR diff --git a/src/libfmt/fmtstr.c b/src/libfmt/fmtstr.c
t@@ -0,0 +1,65 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include <stdlib.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+static int
+fmtStrFlush(Fmt *f)
+{
+ char *s;
+ int n;
+
+ n = (int)f->farg;
+ n += 256;
+ f->farg = (void*)n;
+ s = (char*)f->start;
+ f->start = realloc(s, n);
+ if(f->start == nil){
+ f->start = s;
+ return 0;
+ }
+ f->to = (char*)f->start + ((char*)f->to - s);
+ f->stop = (char*)f->start + n - 1;
+ return 1;
+}
+
+int
+fmtstrinit(Fmt *f)
+{
+ int n;
+
+ f->runes = 0;
+ n = 32;
+ f->start = malloc(n);
+ if(f->start == nil)
+ return -1;
+ f->to = f->start;
+ f->stop = (char*)f->start + n - 1;
+ f->flush = fmtStrFlush;
+ f->farg = (void*)n;
+ f->nfmt = 0;
+ return 0;
+}
+
+char*
+fmtstrflush(Fmt *f)
+{
+ *(char*)f->to = '\0';
+ f->to = f->start;
+ return (char*)f->start;
+}
DIR diff --git a/src/libfmt/fmtvprint.c b/src/libfmt/fmtvprint.c
t@@ -0,0 +1,46 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+
+/*
+ * format a string into the output buffer
+ * designed for formats which themselves call fmt,
+ * but ignore any width flags
+ */
+int
+fmtvprint(Fmt *f, char *fmt, va_list args)
+{
+ va_list va;
+ int n;
+
+ f->flags = 0;
+ f->width = 0;
+ f->prec = 0;
+ va = f->args;
+ f->args = args;
+ n = dofmt(f, fmt);
+ f->flags = 0;
+ f->width = 0;
+ f->prec = 0;
+ f->args = va;
+ if(n >= 0)
+ return 0;
+ return n;
+}
+
DIR diff --git a/src/libfmt/fprint.c b/src/libfmt/fprint.c
t@@ -0,0 +1,28 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "utf.h"
+#include "fmt.h"
+
+int
+fprint(int fd, char *fmt, ...)
+{
+ int n;
+ va_list args;
+
+ va_start(args, fmt);
+ n = vfprint(fd, fmt, args);
+ va_end(args);
+ return n;
+}
DIR diff --git a/src/libfmt/mkfile b/src/libfmt/mkfile
t@@ -0,0 +1 @@
+<../libutf/mkfile
DIR diff --git a/src/libfmt/nan.h b/src/libfmt/nan.h
t@@ -0,0 +1,4 @@
+extern double __NaN(void);
+extern double __Inf(int);
+extern int __isNaN(double);
+extern int __isInf(double, int);
DIR diff --git a/src/libfmt/nan64.c b/src/libfmt/nan64.c
t@@ -0,0 +1,72 @@
+/*
+ * 64-bit IEEE not-a-number routines.
+ * This is big/little-endian portable assuming that
+ * the 64-bit doubles and 64-bit integers have the
+ * same byte ordering.
+ */
+
+#include "nan.h"
+
+typedef unsigned long long uvlong;
+typedef unsigned long ulong;
+
+#ifdef NEEDLL
+static uvlong uvnan = 0x7FF0000000000001LL;
+static uvlong uvinf = 0x7FF0000000000000LL;
+static uvlong uvneginf = 0xFFF0000000000000LL;
+#else
+static uvlong uvnan = 0x7FF0000000000001;
+static uvlong uvinf = 0x7FF0000000000000;
+static uvlong uvneginf = 0xFFF0000000000000;
+#endif
+
+double
+__NaN(void)
+{
+ uvlong *p;
+
+ /* gcc complains about "return *(double*)&uvnan;" */
+ p = &uvnan;
+ return *(double*)p;
+}
+
+int
+__isNaN(double d)
+{
+ uvlong x;
+ double *p;
+
+ p = &d;
+ x = *(uvlong*)p;
+ return (ulong)(x>>32)==0x7FF00000 && !__isInf(d, 0);
+}
+
+double
+__Inf(int sign)
+{
+ uvlong *p;
+
+ if(sign < 0)
+ p = &uvinf;
+ else
+ p = &uvneginf;
+ return *(double*)p;
+}
+
+int
+__isInf(double d, int sign)
+{
+ uvlong x;
+ double *p;
+
+ p = &d;
+ x = *(uvlong*)p;
+ if(sign == 0)
+ return x==uvinf || x==uvneginf;
+ else if(sign > 0)
+ return x==uvinf;
+ else
+ return x==uvneginf;
+}
+
+
DIR diff --git a/src/libfmt/pow10.c b/src/libfmt/pow10.c
t@@ -0,0 +1,57 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * this table might overflow 127-bit exponent representations.
+ * in that case, truncate it after 1.0e38.
+ * it is important to get all one can from this
+ * routine since it is used in atof to scale numbers.
+ * the presumption is that C converts fp numbers better
+ * than multipication of lower powers of 10.
+ */
+
+static
+double tab[] =
+{
+ 1.0e0, 1.0e1, 1.0e2, 1.0e3, 1.0e4, 1.0e5, 1.0e6, 1.0e7, 1.0e8, 1.0e9,
+ 1.0e10,1.0e11,1.0e12,1.0e13,1.0e14,1.0e15,1.0e16,1.0e17,1.0e18,1.0e19,
+ 1.0e20,1.0e21,1.0e22,1.0e23,1.0e24,1.0e25,1.0e26,1.0e27,1.0e28,1.0e29,
+ 1.0e30,1.0e31,1.0e32,1.0e33,1.0e34,1.0e35,1.0e36,1.0e37,1.0e38,1.0e39,
+ 1.0e40,1.0e41,1.0e42,1.0e43,1.0e44,1.0e45,1.0e46,1.0e47,1.0e48,1.0e49,
+ 1.0e50,1.0e51,1.0e52,1.0e53,1.0e54,1.0e55,1.0e56,1.0e57,1.0e58,1.0e59,
+ 1.0e60,1.0e61,1.0e62,1.0e63,1.0e64,1.0e65,1.0e66,1.0e67,1.0e68,1.0e69,
+};
+
+double
+__fmtpow10(int n)
+{
+ int m;
+
+ if(n < 0) {
+ n = -n;
+ if(n < (int)(sizeof(tab)/sizeof(tab[0])))
+ return 1/tab[n];
+ m = n/2;
+ return __fmtpow10(-m) * __fmtpow10(m-n);
+ }
+ if(n < (int)(sizeof(tab)/sizeof(tab[0])))
+ return tab[n];
+ m = n/2;
+ return __fmtpow10(m) * __fmtpow10(n-m);
+}
DIR diff --git a/src/libfmt/print.3 b/src/libfmt/print.3
t@@ -0,0 +1,469 @@
+.TH PRINT 3
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.SH NAME
+print, fprint, sprint, snprint, seprint, smprint, vfprint, vsnprint, vseprint, vsmprint \- print formatted output
+.SH SYNOPSIS
+.B #include <utf.h>
+.PP
+.B #include <fmt.h>
+.PP
+.ta \w'\fLchar* 'u
+.B
+int print(char *format, ...)
+.PP
+.B
+int fprint(int fd, char *format, ...)
+.PP
+.B
+int sprint(char *s, char *format, ...)
+.PP
+.B
+int snprint(char *s, int len, char *format, ...)
+.PP
+.B
+char* seprint(char *s, char *e, char *format, ...)
+.PP
+.B
+char* smprint(char *format, ...)
+.PP
+.B
+int runesprint(Rune *s, char *format, ...)
+.PP
+.B
+int runesnprint(Rune *s, int len, char *format, ...)
+.PP
+.B
+Rune* runeseprint(Rune *s, Rune *e, char *format, ...)
+.PP
+.B
+Rune* runesmprint(char *format, ...)
+.PP
+.B
+int vfprint(int fd, char *format, va_list v)
+.PP
+.B
+int vsnprint(char *s, int len, char *format, va_list v)
+.PP
+.B
+char* vseprint(char *s, char *e, char *format, va_list v)
+.PP
+.B
+char* vsmprint(char *format, va_list v)
+.PP
+.B
+int runevsnprint(Rune *s, int len, char *format, va_list v)
+.PP
+.B
+Rune* runevseprint(Rune *s, Rune *e, char *format, va_list v)
+.PP
+.B
+Rune* runevsmprint(Rune *format, va_list v)
+.PP
+.B
+.SH DESCRIPTION
+.I Print
+writes text to the standard output.
+.I Fprint
+writes to the named output
+file descriptor.
+.I Sprint
+places text
+followed by the NUL character
+.RB ( \e0 )
+in consecutive bytes starting at
+.IR s ;
+it is the user's responsibility to ensure that
+enough storage is available.
+Each function returns the number of bytes
+transmitted (not including the NUL
+in the case of
+.IR sprint ),
+or
+a negative value if an output error was encountered.
+.PP
+.I Snprint
+is like
+.IR sprint ,
+but will not place more than
+.I len
+bytes in
+.IR s .
+Its result is always NUL-terminated and holds the maximal
+number of characters that can fit.
+.I Seprint
+is like
+.IR snprint ,
+except that the end is indicated by a pointer
+.I e
+rather than a count and the return value points to the terminating NUL of the
+resulting string.
+.I Smprint
+is like
+.IR sprint ,
+except that it prints into and returns a string of the required length, which is
+allocated by
+.IR malloc (3).
+.PP
+The routines
+.IR runesprint ,
+.IR runesnprint ,
+.IR runeseprint ,
+and
+.I runesmprint
+are the same as
+.IR sprint ,
+.IR snprint ,
+.IR seprint
+and
+.I smprint
+except that their output is rune strings instead of byte strings.
+.PP
+Finally, the routines
+.IR vfprint ,
+.IR vsnprint ,
+.IR vseprint ,
+.IR vsmprint ,
+.IR runevsnprint ,
+.IR runevseprint ,
+and
+.I runevsmprint
+are like their
+.BR v-less
+relatives except they take as arguments a
+.B va_list
+parameter, so they can be called within a variadic function.
+The Example section shows a representative usage.
+.PP
+Each of these functions
+converts, formats, and prints its
+trailing arguments
+under control of a
+.IR format
+string.
+The
+format
+contains two types of objects:
+plain characters, which are simply copied to the
+output stream,
+and conversion specifications,
+each of which results in fetching of
+zero or more
+arguments.
+The results are undefined if there are arguments of the
+wrong type or too few
+arguments for the format.
+If the format is exhausted while
+arguments remain, the excess
+is ignored.
+.PP
+Each conversion specification has the following format:
+.IP
+.B "% [flags] verb
+.PP
+The verb is a single character and each flag is a single character or a
+(decimal) numeric string.
+Up to two numeric strings may be used;
+the first is called
+.IR width ,
+the second
+.IR precision .
+A period can be used to separate them, and if the period is
+present then
+.I width
+and
+.I precision
+are taken to be zero if missing, otherwise they are `omitted'.
+Either or both of the numbers may be replaced with the character
+.BR * ,
+meaning that the actual number will be obtained from the argument list
+as an integer.
+The flags and numbers are arguments to
+the
+.I verb
+described below.
+.PP
+The numeric verbs
+.BR d ,
+.BR i ,
+.BR u ,
+.BR o ,
+.BR b ,
+.BR x ,
+and
+.B X
+format their arguments in decimal, decimal,
+unsigned decimal, octal, binary, hexadecimal, and upper case hexadecimal.
+Each interprets the flags
+.BR 0 ,
+.BR h ,
+.BR hh ,
+.BR l ,
+.BR + ,
+.BR - ,
+.BR , ,
+and
+.B #
+to mean pad with zeros,
+short, byte, long, always print a sign, left justified, commas every three digits,
+and alternate format.
+Also, a space character in the flag
+position is like
+.BR + ,
+but prints a space instead of a plus sign for non-negative values.
+If neither
+short nor long is specified,
+then the argument is an
+.BR int .
+If an unsigned verb is specified,
+then the argument is interpreted as a
+positive number and no sign is output;
+space and
+.B +
+flags are ignored for unsigned verbs.
+If two
+.B l
+flags are given,
+then the argument is interpreted as a
+.B vlong
+(usually an 8-byte, sometimes a 4-byte integer).
+If
+.I precision
+is not omitted, the number is padded on the left with zeros
+until at least
+.I precision
+digits appear.
+If
+.I precision
+is explicitly 0, and the number is 0,
+no digits are generated, and alternate formatting
+does not apply.
+Then, if alternate format is specified,
+for
+.B o
+conversion, the number is preceded by a
+.B 0
+if it doesn't already begin with one.
+For non-zero numbers and
+.B x
+conversion, the number is preceded by
+.BR 0x ;
+for
+.B X
+conversion, the number is preceded by
+.BR 0X .
+Finally, if
+.I width
+is not omitted, the number is padded on the left (or right, if
+left justification is specified) with enough blanks to
+make the field at least
+.I width
+characters long.
+.PP
+The floating point verbs
+.BR f ,
+.BR e ,
+.BR E ,
+.BR g ,
+and
+.B G
+take a
+.B double
+argument.
+Each interprets the flags
+.BR 0 ,
+.BR L
+.BR + ,
+.BR - ,
+and
+.B #
+to mean pad with zeros,
+long double argument,
+always print a sign,
+left justified,
+and
+alternate format.
+.I Width
+is the minimum field width and,
+if the converted value takes up less than
+.I width
+characters, it is padded on the left (or right, if `left justified')
+with spaces.
+.I Precision
+is the number of digits that are converted after the decimal place for
+.BR e ,
+.BR E ,
+and
+.B f
+conversions,
+and
+.I precision
+is the maximum number of significant digits for
+.B g
+and
+.B G
+conversions.
+The
+.B f
+verb produces output of the form
+.RB [ - ] digits [ .digits\fR].
+.B E
+conversion appends an exponent
+.BR E [ - ] digits ,
+and
+.B e
+conversion appends an exponent
+.BR e [ - ] digits .
+The
+.B g
+verb will output the argument in either
+.B e
+or
+.B f
+with the goal of producing the smallest output.
+Also, trailing zeros are omitted from the fraction part of
+the output, and a trailing decimal point appears only if it is followed
+by a digit.
+The
+.B G
+verb is similar, but uses
+.B E
+format instead of
+.BR e .
+When alternate format is specified, the result will always contain a decimal point,
+and for
+.B g
+and
+.B G
+conversions, trailing zeros are not removed.
+.PP
+The
+.B s
+verb copies a string
+(pointer to
+.BR char )
+to the output.
+The number of characters copied
+.RI ( n )
+is the minimum
+of the size of the string and
+.IR precision .
+These
+.I n
+characters are justified within a field of
+.I width
+characters as described above.
+If a
+.I precision
+is given, it is safe for the string not to be nul-terminated
+as long as it is at least
+.I precision
+characters (not bytes!) long.
+The
+.B S
+verb is similar, but it interprets its pointer as an array
+of runes (see
+.IR utf (7));
+the runes are converted to
+.SM UTF
+before output.
+.PP
+The
+.B c
+verb copies a single
+.B char
+(promoted to
+.BR int )
+justified within a field of
+.I width
+characters as described above.
+The
+.B C
+verb is similar, but works on runes.
+.PP
+The
+.B p
+verb formats a pointer value.
+At the moment, it is a synonym for
+.BR x ,
+but that will change if pointers and integers are different sizes.
+.PP
+The
+.B r
+verb takes no arguments; it copies the error string returned by a call to
+.IR strerror (3)
+with an argument of
+.IR errno.
+.PP
+Custom verbs may be installed using
+.IR fmtinstall (3).
+.SH EXAMPLE
+This function prints an error message with a variable
+number of arguments and then quits.
+.IP
+.EX
+.ta 6n +6n +6n
+void fatal(char *msg, ...)
+{
+ char buf[1024], *out;
+ va_list arg;
+
+ out = vseprint(buf, buf+sizeof buf, "Fatal error: ");
+ va_start(arg, msg);
+ out = vseprint(out, buf+sizeof buf, msg, arg);
+ va_end(arg);
+ write(2, buf, out-buf);
+ exit(1);
+}
+.EE
+.SH SEE ALSO
+.IR fmtinstall (3),
+.IR fprintf (3),
+.IR utf (7)
+.SH DIAGNOSTICS
+Routines that write to a file descriptor or call
+.IR malloc
+set
+.IR errstr .
+.SH BUGS
+The formatting is close to that specified for ANSI
+.IR fprintf (3);
+the main difference is that
+.B b
+and
+.B r
+are not in ANSI and some
+.B C9X
+verbs are missing.
+Also, and distinctly not a bug,
+.I print
+and friends generate
+.SM UTF
+rather than
+.SM ASCII.
+.PP
+There is no
+.BR runeprint ,
+.BR runefprint ,
+etc. because runes are byte-order dependent and should not be written directly to a file; use the
+UTF output of
+.I print
+or
+.I fprint
+instead.
+Also,
+.I sprint
+is deprecated for safety reasons; use
+.IR snprint ,
+.IR seprint ,
+or
+.I smprint
+instead.
+Safety also precludes the existence of
+.IR runesprint .
DIR diff --git a/src/libfmt/print.c b/src/libfmt/print.c
t@@ -0,0 +1,28 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "utf.h"
+#include "fmt.h"
+
+int
+print(char *fmt, ...)
+{
+ int n;
+ va_list args;
+
+ va_start(args, fmt);
+ n = vfprint(1, fmt, args);
+ va_end(args);
+ return n;
+}
DIR diff --git a/src/libfmt/rpm.spec b/src/libfmt/rpm.spec
t@@ -0,0 +1,34 @@
+Summary: Extensible formatted print library. (Printf with user-defined verbs.)
+Name: libfmt
+Version: 2.0
+Release: 1
+Group: Development/C
+Copyright: BSD-like
+Packager: Russ Cox <rsc@post.harvard.edu>
+Source: http://pdos.lcs.mit.edu/~rsc/software/libfmt-2.0.tgz
+URL: http://pdos.lcs.mit.edu/~rsc/software/#libfmt
+Requires: libutf
+
+%description
+Libfmt is a port of Plan 9's formatted print library.
+As a base it provides all the syntax of ANSI printf
+but adds the ability for client programs to install
+new print verbs. One such print verb (installed by
+default) is %r, which prints the system error string.
+Instead of perror("foo"), you can write fprint(2, "foo: %r\n").
+This is especially nice when you write verbs to format
+the data structures used by your particular program.
+%prep
+%setup
+
+%build
+make
+
+%install
+make install
+
+%files
+/usr/local/include/fmt.h
+/usr/local/lib/libfmt.a
+/usr/local/man/man3/print.3
+/usr/local/man/man3/fmtinstall.3
DIR diff --git a/src/libfmt/runefmtstr.c b/src/libfmt/runefmtstr.c
t@@ -0,0 +1,65 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include <stdlib.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+static int
+runeFmtStrFlush(Fmt *f)
+{
+ Rune *s;
+ int n;
+
+ n = (int)f->farg;
+ n += 256;
+ f->farg = (void*)n;
+ s = (Rune*)f->start;
+ f->start = realloc(s, sizeof(Rune)*n);
+ if(f->start == nil){
+ f->start = s;
+ return 0;
+ }
+ f->to = (Rune*)f->start + ((Rune*)f->to - s);
+ f->stop = (Rune*)f->start + n - 1;
+ return 1;
+}
+
+int
+runefmtstrinit(Fmt *f)
+{
+ int n;
+
+ f->runes = 1;
+ n = 32;
+ f->start = malloc(sizeof(Rune)*n);
+ if(f->start == nil)
+ return -1;
+ f->to = f->start;
+ f->stop = (Rune*)f->start + n - 1;
+ f->flush = runeFmtStrFlush;
+ f->farg = (void*)n;
+ f->nfmt = 0;
+ return 0;
+}
+
+Rune*
+runefmtstrflush(Fmt *f)
+{
+ *(Rune*)f->to = '\0';
+ f->to = f->start;
+ return f->start;
+}
DIR diff --git a/src/libfmt/runeseprint.c b/src/libfmt/runeseprint.c
t@@ -0,0 +1,30 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+Rune*
+runeseprint(Rune *buf, Rune *e, char *fmt, ...)
+{
+ Rune *p;
+ va_list args;
+
+ va_start(args, fmt);
+ p = runevseprint(buf, e, fmt, args);
+ va_end(args);
+ return p;
+}
DIR diff --git a/src/libfmt/runesmprint.c b/src/libfmt/runesmprint.c
t@@ -0,0 +1,30 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+Rune*
+runesmprint(char *fmt, ...)
+{
+ va_list args;
+ Rune *p;
+
+ va_start(args, fmt);
+ p = runevsmprint(fmt, args);
+ va_end(args);
+ return p;
+}
DIR diff --git a/src/libfmt/runesnprint.c b/src/libfmt/runesnprint.c
t@@ -0,0 +1,31 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+runesnprint(Rune *buf, int len, char *fmt, ...)
+{
+ int n;
+ va_list args;
+
+ va_start(args, fmt);
+ n = runevsnprint(buf, len, fmt, args);
+ va_end(args);
+ return n;
+}
+
DIR diff --git a/src/libfmt/runesprint.c b/src/libfmt/runesprint.c
t@@ -0,0 +1,30 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+runesprint(Rune *buf, char *fmt, ...)
+{
+ int n;
+ va_list args;
+
+ va_start(args, fmt);
+ n = runevsnprint(buf, 256, fmt, args);
+ va_end(args);
+ return n;
+}
DIR diff --git a/src/libfmt/runevseprint.c b/src/libfmt/runevseprint.c
t@@ -0,0 +1,39 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+Rune*
+runevseprint(Rune *buf, Rune *e, char *fmt, va_list args)
+{
+ Fmt f;
+
+ if(e <= buf)
+ return nil;
+ f.runes = 1;
+ f.start = buf;
+ f.to = buf;
+ f.stop = e - 1;
+ f.flush = nil;
+ f.farg = nil;
+ f.nfmt = 0;
+ f.args = args;
+ dofmt(&f, fmt);
+ *(Rune*)f.to = '\0';
+ return (Rune*)f.to;
+}
+
DIR diff --git a/src/libfmt/runevsmprint.c b/src/libfmt/runevsmprint.c
t@@ -0,0 +1,37 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <stdlib.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * print into an allocated string buffer
+ */
+Rune*
+runevsmprint(char *fmt, va_list args)
+{
+ Fmt f;
+ int n;
+
+ if(runefmtstrinit(&f) < 0)
+ return nil;
+ f.args = args;
+ n = dofmt(&f, fmt);
+ if(n < 0)
+ return nil;
+ *(Rune*)f.to = '\0';
+ return (Rune*)f.start;
+}
DIR diff --git a/src/libfmt/runevsnprint.c b/src/libfmt/runevsnprint.c
t@@ -0,0 +1,38 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+runevsnprint(Rune *buf, int len, char *fmt, va_list args)
+{
+ Fmt f;
+
+ if(len <= 0)
+ return -1;
+ f.runes = 1;
+ f.start = buf;
+ f.to = buf;
+ f.stop = buf + len - 1;
+ f.flush = nil;
+ f.farg = nil;
+ f.nfmt = 0;
+ f.args = args;
+ dofmt(&f, fmt);
+ *(Rune*)f.to = '\0';
+ return (Rune*)f.to - buf;
+}
DIR diff --git a/src/libfmt/seprint.c b/src/libfmt/seprint.c
t@@ -0,0 +1,27 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+
+char*
+seprint(char *buf, char *e, char *fmt, ...)
+{
+ char *p;
+ va_list args;
+
+ va_start(args, fmt);
+ p = vseprint(buf, e, fmt, args);
+ va_end(args);
+ return p;
+}
DIR diff --git a/src/libfmt/smprint.c b/src/libfmt/smprint.c
t@@ -0,0 +1,27 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+
+char*
+smprint(char *fmt, ...)
+{
+ va_list args;
+ char *p;
+
+ va_start(args, fmt);
+ p = vsmprint(fmt, args);
+ va_end(args);
+ return p;
+}
DIR diff --git a/src/libfmt/snprint.c b/src/libfmt/snprint.c
t@@ -0,0 +1,28 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+
+int
+snprint(char *buf, int len, char *fmt, ...)
+{
+ int n;
+ va_list args;
+
+ va_start(args, fmt);
+ n = vsnprint(buf, len, fmt, args);
+ va_end(args);
+ return n;
+}
+
DIR diff --git a/src/libfmt/sprint.c b/src/libfmt/sprint.c
t@@ -0,0 +1,27 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+
+int
+sprint(char *buf, char *fmt, ...)
+{
+ int n;
+ va_list args;
+
+ va_start(args, fmt);
+ n = vsnprint(buf, 65536, fmt, args); /* big number, but sprint is deprecated anyway */
+ va_end(args);
+ return n;
+}
DIR diff --git a/src/libfmt/strtod.c b/src/libfmt/strtod.c
t@@ -0,0 +1,539 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdlib.h>
+#include <math.h>
+#include <ctype.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include "fmt.h"
+#include "nan.h"
+
+#ifndef nelem
+#define nelem(x) (sizeof(x)/sizeof *(x))
+#endif
+#define nil ((void*)0)
+#define ulong _fmtulong
+typedef unsigned long ulong;
+
+static ulong
+umuldiv(ulong a, ulong b, ulong c)
+{
+ double d;
+
+ d = ((double)a * (double)b) / (double)c;
+ if(d >= 4294967295.)
+ d = 4294967295.;
+ return (ulong)d;
+}
+
+/*
+ * This routine will convert to arbitrary precision
+ * floating point entirely in multi-precision fixed.
+ * The answer is the closest floating point number to
+ * the given decimal number. Exactly half way are
+ * rounded ala ieee rules.
+ * Method is to scale input decimal between .500 and .999...
+ * with external power of 2, then binary search for the
+ * closest mantissa to this decimal number.
+ * Nmant is is the required precision. (53 for ieee dp)
+ * Nbits is the max number of bits/word. (must be <= 28)
+ * Prec is calculated - the number of words of fixed mantissa.
+ */
+enum
+{
+ Nbits = 28, /* bits safely represented in a ulong */
+ Nmant = 53, /* bits of precision required */
+ Prec = (Nmant+Nbits+1)/Nbits, /* words of Nbits each to represent mantissa */
+ Sigbit = 1<<(Prec*Nbits-Nmant), /* first significant bit of Prec-th word */
+ Ndig = 1500,
+ One = (ulong)(1<<Nbits),
+ Half = (ulong)(One>>1),
+ Maxe = 310,
+
+ Fsign = 1<<0, /* found - */
+ Fesign = 1<<1, /* found e- */
+ Fdpoint = 1<<2, /* found . */
+
+ S0 = 0, /* _ _S0 +S1 #S2 .S3 */
+ S1, /* _+ #S2 .S3 */
+ S2, /* _+# #S2 .S4 eS5 */
+ S3, /* _+. #S4 */
+ S4, /* _+#.# #S4 eS5 */
+ S5, /* _+#.#e +S6 #S7 */
+ S6, /* _+#.#e+ #S7 */
+ S7, /* _+#.#e+# #S7 */
+};
+
+static int xcmp(char*, char*);
+static int fpcmp(char*, ulong*);
+static void frnorm(ulong*);
+static void divascii(char*, int*, int*, int*);
+static void mulascii(char*, int*, int*, int*);
+
+typedef struct Tab Tab;
+struct Tab
+{
+ int bp;
+ int siz;
+ char* cmp;
+};
+
+double
+fmtstrtod(const char *as, char **aas)
+{
+ int na, ex, dp, bp, c, i, flag, state;
+ ulong low[Prec], hig[Prec], mid[Prec];
+ double d;
+ char *s, a[Ndig];
+
+ flag = 0; /* Fsign, Fesign, Fdpoint */
+ na = 0; /* number of digits of a[] */
+ dp = 0; /* na of decimal point */
+ ex = 0; /* exonent */
+
+ state = S0;
+ for(s=(char*)as;; s++) {
+ c = *s;
+ if(c >= '0' && c <= '9') {
+ switch(state) {
+ case S0:
+ case S1:
+ case S2:
+ state = S2;
+ break;
+ case S3:
+ case S4:
+ state = S4;
+ break;
+
+ case S5:
+ case S6:
+ case S7:
+ state = S7;
+ ex = ex*10 + (c-'0');
+ continue;
+ }
+ if(na == 0 && c == '0') {
+ dp--;
+ continue;
+ }
+ if(na < Ndig-50)
+ a[na++] = c;
+ continue;
+ }
+ switch(c) {
+ case '\t':
+ case '\n':
+ case '\v':
+ case '\f':
+ case '\r':
+ case ' ':
+ if(state == S0)
+ continue;
+ break;
+ case '-':
+ if(state == S0)
+ flag |= Fsign;
+ else
+ flag |= Fesign;
+ case '+':
+ if(state == S0)
+ state = S1;
+ else
+ if(state == S5)
+ state = S6;
+ else
+ break; /* syntax */
+ continue;
+ case '.':
+ flag |= Fdpoint;
+ dp = na;
+ if(state == S0 || state == S1) {
+ state = S3;
+ continue;
+ }
+ if(state == S2) {
+ state = S4;
+ continue;
+ }
+ break;
+ case 'e':
+ case 'E':
+ if(state == S2 || state == S4) {
+ state = S5;
+ continue;
+ }
+ break;
+ }
+ break;
+ }
+
+ /*
+ * clean up return char-pointer
+ */
+ switch(state) {
+ case S0:
+ if(xcmp(s, "nan") == 0) {
+ if(aas != nil)
+ *aas = s+3;
+ goto retnan;
+ }
+ case S1:
+ if(xcmp(s, "infinity") == 0) {
+ if(aas != nil)
+ *aas = s+8;
+ goto retinf;
+ }
+ if(xcmp(s, "inf") == 0) {
+ if(aas != nil)
+ *aas = s+3;
+ goto retinf;
+ }
+ case S3:
+ if(aas != nil)
+ *aas = (char*)as;
+ goto ret0; /* no digits found */
+ case S6:
+ s--; /* back over +- */
+ case S5:
+ s--; /* back over e */
+ break;
+ }
+ if(aas != nil)
+ *aas = s;
+
+ if(flag & Fdpoint)
+ while(na > 0 && a[na-1] == '0')
+ na--;
+ if(na == 0)
+ goto ret0; /* zero */
+ a[na] = 0;
+ if(!(flag & Fdpoint))
+ dp = na;
+ if(flag & Fesign)
+ ex = -ex;
+ dp += ex;
+ if(dp < -Maxe){
+ errno = ERANGE;
+ goto ret0; /* underflow by exp */
+ } else
+ if(dp > +Maxe)
+ goto retinf; /* overflow by exp */
+
+ /*
+ * normalize the decimal ascii number
+ * to range .[5-9][0-9]* e0
+ */
+ bp = 0; /* binary exponent */
+ while(dp > 0)
+ divascii(a, &na, &dp, &bp);
+ while(dp < 0 || a[0] < '5')
+ mulascii(a, &na, &dp, &bp);
+
+ /* close approx by naive conversion */
+ mid[0] = 0;
+ mid[1] = 1;
+ for(i=0; c=a[i]; i++) {
+ mid[0] = mid[0]*10 + (c-'0');
+ mid[1] = mid[1]*10;
+ if(i >= 8)
+ break;
+ }
+ low[0] = umuldiv(mid[0], One, mid[1]);
+ hig[0] = umuldiv(mid[0]+1, One, mid[1]);
+ for(i=1; i<Prec; i++) {
+ low[i] = 0;
+ hig[i] = One-1;
+ }
+
+ /* binary search for closest mantissa */
+ for(;;) {
+ /* mid = (hig + low) / 2 */
+ c = 0;
+ for(i=0; i<Prec; i++) {
+ mid[i] = hig[i] + low[i];
+ if(c)
+ mid[i] += One;
+ c = mid[i] & 1;
+ mid[i] >>= 1;
+ }
+ frnorm(mid);
+
+ /* compare */
+ c = fpcmp(a, mid);
+ if(c > 0) {
+ c = 1;
+ for(i=0; i<Prec; i++)
+ if(low[i] != mid[i]) {
+ c = 0;
+ low[i] = mid[i];
+ }
+ if(c)
+ break; /* between mid and hig */
+ continue;
+ }
+ if(c < 0) {
+ for(i=0; i<Prec; i++)
+ hig[i] = mid[i];
+ continue;
+ }
+
+ /* only hard part is if even/odd roundings wants to go up */
+ c = mid[Prec-1] & (Sigbit-1);
+ if(c == Sigbit/2 && (mid[Prec-1]&Sigbit) == 0)
+ mid[Prec-1] -= c;
+ break; /* exactly mid */
+ }
+
+ /* normal rounding applies */
+ c = mid[Prec-1] & (Sigbit-1);
+ mid[Prec-1] -= c;
+ if(c >= Sigbit/2) {
+ mid[Prec-1] += Sigbit;
+ frnorm(mid);
+ }
+ goto out;
+
+ret0:
+ return 0;
+
+retnan:
+ return __NaN();
+
+retinf:
+ /*
+ * Unix strtod requires these. Plan 9 would return Inf(0) or Inf(-1). */
+ errno = ERANGE;
+ if(flag & Fsign)
+ return -HUGE_VAL;
+ return HUGE_VAL;
+
+out:
+ d = 0;
+ for(i=0; i<Prec; i++)
+ d = d*One + mid[i];
+ if(flag & Fsign)
+ d = -d;
+ d = ldexp(d, bp - Prec*Nbits);
+ if(d == 0){ /* underflow */
+ errno = ERANGE;
+ }
+ return d;
+}
+
+static void
+frnorm(ulong *f)
+{
+ int i, c;
+
+ c = 0;
+ for(i=Prec-1; i>0; i--) {
+ f[i] += c;
+ c = f[i] >> Nbits;
+ f[i] &= One-1;
+ }
+ f[0] += c;
+}
+
+static int
+fpcmp(char *a, ulong* f)
+{
+ ulong tf[Prec];
+ int i, d, c;
+
+ for(i=0; i<Prec; i++)
+ tf[i] = f[i];
+
+ for(;;) {
+ /* tf *= 10 */
+ for(i=0; i<Prec; i++)
+ tf[i] = tf[i]*10;
+ frnorm(tf);
+ d = (tf[0] >> Nbits) + '0';
+ tf[0] &= One-1;
+
+ /* compare next digit */
+ c = *a;
+ if(c == 0) {
+ if('0' < d)
+ return -1;
+ if(tf[0] != 0)
+ goto cont;
+ for(i=1; i<Prec; i++)
+ if(tf[i] != 0)
+ goto cont;
+ return 0;
+ }
+ if(c > d)
+ return +1;
+ if(c < d)
+ return -1;
+ a++;
+ cont:;
+ }
+ return 0;
+}
+
+static void
+divby(char *a, int *na, int b)
+{
+ int n, c;
+ char *p;
+
+ p = a;
+ n = 0;
+ while(n>>b == 0) {
+ c = *a++;
+ if(c == 0) {
+ while(n) {
+ c = n*10;
+ if(c>>b)
+ break;
+ n = c;
+ }
+ goto xx;
+ }
+ n = n*10 + c-'0';
+ (*na)--;
+ }
+ for(;;) {
+ c = n>>b;
+ n -= c<<b;
+ *p++ = c + '0';
+ c = *a++;
+ if(c == 0)
+ break;
+ n = n*10 + c-'0';
+ }
+ (*na)++;
+xx:
+ while(n) {
+ n = n*10;
+ c = n>>b;
+ n -= c<<b;
+ *p++ = c + '0';
+ (*na)++;
+ }
+ *p = 0;
+}
+
+static Tab tab1[] =
+{
+ 1, 0, "",
+ 3, 1, "7",
+ 6, 2, "63",
+ 9, 3, "511",
+ 13, 4, "8191",
+ 16, 5, "65535",
+ 19, 6, "524287",
+ 23, 7, "8388607",
+ 26, 8, "67108863",
+ 27, 9, "134217727",
+};
+
+static void
+divascii(char *a, int *na, int *dp, int *bp)
+{
+ int b, d;
+ Tab *t;
+
+ d = *dp;
+ if(d >= (int)(nelem(tab1)))
+ d = (int)(nelem(tab1))-1;
+ t = tab1 + d;
+ b = t->bp;
+ if(memcmp(a, t->cmp, t->siz) > 0)
+ d--;
+ *dp -= d;
+ *bp += b;
+ divby(a, na, b);
+}
+
+static void
+mulby(char *a, char *p, char *q, int b)
+{
+ int n, c;
+
+ n = 0;
+ *p = 0;
+ for(;;) {
+ q--;
+ if(q < a)
+ break;
+ c = *q - '0';
+ c = (c<<b) + n;
+ n = c/10;
+ c -= n*10;
+ p--;
+ *p = c + '0';
+ }
+ while(n) {
+ c = n;
+ n = c/10;
+ c -= n*10;
+ p--;
+ *p = c + '0';
+ }
+}
+
+static Tab tab2[] =
+{
+ 1, 1, "", /* dp = 0-0 */
+ 3, 3, "125",
+ 6, 5, "15625",
+ 9, 7, "1953125",
+ 13, 10, "1220703125",
+ 16, 12, "152587890625",
+ 19, 14, "19073486328125",
+ 23, 17, "11920928955078125",
+ 26, 19, "1490116119384765625",
+ 27, 19, "7450580596923828125", /* dp 8-9 */
+};
+
+static void
+mulascii(char *a, int *na, int *dp, int *bp)
+{
+ char *p;
+ int d, b;
+ Tab *t;
+
+ d = -*dp;
+ if(d >= (int)(nelem(tab2)))
+ d = (int)(nelem(tab2))-1;
+ t = tab2 + d;
+ b = t->bp;
+ if(memcmp(a, t->cmp, t->siz) < 0)
+ d--;
+ p = a + *na;
+ *bp -= b;
+ *dp += d;
+ *na += d;
+ mulby(a, p+d, p, b);
+}
+
+static int
+xcmp(char *a, char *b)
+{
+ int c1, c2;
+
+ while(c1 = *b++) {
+ c2 = *a++;
+ if(isupper(c2))
+ c2 = tolower(c2);
+ if(c1 != c2)
+ return 1;
+ }
+ return 0;
+}
DIR diff --git a/src/libfmt/strtod.h b/src/libfmt/strtod.h
t@@ -0,0 +1,4 @@
+extern double __NaN(void);
+extern double __Inf(int);
+extern double __isNaN(double);
+extern double __isInf(double, int);
DIR diff --git a/src/libfmt/test.c b/src/libfmt/test.c
t@@ -0,0 +1,39 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <utf.h>
+#include "fmt.h"
+
+int
+main(int argc, char *argv[])
+{
+ quotefmtinstall();
+ print("hello world\n");
+ print("x: %x\n", 0x87654321);
+ print("u: %u\n", 0x87654321);
+ print("d: %d\n", 0x87654321);
+ print("s: %s\n", "hi there");
+ print("q: %q\n", "hi i'm here");
+ print("c: %c\n", '!');
+ print("g: %g %g %g\n", 3.14159, 3.14159e10, 3.14159e-10);
+ print("e: %e %e %e\n", 3.14159, 3.14159e10, 3.14159e-10);
+ print("f: %f %f %f\n", 3.14159, 3.14159e10, 3.14159e-10);
+ print("smiley: %C\n", (Rune)0x263a);
+ print("%g %.18\n", 2e25, 2e25);
+ print("%2.18g\n", 1.0);
+ print("%f\n", 3.1415927/4);
+ print("%d\n", 23);
+ print("%i\n", 23);
+ return 0;
+}
DIR diff --git a/src/libfmt/vfprint.c b/src/libfmt/vfprint.c
t@@ -0,0 +1,31 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+vfprint(int fd, char *fmt, va_list args)
+{
+ Fmt f;
+ char buf[256];
+ int n;
+
+ fmtfdinit(&f, fd, buf, sizeof(buf));
+ f.args = args;
+ n = dofmt(&f, fmt);
+ if(n > 0 && __fmtFdFlush(&f) == 0)
+ return -1;
+ return n;
+}
DIR diff --git a/src/libfmt/vseprint.c b/src/libfmt/vseprint.c
t@@ -0,0 +1,37 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include "fmt.h"
+#include "fmtdef.h"
+
+char*
+vseprint(char *buf, char *e, char *fmt, va_list args)
+{
+ Fmt f;
+
+ if(e <= buf)
+ return nil;
+ f.runes = 0;
+ f.start = buf;
+ f.to = buf;
+ f.stop = e - 1;
+ f.flush = 0;
+ f.farg = nil;
+ f.nfmt = 0;
+ f.args = args;
+ dofmt(&f, fmt);
+ *(char*)f.to = '\0';
+ return (char*)f.to;
+}
+
DIR diff --git a/src/libfmt/vsmprint.c b/src/libfmt/vsmprint.c
t@@ -0,0 +1,36 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdlib.h>
+#include <stdarg.h>
+#include "fmt.h"
+#include "fmtdef.h"
+
+/*
+ * print into an allocated string buffer
+ */
+char*
+vsmprint(char *fmt, va_list args)
+{
+ Fmt f;
+ int n;
+
+ if(fmtstrinit(&f) < 0)
+ return nil;
+ f.args = args;
+ n = dofmt(&f, fmt);
+ if(n < 0)
+ return nil;
+ *(char*)f.to = '\0';
+ return (char*)f.start;
+}
DIR diff --git a/src/libfmt/vsnprint.c b/src/libfmt/vsnprint.c
t@@ -0,0 +1,37 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdlib.h>
+#include <stdarg.h>
+#include "fmt.h"
+#include "fmtdef.h"
+
+int
+vsnprint(char *buf, int len, char *fmt, va_list args)
+{
+ Fmt f;
+
+ if(len <= 0)
+ return -1;
+ f.runes = 0;
+ f.start = buf;
+ f.to = buf;
+ f.stop = buf + len - 1;
+ f.flush = 0;
+ f.farg = nil;
+ f.nfmt = 0;
+ f.args = args;
+ dofmt(&f, fmt);
+ *(char*)f.to = '\0';
+ return (char*)f.to - buf;
+}
DIR diff --git a/src/libregexp/LICENSE b/src/libregexp/LICENSE
t@@ -0,0 +1,258 @@
+The Plan 9 software is provided under the terms of the
+Lucent Public License, Version 1.02, reproduced below,
+with the following exceptions:
+
+1. No right is granted to create derivative works of or
+ to redistribute (other than with the Plan 9 Operating System)
+ the screen imprinter fonts identified in subdirectory
+ /lib/font/bit/lucida and printer fonts (Lucida Sans Unicode, Lucida
+ Sans Italic, Lucida Sans Demibold, Lucida Typewriter, Lucida Sans
+ Typewriter83), identified in subdirectory /sys/lib/postscript/font.
+ These directories contain material copyrights by B&H Inc. and Y&Y Inc.
+
+2. The printer fonts identified in subdirectory /sys/lib/ghostscript/font
+ are subject to the GNU GPL, reproduced in the file /LICENSE.gpl.
+
+3. The ghostscript program in the subdirectory /sys/src/cmd/gs is
+ covered by the Aladdin Free Public License, reproduced in the file
+ /LICENSE.afpl.
+
+===================================================================
+
+Lucent Public License Version 1.02
+
+THE ACCOMPANYING PROGRAM IS PROVIDED UNDER THE TERMS OF THIS PUBLIC
+LICENSE ("AGREEMENT"). ANY USE, REPRODUCTION OR DISTRIBUTION OF THE
+PROGRAM CONSTITUTES RECIPIENT'S ACCEPTANCE OF THIS AGREEMENT.
+
+1. DEFINITIONS
+
+"Contribution" means:
+
+ a. in the case of Lucent Technologies Inc. ("LUCENT"), the Original
+ Program, and
+ b. in the case of each Contributor,
+
+ i. changes to the Program, and
+ ii. additions to the Program;
+
+ where such changes and/or additions to the Program were added to the
+ Program by such Contributor itself or anyone acting on such
+ Contributor's behalf, and the Contributor explicitly consents, in
+ accordance with Section 3C, to characterization of the changes and/or
+ additions as Contributions.
+
+"Contributor" means LUCENT and any other entity that has Contributed a
+Contribution to the Program.
+
+"Distributor" means a Recipient that distributes the Program,
+modifications to the Program, or any part thereof.
+
+"Licensed Patents" mean patent claims licensable by a Contributor
+which are necessarily infringed by the use or sale of its Contribution
+alone or when combined with the Program.
+
+"Original Program" means the original version of the software
+accompanying this Agreement as released by LUCENT, including source
+code, object code and documentation, if any.
+
+"Program" means the Original Program and Contributions or any part
+thereof
+
+"Recipient" means anyone who receives the Program under this
+Agreement, including all Contributors.
+
+2. GRANT OF RIGHTS
+
+ a. Subject to the terms of this Agreement, each Contributor hereby
+ grants Recipient a non-exclusive, worldwide, royalty-free copyright
+ license to reproduce, prepare derivative works of, publicly display,
+ publicly perform, distribute and sublicense the Contribution of such
+ Contributor, if any, and such derivative works, in source code and
+ object code form.
+
+ b. Subject to the terms of this Agreement, each Contributor hereby
+ grants Recipient a non-exclusive, worldwide, royalty-free patent
+ license under Licensed Patents to make, use, sell, offer to sell,
+ import and otherwise transfer the Contribution of such Contributor, if
+ any, in source code and object code form. The patent license granted
+ by a Contributor shall also apply to the combination of the
+ Contribution of that Contributor and the Program if, at the time the
+ Contribution is added by the Contributor, such addition of the
+ Contribution causes such combination to be covered by the Licensed
+ Patents. The patent license granted by a Contributor shall not apply
+ to (i) any other combinations which include the Contribution, nor to
+ (ii) Contributions of other Contributors. No hardware per se is
+ licensed hereunder.
+
+ c. Recipient understands that although each Contributor grants the
+ licenses to its Contributions set forth herein, no assurances are
+ provided by any Contributor that the Program does not infringe the
+ patent or other intellectual property rights of any other entity. Each
+ Contributor disclaims any liability to Recipient for claims brought by
+ any other entity based on infringement of intellectual property rights
+ or otherwise. As a condition to exercising the rights and licenses
+ granted hereunder, each Recipient hereby assumes sole responsibility
+ to secure any other intellectual property rights needed, if any. For
+ example, if a third party patent license is required to allow
+ Recipient to distribute the Program, it is Recipient's responsibility
+ to acquire that license before distributing the Program.
+
+ d. Each Contributor represents that to its knowledge it has sufficient
+ copyright rights in its Contribution, if any, to grant the copyright
+ license set forth in this Agreement.
+
+3. REQUIREMENTS
+
+A. Distributor may choose to distribute the Program in any form under
+this Agreement or under its own license agreement, provided that:
+
+ a. it complies with the terms and conditions of this Agreement;
+
+ b. if the Program is distributed in source code or other tangible
+ form, a copy of this Agreement or Distributor's own license agreement
+ is included with each copy of the Program; and
+
+ c. if distributed under Distributor's own license agreement, such
+ license agreement:
+
+ i. effectively disclaims on behalf of all Contributors all warranties
+ and conditions, express and implied, including warranties or
+ conditions of title and non-infringement, and implied warranties or
+ conditions of merchantability and fitness for a particular purpose;
+ ii. effectively excludes on behalf of all Contributors all liability
+ for damages, including direct, indirect, special, incidental and
+ consequential damages, such as lost profits; and
+ iii. states that any provisions which differ from this Agreement are
+ offered by that Contributor alone and not by any other party.
+
+B. Each Distributor must include the following in a conspicuous
+ location in the Program:
+
+ Copyright (C) 2003, Lucent Technologies Inc. and others. All Rights
+ Reserved.
+
+C. In addition, each Contributor must identify itself as the
+originator of its Contribution in a manner that reasonably allows
+subsequent Recipients to identify the originator of the Contribution.
+Also, each Contributor must agree that the additions and/or changes
+are intended to be a Contribution. Once a Contribution is contributed,
+it may not thereafter be revoked.
+
+4. COMMERCIAL DISTRIBUTION
+
+Commercial distributors of software may accept certain
+responsibilities with respect to end users, business partners and the
+like. While this license is intended to facilitate the commercial use
+of the Program, the Distributor who includes the Program in a
+commercial product offering should do so in a manner which does not
+create potential liability for Contributors. Therefore, if a
+Distributor includes the Program in a commercial product offering,
+such Distributor ("Commercial Distributor") hereby agrees to defend
+and indemnify every Contributor ("Indemnified Contributor") against
+any losses, damages and costs (collectively"Losses") arising from
+claims, lawsuits and other legal actions brought by a third party
+against the Indemnified Contributor to the extent caused by the acts
+or omissions of such Commercial Distributor in connection with its
+distribution of the Program in a commercial product offering. The
+obligations in this section do not apply to any claims or Losses
+relating to any actual or alleged intellectual property infringement.
+In order to qualify, an Indemnified Contributor must: a) promptly
+notify the Commercial Distributor in writing of such claim, and b)
+allow the Commercial Distributor to control, and cooperate with the
+Commercial Distributor in, the defense and any related settlement
+negotiations. The Indemnified Contributor may participate in any such
+claim at its own expense.
+
+For example, a Distributor might include the Program in a commercial
+product offering, Product X. That Distributor is then a Commercial
+Distributor. If that Commercial Distributor then makes performance
+claims, or offers warranties related to Product X, those performance
+claims and warranties are such Commercial Distributor's responsibility
+alone. Under this section, the Commercial Distributor would have to
+defend claims against the Contributors related to those performance
+claims and warranties, and if a court requires any Contributor to pay
+any damages as a result, the Commercial Distributor must pay those
+damages.
+
+5. NO WARRANTY
+
+EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, THE PROGRAM IS
+PROVIDED ON AN"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT LIMITATION, ANY
+WARRANTIES OR CONDITIONS OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY
+OR FITNESS FOR A PARTICULAR PURPOSE. Each Recipient is solely
+responsible for determining the appropriateness of using and
+distributing the Program and assumes all risks associated with its
+exercise of rights under this Agreement, including but not limited to
+the risks and costs of program errors, compliance with applicable
+laws, damage to or loss of data, programs or equipment, and
+unavailability or interruption of operations.
+
+6. DISCLAIMER OF LIABILITY
+
+EXCEPT AS EXPRESSLY SET FORTH IN THIS AGREEMENT, NEITHER RECIPIENT NOR
+ANY CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING
+WITHOUT LIMITATION LOST PROFITS), HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OR
+DISTRIBUTION OF THE PROGRAM OR THE EXERCISE OF ANY RIGHTS GRANTED
+HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+7. EXPORT CONTROL
+
+Recipient agrees that Recipient alone is responsible for compliance
+with the United States export administration regulations (and the
+export control laws and regulation of any other countries).
+
+8. GENERAL
+
+If any provision of this Agreement is invalid or unenforceable under
+applicable law, it shall not affect the validity or enforceability of
+the remainder of the terms of this Agreement, and without further
+action by the parties hereto, such provision shall be reformed to the
+minimum extent necessary to make such provision valid and enforceable.
+
+If Recipient institutes patent litigation against a Contributor with
+respect to a patent applicable to software (including a cross-claim or
+counterclaim in a lawsuit), then any patent licenses granted by that
+Contributor to such Recipient under this Agreement shall terminate as
+of the date such litigation is filed. In addition, if Recipient
+institutes patent litigation against any entity (including a
+cross-claim or counterclaim in a lawsuit) alleging that the Program
+itself (excluding combinations of the Program with other software or
+hardware) infringes such Recipient's patent(s), then such Recipient's
+rights granted under Section 2(b) shall terminate as of the date such
+litigation is filed.
+
+All Recipient's rights under this Agreement shall terminate if it
+fails to comply with any of the material terms or conditions of this
+Agreement and does not cure such failure in a reasonable period of
+time after becoming aware of such noncompliance. If all Recipient's
+rights under this Agreement terminate, Recipient agrees to cease use
+and distribution of the Program as soon as reasonably practicable.
+However, Recipient's obligations under this Agreement and any licenses
+granted by Recipient relating to the Program shall continue and
+survive.
+
+LUCENT may publish new versions (including revisions) of this
+Agreement from time to time. Each new version of the Agreement will be
+given a distinguishing version number. The Program (including
+Contributions) may always be distributed subject to the version of the
+Agreement under which it was received. In addition, after a new
+version of the Agreement is published, Contributor may elect to
+distribute the Program (including its Contributions) under the new
+version. No one other than LUCENT has the right to modify this
+Agreement. Except as expressly stated in Sections 2(a) and 2(b) above,
+Recipient receives no rights or licenses to the intellectual property
+of any Contributor under this Agreement, whether expressly, by
+implication, estoppel or otherwise. All rights in the Program not
+expressly granted under this Agreement are reserved.
+
+This Agreement is governed by the laws of the State of New York and
+the intellectual property laws of the United States of America. No
+party to this Agreement will bring a legal action under this Agreement
+more than one year after the cause of action arose. Each party waives
+its rights to a jury trial in any resulting litigation.
+
DIR diff --git a/src/libregexp/Make.Darwin-PowerMacintosh b/src/libregexp/Make.Darwin-PowerMacintosh
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I${PREFIX}/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.FreeBSD-386 b/src/libregexp/Make.FreeBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.HP-UX-9000 b/src/libregexp/Make.HP-UX-9000
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS=-O -c -Ae -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.Linux-386 b/src/libregexp/Make.Linux-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.NetBSD-386 b/src/libregexp/Make.NetBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.OSF1-alpha b/src/libregexp/Make.OSF1-alpha
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.SunOS-sun4u b/src/libregexp/Make.SunOS-sun4u
t@@ -0,0 +1,2 @@
+include Make.SunOS-sun4u-$(CC)
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.SunOS-sun4u-cc b/src/libregexp/Make.SunOS-sun4u-cc
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I. -O
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Make.SunOS-sun4u-gcc b/src/libregexp/Make.SunOS-sun4u-gcc
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libregexp/Makefile b/src/libregexp/Makefile
t@@ -0,0 +1,106 @@
+
+# this works in gnu make
+SYSNAME:=${shell uname}
+OBJTYPE:=${shell uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'}
+
+# this works in bsd make
+SYSNAME!=uname
+OBJTYPE!=uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'
+
+# the gnu rules will mess up bsd but not vice versa,
+# hence the gnu rules come first.
+
+include Make.$(SYSNAME)-$(OBJTYPE)
+
+PREFIX=/usr/local
+
+NUKEFILES=
+
+TGZFILES=
+
+LIB=libregexp9.a
+VERSION=2.0
+PORTPLACE=devel/libregexp9
+NAME=libregexp9
+
+OFILES=\
+ regcomp.$O\
+ regerror.$O\
+ regexec.$O\
+ regsub.$O\
+ regaux.$O\
+ rregsub.$O\
+ rregaux.$O\
+ rregexec.$O\
+
+HFILES=\
+ regexp9.h\
+ regcomp.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ test -d $(PREFIX)/man/man7 || mkdir $(PREFIX)/man/man7
+ install -m 0644 regexp9.3 $(PREFIX)/man/man3/regexp9.3
+ install -m 0644 regexp9.7 $(PREFIX)/man/man7/regexp9.7
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+ install -m 0644 regexp9.h $(PREFIX)/include/regexp9.h
+
+test: test.$O $(LIB)
+ $(CC) -o test test.$O $(LIB) -L/usr/local/lib -lfmt -lutf
+
+test2: test2.$O $(LIB)
+ $(CC) -o test2 test2.$O $(LIB) -L/usr/local/lib -lfmt -lutf
+
+$(LIB): $(OFILES)
+ $(AR) $(ARFLAGS) $(LIB) $(OFILES)
+
+NUKEFILES+=$(LIB)
+.c.$O:
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+%.$O: %.c
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+
+$(OFILES): $(HFILES)
+
+tgz:
+ rm -rf $(NAME)-$(VERSION)
+ mkdir $(NAME)-$(VERSION)
+ cp Makefile Make.* README LICENSE NOTICE *.[ch137] rpm.spec bundle.ports $(TGZFILES) $(NAME)-$(VERSION)
+ tar cf - $(NAME)-$(VERSION) | gzip >$(NAME)-$(VERSION).tgz
+ rm -rf $(NAME)-$(VERSION)
+
+clean:
+ rm -f $(OFILES) $(LIB)
+
+nuke:
+ rm -f $(OFILES) *.tgz *.rpm $(NUKEFILES)
+
+rpm:
+ make tgz
+ cp $(NAME)-$(VERSION).tgz /usr/src/RPM/SOURCES
+ rpm -ba rpm.spec
+ cp /usr/src/RPM/SRPMS/$(NAME)-$(VERSION)-1.src.rpm .
+ cp /usr/src/RPM/RPMS/i586/$(NAME)-$(VERSION)-1.i586.rpm .
+ scp *.rpm rsc@amsterdam.lcs.mit.edu:public_html/software
+
+PORTDIR=/usr/ports/$(PORTPLACE)
+
+ports:
+ make tgz
+ rm -rf $(PORTDIR)
+ mkdir $(PORTDIR)
+ cp $(NAME)-$(VERSION).tgz /usr/ports/distfiles
+ cat bundle.ports | (cd $(PORTDIR) && awk '$$1=="---" && $$3=="---" { ofile=$$2; next} {if(ofile) print >ofile}')
+ (cd $(PORTDIR); make makesum)
+ (cd $(PORTDIR); make)
+ (cd $(PORTDIR); /usr/local/bin/portlint)
+ rm -rf $(PORTDIR)/work
+ shar `find $(PORTDIR)` > ports.shar
+ (cd $(PORTDIR); tar cf - *) | gzip >$(NAME)-$(VERSION)-ports.tgz
+ scp *.tgz rsc@amsterdam.lcs.mit.edu:public_html/software
+
+.phony: all clean nuke install tgz rpm ports
DIR diff --git a/src/libregexp/Makefile.MID b/src/libregexp/Makefile.MID
t@@ -0,0 +1,34 @@
+LIB=libregexp9.a
+VERSION=2.0
+PORTPLACE=devel/libregexp9
+NAME=libregexp9
+
+OFILES=\
+ regcomp.$O\
+ regerror.$O\
+ regexec.$O\
+ regsub.$O\
+ regaux.$O\
+ rregsub.$O\
+ rregaux.$O\
+
+HFILES=\
+ regexp9.h\
+ regcomp.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ test -d $(PREFIX)/man/man7 || mkdir $(PREFIX)/man/man7
+ install -m 0644 regexp9.3 $(PREFIX)/man/man3/regexp9.3
+ install -m 0644 regexp9.7 $(PREFIX)/man/man7/regexp9.7
+ install -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+ install -m 0644 regexp9.h $(PREFIX)/include/regexp9.h
+
+test: test.$O $(LIB)
+ $(CC) -o test test.$O $(LIB) -L/usr/local/lib -lfmt -lutf
+
+test2: test2.$O $(LIB)
+ $(CC) -o test2 test2.$O $(LIB) -L/usr/local/lib -lfmt -lutf
+
DIR diff --git a/src/libregexp/NOTICE b/src/libregexp/NOTICE
t@@ -0,0 +1,25 @@
+Copyright © 1994-1999 Lucent Technologies Inc. All rights reserved.
+Portions Copyright © 2000-2002 Vita Nuova Holdings Limited (www.vitanuova.com). All rights reserved.
+
+Under a licence agreement with Lucent Technologies Inc. effective 1st March 2000,
+Vita Nuova Holdings Limited has the right to determine (within a specified scope)
+the form and content of sublicences for this software.
+
+Vita Nuova Holdings Limited now makes this software available as Free
+Software under the terms of the `GNU Lesser Public License, Version 2.1'
+(see the file LICENCE or http://www.fsf.org/copyleft/lesser.html for
+the full terms and conditions). One of the conditions of that licence
+is that you must keep intact all notices that refer to that licence and to the absence of
+of any warranty: for this software, note that includes this NOTICE file in particular.
+
+This suite of programs is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+`GNU General Public License' for more details.
+
+This copyright NOTICE applies to all files in this directory and
+subdirectories, unless another copyright notice appears in a given
+file or subdirectory. If you take code from this software to use in
+other programs, you must somehow include with it an appropriate
+copyright notice that includes the copyright notice and the other
+notices above.
DIR diff --git a/src/libregexp/README b/src/libregexp/README
t@@ -0,0 +1,7 @@
+This is a Unix port of the Plan 9 regular expression library,
+originally done for the Inferno operating system.
+
+Russ Cox repackaged this to build as a standalone
+Unix library. Send comments about packaging to
+Russ Cox <rsc@post.harvard.edu>
+
DIR diff --git a/src/libregexp/bundle.ports b/src/libregexp/bundle.ports
t@@ -0,0 +1,51 @@
+--- Makefile ---
+# New ports collection makefile for: libbio
+# Date Created: 11 Feb 2003
+# Whom: rsc
+#
+# THIS LINE NEEDS REPLACING. IT'S HERE TO GET BY PORTLINT
+# $FreeBSD: ports/devel/libbio/Makefile,v 1.1 2003/02/12 00:51:22 rsc Exp $
+
+PORTNAME= libregexp9
+PORTVERSION= 2.0
+CATEGORIES= devel
+MASTER_SITES= http://pdos.lcs.mit.edu/~rsc/software/
+EXTRACT_SUFX= .tgz
+
+MAINTAINER= rsc@post.harvard.edu
+
+DEPENDS= ${PORTSDIR}/devel/libfmt ${PORTSDIR}/devel/libutf
+
+MAN3= regexp9.3
+MAN7= regexp9.7
+USE_REINPLACE= yes
+
+.include <bsd.port.pre.mk>
+
+post-patch:
+ ${REINPLACE_CMD} -e 's,$$(PREFIX),${PREFIX},g' ${WRKSRC}/Makefile
+
+.include <bsd.port.post.mk>
+--- pkg-comment ---
+Simple regular expression library from Plan 9
+--- pkg-descr ---
+Libregexp9 is a port of Plan 9's regexp library.
+It is small and simple and provides the traditional
+extended regular expressions (as opposed to the
+current extended regular expressions, which add {}
+and various \x character classes, among other
+complications).
+
+It handles Unicode in wide character or UTF8 format!
+
+WWW: http://pdos.lcs.mit.edu/~rsc/software/
+http://plan9.bell-labs.com/magic/man2html/2/regexp
+
+Russ Cox
+rsc@post.harvard.edu
+--- pkg-plist ---
+lib/libregexp9.a
+include/regex9.h
+--- /dev/null ---
+This is just a way to make sure blank lines don't
+creep into pkg-plist.
DIR diff --git a/src/libregexp/lib9.h b/src/libregexp/lib9.h
t@@ -0,0 +1,6 @@
+#include <fmt.h>
+#include <setjmp.h>
+#include <string.h>
+#include <stdlib.h>
+#include <unistd.h>
+
DIR diff --git a/src/libregexp/mkfile b/src/libregexp/mkfile
t@@ -0,0 +1 @@
+<../libutf/mkfile
DIR diff --git a/src/libregexp/regaux.c b/src/libregexp/regaux.c
t@@ -0,0 +1,76 @@
+#include "lib9.h"
+#include "regexp9.h"
+#include "regcomp.h"
+
+
+/*
+ * save a new match in mp
+ */
+extern void
+_renewmatch(Resub *mp, int ms, Resublist *sp)
+{
+ int i;
+
+ if(mp==0 || ms<=0)
+ return;
+ if(mp[0].s.sp==0 || sp->m[0].s.sp<mp[0].s.sp ||
+ (sp->m[0].s.sp==mp[0].s.sp && sp->m[0].e.ep>mp[0].e.ep)){
+ for(i=0; i<ms && i<NSUBEXP; i++)
+ mp[i] = sp->m[i];
+ for(; i<ms; i++)
+ mp[i].s.sp = mp[i].e.ep = 0;
+ }
+}
+
+/*
+ * Note optimization in _renewthread:
+ * *lp must be pending when _renewthread called; if *l has been looked
+ * at already, the optimization is a bug.
+ */
+extern Relist*
+_renewthread(Relist *lp, /* _relist to add to */
+ Reinst *ip, /* instruction to add */
+ Resublist *sep) /* pointers to subexpressions */
+{
+ Relist *p;
+
+ for(p=lp; p->inst; p++){
+ if(p->inst == ip){
+ if((sep)->m[0].s.sp < p->se.m[0].s.sp)
+ p->se = *sep;
+ return 0;
+ }
+ }
+ p->inst = ip;
+ p->se = *sep;
+ (++p)->inst = 0;
+ return p;
+}
+
+/*
+ * same as renewthread, but called with
+ * initial empty start pointer.
+ */
+extern Relist*
+_renewemptythread(Relist *lp, /* _relist to add to */
+ Reinst *ip, /* instruction to add */
+ char *sp) /* pointers to subexpressions */
+{
+ Relist *p;
+
+ for(p=lp; p->inst; p++){
+ if(p->inst == ip){
+ if(sp < p->se.m[0].s.sp) {
+ memset((void *)&p->se, 0, sizeof(p->se));
+ p->se.m[0].s.sp = sp;
+ }
+ return 0;
+ }
+ }
+ p->inst = ip;
+ memset((void *)&p->se, 0, sizeof(p->se));
+ p->se.m[0].s.sp = sp;
+ (++p)->inst = 0;
+ return p;
+}
+
DIR diff --git a/src/libregexp/regcomp.c b/src/libregexp/regcomp.c
t@@ -0,0 +1,557 @@
+#include "lib9.h"
+#include "regexp9.h"
+#include "regcomp.h"
+
+#define TRUE 1
+#define FALSE 0
+
+/*
+ * Parser Information
+ */
+typedef
+struct Node
+{
+ Reinst* first;
+ Reinst* last;
+}Node;
+
+Reprog RePrOg;
+
+#define NSTACK 20
+static Node andstack[NSTACK];
+static Node *andp;
+static int atorstack[NSTACK];
+static int* atorp;
+static int cursubid; /* id of current subexpression */
+static int subidstack[NSTACK]; /* parallel to atorstack */
+static int* subidp;
+static int lastwasand; /* Last token was operand */
+static int nbra;
+static char* exprp; /* pointer to next character in source expression */
+static int lexdone;
+static int nclass;
+static Reclass*classp;
+static Reinst* freep;
+static int errors;
+static Rune yyrune; /* last lex'd rune */
+static Reclass*yyclassp; /* last lex'd class */
+
+/* predeclared crap */
+static void operator(int);
+static void pushand(Reinst*, Reinst*);
+static void pushator(int);
+static void evaluntil(int);
+static int bldcclass(void);
+
+static jmp_buf regkaboom;
+
+static void
+rcerror(char *s)
+{
+ errors++;
+ regerror(s);
+ longjmp(regkaboom, 1);
+}
+
+static Reinst*
+newinst(int t)
+{
+ freep->type = t;
+ freep->u2.left = 0;
+ freep->u1.right = 0;
+ return freep++;
+}
+
+static void
+operand(int t)
+{
+ Reinst *i;
+
+ if(lastwasand)
+ operator(CAT); /* catenate is implicit */
+ i = newinst(t);
+
+ if(t == CCLASS || t == NCCLASS)
+ i->u1.cp = yyclassp;
+ if(t == RUNE)
+ i->u1.r = yyrune;
+
+ pushand(i, i);
+ lastwasand = TRUE;
+}
+
+static void
+operator(int t)
+{
+ if(t==RBRA && --nbra<0)
+ rcerror("unmatched right paren");
+ if(t==LBRA){
+ if(++cursubid >= NSUBEXP)
+ rcerror ("too many subexpressions");
+ nbra++;
+ if(lastwasand)
+ operator(CAT);
+ } else
+ evaluntil(t);
+ if(t != RBRA)
+ pushator(t);
+ lastwasand = FALSE;
+ if(t==STAR || t==QUEST || t==PLUS || t==RBRA)
+ lastwasand = TRUE; /* these look like operands */
+}
+
+static void
+regerr2(char *s, int c)
+{
+ char buf[100];
+ char *cp = buf;
+ while(*s)
+ *cp++ = *s++;
+ *cp++ = c;
+ *cp = '\0';
+ rcerror(buf);
+}
+
+static void
+cant(char *s)
+{
+ char buf[100];
+ strcpy(buf, "can't happen: ");
+ strcat(buf, s);
+ rcerror(buf);
+}
+
+static void
+pushand(Reinst *f, Reinst *l)
+{
+ if(andp >= &andstack[NSTACK])
+ cant("operand stack overflow");
+ andp->first = f;
+ andp->last = l;
+ andp++;
+}
+
+static void
+pushator(int t)
+{
+ if(atorp >= &atorstack[NSTACK])
+ cant("operator stack overflow");
+ *atorp++ = t;
+ *subidp++ = cursubid;
+}
+
+static Node*
+popand(int op)
+{
+ Reinst *inst;
+
+ if(andp <= &andstack[0]){
+ regerr2("missing operand for ", op);
+ inst = newinst(NOP);
+ pushand(inst,inst);
+ }
+ return --andp;
+}
+
+static int
+popator(void)
+{
+ if(atorp <= &atorstack[0])
+ cant("operator stack underflow");
+ --subidp;
+ return *--atorp;
+}
+
+static void
+evaluntil(int pri)
+{
+ Node *op1, *op2;
+ Reinst *inst1, *inst2;
+
+ while(pri==RBRA || atorp[-1]>=pri){
+ switch(popator()){
+ default:
+ rcerror("unknown operator in evaluntil");
+ break;
+ case LBRA: /* must have been RBRA */
+ op1 = popand('(');
+ inst2 = newinst(RBRA);
+ inst2->u1.subid = *subidp;
+ op1->last->u2.next = inst2;
+ inst1 = newinst(LBRA);
+ inst1->u1.subid = *subidp;
+ inst1->u2.next = op1->first;
+ pushand(inst1, inst2);
+ return;
+ case OR:
+ op2 = popand('|');
+ op1 = popand('|');
+ inst2 = newinst(NOP);
+ op2->last->u2.next = inst2;
+ op1->last->u2.next = inst2;
+ inst1 = newinst(OR);
+ inst1->u1.right = op1->first;
+ inst1->u2.left = op2->first;
+ pushand(inst1, inst2);
+ break;
+ case CAT:
+ op2 = popand(0);
+ op1 = popand(0);
+ op1->last->u2.next = op2->first;
+ pushand(op1->first, op2->last);
+ break;
+ case STAR:
+ op2 = popand('*');
+ inst1 = newinst(OR);
+ op2->last->u2.next = inst1;
+ inst1->u1.right = op2->first;
+ pushand(inst1, inst1);
+ break;
+ case PLUS:
+ op2 = popand('+');
+ inst1 = newinst(OR);
+ op2->last->u2.next = inst1;
+ inst1->u1.right = op2->first;
+ pushand(op2->first, inst1);
+ break;
+ case QUEST:
+ op2 = popand('?');
+ inst1 = newinst(OR);
+ inst2 = newinst(NOP);
+ inst1->u2.left = inst2;
+ inst1->u1.right = op2->first;
+ op2->last->u2.next = inst2;
+ pushand(inst1, inst2);
+ break;
+ }
+ }
+}
+
+static Reprog*
+optimize(Reprog *pp)
+{
+ Reinst *inst, *target;
+ int size;
+ Reprog *npp;
+ Reclass *cl;
+ int diff;
+
+ /*
+ * get rid of NOOP chains
+ */
+ for(inst=pp->firstinst; inst->type!=END; inst++){
+ target = inst->u2.next;
+ while(target->type == NOP)
+ target = target->u2.next;
+ inst->u2.next = target;
+ }
+
+ /*
+ * The original allocation is for an area larger than
+ * necessary. Reallocate to the actual space used
+ * and then relocate the code.
+ */
+ size = sizeof(Reprog) + (freep - pp->firstinst)*sizeof(Reinst);
+ npp = (Reprog *)realloc(pp, size);
+ if(npp==0 || npp==pp)
+ return pp;
+ diff = (char *)npp - (char *)pp;
+ freep = (Reinst *)((char *)freep + diff);
+ for(inst=npp->firstinst; inst<freep; inst++){
+ switch(inst->type){
+ case OR:
+ case STAR:
+ case PLUS:
+ case QUEST:
+ *(char **)&inst->u1.right += diff;
+ break;
+ case CCLASS:
+ case NCCLASS:
+ *(char **)&inst->u1.right += diff;
+ cl = inst->u1.cp;
+ *(char **)&cl->end += diff;
+ break;
+ }
+ *(char **)&inst->u2.left += diff;
+ }
+ *(char **)&npp->startinst += diff;
+ return npp;
+}
+
+#ifdef DEBUG
+static void
+dumpstack(void){
+ Node *stk;
+ int *ip;
+
+ print("operators\n");
+ for(ip=atorstack; ip<atorp; ip++)
+ print("0%o\n", *ip);
+ print("operands\n");
+ for(stk=andstack; stk<andp; stk++)
+ print("0%o\t0%o\n", stk->first->type, stk->last->type);
+}
+
+static void
+dump(Reprog *pp)
+{
+ Reinst *l;
+ Rune *p;
+
+ l = pp->firstinst;
+ do{
+ print("%d:\t0%o\t%d\t%d", l-pp->firstinst, l->type,
+ l->u2.left-pp->firstinst, l->u1.right-pp->firstinst);
+ if(l->type == RUNE)
+ print("\t%C\n", l->r);
+ else if(l->type == CCLASS || l->type == NCCLASS){
+ print("\t[");
+ if(l->type == NCCLASS)
+ print("^");
+ for(p = l->cp->spans; p < l->cp->end; p += 2)
+ if(p[0] == p[1])
+ print("%C", p[0]);
+ else
+ print("%C-%C", p[0], p[1]);
+ print("]\n");
+ } else
+ print("\n");
+ }while(l++->type);
+}
+#endif
+
+static Reclass*
+newclass(void)
+{
+ if(nclass >= NCLASS)
+ regerr2("too many character classes; limit", NCLASS+'0');
+ return &(classp[nclass++]);
+}
+
+static int
+nextc(Rune *rp)
+{
+ if(lexdone){
+ *rp = 0;
+ return 1;
+ }
+ exprp += chartorune(rp, exprp);
+ if(*rp == L'\\'){
+ exprp += chartorune(rp, exprp);
+ return 1;
+ }
+ if(*rp == 0)
+ lexdone = 1;
+ return 0;
+}
+
+static int
+lex(int literal, int dot_type)
+{
+ int quoted;
+
+ quoted = nextc(&yyrune);
+ if(literal || quoted){
+ if(yyrune == 0)
+ return END;
+ return RUNE;
+ }
+
+ switch(yyrune){
+ case 0:
+ return END;
+ case L'*':
+ return STAR;
+ case L'?':
+ return QUEST;
+ case L'+':
+ return PLUS;
+ case L'|':
+ return OR;
+ case L'.':
+ return dot_type;
+ case L'(':
+ return LBRA;
+ case L')':
+ return RBRA;
+ case L'^':
+ return BOL;
+ case L'$':
+ return EOL;
+ case L'[':
+ return bldcclass();
+ }
+ return RUNE;
+}
+
+static int
+bldcclass(void)
+{
+ int type;
+ Rune r[NCCRUNE];
+ Rune *p, *ep, *np;
+ Rune rune;
+ int quoted;
+
+ /* we have already seen the '[' */
+ type = CCLASS;
+ yyclassp = newclass();
+
+ /* look ahead for negation */
+ /* SPECIAL CASE!!! negated classes don't match \n */
+ ep = r;
+ quoted = nextc(&rune);
+ if(!quoted && rune == L'^'){
+ type = NCCLASS;
+ quoted = nextc(&rune);
+ *ep++ = L'\n';
+ *ep++ = L'\n';
+ }
+
+ /* parse class into a set of spans */
+ for(; ep<&r[NCCRUNE];){
+ if(rune == 0){
+ rcerror("malformed '[]'");
+ return 0;
+ }
+ if(!quoted && rune == L']')
+ break;
+ if(!quoted && rune == L'-'){
+ if(ep == r){
+ rcerror("malformed '[]'");
+ return 0;
+ }
+ quoted = nextc(&rune);
+ if((!quoted && rune == L']') || rune == 0){
+ rcerror("malformed '[]'");
+ return 0;
+ }
+ *(ep-1) = rune;
+ } else {
+ *ep++ = rune;
+ *ep++ = rune;
+ }
+ quoted = nextc(&rune);
+ }
+
+ /* sort on span start */
+ for(p = r; p < ep; p += 2){
+ for(np = p; np < ep; np += 2)
+ if(*np < *p){
+ rune = np[0];
+ np[0] = p[0];
+ p[0] = rune;
+ rune = np[1];
+ np[1] = p[1];
+ p[1] = rune;
+ }
+ }
+
+ /* merge spans */
+ np = yyclassp->spans;
+ p = r;
+ if(r == ep)
+ yyclassp->end = np;
+ else {
+ np[0] = *p++;
+ np[1] = *p++;
+ for(; p < ep; p += 2)
+ if(p[0] <= np[1]){
+ if(p[1] > np[1])
+ np[1] = p[1];
+ } else {
+ np += 2;
+ np[0] = p[0];
+ np[1] = p[1];
+ }
+ yyclassp->end = np+2;
+ }
+
+ return type;
+}
+
+static Reprog*
+regcomp1(char *s, int literal, int dot_type)
+{
+ int token;
+ Reprog *pp;
+
+ /* get memory for the program */
+ pp = (Reprog *)malloc(sizeof(Reprog) + 6*sizeof(Reinst)*strlen(s));
+ if(pp == 0){
+ regerror("out of memory");
+ return 0;
+ }
+ freep = pp->firstinst;
+ classp = pp->class;
+ errors = 0;
+
+ if(setjmp(regkaboom))
+ goto out;
+
+ /* go compile the sucker */
+ lexdone = 0;
+ exprp = s;
+ nclass = 0;
+ nbra = 0;
+ atorp = atorstack;
+ andp = andstack;
+ subidp = subidstack;
+ lastwasand = FALSE;
+ cursubid = 0;
+
+ /* Start with a low priority operator to prime parser */
+ pushator(START-1);
+ while((token = lex(literal, dot_type)) != END){
+ if((token&0300) == OPERATOR)
+ operator(token);
+ else
+ operand(token);
+ }
+
+ /* Close with a low priority operator */
+ evaluntil(START);
+
+ /* Force END */
+ operand(END);
+ evaluntil(START);
+#ifdef DEBUG
+ dumpstack();
+#endif
+ if(nbra)
+ rcerror("unmatched left paren");
+ --andp; /* points to first and only operand */
+ pp->startinst = andp->first;
+#ifdef DEBUG
+ dump(pp);
+#endif
+ pp = optimize(pp);
+#ifdef DEBUG
+ print("start: %d\n", andp->first-pp->firstinst);
+ dump(pp);
+#endif
+out:
+ if(errors){
+ free(pp);
+ pp = 0;
+ }
+ return pp;
+}
+
+extern Reprog*
+regcomp(char *s)
+{
+ return regcomp1(s, 0, ANY);
+}
+
+extern Reprog*
+regcomplit(char *s)
+{
+ return regcomp1(s, 1, ANY);
+}
+
+extern Reprog*
+regcompnl(char *s)
+{
+ return regcomp1(s, 0, ANYNL);
+}
DIR diff --git a/src/libregexp/regcomp.h b/src/libregexp/regcomp.h
t@@ -0,0 +1,74 @@
+/*
+ * substitution list
+ */
+#define uchar __reuchar
+typedef unsigned char uchar;
+#define nelem(x) (sizeof(x)/sizeof((x)[0]))
+
+#define NSUBEXP 32
+typedef struct Resublist Resublist;
+struct Resublist
+{
+ Resub m[NSUBEXP];
+};
+
+/* max character classes per program */
+extern Reprog RePrOg;
+#define NCLASS (sizeof(RePrOg.class)/sizeof(Reclass))
+
+/* max rune ranges per character class */
+#define NCCRUNE (sizeof(Reclass)/sizeof(Rune))
+
+/*
+ * Actions and Tokens (Reinst types)
+ *
+ * 02xx are operators, value == precedence
+ * 03xx are tokens, i.e. operands for operators
+ */
+#define RUNE 0177
+#define OPERATOR 0200 /* Bitmask of all operators */
+#define START 0200 /* Start, used for marker on stack */
+#define RBRA 0201 /* Right bracket, ) */
+#define LBRA 0202 /* Left bracket, ( */
+#define OR 0203 /* Alternation, | */
+#define CAT 0204 /* Concatentation, implicit operator */
+#define STAR 0205 /* Closure, * */
+#define PLUS 0206 /* a+ == aa* */
+#define QUEST 0207 /* a? == a|nothing, i.e. 0 or 1 a's */
+#define ANY 0300 /* Any character except newline, . */
+#define ANYNL 0301 /* Any character including newline, . */
+#define NOP 0302 /* No operation, internal use only */
+#define BOL 0303 /* Beginning of line, ^ */
+#define EOL 0304 /* End of line, $ */
+#define CCLASS 0305 /* Character class, [] */
+#define NCCLASS 0306 /* Negated character class, [] */
+#define END 0377 /* Terminate: match found */
+
+/*
+ * regexec execution lists
+ */
+#define LISTSIZE 10
+#define BIGLISTSIZE (10*LISTSIZE)
+typedef struct Relist Relist;
+struct Relist
+{
+ Reinst* inst; /* Reinstruction of the thread */
+ Resublist se; /* matched subexpressions in this thread */
+};
+typedef struct Reljunk Reljunk;
+struct Reljunk
+{
+ Relist* relist[2];
+ Relist* reliste[2];
+ int starttype;
+ Rune startchar;
+ char* starts;
+ char* eol;
+ Rune* rstarts;
+ Rune* reol;
+};
+
+extern Relist* _renewthread(Relist*, Reinst*, Resublist*);
+extern void _renewmatch(Resub*, int, Resublist*);
+extern Relist* _renewemptythread(Relist*, Reinst*, char*);
+extern Relist* _rrenewemptythread(Relist*, Reinst*, Rune*);
DIR diff --git a/src/libregexp/regerror.c b/src/libregexp/regerror.c
t@@ -0,0 +1,14 @@
+#include "lib9.h"
+#include "regexp9.h"
+
+void
+regerror(char *s)
+{
+ char buf[132];
+
+ strcpy(buf, "regerror: ");
+ strcat(buf, s);
+ strcat(buf, "\n");
+ write(2, buf, strlen(buf));
+ exit(1);
+}
DIR diff --git a/src/libregexp/regexec.c b/src/libregexp/regexec.c
t@@ -0,0 +1,219 @@
+#include "lib9.h"
+#include "regexp9.h"
+#include "regcomp.h"
+
+
+/*
+ * return 0 if no match
+ * >0 if a match
+ * <0 if we ran out of _relist space
+ */
+static int
+regexec1(Reprog *progp, /* program to run */
+ char *bol, /* string to run machine on */
+ Resub *mp, /* subexpression elements */
+ int ms, /* number of elements at mp */
+ Reljunk *j
+)
+{
+ int flag=0;
+ Reinst *inst;
+ Relist *tlp;
+ char *s;
+ int i, checkstart;
+ Rune r, *rp, *ep;
+ int n;
+ Relist* tl; /* This list, next list */
+ Relist* nl;
+ Relist* tle; /* ends of this and next list */
+ Relist* nle;
+ int match;
+ char *p;
+
+ match = 0;
+ checkstart = j->starttype;
+ if(mp)
+ for(i=0; i<ms; i++) {
+ mp[i].s.sp = 0;
+ mp[i].e.ep = 0;
+ }
+ j->relist[0][0].inst = 0;
+ j->relist[1][0].inst = 0;
+
+ /* Execute machine once for each character, including terminal NUL */
+ s = j->starts;
+ do{
+ /* fast check for first char */
+ if(checkstart) {
+ switch(j->starttype) {
+ case RUNE:
+ p = utfrune(s, j->startchar);
+ if(p == 0)
+ return match;
+ s = p;
+ break;
+ case BOL:
+ if(s == bol)
+ break;
+ p = utfrune(s, '\n');
+ if(p == 0)
+ return match;
+ s = p;
+ break;
+ }
+ }
+ r = *(uchar*)s;
+ if(r < (Rune)Runeself)
+ n = 1;
+ else
+ n = chartorune(&r, s);
+
+ /* switch run lists */
+ tl = j->relist[flag];
+ tle = j->reliste[flag];
+ nl = j->relist[flag^=1];
+ nle = j->reliste[flag];
+ nl->inst = 0;
+
+ /* Add first instruction to current list */
+ if(match == 0)
+ _renewemptythread(tl, progp->startinst, s);
+
+ /* Execute machine until current list is empty */
+ for(tlp=tl; tlp->inst; tlp++){ /* assignment = */
+ for(inst = tlp->inst; ; inst = inst->u2.next){
+ switch(inst->type){
+ case RUNE: /* regular character */
+ if(inst->u1.r == r){
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ }
+ break;
+ case LBRA:
+ tlp->se.m[inst->u1.subid].s.sp = s;
+ continue;
+ case RBRA:
+ tlp->se.m[inst->u1.subid].e.ep = s;
+ continue;
+ case ANY:
+ if(r != '\n')
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case ANYNL:
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case BOL:
+ if(s == bol || *(s-1) == '\n')
+ continue;
+ break;
+ case EOL:
+ if(s == j->eol || r == 0 || r == '\n')
+ continue;
+ break;
+ case CCLASS:
+ ep = inst->u1.cp->end;
+ for(rp = inst->u1.cp->spans; rp < ep; rp += 2)
+ if(r >= rp[0] && r <= rp[1]){
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ }
+ break;
+ case NCCLASS:
+ ep = inst->u1.cp->end;
+ for(rp = inst->u1.cp->spans; rp < ep; rp += 2)
+ if(r >= rp[0] && r <= rp[1])
+ break;
+ if(rp == ep)
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case OR:
+ /* evaluate right choice later */
+ if(_renewthread(tlp, inst->u1.right, &tlp->se) == tle)
+ return -1;
+ /* efficiency: advance and re-evaluate */
+ continue;
+ case END: /* Match! */
+ match = 1;
+ tlp->se.m[0].e.ep = s;
+ if(mp != 0)
+ _renewmatch(mp, ms, &tlp->se);
+ break;
+ }
+ break;
+ }
+ }
+ if(s == j->eol)
+ break;
+ checkstart = j->starttype && nl->inst==0;
+ s += n;
+ }while(r);
+ return match;
+}
+
+static int
+regexec2(Reprog *progp, /* program to run */
+ char *bol, /* string to run machine on */
+ Resub *mp, /* subexpression elements */
+ int ms, /* number of elements at mp */
+ Reljunk *j
+)
+{
+ Relist relist0[BIGLISTSIZE], relist1[BIGLISTSIZE];
+
+ /* mark space */
+ j->relist[0] = relist0;
+ j->relist[1] = relist1;
+ j->reliste[0] = relist0 + nelem(relist0) - 2;
+ j->reliste[1] = relist1 + nelem(relist1) - 2;
+
+ return regexec1(progp, bol, mp, ms, j);
+}
+
+extern int
+regexec(Reprog *progp, /* program to run */
+ char *bol, /* string to run machine on */
+ Resub *mp, /* subexpression elements */
+ int ms) /* number of elements at mp */
+{
+ Reljunk j;
+ Relist relist0[LISTSIZE], relist1[LISTSIZE];
+ int rv;
+
+ /*
+ * use user-specified starting/ending location if specified
+ */
+ j.starts = bol;
+ j.eol = 0;
+ if(mp && ms>0){
+ if(mp->s.sp)
+ j.starts = mp->s.sp;
+ if(mp->e.ep)
+ j.eol = mp->e.ep;
+ }
+ j.starttype = 0;
+ j.startchar = 0;
+ if(progp->startinst->type == RUNE && progp->startinst->u1.r < (Rune)Runeself) {
+ j.starttype = RUNE;
+ j.startchar = progp->startinst->u1.r;
+ }
+ if(progp->startinst->type == BOL)
+ j.starttype = BOL;
+
+ /* mark space */
+ j.relist[0] = relist0;
+ j.relist[1] = relist1;
+ j.reliste[0] = relist0 + nelem(relist0) - 2;
+ j.reliste[1] = relist1 + nelem(relist1) - 2;
+
+ rv = regexec1(progp, bol, mp, ms, &j);
+ if(rv >= 0)
+ return rv;
+ rv = regexec2(progp, bol, mp, ms, &j);
+ if(rv >= 0)
+ return rv;
+ return -1;
+}
DIR diff --git a/src/libregexp/regexp9.3 b/src/libregexp/regexp9.3
t@@ -0,0 +1,227 @@
+.TH REGEXP9 3
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.de LR
+.if t .BR \\$1 \\$2
+.if n .RB ` \\$1 '\\$2
+..
+.de L
+.nh
+.if t .B \\$1
+.if n .RB ` \\$1 '
+..
+.SH NAME
+regcomp, regcomplit, regcompnl, regexec, regsub, regerror \- Plan 9 regular expression library
+.SH SYNOPSIS
+.B #include <regexp9.h>
+.PP
+.ta \w'\fLRegprog 'u
+.B
+Reprog *regcomp(char *exp)
+.PP
+.B
+Reprog *regcomplit(char *exp)
+.PP
+.B
+Reprog *regcompnl(char *exp)
+.PP
+.nf
+.B
+int regexec(Reprog *prog, char *string, Resub *match, int msize)
+.PP
+.nf
+.B
+void regsub(char *source, char *dest, int dlen, Resub *match, int msize)
+.PP
+.nf
+.B
+int rregexec(Reprog *prog, Rune *string, Resub *match, int msize)
+.PP
+.nf
+.B
+void rregsub(Rune *source, Rune *dest, int dlen, Resub *match, int msize)
+.PP
+.B
+void regerror(char *msg)
+.SH DESCRIPTION
+.I Regcomp
+compiles a
+regular expression and returns
+a pointer to the generated description.
+The space is allocated by
+.IR malloc (3)
+and may be released by
+.IR free .
+Regular expressions are exactly as in
+.IR regexp9 (7).
+.PP
+.I Regcomplit
+is like
+.I regcomp
+except that all characters are treated literally.
+.I Regcompnl
+is like
+.I regcomp
+except that the
+.B .
+metacharacter matches all characters, including newlines.
+.PP
+.I Regexec
+matches a null-terminated
+.I string
+against the compiled regular expression in
+.IR prog .
+If it matches,
+.I regexec
+returns
+.B 1
+and fills in the array
+.I match
+with character pointers to the substrings of
+.I string
+that correspond to the
+parenthesized subexpressions of
+.IR exp :
+.BI match[ i ].sp
+points to the beginning and
+.BI match[ i ].ep
+points just beyond
+the end of the
+.IR i th
+substring.
+(Subexpression
+.I i
+begins at the
+.IR i th
+left parenthesis, counting from 1.)
+Pointers in
+.B match[0]
+pick out the substring that corresponds to
+the whole regular expression.
+Unused elements of
+.I match
+are filled with zeros.
+Matches involving
+.LR * ,
+.LR + ,
+and
+.L ?
+are extended as far as possible.
+The number of array elements in
+.I match
+is given by
+.IR msize .
+The structure of elements of
+.I match
+is:
+.IP
+.EX
+typedef struct {
+ union {
+ char *sp;
+ Rune *rsp;
+ } s;
+ union {
+ char *ep;
+ Rune *rep;
+ } e;
+} Resub;
+.EE
+.LP
+If
+.B match[0].s.sp
+is nonzero on entry,
+.I regexec
+starts matching at that point within
+.IR string .
+If
+.B match[0].e.ep
+is nonzero on entry,
+the last character matched is the one
+preceding that point.
+.PP
+.I Regsub
+places in
+.I dest
+a substitution instance of
+.I source
+in the context of the last
+.I regexec
+performed using
+.IR match .
+Each instance of
+.BI \e n\f1,
+where
+.I n
+is a digit, is replaced by the
+string delimited by
+.BI match[ n ].s.sp
+and
+.BI match[ n ].e.ep\f1.
+Each instance of
+.L &
+is replaced by the string delimited by
+.B match[0].s.sp
+and
+.BR match[0].e.ep .
+The substitution will always be null terminated and
+trimmed to fit into dlen bytes.
+.PP
+.IR Regerror ,
+called whenever an error is detected in
+.IR regcomp ,
+writes the string
+.I msg
+on the standard error file and exits.
+.I Regerror
+can be replaced to perform
+special error processing.
+If the user supplied
+.I regerror
+returns rather than exits,
+.I regcomp
+will return 0.
+.PP
+.I Rregexec
+and
+.I rregsub
+are variants of
+.I regexec
+and
+.I regsub
+that use strings of
+.B Runes
+instead of strings of
+.BR chars .
+With these routines, the
+.I rsp
+and
+.I rep
+fields of the
+.I match
+array elements should be used.
+.SH "SEE ALSO"
+.IR grep (1),
+.IR regexp9 (7)
+.SH DIAGNOSTICS
+.I Regcomp
+returns
+.B 0
+for an illegal expression
+or other failure.
+.I Regexec
+returns 0
+if
+.I string
+is not matched.
+.SH HISTORY
+This particular regular expression was first written by Rob Pike for Plan 9.
+It has also appeared as part of the Inferno operating system.
+.SH BUGS
+There is no way to specify or match a NUL character; NULs terminate patterns and strings.
DIR diff --git a/src/libregexp/regexp9.7 b/src/libregexp/regexp9.7
t@@ -0,0 +1,150 @@
+.TH REGEXP9 7
+.de EX
+.nf
+.ft B
+..
+.de EE
+.fi
+.ft R
+..
+.de LR
+.if t .BR \\$1 \\$2
+.if n .RB ` \\$1 '\\$2
+..
+.de L
+.nh
+.if t .B \\$1
+.if n .RB ` \\$1 '
+..
+.SH NAME
+regexp9 \- Plan 9 regular expression notation
+.SH DESCRIPTION
+This manual page describes the regular expression
+syntax used by the Plan 9 regular expression library
+.IR regexp9 (3).
+It is the form used by
+.IR egrep (1)
+before
+.I egrep
+got complicated.
+.PP
+A
+.I "regular expression"
+specifies
+a set of strings of characters.
+A member of this set of strings is said to be
+.I matched
+by the regular expression. In many applications
+a delimiter character, commonly
+.LR / ,
+bounds a regular expression.
+In the following specification for regular expressions
+the word `character' means any character (rune) but newline.
+.PP
+The syntax for a regular expression
+.B e0
+is
+.IP
+.EX
+e3: literal | charclass | '.' | '^' | '$' | '(' e0 ')'
+
+e2: e3
+ | e2 REP
+
+REP: '*' | '+' | '?'
+
+e1: e2
+ | e1 e2
+
+e0: e1
+ | e0 '|' e1
+.EE
+.PP
+A
+.B literal
+is any non-metacharacter, or a metacharacter
+(one of
+.BR .*+?[]()|\e^$ ),
+or the delimiter
+preceded by
+.LR \e .
+.PP
+A
+.B charclass
+is a nonempty string
+.I s
+bracketed
+.BI [ \|s\| ]
+(or
+.BI [^ s\| ]\fR);
+it matches any character in (or not in)
+.IR s .
+A negated character class never
+matches newline.
+A substring
+.IB a - b\f1,
+with
+.I a
+and
+.I b
+in ascending
+order, stands for the inclusive
+range of
+characters between
+.I a
+and
+.IR b .
+In
+.IR s ,
+the metacharacters
+.LR - ,
+.LR ] ,
+an initial
+.LR ^ ,
+and the regular expression delimiter
+must be preceded by a
+.LR \e ;
+other metacharacters
+have no special meaning and
+may appear unescaped.
+.PP
+A
+.L .
+matches any character.
+.PP
+A
+.L ^
+matches the beginning of a line;
+.L $
+matches the end of the line.
+.PP
+The
+.B REP
+operators match zero or more
+.RB ( * ),
+one or more
+.RB ( + ),
+zero or one
+.RB ( ? ),
+instances respectively of the preceding regular expression
+.BR e2 .
+.PP
+A concatenated regular expression,
+.BR "e1\|e2" ,
+matches a match to
+.B e1
+followed by a match to
+.BR e2 .
+.PP
+An alternative regular expression,
+.BR "e0\||\|e1" ,
+matches either a match to
+.B e0
+or a match to
+.BR e1 .
+.PP
+A match to any part of a regular expression
+extends as far as possible without preventing
+a match to the remainder of the regular expression.
+.SH "SEE ALSO"
+.IR regexp9 (3)
DIR diff --git a/src/libregexp/regexp9.h b/src/libregexp/regexp9.h
t@@ -0,0 +1,71 @@
+#ifndef _REGEXP9H_
+
+#define _REGEXP9H_ 1
+#include <utf.h>
+
+typedef struct Resub Resub;
+typedef struct Reclass Reclass;
+typedef struct Reinst Reinst;
+typedef struct Reprog Reprog;
+
+/*
+ * Sub expression matches
+ */
+struct Resub{
+ union
+ {
+ char *sp;
+ Rune *rsp;
+ }s;
+ union
+ {
+ char *ep;
+ Rune *rep;
+ }e;
+};
+
+/*
+ * character class, each pair of rune's defines a range
+ */
+struct Reclass{
+ Rune *end;
+ Rune spans[64];
+};
+
+/*
+ * Machine instructions
+ */
+struct Reinst{
+ int type;
+ union {
+ Reclass *cp; /* class pointer */
+ Rune r; /* character */
+ int subid; /* sub-expression id for RBRA and LBRA */
+ Reinst *right; /* right child of OR */
+ }u1;
+ union { /* regexp relies on these two being in the same union */
+ Reinst *left; /* left child of OR */
+ Reinst *next; /* next instruction for CAT & LBRA */
+ }u2;
+};
+
+/*
+ * Reprogram definition
+ */
+struct Reprog{
+ Reinst *startinst; /* start pc */
+ Reclass class[16]; /* .data */
+ Reinst firstinst[5]; /* .text */
+};
+
+extern Reprog *regcomp(char*);
+extern Reprog *regcomplit(char*);
+extern Reprog *regcompnl(char*);
+extern void regerror(char*);
+extern int regexec(Reprog*, char*, Resub*, int);
+extern void regsub(char*, char*, int, Resub*, int);
+
+extern int rregexec(Reprog*, Rune*, Resub*, int);
+extern void rregsub(Rune*, Rune*, Resub*, int);
+
+#endif
DIR diff --git a/src/libregexp/regsub.c b/src/libregexp/regsub.c
t@@ -0,0 +1,62 @@
+#include "lib9.h"
+#include "regexp9.h"
+
+/* substitute into one string using the matches from the last regexec() */
+extern void
+regsub(char *sp, /* source string */
+ char *dp, /* destination string */
+ int dlen,
+ Resub *mp, /* subexpression elements */
+ int ms) /* number of elements pointed to by mp */
+{
+ char *ssp, *ep;
+ int i;
+
+ ep = dp+dlen-1;
+ while(*sp != '\0'){
+ if(*sp == '\\'){
+ switch(*++sp){
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ i = *sp-'0';
+ if(mp[i].s.sp != 0 && mp!=0 && ms>i)
+ for(ssp = mp[i].s.sp;
+ ssp < mp[i].e.ep;
+ ssp++)
+ if(dp < ep)
+ *dp++ = *ssp;
+ break;
+ case '\\':
+ if(dp < ep)
+ *dp++ = '\\';
+ break;
+ case '\0':
+ sp--;
+ break;
+ default:
+ if(dp < ep)
+ *dp++ = *sp;
+ break;
+ }
+ }else if(*sp == '&'){
+ if(mp[0].s.sp != 0 && mp!=0 && ms>0)
+ if(mp[0].s.sp != 0)
+ for(ssp = mp[0].s.sp;
+ ssp < mp[0].e.ep; ssp++)
+ if(dp < ep)
+ *dp++ = *ssp;
+ }else
+ if(dp < ep)
+ *dp++ = *sp;
+ sp++;
+ }
+ *dp = '\0';
+}
DIR diff --git a/src/libregexp/rpm.spec b/src/libregexp/rpm.spec
t@@ -0,0 +1,34 @@
+Summary: Simple regular expression library from Plan 9
+Name: libregexp9
+Version: 2.0
+Release: 1
+Group: Development/C
+Copyright: Public Domain
+Packager: Russ Cox <rsc@post.harvard.edu>
+Source: http://pdos.lcs.mit.edu/~rsc/software/libregexp9-2.0.tgz
+URL: http://pdos.lcs.mit.edu/~rsc/software/#libregexp9
+Requires: libfmt libutf
+
+%description
+Libregexp9 is a port of Plan 9's regexp library.
+It is small and simple and provides the traditional
+extended regular expressions (as opposed to the
+current extended regular expressions, which add {}
+and various \x character classes, among other
+complications).
+
+http://plan9.bell-labs.com/magic/man2html/2/regexp
+%prep
+%setup
+
+%build
+make
+
+%install
+make install
+
+%files
+/usr/local/include/regexp9.h
+/usr/local/lib/libregexp9.a
+/usr/local/man/man3/regexp9.3
+/usr/local/man/man7/regexp9.7
DIR diff --git a/src/libregexp/rregaux.c b/src/libregexp/rregaux.c
t@@ -0,0 +1,26 @@
+#include "lib9.h"
+#include "regexp9.h"
+#include "regcomp.h"
+
+extern Relist*
+_rrenewemptythread(Relist *lp, /* _relist to add to */
+ Reinst *ip, /* instruction to add */
+ Rune *rsp) /* pointers to subexpressions */
+{
+ Relist *p;
+
+ for(p=lp; p->inst; p++){
+ if(p->inst == ip){
+ if(rsp < p->se.m[0].s.rsp) {
+ memset((void *)&p->se, 0, sizeof(p->se));
+ p->se.m[0].s.rsp = rsp;
+ }
+ return 0;
+ }
+ }
+ p->inst = ip;
+ memset((void *)&p->se, 0, sizeof(p->se));
+ p->se.m[0].s.rsp = rsp;
+ (++p)->inst = 0;
+ return p;
+}
DIR diff --git a/src/libregexp/rregexec.c b/src/libregexp/rregexec.c
t@@ -0,0 +1,213 @@
+#include "lib9.h"
+#include "regexp9.h"
+#include "regcomp.h"
+
+/*
+ * return 0 if no match
+ * >0 if a match
+ * <0 if we ran out of _relist space
+ */
+static int
+rregexec1(Reprog *progp, /* program to run */
+ Rune *bol, /* string to run machine on */
+ Resub *mp, /* subexpression elements */
+ int ms, /* number of elements at mp */
+ Reljunk *j)
+{
+ int flag=0;
+ Reinst *inst;
+ Relist *tlp;
+ Rune *s;
+ int i, checkstart;
+ Rune r, *rp, *ep;
+ Relist* tl; /* This list, next list */
+ Relist* nl;
+ Relist* tle; /* ends of this and next list */
+ Relist* nle;
+ int match;
+
+ match = 0;
+ checkstart = j->startchar;
+ if(mp)
+ for(i=0; i<ms; i++) {
+ mp[i].s.rsp = 0;
+ mp[i].e.rep = 0;
+ }
+ j->relist[0][0].inst = 0;
+ j->relist[1][0].inst = 0;
+
+ /* Execute machine once for each character, including terminal NUL */
+ s = j->rstarts;
+ do{
+
+ /* fast check for first char */
+ if(checkstart) {
+ switch(j->starttype) {
+ case RUNE:
+ while(*s != j->startchar) {
+ if(*s == 0)
+ return match;
+ s++;
+ }
+ break;
+ case BOL:
+ if(s == bol)
+ break;
+ while(*s != '\n') {
+ if(*s == 0)
+ return match;
+ s++;
+ }
+ break;
+ }
+ }
+
+ r = *s;
+
+ /* switch run lists */
+ tl = j->relist[flag];
+ tle = j->reliste[flag];
+ nl = j->relist[flag^=1];
+ nle = j->reliste[flag];
+ nl->inst = 0;
+
+ /* Add first instruction to current list */
+ _rrenewemptythread(tl, progp->startinst, s);
+
+ /* Execute machine until current list is empty */
+ for(tlp=tl; tlp->inst; tlp++){
+ for(inst=tlp->inst; ; inst = inst->u2.next){
+ switch(inst->type){
+ case RUNE: /* regular character */
+ if(inst->u1.r == r)
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case LBRA:
+ tlp->se.m[inst->u1.subid].s.rsp = s;
+ continue;
+ case RBRA:
+ tlp->se.m[inst->u1.subid].e.rep = s;
+ continue;
+ case ANY:
+ if(r != '\n')
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case ANYNL:
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case BOL:
+ if(s == bol || *(s-1) == '\n')
+ continue;
+ break;
+ case EOL:
+ if(s == j->reol || r == 0 || r == '\n')
+ continue;
+ break;
+ case CCLASS:
+ ep = inst->u1.cp->end;
+ for(rp = inst->u1.cp->spans; rp < ep; rp += 2)
+ if(r >= rp[0] && r <= rp[1]){
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ }
+ break;
+ case NCCLASS:
+ ep = inst->u1.cp->end;
+ for(rp = inst->u1.cp->spans; rp < ep; rp += 2)
+ if(r >= rp[0] && r <= rp[1])
+ break;
+ if(rp == ep)
+ if(_renewthread(nl, inst->u2.next, &tlp->se)==nle)
+ return -1;
+ break;
+ case OR:
+ /* evaluate right choice later */
+ if(_renewthread(tlp, inst->u1.right, &tlp->se) == tle)
+ return -1;
+ /* efficiency: advance and re-evaluate */
+ continue;
+ case END: /* Match! */
+ match = 1;
+ tlp->se.m[0].e.rep = s;
+ if(mp != 0)
+ _renewmatch(mp, ms, &tlp->se);
+ break;
+ }
+ break;
+ }
+ }
+ if(s == j->reol)
+ break;
+ checkstart = j->startchar && nl->inst==0;
+ s++;
+ }while(r);
+ return match;
+}
+
+static int
+rregexec2(Reprog *progp, /* program to run */
+ Rune *bol, /* string to run machine on */
+ Resub *mp, /* subexpression elements */
+ int ms, /* number of elements at mp */
+ Reljunk *j
+)
+{
+ Relist relist0[5*LISTSIZE], relist1[5*LISTSIZE];
+
+ /* mark space */
+ j->relist[0] = relist0;
+ j->relist[1] = relist1;
+ j->reliste[0] = relist0 + nelem(relist0) - 2;
+ j->reliste[1] = relist1 + nelem(relist1) - 2;
+
+ return rregexec1(progp, bol, mp, ms, j);
+}
+
+extern int
+rregexec(Reprog *progp, /* program to run */
+ Rune *bol, /* string to run machine on */
+ Resub *mp, /* subexpression elements */
+ int ms) /* number of elements at mp */
+{
+ Reljunk j;
+ Relist relist0[LISTSIZE], relist1[LISTSIZE];
+ int rv;
+
+ /*
+ * use user-specified starting/ending location if specified
+ */
+ j.rstarts = bol;
+ j.reol = 0;
+ if(mp && ms>0){
+ if(mp->s.sp)
+ j.rstarts = mp->s.rsp;
+ if(mp->e.ep)
+ j.reol = mp->e.rep;
+ }
+ j.starttype = 0;
+ j.startchar = 0;
+ if(progp->startinst->type == RUNE && progp->startinst->u1.r < (Rune)Runeself) {
+ j.starttype = RUNE;
+ j.startchar = progp->startinst->u1.r;
+ }
+ if(progp->startinst->type == BOL)
+ j.starttype = BOL;
+
+ /* mark space */
+ j.relist[0] = relist0;
+ j.relist[1] = relist1;
+ j.reliste[0] = relist0 + nelem(relist0) - 2;
+ j.reliste[1] = relist1 + nelem(relist1) - 2;
+
+ rv = rregexec1(progp, bol, mp, ms, &j);
+ if(rv >= 0)
+ return rv;
+ rv = rregexec2(progp, bol, mp, ms, &j);
+ if(rv >= 0)
+ return rv;
+ return -1;
+}
DIR diff --git a/src/libregexp/rregsub.c b/src/libregexp/rregsub.c
t@@ -0,0 +1,55 @@
+#include "lib9.h"
+#include "regexp9.h"
+
+/* substitute into one string using the matches from the last regexec() */
+extern void
+rregsub(Rune *sp, /* source string */
+ Rune *dp, /* destination string */
+ Resub *mp, /* subexpression elements */
+ int ms) /* number of elements pointed to by mp */
+{
+ Rune *ssp;
+ int i;
+
+ while(*sp != '\0'){
+ if(*sp == '\\'){
+ switch(*++sp){
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ i = *sp-'0';
+ if(mp[i].s.rsp != 0 && mp!=0 && ms>i)
+ for(ssp = mp[i].s.rsp;
+ ssp < mp[i].e.rep;
+ ssp++)
+ *dp++ = *ssp;
+ break;
+ case '\\':
+ *dp++ = '\\';
+ break;
+ case '\0':
+ sp--;
+ break;
+ default:
+ *dp++ = *sp;
+ break;
+ }
+ }else if(*sp == '&'){
+ if(mp[0].s.rsp != 0 && mp!=0 && ms>0)
+ if(mp[0].s.rsp != 0)
+ for(ssp = mp[0].s.rsp;
+ ssp < mp[0].e.rep; ssp++)
+ *dp++ = *ssp;
+ }else
+ *dp++ = *sp;
+ sp++;
+ }
+ *dp = '\0';
+}
DIR diff --git a/src/libregexp/test.c b/src/libregexp/test.c
t@@ -0,0 +1,46 @@
+#include "lib9.h"
+#include <regexp9.h>
+
+struct x
+{
+ char *re;
+ char *s;
+ Reprog *p;
+};
+
+struct x t[] = {
+ { "^[^!@]+$", "/bin/upas/aliasmail '&'", 0 },
+ { "^local!(.*)$", "/mail/box/\\1/mbox", 0 },
+ { "^plan9!(.*)$", "\\1", 0 },
+ { "^helix!(.*)$", "\\1", 0 },
+ { "^([^!]+)@([^!@]+)$", "\\2!\\1", 0 },
+ { "^(uk\\.[^!]*)(!.*)$", "/bin/upas/uk2uk '\\1' '\\2'", 0 },
+ { "^[^!]*\\.[^!]*!.*$", "inet!&", 0 },
+ { "^\xE2\x98\xBA$", "smiley", 0 },
+ { "^(coma|research|pipe|pyxis|inet|hunny|gauss)!(.*)$", "/mail/lib/qmail '\\s' 'net!\\1' '\\2'", 0 },
+ { "^.*$", "/mail/lib/qmail '\\s' 'net!research' '&'", 0 },
+ { 0, 0, 0 },
+};
+
+main(int ac, char **av)
+{
+ Resub rs[10];
+ char dst[128];
+ int n;
+ struct x *tp;
+
+ for(tp = t; tp->re; tp++)
+ tp->p = regcomp(tp->re);
+
+
+ for(tp = t; tp->re; tp++){
+ print("%s VIA %s", av[1], tp->re);
+ memset(rs, 0, sizeof rs);
+ if(regexec(tp->p, av[1], rs, 10)){
+ regsub(tp->s, dst, sizeof dst, rs, 10);
+ print(" sub %s -> %s", tp->s, dst);
+ }
+ print("\n");
+ }
+ exit(0);
+}
DIR diff --git a/src/libregexp/test2.c b/src/libregexp/test2.c
t@@ -0,0 +1,20 @@
+#include "lib9.h"
+#include <regexp9.h>
+
+
+main(int ac, char **av)
+{
+ Resub rs[10];
+ Reprog *p;
+ char *s;
+ int i;
+
+ p = regcomp("[^a-z]");
+ s = "\n";
+ if(regexec(p, s, rs, 10))
+ print("%s %lux %lux %lux\n", s, s, rs[0].sp, rs[0].ep);
+ s = "0";
+ if(regexec(p, s, rs, 10))
+ print("%s %lux %lux %lux\n", s, s, rs[0].sp, rs[0].ep);
+ exit(0);
+}
DIR diff --git a/src/libutf/LICENSE b/src/libutf/LICENSE
t@@ -0,0 +1,13 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 1998-2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
DIR diff --git a/src/libutf/Make.Darwin-PowerMacintosh b/src/libutf/Make.Darwin-PowerMacintosh
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I${PREFIX}/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.FreeBSD-386 b/src/libutf/Make.FreeBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.HP-UX-9000 b/src/libutf/Make.HP-UX-9000
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS=-O -c -Ae -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.Linux-386 b/src/libutf/Make.Linux-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.NetBSD-386 b/src/libutf/Make.NetBSD-386
t@@ -0,0 +1,7 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c -I. -I$(PREFIX)/include
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O # default, can be overriden by Make.$(SYSNAME)
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.OSF1-alpha b/src/libutf/Make.OSF1-alpha
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I.
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.SunOS-sun4u b/src/libutf/Make.SunOS-sun4u
t@@ -0,0 +1,2 @@
+include Make.SunOS-sun4u-$(CC)
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.SunOS-sun4u-cc b/src/libutf/Make.SunOS-sun4u-cc
t@@ -0,0 +1,6 @@
+CC=cc
+CFLAGS+=-g -c -I. -O
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libutf/Make.SunOS-sun4u-gcc b/src/libutf/Make.SunOS-sun4u-gcc
t@@ -0,0 +1,6 @@
+CC=gcc
+CFLAGS+=-Wall -Wno-missing-braces -Wno-parentheses -Wno-switch -O2 -g -c
+O=o
+AR=ar
+ARFLAGS=rvc
+NAN=nan64.$O
DIR diff --git a/src/libutf/Makefile b/src/libutf/Makefile
t@@ -0,0 +1,112 @@
+
+# this works in gnu make
+SYSNAME:=${shell uname}
+OBJTYPE:=${shell uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'}
+
+# this works in bsd make
+SYSNAME!=uname
+OBJTYPE!=uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'
+
+# the gnu rules will mess up bsd but not vice versa,
+# hence the gnu rules come first.
+
+include Make.$(SYSNAME)-$(OBJTYPE)
+
+PREFIX=/usr/local
+
+NUKEFILES=
+
+TGZFILES=
+
+LIB=libutf.a
+VERSION=2.0
+PORTPLACE=devel/libutf
+NAME=libutf
+
+OFILES=\
+ rune.$O\
+ runestrcat.$O\
+ runestrchr.$O\
+ runestrcmp.$O\
+ runestrcpy.$O\
+ runestrdup.$O\
+ runestrlen.$O\
+ runestrecpy.$O\
+ runestrncat.$O\
+ runestrncmp.$O\
+ runestrncpy.$O\
+ runestrrchr.$O\
+ runestrstr.$O\
+ runetype.$O\
+ utfecpy.$O\
+ utflen.$O\
+ utfnlen.$O\
+ utfrrune.$O\
+ utfrune.$O\
+ utfutf.$O\
+
+HFILES=\
+ utf.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ install -c -m 0644 isalpharune.3 $(PREFIX)/man/man3/isalpharune.3
+ install -c -m 0644 utf.7 $(PREFIX)/man/man7/utf.7
+ install -c -m 0644 rune.3 $(PREFIX)/man/man3/rune.3
+ install -c -m 0644 runestrcat.3 $(PREFIX)/man/man3/runestrcat.3
+ install -c -m 0644 utf.h $(PREFIX)/include/utf.h
+ install -c -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
+$(LIB): $(OFILES)
+ $(AR) $(ARFLAGS) $(LIB) $(OFILES)
+
+NUKEFILES+=$(LIB)
+.c.$O:
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+%.$O: %.c
+ $(CC) $(CFLAGS) -I$(PREFIX)/include $*.c
+
+
+$(OFILES): $(HFILES)
+
+tgz:
+ rm -rf $(NAME)-$(VERSION)
+ mkdir $(NAME)-$(VERSION)
+ cp Makefile Make.* README LICENSE NOTICE *.[ch137] rpm.spec bundle.ports $(TGZFILES) $(NAME)-$(VERSION)
+ tar cf - $(NAME)-$(VERSION) | gzip >$(NAME)-$(VERSION).tgz
+ rm -rf $(NAME)-$(VERSION)
+
+clean:
+ rm -f $(OFILES) $(LIB)
+
+nuke:
+ rm -f $(OFILES) *.tgz *.rpm $(NUKEFILES)
+
+rpm:
+ make tgz
+ cp $(NAME)-$(VERSION).tgz /usr/src/RPM/SOURCES
+ rpm -ba rpm.spec
+ cp /usr/src/RPM/SRPMS/$(NAME)-$(VERSION)-1.src.rpm .
+ cp /usr/src/RPM/RPMS/i586/$(NAME)-$(VERSION)-1.i586.rpm .
+ scp *.rpm rsc@amsterdam.lcs.mit.edu:public_html/software
+
+PORTDIR=/usr/ports/$(PORTPLACE)
+
+ports:
+ make tgz
+ rm -rf $(PORTDIR)
+ mkdir $(PORTDIR)
+ cp $(NAME)-$(VERSION).tgz /usr/ports/distfiles
+ cat bundle.ports | (cd $(PORTDIR) && awk '$$1=="---" && $$3=="---" { ofile=$$2; next} {if(ofile) print >ofile}')
+ (cd $(PORTDIR); make makesum)
+ (cd $(PORTDIR); make)
+ (cd $(PORTDIR); /usr/local/bin/portlint)
+ rm -rf $(PORTDIR)/work
+ shar `find $(PORTDIR)` > ports.shar
+ (cd $(PORTDIR); tar cf - *) | gzip >$(NAME)-$(VERSION)-ports.tgz
+ scp *.tgz rsc@amsterdam.lcs.mit.edu:public_html/software
+
+.phony: all clean nuke install tgz rpm ports
DIR diff --git a/src/libutf/Makefile.BOT b/src/libutf/Makefile.BOT
t@@ -0,0 +1,47 @@
+.c.$O:
+ $(CC) $(CFLAGS) -I/usr/X11R6/include -I../sam -I$(PREFIX)/include $*.c
+
+%.$O: %.c
+ $(CC) $(CFLAGS) -I/usr/X11R6/include -I../sam -I$(PREFIX)/include $*.c
+
+
+$(OFILES): $(HFILES)
+
+tgz:
+ rm -rf $(NAME)-$(VERSION)
+ mkdir $(NAME)-$(VERSION)
+ cp Makefile Make.* README LICENSE NOTICE *.[ch137] rpm.spec bundle.ports $(TGZFILES) $(NAME)-$(VERSION)
+ tar cf - $(NAME)-$(VERSION) | gzip >$(NAME)-$(VERSION).tgz
+ rm -rf $(NAME)-$(VERSION)
+
+clean:
+ rm -f $(OFILES) $(LIB)
+
+nuke:
+ rm -f $(OFILES) *.tgz *.rpm $(NUKEFILES)
+
+rpm:
+ make tgz
+ cp $(NAME)-$(VERSION).tgz /usr/src/RPM/SOURCES
+ rpm -ba rpm.spec
+ cp /usr/src/RPM/SRPMS/$(NAME)-$(VERSION)-1.src.rpm .
+ cp /usr/src/RPM/RPMS/i586/$(NAME)-$(VERSION)-1.i586.rpm .
+ scp *.rpm rsc@amsterdam.lcs.mit.edu:public_html/software
+
+PORTDIR=/usr/ports/$(PORTPLACE)
+
+ports:
+ make tgz
+ rm -rf $(PORTDIR)
+ mkdir $(PORTDIR)
+ cp $(NAME)-$(VERSION).tgz /usr/ports/distfiles
+ cat bundle.ports | (cd $(PORTDIR) && awk '$$1=="---" && $$3=="---" { ofile=$$2; next} {if(ofile) print >ofile}')
+ (cd $(PORTDIR); make makesum)
+ (cd $(PORTDIR); make)
+ (cd $(PORTDIR); /usr/local/bin/portlint)
+ rm -rf $(PORTDIR)/work
+ shar `find $(PORTDIR)` > ports.shar
+ (cd $(PORTDIR); tar cf - *) | gzip >$(NAME)-$(VERSION)-ports.tgz
+ scp *.tgz rsc@amsterdam.lcs.mit.edu:public_html/software
+
+.phony: all clean nuke install tgz rpm ports
DIR diff --git a/src/libutf/Makefile.CMD b/src/libutf/Makefile.CMD
t@@ -0,0 +1,5 @@
+
+$(TARG): $(OFILES)
+ $(CC) -o $(TARG) $(OFILES) -L$(PREFIX)/lib -lframe -ldraw -lthread -l9 -lregexp9 -lbio -lfmt -lutf -L/usr/X11R6/lib -lX11 -lm
+
+
DIR diff --git a/src/libutf/Makefile.LIB b/src/libutf/Makefile.LIB
t@@ -0,0 +1,4 @@
+$(LIB): $(OFILES)
+ $(AR) $(ARFLAGS) $(LIB) $(OFILES)
+
+NUKEFILES+=$(LIB)
DIR diff --git a/src/libutf/Makefile.MID b/src/libutf/Makefile.MID
t@@ -0,0 +1,41 @@
+LIB=libutf.a
+VERSION=2.0
+PORTPLACE=devel/libutf
+NAME=libutf
+
+OFILES=\
+ rune.$O\
+ runestrcat.$O\
+ runestrchr.$O\
+ runestrcmp.$O\
+ runestrcpy.$O\
+ runestrdup.$O\
+ runestrlen.$O\
+ runestrecpy.$O\
+ runestrncat.$O\
+ runestrncmp.$O\
+ runestrncpy.$O\
+ runestrrchr.$O\
+ runestrstr.$O\
+ runetype.$O\
+ utfecpy.$O\
+ utflen.$O\
+ utfnlen.$O\
+ utfrrune.$O\
+ utfrune.$O\
+ utfutf.$O\
+
+HFILES=\
+ utf.h\
+
+all: $(LIB)
+
+install: $(LIB)
+ test -d $(PREFIX)/man/man3 || mkdir $(PREFIX)/man/man3
+ install -c -m 0644 isalpharune.3 $(PREFIX)/man/man3/isalpharune.3
+ install -c -m 0644 utf.7 $(PREFIX)/man/man7/utf.7
+ install -c -m 0644 rune.3 $(PREFIX)/man/man3/rune.3
+ install -c -m 0644 runestrcat.3 $(PREFIX)/man/man3/runestrcat.3
+ install -c -m 0644 utf.h $(PREFIX)/include/utf.h
+ install -c -m 0644 $(LIB) $(PREFIX)/lib/$(LIB)
+
DIR diff --git a/src/libutf/Makefile.TOP b/src/libutf/Makefile.TOP
t@@ -0,0 +1,20 @@
+
+# this works in gnu make
+SYSNAME:=${shell uname}
+OBJTYPE:=${shell uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'}
+
+# this works in bsd make
+SYSNAME!=uname
+OBJTYPE!=uname -m | sed 's;i.86;386;; s;/.*;;; s; ;;g'
+
+# the gnu rules will mess up bsd but not vice versa,
+# hence the gnu rules come first.
+
+include Make.$(SYSNAME)-$(OBJTYPE)
+
+PREFIX=/usr/local
+
+NUKEFILES=
+
+TGZFILES=
+
DIR diff --git a/src/libutf/NOTICE b/src/libutf/NOTICE
t@@ -0,0 +1,13 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 1998-2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
DIR diff --git a/src/libutf/README b/src/libutf/README
t@@ -0,0 +1,13 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 1998-2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
DIR diff --git a/src/libutf/bundle.ports b/src/libutf/bundle.ports
t@@ -0,0 +1,43 @@
+--- Makefile ---
+# New ports collection makefile for: libutf
+# Date Created: 11 Feb 2003
+# Whom: rsc
+#
+# THIS LINE NEEDS REPLACING. IT'S HERE TO GET BY PORTLINT
+# $FreeBSD: ports/devel/libfmt/Makefile,v 1.1 2003/02/12 00:51:22 rsc Exp $
+
+PORTNAME= libutf
+PORTVERSION= 2.0
+CATEGORIES= devel
+MASTER_SITES= http://pdos.lcs.mit.edu/~rsc/software/
+EXTRACT_SUFX= .tgz
+
+MAINTAINER= rsc@post.harvard.edu
+
+MAN3= rune.3 runestrcat.3 isalpharune.3
+MAN7= utf.7
+
+USE_REINPLACE=yes
+
+.include <bsd.port.pre.mk>
+
+post-patch:
+ ${REINPLACE_CMD} -e 's,$$(PREFIX),${PREFIX},g' ${WRKSRC}/Makefile
+
+.include <bsd.port.post.mk>
+--- pkg-comment ---
+UTF8 support library from Plan 9
+--- pkg-descr ---
+UTF8 support library from Plan 9.
+
+WWW: http://pdos.lcs.mit.edu/~rsc/software/#libutf
+http://plan9.bell-labs.com/magic/man2html/3/rune
+
+Russ Cox
+rsc@post.harvard.edu
+--- pkg-plist ---
+lib/libutf.a
+include/utf.h
+--- /dev/null ---
+This is just a way to make sure blank lines don't
+creep into pkg-plist.
DIR diff --git a/src/libutf/isalpharune.3 b/src/libutf/isalpharune.3
t@@ -0,0 +1,47 @@
+.TH ISALPHARUNE 3
+.SH NAME
+isalpharune, islowerrune, isspacerune, istitlerune, isupperrune, tolowerrune, totitlerune, toupperrune \- Unicode character classes and cases
+.SH SYNOPSIS
+.B #include <utf.h>
+.PP
+.B
+int isalpharune(Rune c)
+.PP
+.B
+int islowerrune(Rune c)
+.PP
+.B
+int isspacerune(Rune c)
+.PP
+.B
+int istitlerune(Rune c)
+.PP
+.B
+int isupperrune(Rune c)
+.PP
+.B
+Rune tolowerrune(Rune c)
+.PP
+.B
+Rune totitlerune(Rune c)
+.PP
+.B
+Rune toupperrune(Rune c)
+.SH DESCRIPTION
+These routines examine and operate on Unicode characters,
+in particular a subset of their properties as defined in the Unicode standard.
+Unicode defines some characters as alphabetic and specifies three cases:
+upper, lower, and title.
+Analogously to
+.IR ctype (3)
+for
+.SM ASCII\c
+,
+these routines
+test types and modify cases for Unicode characters.
+The names are self-explanatory.
+.PP
+The case-conversion routines return the character unchanged if it has no case.
+.SH "SEE ALSO
+.IR ctype (3) ,
+.IR "The Unicode Standard" .
DIR diff --git a/src/libutf/lib9.h b/src/libutf/lib9.h
t@@ -0,0 +1,17 @@
+#include <string.h>
+#include "utf.h"
+
+#define nil ((void*)0)
+
+#define uchar _fmtuchar
+#define ushort _fmtushort
+#define uint _fmtuint
+#define ulong _fmtulong
+#define vlong _fmtvlong
+#define uvlong _fmtuvlong
+
+typedef unsigned char uchar;
+typedef unsigned short ushort;
+typedef unsigned int uint;
+typedef unsigned long ulong;
+
DIR diff --git a/src/libutf/mkfile b/src/libutf/mkfile
t@@ -0,0 +1,9 @@
+all:V: Makefile Make.FreeBSD-386 Make.Linux-386 Make.NetBSD-386 Make.HP-UX-9000 Make.OSF1-alpha \
+ Make.SunOS-sun4u Make.SunOS-sun4u-cc Make.SunOS-sun4u-gcc \
+ Make.Darwin-PowerMacintosh
+
+Makefile:D: ../libutf/Makefile.TOP Makefile.MID ../libutf/Makefile.LIB ../libutf/Makefile.BOT
+ cat $prereq >$target
+
+Make.%: ../libutf/Make.%
+ cp $prereq $target
DIR diff --git a/src/libutf/rpm.spec b/src/libutf/rpm.spec
t@@ -0,0 +1,28 @@
+Summary: Port of Plan 9's UTF8 support functions
+Name: libutf
+Version: 2.0
+Release: 1
+Group: Development/C
+Copyright: Public Domain
+Packager: Russ Cox <rsc@post.harvard.edu>
+Source: http://pdos.lcs.mit.edu/~rsc/software/libutf-2.0.tgz
+URL: http://pdos.lcs.mit.edu/~rsc/software/#libutf
+
+%description
+Libutf is a port of Plan 9's UTF8 support functions.
+%prep
+%setup
+
+%build
+make
+
+%install
+make install
+
+%files
+/usr/local/include/utf.h
+/usr/local/lib/libutf.a
+/usr/local/man/man3/runestrcat.3
+/usr/local/man/man3/isalpharune.3
+/usr/local/man/man3/rune.3
+/usr/local/man/man7/utf.7
DIR diff --git a/src/libutf/rune.3 b/src/libutf/rune.3
t@@ -0,0 +1,187 @@
+.TH RUNE 3
+.SH NAME
+runetochar, chartorune, runelen, runenlen, fullrune, utfecpy, utflen, utfnlen, utfrune, utfrrune, utfutf \- rune/UTF conversion
+.SH SYNOPSIS
+.ta \w'\fLchar*xx'u
+.B #include <utf.h>
+.PP
+.B
+int runetochar(char *s, Rune *r)
+.PP
+.B
+int chartorune(Rune *r, char *s)
+.PP
+.B
+int runelen(long r)
+.PP
+.B
+int runenlen(Rune *r, int n)
+.PP
+.B
+int fullrune(char *s, int n)
+.PP
+.B
+char* utfecpy(char *s1, char *es1, char *s2)
+.PP
+.B
+int utflen(char *s)
+.PP
+.B
+int utfnlen(char *s, long n)
+.PP
+.B
+char* utfrune(char *s, long c)
+.PP
+.B
+char* utfrrune(char *s, long c)
+.PP
+.B
+char* utfutf(char *s1, char *s2)
+.SH DESCRIPTION
+These routines convert to and from a
+.SM UTF
+byte stream and runes.
+.PP
+.I Runetochar
+copies one rune at
+.I r
+to at most
+.B UTFmax
+bytes starting at
+.I s
+and returns the number of bytes copied.
+.BR UTFmax ,
+defined as
+.B 3
+in
+.BR <libc.h> ,
+is the maximum number of bytes required to represent a rune.
+.PP
+.I Chartorune
+copies at most
+.B UTFmax
+bytes starting at
+.I s
+to one rune at
+.I r
+and returns the number of bytes copied.
+If the input is not exactly in
+.SM UTF
+format,
+.I chartorune
+will convert to 0x80 and return 1.
+.PP
+.I Runelen
+returns the number of bytes
+required to convert
+.I r
+into
+.SM UTF.
+.PP
+.I Runenlen
+returns the number of bytes
+required to convert the
+.I n
+runes pointed to by
+.I r
+into
+.SM UTF.
+.PP
+.I Fullrune
+returns 1 if the string
+.I s
+of length
+.I n
+is long enough to be decoded by
+.I chartorune
+and 0 otherwise.
+This does not guarantee that the string
+contains a legal
+.SM UTF
+encoding.
+This routine is used by programs that
+obtain input a byte at
+a time and need to know when a full rune
+has arrived.
+.PP
+The following routines are analogous to the
+corresponding string routines with
+.B utf
+substituted for
+.B str
+and
+.B rune
+substituted for
+.BR chr .
+.PP
+.I Utfecpy
+copies UTF sequences until a null sequence has been copied, but writes no
+sequences beyond
+.IR es1 .
+If any sequences are copied,
+.I s1
+is terminated by a null sequence, and a pointer to that sequence is returned.
+Otherwise, the original
+.I s1
+is returned.
+.PP
+.I Utflen
+returns the number of runes that
+are represented by the
+.SM UTF
+string
+.IR s .
+.PP
+.I Utfnlen
+returns the number of complete runes that
+are represented by the first
+.I n
+bytes of
+.SM UTF
+string
+.IR s .
+If the last few bytes of the string contain an incompletely coded rune,
+.I utfnlen
+will not count them; in this way, it differs from
+.IR utflen ,
+which includes every byte of the string.
+.PP
+.I Utfrune
+.RI ( utfrrune )
+returns a pointer to the first (last)
+occurrence of rune
+.I c
+in the
+.SM UTF
+string
+.IR s ,
+or 0 if
+.I c
+does not occur in the string.
+The NUL byte terminating a string is considered to
+be part of the string
+.IR s .
+.PP
+.I Utfutf
+returns a pointer to the first occurrence of
+the
+.SM UTF
+string
+.I s2
+as a
+.SM UTF
+substring of
+.IR s1 ,
+or 0 if there is none.
+If
+.I s2
+is the null string,
+.I utfutf
+returns
+.IR s1 .
+.SH HISTORY
+These routines were written by Rob Pike and Ken Thompson
+and first appeared in Plan 9.
+.SH SEE ALSO
+.IR utf (7),
+.IR tcs (1)
DIR diff --git a/src/libutf/rune.c b/src/libutf/rune.c
t@@ -0,0 +1,177 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+enum
+{
+ Bit1 = 7,
+ Bitx = 6,
+ Bit2 = 5,
+ Bit3 = 4,
+ Bit4 = 3,
+
+ T1 = ((1<<(Bit1+1))-1) ^ 0xFF, /* 0000 0000 */
+ Tx = ((1<<(Bitx+1))-1) ^ 0xFF, /* 1000 0000 */
+ T2 = ((1<<(Bit2+1))-1) ^ 0xFF, /* 1100 0000 */
+ T3 = ((1<<(Bit3+1))-1) ^ 0xFF, /* 1110 0000 */
+ T4 = ((1<<(Bit4+1))-1) ^ 0xFF, /* 1111 0000 */
+
+ Rune1 = (1<<(Bit1+0*Bitx))-1, /* 0000 0000 0111 1111 */
+ Rune2 = (1<<(Bit2+1*Bitx))-1, /* 0000 0111 1111 1111 */
+ Rune3 = (1<<(Bit3+2*Bitx))-1, /* 1111 1111 1111 1111 */
+
+ Maskx = (1<<Bitx)-1, /* 0011 1111 */
+ Testx = Maskx ^ 0xFF, /* 1100 0000 */
+
+ Bad = Runeerror,
+};
+
+int
+chartorune(Rune *rune, char *str)
+{
+ int c, c1, c2;
+ long l;
+
+ /*
+ * one character sequence
+ * 00000-0007F => T1
+ */
+ c = *(uchar*)str;
+ if(c < Tx) {
+ *rune = c;
+ return 1;
+ }
+
+ /*
+ * two character sequence
+ * 0080-07FF => T2 Tx
+ */
+ c1 = *(uchar*)(str+1) ^ Tx;
+ if(c1 & Testx)
+ goto bad;
+ if(c < T3) {
+ if(c < T2)
+ goto bad;
+ l = ((c << Bitx) | c1) & Rune2;
+ if(l <= Rune1)
+ goto bad;
+ *rune = l;
+ return 2;
+ }
+
+ /*
+ * three character sequence
+ * 0800-FFFF => T3 Tx Tx
+ */
+ c2 = *(uchar*)(str+2) ^ Tx;
+ if(c2 & Testx)
+ goto bad;
+ if(c < T4) {
+ l = ((((c << Bitx) | c1) << Bitx) | c2) & Rune3;
+ if(l <= Rune2)
+ goto bad;
+ *rune = l;
+ return 3;
+ }
+
+ /*
+ * bad decoding
+ */
+bad:
+ *rune = Bad;
+ return 1;
+}
+
+int
+runetochar(char *str, Rune *rune)
+{
+ long c;
+
+ /*
+ * one character sequence
+ * 00000-0007F => 00-7F
+ */
+ c = *rune;
+ if(c <= Rune1) {
+ str[0] = c;
+ return 1;
+ }
+
+ /*
+ * two character sequence
+ * 0080-07FF => T2 Tx
+ */
+ if(c <= Rune2) {
+ str[0] = T2 | (c >> 1*Bitx);
+ str[1] = Tx | (c & Maskx);
+ return 2;
+ }
+
+ /*
+ * three character sequence
+ * 0800-FFFF => T3 Tx Tx
+ */
+ str[0] = T3 | (c >> 2*Bitx);
+ str[1] = Tx | ((c >> 1*Bitx) & Maskx);
+ str[2] = Tx | (c & Maskx);
+ return 3;
+}
+
+int
+runelen(long c)
+{
+ Rune rune;
+ char str[10];
+
+ rune = c;
+ return runetochar(str, &rune);
+}
+
+int
+runenlen(Rune *r, int nrune)
+{
+ int nb, c;
+
+ nb = 0;
+ while(nrune--) {
+ c = *r++;
+ if(c <= Rune1)
+ nb++;
+ else
+ if(c <= Rune2)
+ nb += 2;
+ else
+ nb += 3;
+ }
+ return nb;
+}
+
+int
+fullrune(char *str, int n)
+{
+ int c;
+
+ if(n > 0) {
+ c = *(uchar*)str;
+ if(c < Tx)
+ return 1;
+ if(n > 1)
+ if(c < T3 || n > 2)
+ return 1;
+ }
+ return 0;
+}
DIR diff --git a/src/libutf/runestrcat.3 b/src/libutf/runestrcat.3
t@@ -0,0 +1,65 @@
+.TH RUNESTRCAT 3
+.SH NAME
+runestrcat,
+runestrncat,
+runestrcmp,
+runestrncmp,
+runestrcpy,
+runestrncpy,
+runestrecpy,
+runestrlen,
+runestrchr,
+runestrrchr,
+runestrdup,
+runestrstr \- rune string operations
+.SH SYNOPSIS
+.B #include <utf.h>
+.PP
+.ta \w'\fLRune* \fP'u
+.B
+Rune* runestrcat(Rune *s1, Rune *s2)
+.PP
+.B
+Rune* runestrncat(Rune *s1, Rune *s2, long n)
+.PP
+.B
+int runestrcmp(Rune *s1, Rune *s2)
+.PP
+.B
+int runestrncmp(Rune *s1, Rune *s2, long n)
+.PP
+.B
+Rune* runestrcpy(Rune *s1, Rune *s2)
+.PP
+.B
+Rune* runestrncpy(Rune *s1, Rune *s2, long n)
+.PP
+.B
+Rune* runestrecpy(Rune *s1, Rune *es1, Rune *s2)
+.PP
+.B
+long runestrlen(Rune *s)
+.PP
+.B
+Rune* runestrchr(Rune *s, Rune c)
+.PP
+.B
+Rune* runestrrchr(Rune *s, Rune c)
+.PP
+.B
+Rune* runestrdup(Rune *s)
+.PP
+.B
+Rune* runestrstr(Rune *s1, Rune *s2)
+.SH DESCRIPTION
+These functions are rune string analogues of
+the corresponding functions in
+.IR strcat (3).
+.SH HISTORY
+These routines first appeared in Plan 9.
+.SH SEE ALSO
+.IR memmove (3),
+.IR rune (3),
+.IR strcat (2)
+.SH BUGS
+The outcome of overlapping moves varies among implementations.
DIR diff --git a/src/libutf/runestrcat.c b/src/libutf/runestrcat.c
t@@ -0,0 +1,25 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrcat(Rune *s1, Rune *s2)
+{
+
+ runestrcpy(runestrchr(s1, 0), s2);
+ return s1;
+}
DIR diff --git a/src/libutf/runestrchr.c b/src/libutf/runestrchr.c
t@@ -0,0 +1,35 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrchr(Rune *s, Rune c)
+{
+ Rune c0 = c;
+ Rune c1;
+
+ if(c == 0) {
+ while(*s++)
+ ;
+ return s-1;
+ }
+
+ while(c1 = *s++)
+ if(c1 == c0)
+ return s-1;
+ return 0;
+}
DIR diff --git a/src/libutf/runestrcmp.c b/src/libutf/runestrcmp.c
t@@ -0,0 +1,35 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+int
+runestrcmp(Rune *s1, Rune *s2)
+{
+ Rune c1, c2;
+
+ for(;;) {
+ c1 = *s1++;
+ c2 = *s2++;
+ if(c1 != c2) {
+ if(c1 > c2)
+ return 1;
+ return -1;
+ }
+ if(c1 == 0)
+ return 0;
+ }
+}
DIR diff --git a/src/libutf/runestrcpy.c b/src/libutf/runestrcpy.c
t@@ -0,0 +1,28 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrcpy(Rune *s1, Rune *s2)
+{
+ Rune *os1;
+
+ os1 = s1;
+ while(*s1++ = *s2++)
+ ;
+ return os1;
+}
DIR diff --git a/src/libutf/runestrdup.c b/src/libutf/runestrdup.c
t@@ -0,0 +1,30 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include <stdlib.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrdup(Rune *s)
+{
+ Rune *ns;
+
+ ns = malloc(sizeof(Rune)*(runestrlen(s) + 1));
+ if(ns == 0)
+ return 0;
+
+ return runestrcpy(ns, s);
+}
DIR diff --git a/src/libutf/runestrecpy.c b/src/libutf/runestrecpy.c
t@@ -0,0 +1,32 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrecpy(Rune *s1, Rune *es1, Rune *s2)
+{
+ if(s1 >= es1)
+ return s1;
+
+ while(*s1++ = *s2++){
+ if(s1 == es1){
+ *--s1 = '\0';
+ break;
+ }
+ }
+ return s1;
+}
DIR diff --git a/src/libutf/runestrlen.c b/src/libutf/runestrlen.c
t@@ -0,0 +1,24 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+long
+runestrlen(Rune *s)
+{
+
+ return runestrchr(s, 0) - s;
+}
DIR diff --git a/src/libutf/runestrncat.c b/src/libutf/runestrncat.c
t@@ -0,0 +1,32 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrncat(Rune *s1, Rune *s2, long n)
+{
+ Rune *os1;
+
+ os1 = s1;
+ s1 = runestrchr(s1, 0);
+ while(*s1++ = *s2++)
+ if(--n < 0) {
+ s1[-1] = 0;
+ break;
+ }
+ return os1;
+}
DIR diff --git a/src/libutf/runestrncmp.c b/src/libutf/runestrncmp.c
t@@ -0,0 +1,37 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+int
+runestrncmp(Rune *s1, Rune *s2, long n)
+{
+ Rune c1, c2;
+
+ while(n > 0) {
+ c1 = *s1++;
+ c2 = *s2++;
+ n--;
+ if(c1 != c2) {
+ if(c1 > c2)
+ return 1;
+ return -1;
+ }
+ if(c1 == 0)
+ break;
+ }
+ return 0;
+}
DIR diff --git a/src/libutf/runestrncpy.c b/src/libutf/runestrncpy.c
t@@ -0,0 +1,33 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrncpy(Rune *s1, Rune *s2, long n)
+{
+ int i;
+ Rune *os1;
+
+ os1 = s1;
+ for(i = 0; i < n; i++)
+ if((*s1++ = *s2++) == 0) {
+ while(++i < n)
+ *s1++ = 0;
+ return os1;
+ }
+ return os1;
+}
DIR diff --git a/src/libutf/runestrrchr.c b/src/libutf/runestrrchr.c
t@@ -0,0 +1,30 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+Rune*
+runestrrchr(Rune *s, Rune c)
+{
+ Rune *r;
+
+ if(c == 0)
+ return runestrchr(s, 0);
+ r = 0;
+ while(s = runestrchr(s, c))
+ r = s++;
+ return r;
+}
DIR diff --git a/src/libutf/runestrstr.c b/src/libutf/runestrstr.c
t@@ -0,0 +1,44 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+/*
+ * Return pointer to first occurrence of s2 in s1,
+ * 0 if none
+ */
+Rune*
+runestrstr(Rune *s1, Rune *s2)
+{
+ Rune *p, *pa, *pb;
+ int c0, c;
+
+ c0 = *s2;
+ if(c0 == 0)
+ return s1;
+ s2++;
+ for(p=runestrchr(s1, c0); p; p=runestrchr(p+1, c0)) {
+ pa = p;
+ for(pb=s2;; pb++) {
+ c = *pb;
+ if(c == 0)
+ return p;
+ if(c != *++pa)
+ break;
+ }
+ }
+ return 0;
+}
DIR diff --git a/src/libutf/runetype.c b/src/libutf/runetype.c
t@@ -0,0 +1,1152 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+/*
+ * alpha ranges -
+ * only covers ranges not in lower||upper
+ */
+static
+Rune __alpha2[] =
+{
+ 0x00d8, 0x00f6, /* Ø - ö */
+ 0x00f8, 0x01f5, /* ø - ǵ */
+ 0x0250, 0x02a8, /* ɐ - ʨ */
+ 0x038e, 0x03a1, /* Ύ - Ρ */
+ 0x03a3, 0x03ce, /* Σ - ώ */
+ 0x03d0, 0x03d6, /* ϐ - ϖ */
+ 0x03e2, 0x03f3, /* Ϣ - ϳ */
+ 0x0490, 0x04c4, /* Ґ - ӄ */
+ 0x0561, 0x0587, /* ա - և */
+ 0x05d0, 0x05ea, /* א - ת */
+ 0x05f0, 0x05f2, /* װ - ײ */
+ 0x0621, 0x063a, /* ء - غ */
+ 0x0640, 0x064a, /* ـ - ي */
+ 0x0671, 0x06b7, /* ٱ - ڷ */
+ 0x06ba, 0x06be, /* ں - ھ */
+ 0x06c0, 0x06ce, /* ۀ - ێ */
+ 0x06d0, 0x06d3, /* ې - ۓ */
+ 0x0905, 0x0939, /* अ - ह */
+ 0x0958, 0x0961, /* क़ - ॡ */
+ 0x0985, 0x098c, /* অ - ঌ */
+ 0x098f, 0x0990, /* এ - ঐ */
+ 0x0993, 0x09a8, /* ও - ন */
+ 0x09aa, 0x09b0, /* প - র */
+ 0x09b6, 0x09b9, /* শ - হ */
+ 0x09dc, 0x09dd, /* ড় - ঢ় */
+ 0x09df, 0x09e1, /* য় - ৡ */
+ 0x09f0, 0x09f1, /* ৰ - ৱ */
+ 0x0a05, 0x0a0a, /* ਅ - ਊ */
+ 0x0a0f, 0x0a10, /* ਏ - ਐ */
+ 0x0a13, 0x0a28, /* ਓ - ਨ */
+ 0x0a2a, 0x0a30, /* ਪ - ਰ */
+ 0x0a32, 0x0a33, /* ਲ - ਲ਼ */
+ 0x0a35, 0x0a36, /* ਵ - ਸ਼ */
+ 0x0a38, 0x0a39, /* ਸ - ਹ */
+ 0x0a59, 0x0a5c, /* ਖ਼ - ੜ */
+ 0x0a85, 0x0a8b, /* અ - ઋ */
+ 0x0a8f, 0x0a91, /* એ - ઑ */
+ 0x0a93, 0x0aa8, /* ઓ - ન */
+ 0x0aaa, 0x0ab0, /* પ - ર */
+ 0x0ab2, 0x0ab3, /* લ - ળ */
+ 0x0ab5, 0x0ab9, /* વ - હ */
+ 0x0b05, 0x0b0c, /* ଅ - ଌ */
+ 0x0b0f, 0x0b10, /* ଏ - ଐ */
+ 0x0b13, 0x0b28, /* ଓ - ନ */
+ 0x0b2a, 0x0b30, /* ପ - ର */
+ 0x0b32, 0x0b33, /* ଲ - ଳ */
+ 0x0b36, 0x0b39, /* ଶ - ହ */
+ 0x0b5c, 0x0b5d, /* ଡ଼ - ଢ଼ */
+ 0x0b5f, 0x0b61, /* ୟ - ୡ */
+ 0x0b85, 0x0b8a, /* அ - ஊ */
+ 0x0b8e, 0x0b90, /* எ - ஐ */
+ 0x0b92, 0x0b95, /* ஒ - க */
+ 0x0b99, 0x0b9a, /* ங - ச */
+ 0x0b9e, 0x0b9f, /* ஞ - ட */
+ 0x0ba3, 0x0ba4, /* ண - த */
+ 0x0ba8, 0x0baa, /* ந - ப */
+ 0x0bae, 0x0bb5, /* ம - வ */
+ 0x0bb7, 0x0bb9, /* ஷ - ஹ */
+ 0x0c05, 0x0c0c, /* అ - ఌ */
+ 0x0c0e, 0x0c10, /* ఎ - ఐ */
+ 0x0c12, 0x0c28, /* ఒ - న */
+ 0x0c2a, 0x0c33, /* ప - ళ */
+ 0x0c35, 0x0c39, /* వ - హ */
+ 0x0c60, 0x0c61, /* ౠ - ౡ */
+ 0x0c85, 0x0c8c, /* ಅ - ಌ */
+ 0x0c8e, 0x0c90, /* ಎ - ಐ */
+ 0x0c92, 0x0ca8, /* ಒ - ನ */
+ 0x0caa, 0x0cb3, /* ಪ - ಳ */
+ 0x0cb5, 0x0cb9, /* ವ - ಹ */
+ 0x0ce0, 0x0ce1, /* ೠ - ೡ */
+ 0x0d05, 0x0d0c, /* അ - ഌ */
+ 0x0d0e, 0x0d10, /* എ - ഐ */
+ 0x0d12, 0x0d28, /* ഒ - ന */
+ 0x0d2a, 0x0d39, /* പ - ഹ */
+ 0x0d60, 0x0d61, /* ൠ - ൡ */
+ 0x0e01, 0x0e30, /* ก - ะ */
+ 0x0e32, 0x0e33, /* า - ำ */
+ 0x0e40, 0x0e46, /* เ - ๆ */
+ 0x0e5a, 0x0e5b, /* ๚ - ๛ */
+ 0x0e81, 0x0e82, /* ກ - ຂ */
+ 0x0e87, 0x0e88, /* ງ - ຈ */
+ 0x0e94, 0x0e97, /* ດ - ທ */
+ 0x0e99, 0x0e9f, /* ນ - ຟ */
+ 0x0ea1, 0x0ea3, /* ມ - ຣ */
+ 0x0eaa, 0x0eab, /* ສ - ຫ */
+ 0x0ead, 0x0eae, /* ອ - ຮ */
+ 0x0eb2, 0x0eb3, /* າ - ຳ */
+ 0x0ec0, 0x0ec4, /* ເ - ໄ */
+ 0x0edc, 0x0edd, /* ໜ - ໝ */
+ 0x0f18, 0x0f19, /* ༘ - ༙ */
+ 0x0f40, 0x0f47, /* ཀ - ཇ */
+ 0x0f49, 0x0f69, /* ཉ - ཀྵ */
+ 0x10d0, 0x10f6, /* ა - ჶ */
+ 0x1100, 0x1159, /* ᄀ - ᅙ */
+ 0x115f, 0x11a2, /* ᅟ - ᆢ */
+ 0x11a8, 0x11f9, /* ᆨ - ᇹ */
+ 0x1e00, 0x1e9b, /* Ḁ - ẛ */
+ 0x1f50, 0x1f57, /* ὐ - ὗ */
+ 0x1f80, 0x1fb4, /* ᾀ - ᾴ */
+ 0x1fb6, 0x1fbc, /* ᾶ - ᾼ */
+ 0x1fc2, 0x1fc4, /* ῂ - ῄ */
+ 0x1fc6, 0x1fcc, /* ῆ - ῌ */
+ 0x1fd0, 0x1fd3, /* ῐ - ΐ */
+ 0x1fd6, 0x1fdb, /* ῖ - Ί */
+ 0x1fe0, 0x1fec, /* ῠ - Ῥ */
+ 0x1ff2, 0x1ff4, /* ῲ - ῴ */
+ 0x1ff6, 0x1ffc, /* ῶ - ῼ */
+ 0x210a, 0x2113, /* ℊ - ℓ */
+ 0x2115, 0x211d, /* ℕ - ℝ */
+ 0x2120, 0x2122, /* ℠ - ™ */
+ 0x212a, 0x2131, /* K - ℱ */
+ 0x2133, 0x2138, /* ℳ - ℸ */
+ 0x3041, 0x3094, /* ぁ - ゔ */
+ 0x30a1, 0x30fa, /* ァ - ヺ */
+ 0x3105, 0x312c, /* ㄅ - ㄬ */
+ 0x3131, 0x318e, /* ㄱ - ㆎ */
+ 0x3192, 0x319f, /* ㆒ - ㆟ */
+ 0x3260, 0x327b, /* ㉠ - ㉻ */
+ 0x328a, 0x32b0, /* ㊊ - ㊰ */
+ 0x32d0, 0x32fe, /* ㋐ - ㋾ */
+ 0x3300, 0x3357, /* ㌀ - ㍗ */
+ 0x3371, 0x3376, /* ㍱ - ㍶ */
+ 0x337b, 0x3394, /* ㍻ - ㎔ */
+ 0x3399, 0x339e, /* ㎙ - ㎞ */
+ 0x33a9, 0x33ad, /* ㎩ - ㎭ */
+ 0x33b0, 0x33c1, /* ㎰ - ㏁ */
+ 0x33c3, 0x33c5, /* ㏃ - ㏅ */
+ 0x33c7, 0x33d7, /* ㏇ - ㏗ */
+ 0x33d9, 0x33dd, /* ㏙ - ㏝ */
+ 0x4e00, 0x9fff, /* 一 - 鿿 */
+ 0xac00, 0xd7a3, /* 가 - 힣 */
+ 0xf900, 0xfb06, /* 豈 - st */
+ 0xfb13, 0xfb17, /* ﬓ - ﬗ */
+ 0xfb1f, 0xfb28, /* ײַ - ﬨ */
+ 0xfb2a, 0xfb36, /* שׁ - זּ */
+ 0xfb38, 0xfb3c, /* טּ - לּ */
+ 0xfb40, 0xfb41, /* נּ - סּ */
+ 0xfb43, 0xfb44, /* ףּ - פּ */
+ 0xfb46, 0xfbb1, /* צּ - ﮱ */
+ 0xfbd3, 0xfd3d, /* ﯓ - ﴽ */
+ 0xfd50, 0xfd8f, /* ﵐ - ﶏ */
+ 0xfd92, 0xfdc7, /* ﶒ - ﷇ */
+ 0xfdf0, 0xfdf9, /* ﷰ - ﷹ */
+ 0xfe70, 0xfe72, /* ﹰ - ﹲ */
+ 0xfe76, 0xfefc, /* ﹶ - ﻼ */
+ 0xff66, 0xff6f, /* ｦ - ｯ */
+ 0xff71, 0xff9d, /* ｱ - ﾝ */
+ 0xffa0, 0xffbe, /* ﾠ - ﾾ */
+ 0xffc2, 0xffc7, /* ￂ - ￇ */
+ 0xffca, 0xffcf, /* ￊ - ￏ */
+ 0xffd2, 0xffd7, /* ￒ - ￗ */
+ 0xffda, 0xffdc, /* ￚ - ￜ */
+};
+
+/*
+ * alpha singlets -
+ * only covers ranges not in lower||upper
+ */
+static
+Rune __alpha1[] =
+{
+ 0x00aa, /* ª */
+ 0x00b5, /* µ */
+ 0x00ba, /* º */
+ 0x03da, /* Ϛ */
+ 0x03dc, /* Ϝ */
+ 0x03de, /* Ϟ */
+ 0x03e0, /* Ϡ */
+ 0x06d5, /* ە */
+ 0x09b2, /* ল */
+ 0x0a5e, /* ਫ਼ */
+ 0x0a8d, /* ઍ */
+ 0x0ae0, /* ૠ */
+ 0x0b9c, /* ஜ */
+ 0x0cde, /* ೞ */
+ 0x0e4f, /* ๏ */
+ 0x0e84, /* ຄ */
+ 0x0e8a, /* ຊ */
+ 0x0e8d, /* ຍ */
+ 0x0ea5, /* ລ */
+ 0x0ea7, /* ວ */
+ 0x0eb0, /* ະ */
+ 0x0ebd, /* ຽ */
+ 0x1fbe, /* ι */
+ 0x207f, /* ⁿ */
+ 0x20a8, /* ₨ */
+ 0x2102, /* ℂ */
+ 0x2107, /* ℇ */
+ 0x2124, /* ℤ */
+ 0x2126, /* Ω */
+ 0x2128, /* ℨ */
+ 0xfb3e, /* מּ */
+ 0xfe74, /* ﹴ */
+};
+
+/*
+ * space ranges
+ */
+static
+Rune __space2[] =
+{
+ 0x0009, 0x000a, /* tab and newline */
+ 0x0020, 0x0020, /* space */
+ 0x00a0, 0x00a0, /* */
+ 0x2000, 0x200b, /* - */
+ 0x2028, 0x2029, /*
-
*/
+ 0x3000, 0x3000, /* */
+ 0xfeff, 0xfeff, /* */
+};
+
+/*
+ * lower case ranges
+ * 3rd col is conversion excess 500
+ */
+static
+Rune __toupper2[] =
+{
+ 0x0061, 0x007a, 468, /* a-z A-Z */
+ 0x00e0, 0x00f6, 468, /* à-ö À-Ö */
+ 0x00f8, 0x00fe, 468, /* ø-þ Ø-Þ */
+ 0x0256, 0x0257, 295, /* ɖ-ɗ Ɖ-Ɗ */
+ 0x0258, 0x0259, 298, /* ɘ-ə Ǝ-Ə */
+ 0x028a, 0x028b, 283, /* ʊ-ʋ Ʊ-Ʋ */
+ 0x03ad, 0x03af, 463, /* έ-ί Έ-Ί */
+ 0x03b1, 0x03c1, 468, /* α-ρ Α-Ρ */
+ 0x03c3, 0x03cb, 468, /* σ-ϋ Σ-Ϋ */
+ 0x03cd, 0x03ce, 437, /* ύ-ώ Ύ-Ώ */
+ 0x0430, 0x044f, 468, /* а-я А-Я */
+ 0x0451, 0x045c, 420, /* ё-ќ Ё-Ќ */
+ 0x045e, 0x045f, 420, /* ў-џ Ў-Џ */
+ 0x0561, 0x0586, 452, /* ա-ֆ Ա-Ֆ */
+ 0x1f00, 0x1f07, 508, /* ἀ-ἇ Ἀ-Ἇ */
+ 0x1f10, 0x1f15, 508, /* ἐ-ἕ Ἐ-Ἕ */
+ 0x1f20, 0x1f27, 508, /* ἠ-ἧ Ἠ-Ἧ */
+ 0x1f30, 0x1f37, 508, /* ἰ-ἷ Ἰ-Ἷ */
+ 0x1f40, 0x1f45, 508, /* ὀ-ὅ Ὀ-Ὅ */
+ 0x1f60, 0x1f67, 508, /* ὠ-ὧ Ὠ-Ὧ */
+ 0x1f70, 0x1f71, 574, /* ὰ-ά Ὰ-Ά */
+ 0x1f72, 0x1f75, 586, /* ὲ-ή Ὲ-Ή */
+ 0x1f76, 0x1f77, 600, /* ὶ-ί Ὶ-Ί */
+ 0x1f78, 0x1f79, 628, /* ὸ-ό Ὸ-Ό */
+ 0x1f7a, 0x1f7b, 612, /* ὺ-ύ Ὺ-Ύ */
+ 0x1f7c, 0x1f7d, 626, /* ὼ-ώ Ὼ-Ώ */
+ 0x1f80, 0x1f87, 508, /* ᾀ-ᾇ ᾈ-ᾏ */
+ 0x1f90, 0x1f97, 508, /* ᾐ-ᾗ ᾘ-ᾟ */
+ 0x1fa0, 0x1fa7, 508, /* ᾠ-ᾧ ᾨ-ᾯ */
+ 0x1fb0, 0x1fb1, 508, /* ᾰ-ᾱ Ᾰ-Ᾱ */
+ 0x1fd0, 0x1fd1, 508, /* ῐ-ῑ Ῐ-Ῑ */
+ 0x1fe0, 0x1fe1, 508, /* ῠ-ῡ Ῠ-Ῡ */
+ 0x2170, 0x217f, 484, /* ⅰ-ⅿ Ⅰ-Ⅿ */
+ 0x24d0, 0x24e9, 474, /* ⓐ-ⓩ Ⓐ-Ⓩ */
+ 0xff41, 0xff5a, 468, /* ａ-ｚ Ａ-Ｚ */
+};
+
+/*
+ * lower case singlets
+ * 2nd col is conversion excess 500
+ */
+static
+Rune __toupper1[] =
+{
+ 0x00ff, 621, /* ÿ Ÿ */
+ 0x0101, 499, /* ā Ā */
+ 0x0103, 499, /* ă Ă */
+ 0x0105, 499, /* ą Ą */
+ 0x0107, 499, /* ć Ć */
+ 0x0109, 499, /* ĉ Ĉ */
+ 0x010b, 499, /* ċ Ċ */
+ 0x010d, 499, /* č Č */
+ 0x010f, 499, /* ď Ď */
+ 0x0111, 499, /* đ Đ */
+ 0x0113, 499, /* ē Ē */
+ 0x0115, 499, /* ĕ Ĕ */
+ 0x0117, 499, /* ė Ė */
+ 0x0119, 499, /* ę Ę */
+ 0x011b, 499, /* ě Ě */
+ 0x011d, 499, /* ĝ Ĝ */
+ 0x011f, 499, /* ğ Ğ */
+ 0x0121, 499, /* ġ Ġ */
+ 0x0123, 499, /* ģ Ģ */
+ 0x0125, 499, /* ĥ Ĥ */
+ 0x0127, 499, /* ħ Ħ */
+ 0x0129, 499, /* ĩ Ĩ */
+ 0x012b, 499, /* ī Ī */
+ 0x012d, 499, /* ĭ Ĭ */
+ 0x012f, 499, /* į Į */
+ 0x0131, 268, /* ı I */
+ 0x0133, 499, /* ij IJ */
+ 0x0135, 499, /* ĵ Ĵ */
+ 0x0137, 499, /* ķ Ķ */
+ 0x013a, 499, /* ĺ Ĺ */
+ 0x013c, 499, /* ļ Ļ */
+ 0x013e, 499, /* ľ Ľ */
+ 0x0140, 499, /* ŀ Ŀ */
+ 0x0142, 499, /* ł Ł */
+ 0x0144, 499, /* ń Ń */
+ 0x0146, 499, /* ņ Ņ */
+ 0x0148, 499, /* ň Ň */
+ 0x014b, 499, /* ŋ Ŋ */
+ 0x014d, 499, /* ō Ō */
+ 0x014f, 499, /* ŏ Ŏ */
+ 0x0151, 499, /* ő Ő */
+ 0x0153, 499, /* œ Œ */
+ 0x0155, 499, /* ŕ Ŕ */
+ 0x0157, 499, /* ŗ Ŗ */
+ 0x0159, 499, /* ř Ř */
+ 0x015b, 499, /* ś Ś */
+ 0x015d, 499, /* ŝ Ŝ */
+ 0x015f, 499, /* ş Ş */
+ 0x0161, 499, /* š Š */
+ 0x0163, 499, /* ţ Ţ */
+ 0x0165, 499, /* ť Ť */
+ 0x0167, 499, /* ŧ Ŧ */
+ 0x0169, 499, /* ũ Ũ */
+ 0x016b, 499, /* ū Ū */
+ 0x016d, 499, /* ŭ Ŭ */
+ 0x016f, 499, /* ů Ů */
+ 0x0171, 499, /* ű Ű */
+ 0x0173, 499, /* ų Ų */
+ 0x0175, 499, /* ŵ Ŵ */
+ 0x0177, 499, /* ŷ Ŷ */
+ 0x017a, 499, /* ź Ź */
+ 0x017c, 499, /* ż Ż */
+ 0x017e, 499, /* ž Ž */
+ 0x017f, 200, /* ſ S */
+ 0x0183, 499, /* ƃ Ƃ */
+ 0x0185, 499, /* ƅ Ƅ */
+ 0x0188, 499, /* ƈ Ƈ */
+ 0x018c, 499, /* ƌ Ƌ */
+ 0x0192, 499, /* ƒ Ƒ */
+ 0x0199, 499, /* ƙ Ƙ */
+ 0x01a1, 499, /* ơ Ơ */
+ 0x01a3, 499, /* ƣ Ƣ */
+ 0x01a5, 499, /* ƥ Ƥ */
+ 0x01a8, 499, /* ƨ Ƨ */
+ 0x01ad, 499, /* ƭ Ƭ */
+ 0x01b0, 499, /* ư Ư */
+ 0x01b4, 499, /* ƴ Ƴ */
+ 0x01b6, 499, /* ƶ Ƶ */
+ 0x01b9, 499, /* ƹ Ƹ */
+ 0x01bd, 499, /* ƽ Ƽ */
+ 0x01c5, 499, /* Dž DŽ */
+ 0x01c6, 498, /* dž DŽ */
+ 0x01c8, 499, /* Lj LJ */
+ 0x01c9, 498, /* lj LJ */
+ 0x01cb, 499, /* Nj NJ */
+ 0x01cc, 498, /* nj NJ */
+ 0x01ce, 499, /* ǎ Ǎ */
+ 0x01d0, 499, /* ǐ Ǐ */
+ 0x01d2, 499, /* ǒ Ǒ */
+ 0x01d4, 499, /* ǔ Ǔ */
+ 0x01d6, 499, /* ǖ Ǖ */
+ 0x01d8, 499, /* ǘ Ǘ */
+ 0x01da, 499, /* ǚ Ǚ */
+ 0x01dc, 499, /* ǜ Ǜ */
+ 0x01df, 499, /* ǟ Ǟ */
+ 0x01e1, 499, /* ǡ Ǡ */
+ 0x01e3, 499, /* ǣ Ǣ */
+ 0x01e5, 499, /* ǥ Ǥ */
+ 0x01e7, 499, /* ǧ Ǧ */
+ 0x01e9, 499, /* ǩ Ǩ */
+ 0x01eb, 499, /* ǫ Ǫ */
+ 0x01ed, 499, /* ǭ Ǭ */
+ 0x01ef, 499, /* ǯ Ǯ */
+ 0x01f2, 499, /* Dz DZ */
+ 0x01f3, 498, /* dz DZ */
+ 0x01f5, 499, /* ǵ Ǵ */
+ 0x01fb, 499, /* ǻ Ǻ */
+ 0x01fd, 499, /* ǽ Ǽ */
+ 0x01ff, 499, /* ǿ Ǿ */
+ 0x0201, 499, /* ȁ Ȁ */
+ 0x0203, 499, /* ȃ Ȃ */
+ 0x0205, 499, /* ȅ Ȅ */
+ 0x0207, 499, /* ȇ Ȇ */
+ 0x0209, 499, /* ȉ Ȉ */
+ 0x020b, 499, /* ȋ Ȋ */
+ 0x020d, 499, /* ȍ Ȍ */
+ 0x020f, 499, /* ȏ Ȏ */
+ 0x0211, 499, /* ȑ Ȑ */
+ 0x0213, 499, /* ȓ Ȓ */
+ 0x0215, 499, /* ȕ Ȕ */
+ 0x0217, 499, /* ȗ Ȗ */
+ 0x0253, 290, /* ɓ Ɓ */
+ 0x0254, 294, /* ɔ Ɔ */
+ 0x025b, 297, /* ɛ Ɛ */
+ 0x0260, 295, /* ɠ Ɠ */
+ 0x0263, 293, /* ɣ Ɣ */
+ 0x0268, 291, /* ɨ Ɨ */
+ 0x0269, 289, /* ɩ Ɩ */
+ 0x026f, 289, /* ɯ Ɯ */
+ 0x0272, 287, /* ɲ Ɲ */
+ 0x0283, 282, /* ʃ Ʃ */
+ 0x0288, 282, /* ʈ Ʈ */
+ 0x0292, 281, /* ʒ Ʒ */
+ 0x03ac, 462, /* ά Ά */
+ 0x03cc, 436, /* ό Ό */
+ 0x03d0, 438, /* ϐ Β */
+ 0x03d1, 443, /* ϑ Θ */
+ 0x03d5, 453, /* ϕ Φ */
+ 0x03d6, 446, /* ϖ Π */
+ 0x03e3, 499, /* ϣ Ϣ */
+ 0x03e5, 499, /* ϥ Ϥ */
+ 0x03e7, 499, /* ϧ Ϧ */
+ 0x03e9, 499, /* ϩ Ϩ */
+ 0x03eb, 499, /* ϫ Ϫ */
+ 0x03ed, 499, /* ϭ Ϭ */
+ 0x03ef, 499, /* ϯ Ϯ */
+ 0x03f0, 414, /* ϰ Κ */
+ 0x03f1, 420, /* ϱ Ρ */
+ 0x0461, 499, /* ѡ Ѡ */
+ 0x0463, 499, /* ѣ Ѣ */
+ 0x0465, 499, /* ѥ Ѥ */
+ 0x0467, 499, /* ѧ Ѧ */
+ 0x0469, 499, /* ѩ Ѩ */
+ 0x046b, 499, /* ѫ Ѫ */
+ 0x046d, 499, /* ѭ Ѭ */
+ 0x046f, 499, /* ѯ Ѯ */
+ 0x0471, 499, /* ѱ Ѱ */
+ 0x0473, 499, /* ѳ Ѳ */
+ 0x0475, 499, /* ѵ Ѵ */
+ 0x0477, 499, /* ѷ Ѷ */
+ 0x0479, 499, /* ѹ Ѹ */
+ 0x047b, 499, /* ѻ Ѻ */
+ 0x047d, 499, /* ѽ Ѽ */
+ 0x047f, 499, /* ѿ Ѿ */
+ 0x0481, 499, /* ҁ Ҁ */
+ 0x0491, 499, /* ґ Ґ */
+ 0x0493, 499, /* ғ Ғ */
+ 0x0495, 499, /* ҕ Ҕ */
+ 0x0497, 499, /* җ Җ */
+ 0x0499, 499, /* ҙ Ҙ */
+ 0x049b, 499, /* қ Қ */
+ 0x049d, 499, /* ҝ Ҝ */
+ 0x049f, 499, /* ҟ Ҟ */
+ 0x04a1, 499, /* ҡ Ҡ */
+ 0x04a3, 499, /* ң Ң */
+ 0x04a5, 499, /* ҥ Ҥ */
+ 0x04a7, 499, /* ҧ Ҧ */
+ 0x04a9, 499, /* ҩ Ҩ */
+ 0x04ab, 499, /* ҫ Ҫ */
+ 0x04ad, 499, /* ҭ Ҭ */
+ 0x04af, 499, /* ү Ү */
+ 0x04b1, 499, /* ұ Ұ */
+ 0x04b3, 499, /* ҳ Ҳ */
+ 0x04b5, 499, /* ҵ Ҵ */
+ 0x04b7, 499, /* ҷ Ҷ */
+ 0x04b9, 499, /* ҹ Ҹ */
+ 0x04bb, 499, /* һ Һ */
+ 0x04bd, 499, /* ҽ Ҽ */
+ 0x04bf, 499, /* ҿ Ҿ */
+ 0x04c2, 499, /* ӂ Ӂ */
+ 0x04c4, 499, /* ӄ Ӄ */
+ 0x04c8, 499, /* ӈ Ӈ */
+ 0x04cc, 499, /* ӌ Ӌ */
+ 0x04d1, 499, /* ӑ Ӑ */
+ 0x04d3, 499, /* ӓ Ӓ */
+ 0x04d5, 499, /* ӕ Ӕ */
+ 0x04d7, 499, /* ӗ Ӗ */
+ 0x04d9, 499, /* ә Ә */
+ 0x04db, 499, /* ӛ Ӛ */
+ 0x04dd, 499, /* ӝ Ӝ */
+ 0x04df, 499, /* ӟ Ӟ */
+ 0x04e1, 499, /* ӡ Ӡ */
+ 0x04e3, 499, /* ӣ Ӣ */
+ 0x04e5, 499, /* ӥ Ӥ */
+ 0x04e7, 499, /* ӧ Ӧ */
+ 0x04e9, 499, /* ө Ө */
+ 0x04eb, 499, /* ӫ Ӫ */
+ 0x04ef, 499, /* ӯ Ӯ */
+ 0x04f1, 499, /* ӱ Ӱ */
+ 0x04f3, 499, /* ӳ Ӳ */
+ 0x04f5, 499, /* ӵ Ӵ */
+ 0x04f9, 499, /* ӹ Ӹ */
+ 0x1e01, 499, /* ḁ Ḁ */
+ 0x1e03, 499, /* ḃ Ḃ */
+ 0x1e05, 499, /* ḅ Ḅ */
+ 0x1e07, 499, /* ḇ Ḇ */
+ 0x1e09, 499, /* ḉ Ḉ */
+ 0x1e0b, 499, /* ḋ Ḋ */
+ 0x1e0d, 499, /* ḍ Ḍ */
+ 0x1e0f, 499, /* ḏ Ḏ */
+ 0x1e11, 499, /* ḑ Ḑ */
+ 0x1e13, 499, /* ḓ Ḓ */
+ 0x1e15, 499, /* ḕ Ḕ */
+ 0x1e17, 499, /* ḗ Ḗ */
+ 0x1e19, 499, /* ḙ Ḙ */
+ 0x1e1b, 499, /* ḛ Ḛ */
+ 0x1e1d, 499, /* ḝ Ḝ */
+ 0x1e1f, 499, /* ḟ Ḟ */
+ 0x1e21, 499, /* ḡ Ḡ */
+ 0x1e23, 499, /* ḣ Ḣ */
+ 0x1e25, 499, /* ḥ Ḥ */
+ 0x1e27, 499, /* ḧ Ḧ */
+ 0x1e29, 499, /* ḩ Ḩ */
+ 0x1e2b, 499, /* ḫ Ḫ */
+ 0x1e2d, 499, /* ḭ Ḭ */
+ 0x1e2f, 499, /* ḯ Ḯ */
+ 0x1e31, 499, /* ḱ Ḱ */
+ 0x1e33, 499, /* ḳ Ḳ */
+ 0x1e35, 499, /* ḵ Ḵ */
+ 0x1e37, 499, /* ḷ Ḷ */
+ 0x1e39, 499, /* ḹ Ḹ */
+ 0x1e3b, 499, /* ḻ Ḻ */
+ 0x1e3d, 499, /* ḽ Ḽ */
+ 0x1e3f, 499, /* ḿ Ḿ */
+ 0x1e41, 499, /* ṁ Ṁ */
+ 0x1e43, 499, /* ṃ Ṃ */
+ 0x1e45, 499, /* ṅ Ṅ */
+ 0x1e47, 499, /* ṇ Ṇ */
+ 0x1e49, 499, /* ṉ Ṉ */
+ 0x1e4b, 499, /* ṋ Ṋ */
+ 0x1e4d, 499, /* ṍ Ṍ */
+ 0x1e4f, 499, /* ṏ Ṏ */
+ 0x1e51, 499, /* ṑ Ṑ */
+ 0x1e53, 499, /* ṓ Ṓ */
+ 0x1e55, 499, /* ṕ Ṕ */
+ 0x1e57, 499, /* ṗ Ṗ */
+ 0x1e59, 499, /* ṙ Ṙ */
+ 0x1e5b, 499, /* ṛ Ṛ */
+ 0x1e5d, 499, /* ṝ Ṝ */
+ 0x1e5f, 499, /* ṟ Ṟ */
+ 0x1e61, 499, /* ṡ Ṡ */
+ 0x1e63, 499, /* ṣ Ṣ */
+ 0x1e65, 499, /* ṥ Ṥ */
+ 0x1e67, 499, /* ṧ Ṧ */
+ 0x1e69, 499, /* ṩ Ṩ */
+ 0x1e6b, 499, /* ṫ Ṫ */
+ 0x1e6d, 499, /* ṭ Ṭ */
+ 0x1e6f, 499, /* ṯ Ṯ */
+ 0x1e71, 499, /* ṱ Ṱ */
+ 0x1e73, 499, /* ṳ Ṳ */
+ 0x1e75, 499, /* ṵ Ṵ */
+ 0x1e77, 499, /* ṷ Ṷ */
+ 0x1e79, 499, /* ṹ Ṹ */
+ 0x1e7b, 499, /* ṻ Ṻ */
+ 0x1e7d, 499, /* ṽ Ṽ */
+ 0x1e7f, 499, /* ṿ Ṿ */
+ 0x1e81, 499, /* ẁ Ẁ */
+ 0x1e83, 499, /* ẃ Ẃ */
+ 0x1e85, 499, /* ẅ Ẅ */
+ 0x1e87, 499, /* ẇ Ẇ */
+ 0x1e89, 499, /* ẉ Ẉ */
+ 0x1e8b, 499, /* ẋ Ẋ */
+ 0x1e8d, 499, /* ẍ Ẍ */
+ 0x1e8f, 499, /* ẏ Ẏ */
+ 0x1e91, 499, /* ẑ Ẑ */
+ 0x1e93, 499, /* ẓ Ẓ */
+ 0x1e95, 499, /* ẕ Ẕ */
+ 0x1ea1, 499, /* ạ Ạ */
+ 0x1ea3, 499, /* ả Ả */
+ 0x1ea5, 499, /* ấ Ấ */
+ 0x1ea7, 499, /* ầ Ầ */
+ 0x1ea9, 499, /* ẩ Ẩ */
+ 0x1eab, 499, /* ẫ Ẫ */
+ 0x1ead, 499, /* ậ Ậ */
+ 0x1eaf, 499, /* ắ Ắ */
+ 0x1eb1, 499, /* ằ Ằ */
+ 0x1eb3, 499, /* ẳ Ẳ */
+ 0x1eb5, 499, /* ẵ Ẵ */
+ 0x1eb7, 499, /* ặ Ặ */
+ 0x1eb9, 499, /* ẹ Ẹ */
+ 0x1ebb, 499, /* ẻ Ẻ */
+ 0x1ebd, 499, /* ẽ Ẽ */
+ 0x1ebf, 499, /* ế Ế */
+ 0x1ec1, 499, /* ề Ề */
+ 0x1ec3, 499, /* ể Ể */
+ 0x1ec5, 499, /* ễ Ễ */
+ 0x1ec7, 499, /* ệ Ệ */
+ 0x1ec9, 499, /* ỉ Ỉ */
+ 0x1ecb, 499, /* ị Ị */
+ 0x1ecd, 499, /* ọ Ọ */
+ 0x1ecf, 499, /* ỏ Ỏ */
+ 0x1ed1, 499, /* ố Ố */
+ 0x1ed3, 499, /* ồ Ồ */
+ 0x1ed5, 499, /* ổ Ổ */
+ 0x1ed7, 499, /* ỗ Ỗ */
+ 0x1ed9, 499, /* ộ Ộ */
+ 0x1edb, 499, /* ớ Ớ */
+ 0x1edd, 499, /* ờ Ờ */
+ 0x1edf, 499, /* ở Ở */
+ 0x1ee1, 499, /* ỡ Ỡ */
+ 0x1ee3, 499, /* ợ Ợ */
+ 0x1ee5, 499, /* ụ Ụ */
+ 0x1ee7, 499, /* ủ Ủ */
+ 0x1ee9, 499, /* ứ Ứ */
+ 0x1eeb, 499, /* ừ Ừ */
+ 0x1eed, 499, /* ử Ử */
+ 0x1eef, 499, /* ữ Ữ */
+ 0x1ef1, 499, /* ự Ự */
+ 0x1ef3, 499, /* ỳ Ỳ */
+ 0x1ef5, 499, /* ỵ Ỵ */
+ 0x1ef7, 499, /* ỷ Ỷ */
+ 0x1ef9, 499, /* ỹ Ỹ */
+ 0x1f51, 508, /* ὑ Ὑ */
+ 0x1f53, 508, /* ὓ Ὓ */
+ 0x1f55, 508, /* ὕ Ὕ */
+ 0x1f57, 508, /* ὗ Ὗ */
+ 0x1fb3, 509, /* ᾳ ᾼ */
+ 0x1fc3, 509, /* ῃ ῌ */
+ 0x1fe5, 507, /* ῥ Ῥ */
+ 0x1ff3, 509, /* ῳ ῼ */
+};
+
+/*
+ * upper case ranges
+ * 3rd col is conversion excess 500
+ */
+static
+Rune __tolower2[] =
+{
+ 0x0041, 0x005a, 532, /* A-Z a-z */
+ 0x00c0, 0x00d6, 532, /* À-Ö à-ö */
+ 0x00d8, 0x00de, 532, /* Ø-Þ ø-þ */
+ 0x0189, 0x018a, 705, /* Ɖ-Ɗ ɖ-ɗ */
+ 0x018e, 0x018f, 702, /* Ǝ-Ə ɘ-ə */
+ 0x01b1, 0x01b2, 717, /* Ʊ-Ʋ ʊ-ʋ */
+ 0x0388, 0x038a, 537, /* Έ-Ί έ-ί */
+ 0x038e, 0x038f, 563, /* Ύ-Ώ ύ-ώ */
+ 0x0391, 0x03a1, 532, /* Α-Ρ α-ρ */
+ 0x03a3, 0x03ab, 532, /* Σ-Ϋ σ-ϋ */
+ 0x0401, 0x040c, 580, /* Ё-Ќ ё-ќ */
+ 0x040e, 0x040f, 580, /* Ў-Џ ў-џ */
+ 0x0410, 0x042f, 532, /* А-Я а-я */
+ 0x0531, 0x0556, 548, /* Ա-Ֆ ա-ֆ */
+ 0x10a0, 0x10c5, 548, /* Ⴀ-Ⴥ ა-ჵ */
+ 0x1f08, 0x1f0f, 492, /* Ἀ-Ἇ ἀ-ἇ */
+ 0x1f18, 0x1f1d, 492, /* Ἐ-Ἕ ἐ-ἕ */
+ 0x1f28, 0x1f2f, 492, /* Ἠ-Ἧ ἠ-ἧ */
+ 0x1f38, 0x1f3f, 492, /* Ἰ-Ἷ ἰ-ἷ */
+ 0x1f48, 0x1f4d, 492, /* Ὀ-Ὅ ὀ-ὅ */
+ 0x1f68, 0x1f6f, 492, /* Ὠ-Ὧ ὠ-ὧ */
+ 0x1f88, 0x1f8f, 492, /* ᾈ-ᾏ ᾀ-ᾇ */
+ 0x1f98, 0x1f9f, 492, /* ᾘ-ᾟ ᾐ-ᾗ */
+ 0x1fa8, 0x1faf, 492, /* ᾨ-ᾯ ᾠ-ᾧ */
+ 0x1fb8, 0x1fb9, 492, /* Ᾰ-Ᾱ ᾰ-ᾱ */
+ 0x1fba, 0x1fbb, 426, /* Ὰ-Ά ὰ-ά */
+ 0x1fc8, 0x1fcb, 414, /* Ὲ-Ή ὲ-ή */
+ 0x1fd8, 0x1fd9, 492, /* Ῐ-Ῑ ῐ-ῑ */
+ 0x1fda, 0x1fdb, 400, /* Ὶ-Ί ὶ-ί */
+ 0x1fe8, 0x1fe9, 492, /* Ῠ-Ῡ ῠ-ῡ */
+ 0x1fea, 0x1feb, 388, /* Ὺ-Ύ ὺ-ύ */
+ 0x1ff8, 0x1ff9, 372, /* Ὸ-Ό ὸ-ό */
+ 0x1ffa, 0x1ffb, 374, /* Ὼ-Ώ ὼ-ώ */
+ 0x2160, 0x216f, 516, /* Ⅰ-Ⅿ ⅰ-ⅿ */
+ 0x24b6, 0x24cf, 526, /* Ⓐ-Ⓩ ⓐ-ⓩ */
+ 0xff21, 0xff3a, 532, /* Ａ-Ｚ ａ-ｚ */
+};
+
+/*
+ * upper case singlets
+ * 2nd col is conversion excess 500
+ */
+static
+Rune __tolower1[] =
+{
+ 0x0100, 501, /* Ā ā */
+ 0x0102, 501, /* Ă ă */
+ 0x0104, 501, /* Ą ą */
+ 0x0106, 501, /* Ć ć */
+ 0x0108, 501, /* Ĉ ĉ */
+ 0x010a, 501, /* Ċ ċ */
+ 0x010c, 501, /* Č č */
+ 0x010e, 501, /* Ď ď */
+ 0x0110, 501, /* Đ đ */
+ 0x0112, 501, /* Ē ē */
+ 0x0114, 501, /* Ĕ ĕ */
+ 0x0116, 501, /* Ė ė */
+ 0x0118, 501, /* Ę ę */
+ 0x011a, 501, /* Ě ě */
+ 0x011c, 501, /* Ĝ ĝ */
+ 0x011e, 501, /* Ğ ğ */
+ 0x0120, 501, /* Ġ ġ */
+ 0x0122, 501, /* Ģ ģ */
+ 0x0124, 501, /* Ĥ ĥ */
+ 0x0126, 501, /* Ħ ħ */
+ 0x0128, 501, /* Ĩ ĩ */
+ 0x012a, 501, /* Ī ī */
+ 0x012c, 501, /* Ĭ ĭ */
+ 0x012e, 501, /* Į į */
+ 0x0130, 301, /* İ i */
+ 0x0132, 501, /* IJ ij */
+ 0x0134, 501, /* Ĵ ĵ */
+ 0x0136, 501, /* Ķ ķ */
+ 0x0139, 501, /* Ĺ ĺ */
+ 0x013b, 501, /* Ļ ļ */
+ 0x013d, 501, /* Ľ ľ */
+ 0x013f, 501, /* Ŀ ŀ */
+ 0x0141, 501, /* Ł ł */
+ 0x0143, 501, /* Ń ń */
+ 0x0145, 501, /* Ņ ņ */
+ 0x0147, 501, /* Ň ň */
+ 0x014a, 501, /* Ŋ ŋ */
+ 0x014c, 501, /* Ō ō */
+ 0x014e, 501, /* Ŏ ŏ */
+ 0x0150, 501, /* Ő ő */
+ 0x0152, 501, /* Œ œ */
+ 0x0154, 501, /* Ŕ ŕ */
+ 0x0156, 501, /* Ŗ ŗ */
+ 0x0158, 501, /* Ř ř */
+ 0x015a, 501, /* Ś ś */
+ 0x015c, 501, /* Ŝ ŝ */
+ 0x015e, 501, /* Ş ş */
+ 0x0160, 501, /* Š š */
+ 0x0162, 501, /* Ţ ţ */
+ 0x0164, 501, /* Ť ť */
+ 0x0166, 501, /* Ŧ ŧ */
+ 0x0168, 501, /* Ũ ũ */
+ 0x016a, 501, /* Ū ū */
+ 0x016c, 501, /* Ŭ ŭ */
+ 0x016e, 501, /* Ů ů */
+ 0x0170, 501, /* Ű ű */
+ 0x0172, 501, /* Ų ų */
+ 0x0174, 501, /* Ŵ ŵ */
+ 0x0176, 501, /* Ŷ ŷ */
+ 0x0178, 379, /* Ÿ ÿ */
+ 0x0179, 501, /* Ź ź */
+ 0x017b, 501, /* Ż ż */
+ 0x017d, 501, /* Ž ž */
+ 0x0181, 710, /* Ɓ ɓ */
+ 0x0182, 501, /* Ƃ ƃ */
+ 0x0184, 501, /* Ƅ ƅ */
+ 0x0186, 706, /* Ɔ ɔ */
+ 0x0187, 501, /* Ƈ ƈ */
+ 0x018b, 501, /* Ƌ ƌ */
+ 0x0190, 703, /* Ɛ ɛ */
+ 0x0191, 501, /* Ƒ ƒ */
+ 0x0193, 705, /* Ɠ ɠ */
+ 0x0194, 707, /* Ɣ ɣ */
+ 0x0196, 711, /* Ɩ ɩ */
+ 0x0197, 709, /* Ɨ ɨ */
+ 0x0198, 501, /* Ƙ ƙ */
+ 0x019c, 711, /* Ɯ ɯ */
+ 0x019d, 713, /* Ɲ ɲ */
+ 0x01a0, 501, /* Ơ ơ */
+ 0x01a2, 501, /* Ƣ ƣ */
+ 0x01a4, 501, /* Ƥ ƥ */
+ 0x01a7, 501, /* Ƨ ƨ */
+ 0x01a9, 718, /* Ʃ ʃ */
+ 0x01ac, 501, /* Ƭ ƭ */
+ 0x01ae, 718, /* Ʈ ʈ */
+ 0x01af, 501, /* Ư ư */
+ 0x01b3, 501, /* Ƴ ƴ */
+ 0x01b5, 501, /* Ƶ ƶ */
+ 0x01b7, 719, /* Ʒ ʒ */
+ 0x01b8, 501, /* Ƹ ƹ */
+ 0x01bc, 501, /* Ƽ ƽ */
+ 0x01c4, 502, /* DŽ dž */
+ 0x01c5, 501, /* Dž dž */
+ 0x01c7, 502, /* LJ lj */
+ 0x01c8, 501, /* Lj lj */
+ 0x01ca, 502, /* NJ nj */
+ 0x01cb, 501, /* Nj nj */
+ 0x01cd, 501, /* Ǎ ǎ */
+ 0x01cf, 501, /* Ǐ ǐ */
+ 0x01d1, 501, /* Ǒ ǒ */
+ 0x01d3, 501, /* Ǔ ǔ */
+ 0x01d5, 501, /* Ǖ ǖ */
+ 0x01d7, 501, /* Ǘ ǘ */
+ 0x01d9, 501, /* Ǚ ǚ */
+ 0x01db, 501, /* Ǜ ǜ */
+ 0x01de, 501, /* Ǟ ǟ */
+ 0x01e0, 501, /* Ǡ ǡ */
+ 0x01e2, 501, /* Ǣ ǣ */
+ 0x01e4, 501, /* Ǥ ǥ */
+ 0x01e6, 501, /* Ǧ ǧ */
+ 0x01e8, 501, /* Ǩ ǩ */
+ 0x01ea, 501, /* Ǫ ǫ */
+ 0x01ec, 501, /* Ǭ ǭ */
+ 0x01ee, 501, /* Ǯ ǯ */
+ 0x01f1, 502, /* DZ dz */
+ 0x01f2, 501, /* Dz dz */
+ 0x01f4, 501, /* Ǵ ǵ */
+ 0x01fa, 501, /* Ǻ ǻ */
+ 0x01fc, 501, /* Ǽ ǽ */
+ 0x01fe, 501, /* Ǿ ǿ */
+ 0x0200, 501, /* Ȁ ȁ */
+ 0x0202, 501, /* Ȃ ȃ */
+ 0x0204, 501, /* Ȅ ȅ */
+ 0x0206, 501, /* Ȇ ȇ */
+ 0x0208, 501, /* Ȉ ȉ */
+ 0x020a, 501, /* Ȋ ȋ */
+ 0x020c, 501, /* Ȍ ȍ */
+ 0x020e, 501, /* Ȏ ȏ */
+ 0x0210, 501, /* Ȑ ȑ */
+ 0x0212, 501, /* Ȓ ȓ */
+ 0x0214, 501, /* Ȕ ȕ */
+ 0x0216, 501, /* Ȗ ȗ */
+ 0x0386, 538, /* Ά ά */
+ 0x038c, 564, /* Ό ό */
+ 0x03e2, 501, /* Ϣ ϣ */
+ 0x03e4, 501, /* Ϥ ϥ */
+ 0x03e6, 501, /* Ϧ ϧ */
+ 0x03e8, 501, /* Ϩ ϩ */
+ 0x03ea, 501, /* Ϫ ϫ */
+ 0x03ec, 501, /* Ϭ ϭ */
+ 0x03ee, 501, /* Ϯ ϯ */
+ 0x0460, 501, /* Ѡ ѡ */
+ 0x0462, 501, /* Ѣ ѣ */
+ 0x0464, 501, /* Ѥ ѥ */
+ 0x0466, 501, /* Ѧ ѧ */
+ 0x0468, 501, /* Ѩ ѩ */
+ 0x046a, 501, /* Ѫ ѫ */
+ 0x046c, 501, /* Ѭ ѭ */
+ 0x046e, 501, /* Ѯ ѯ */
+ 0x0470, 501, /* Ѱ ѱ */
+ 0x0472, 501, /* Ѳ ѳ */
+ 0x0474, 501, /* Ѵ ѵ */
+ 0x0476, 501, /* Ѷ ѷ */
+ 0x0478, 501, /* Ѹ ѹ */
+ 0x047a, 501, /* Ѻ ѻ */
+ 0x047c, 501, /* Ѽ ѽ */
+ 0x047e, 501, /* Ѿ ѿ */
+ 0x0480, 501, /* Ҁ ҁ */
+ 0x0490, 501, /* Ґ ґ */
+ 0x0492, 501, /* Ғ ғ */
+ 0x0494, 501, /* Ҕ ҕ */
+ 0x0496, 501, /* Җ җ */
+ 0x0498, 501, /* Ҙ ҙ */
+ 0x049a, 501, /* Қ қ */
+ 0x049c, 501, /* Ҝ ҝ */
+ 0x049e, 501, /* Ҟ ҟ */
+ 0x04a0, 501, /* Ҡ ҡ */
+ 0x04a2, 501, /* Ң ң */
+ 0x04a4, 501, /* Ҥ ҥ */
+ 0x04a6, 501, /* Ҧ ҧ */
+ 0x04a8, 501, /* Ҩ ҩ */
+ 0x04aa, 501, /* Ҫ ҫ */
+ 0x04ac, 501, /* Ҭ ҭ */
+ 0x04ae, 501, /* Ү ү */
+ 0x04b0, 501, /* Ұ ұ */
+ 0x04b2, 501, /* Ҳ ҳ */
+ 0x04b4, 501, /* Ҵ ҵ */
+ 0x04b6, 501, /* Ҷ ҷ */
+ 0x04b8, 501, /* Ҹ ҹ */
+ 0x04ba, 501, /* Һ һ */
+ 0x04bc, 501, /* Ҽ ҽ */
+ 0x04be, 501, /* Ҿ ҿ */
+ 0x04c1, 501, /* Ӂ ӂ */
+ 0x04c3, 501, /* Ӄ ӄ */
+ 0x04c7, 501, /* Ӈ ӈ */
+ 0x04cb, 501, /* Ӌ ӌ */
+ 0x04d0, 501, /* Ӑ ӑ */
+ 0x04d2, 501, /* Ӓ ӓ */
+ 0x04d4, 501, /* Ӕ ӕ */
+ 0x04d6, 501, /* Ӗ ӗ */
+ 0x04d8, 501, /* Ә ә */
+ 0x04da, 501, /* Ӛ ӛ */
+ 0x04dc, 501, /* Ӝ ӝ */
+ 0x04de, 501, /* Ӟ ӟ */
+ 0x04e0, 501, /* Ӡ ӡ */
+ 0x04e2, 501, /* Ӣ ӣ */
+ 0x04e4, 501, /* Ӥ ӥ */
+ 0x04e6, 501, /* Ӧ ӧ */
+ 0x04e8, 501, /* Ө ө */
+ 0x04ea, 501, /* Ӫ ӫ */
+ 0x04ee, 501, /* Ӯ ӯ */
+ 0x04f0, 501, /* Ӱ ӱ */
+ 0x04f2, 501, /* Ӳ ӳ */
+ 0x04f4, 501, /* Ӵ ӵ */
+ 0x04f8, 501, /* Ӹ ӹ */
+ 0x1e00, 501, /* Ḁ ḁ */
+ 0x1e02, 501, /* Ḃ ḃ */
+ 0x1e04, 501, /* Ḅ ḅ */
+ 0x1e06, 501, /* Ḇ ḇ */
+ 0x1e08, 501, /* Ḉ ḉ */
+ 0x1e0a, 501, /* Ḋ ḋ */
+ 0x1e0c, 501, /* Ḍ ḍ */
+ 0x1e0e, 501, /* Ḏ ḏ */
+ 0x1e10, 501, /* Ḑ ḑ */
+ 0x1e12, 501, /* Ḓ ḓ */
+ 0x1e14, 501, /* Ḕ ḕ */
+ 0x1e16, 501, /* Ḗ ḗ */
+ 0x1e18, 501, /* Ḙ ḙ */
+ 0x1e1a, 501, /* Ḛ ḛ */
+ 0x1e1c, 501, /* Ḝ ḝ */
+ 0x1e1e, 501, /* Ḟ ḟ */
+ 0x1e20, 501, /* Ḡ ḡ */
+ 0x1e22, 501, /* Ḣ ḣ */
+ 0x1e24, 501, /* Ḥ ḥ */
+ 0x1e26, 501, /* Ḧ ḧ */
+ 0x1e28, 501, /* Ḩ ḩ */
+ 0x1e2a, 501, /* Ḫ ḫ */
+ 0x1e2c, 501, /* Ḭ ḭ */
+ 0x1e2e, 501, /* Ḯ ḯ */
+ 0x1e30, 501, /* Ḱ ḱ */
+ 0x1e32, 501, /* Ḳ ḳ */
+ 0x1e34, 501, /* Ḵ ḵ */
+ 0x1e36, 501, /* Ḷ ḷ */
+ 0x1e38, 501, /* Ḹ ḹ */
+ 0x1e3a, 501, /* Ḻ ḻ */
+ 0x1e3c, 501, /* Ḽ ḽ */
+ 0x1e3e, 501, /* Ḿ ḿ */
+ 0x1e40, 501, /* Ṁ ṁ */
+ 0x1e42, 501, /* Ṃ ṃ */
+ 0x1e44, 501, /* Ṅ ṅ */
+ 0x1e46, 501, /* Ṇ ṇ */
+ 0x1e48, 501, /* Ṉ ṉ */
+ 0x1e4a, 501, /* Ṋ ṋ */
+ 0x1e4c, 501, /* Ṍ ṍ */
+ 0x1e4e, 501, /* Ṏ ṏ */
+ 0x1e50, 501, /* Ṑ ṑ */
+ 0x1e52, 501, /* Ṓ ṓ */
+ 0x1e54, 501, /* Ṕ ṕ */
+ 0x1e56, 501, /* Ṗ ṗ */
+ 0x1e58, 501, /* Ṙ ṙ */
+ 0x1e5a, 501, /* Ṛ ṛ */
+ 0x1e5c, 501, /* Ṝ ṝ */
+ 0x1e5e, 501, /* Ṟ ṟ */
+ 0x1e60, 501, /* Ṡ ṡ */
+ 0x1e62, 501, /* Ṣ ṣ */
+ 0x1e64, 501, /* Ṥ ṥ */
+ 0x1e66, 501, /* Ṧ ṧ */
+ 0x1e68, 501, /* Ṩ ṩ */
+ 0x1e6a, 501, /* Ṫ ṫ */
+ 0x1e6c, 501, /* Ṭ ṭ */
+ 0x1e6e, 501, /* Ṯ ṯ */
+ 0x1e70, 501, /* Ṱ ṱ */
+ 0x1e72, 501, /* Ṳ ṳ */
+ 0x1e74, 501, /* Ṵ ṵ */
+ 0x1e76, 501, /* Ṷ ṷ */
+ 0x1e78, 501, /* Ṹ ṹ */
+ 0x1e7a, 501, /* Ṻ ṻ */
+ 0x1e7c, 501, /* Ṽ ṽ */
+ 0x1e7e, 501, /* Ṿ ṿ */
+ 0x1e80, 501, /* Ẁ ẁ */
+ 0x1e82, 501, /* Ẃ ẃ */
+ 0x1e84, 501, /* Ẅ ẅ */
+ 0x1e86, 501, /* Ẇ ẇ */
+ 0x1e88, 501, /* Ẉ ẉ */
+ 0x1e8a, 501, /* Ẋ ẋ */
+ 0x1e8c, 501, /* Ẍ ẍ */
+ 0x1e8e, 501, /* Ẏ ẏ */
+ 0x1e90, 501, /* Ẑ ẑ */
+ 0x1e92, 501, /* Ẓ ẓ */
+ 0x1e94, 501, /* Ẕ ẕ */
+ 0x1ea0, 501, /* Ạ ạ */
+ 0x1ea2, 501, /* Ả ả */
+ 0x1ea4, 501, /* Ấ ấ */
+ 0x1ea6, 501, /* Ầ ầ */
+ 0x1ea8, 501, /* Ẩ ẩ */
+ 0x1eaa, 501, /* Ẫ ẫ */
+ 0x1eac, 501, /* Ậ ậ */
+ 0x1eae, 501, /* Ắ ắ */
+ 0x1eb0, 501, /* Ằ ằ */
+ 0x1eb2, 501, /* Ẳ ẳ */
+ 0x1eb4, 501, /* Ẵ ẵ */
+ 0x1eb6, 501, /* Ặ ặ */
+ 0x1eb8, 501, /* Ẹ ẹ */
+ 0x1eba, 501, /* Ẻ ẻ */
+ 0x1ebc, 501, /* Ẽ ẽ */
+ 0x1ebe, 501, /* Ế ế */
+ 0x1ec0, 501, /* Ề ề */
+ 0x1ec2, 501, /* Ể ể */
+ 0x1ec4, 501, /* Ễ ễ */
+ 0x1ec6, 501, /* Ệ ệ */
+ 0x1ec8, 501, /* Ỉ ỉ */
+ 0x1eca, 501, /* Ị ị */
+ 0x1ecc, 501, /* Ọ ọ */
+ 0x1ece, 501, /* Ỏ ỏ */
+ 0x1ed0, 501, /* Ố ố */
+ 0x1ed2, 501, /* Ồ ồ */
+ 0x1ed4, 501, /* Ổ ổ */
+ 0x1ed6, 501, /* Ỗ ỗ */
+ 0x1ed8, 501, /* Ộ ộ */
+ 0x1eda, 501, /* Ớ ớ */
+ 0x1edc, 501, /* Ờ ờ */
+ 0x1ede, 501, /* Ở ở */
+ 0x1ee0, 501, /* Ỡ ỡ */
+ 0x1ee2, 501, /* Ợ ợ */
+ 0x1ee4, 501, /* Ụ ụ */
+ 0x1ee6, 501, /* Ủ ủ */
+ 0x1ee8, 501, /* Ứ ứ */
+ 0x1eea, 501, /* Ừ ừ */
+ 0x1eec, 501, /* Ử ử */
+ 0x1eee, 501, /* Ữ ữ */
+ 0x1ef0, 501, /* Ự ự */
+ 0x1ef2, 501, /* Ỳ ỳ */
+ 0x1ef4, 501, /* Ỵ ỵ */
+ 0x1ef6, 501, /* Ỷ ỷ */
+ 0x1ef8, 501, /* Ỹ ỹ */
+ 0x1f59, 492, /* Ὑ ὑ */
+ 0x1f5b, 492, /* Ὓ ὓ */
+ 0x1f5d, 492, /* Ὕ ὕ */
+ 0x1f5f, 492, /* Ὗ ὗ */
+ 0x1fbc, 491, /* ᾼ ᾳ */
+ 0x1fcc, 491, /* ῌ ῃ */
+ 0x1fec, 493, /* Ῥ ῥ */
+ 0x1ffc, 491, /* ῼ ῳ */
+};
+
+/*
+ * title characters are those between
+ * upper and lower case. ie DZ Dz dz
+ */
+static
+Rune __totitle1[] =
+{
+ 0x01c4, 501, /* DŽ Dž */
+ 0x01c6, 499, /* dž Dž */
+ 0x01c7, 501, /* LJ Lj */
+ 0x01c9, 499, /* lj Lj */
+ 0x01ca, 501, /* NJ Nj */
+ 0x01cc, 499, /* nj Nj */
+ 0x01f1, 501, /* DZ Dz */
+ 0x01f3, 499, /* dz Dz */
+};
+
+static
+Rune*
+bsearch(Rune c, Rune *t, int n, int ne)
+{
+ Rune *p;
+ int m;
+
+ while(n > 1) {
+ m = n/2;
+ p = t + m*ne;
+ if(c >= p[0]) {
+ t = p;
+ n = n-m;
+ } else
+ n = m;
+ }
+ if(n && c >= t[0])
+ return t;
+ return 0;
+}
+
+Rune
+tolowerrune(Rune c)
+{
+ Rune *p;
+
+ p = bsearch(c, __tolower2, nelem(__tolower2)/3, 3);
+ if(p && c >= p[0] && c <= p[1])
+ return c + p[2] - 500;
+ p = bsearch(c, __tolower1, nelem(__tolower1)/2, 2);
+ if(p && c == p[0])
+ return c + p[1] - 500;
+ return c;
+}
+
+Rune
+toupperrune(Rune c)
+{
+ Rune *p;
+
+ p = bsearch(c, __toupper2, nelem(__toupper2)/3, 3);
+ if(p && c >= p[0] && c <= p[1])
+ return c + p[2] - 500;
+ p = bsearch(c, __toupper1, nelem(__toupper1)/2, 2);
+ if(p && c == p[0])
+ return c + p[1] - 500;
+ return c;
+}
+
+Rune
+totitlerune(Rune c)
+{
+ Rune *p;
+
+ p = bsearch(c, __totitle1, nelem(__totitle1)/2, 2);
+ if(p && c == p[0])
+ return c + p[1] - 500;
+ return c;
+}
+
+int
+islowerrune(Rune c)
+{
+ Rune *p;
+
+ p = bsearch(c, __toupper2, nelem(__toupper2)/3, 3);
+ if(p && c >= p[0] && c <= p[1])
+ return 1;
+ p = bsearch(c, __toupper1, nelem(__toupper1)/2, 2);
+ if(p && c == p[0])
+ return 1;
+ return 0;
+}
+
+int
+isupperrune(Rune c)
+{
+ Rune *p;
+
+ p = bsearch(c, __tolower2, nelem(__tolower2)/3, 3);
+ if(p && c >= p[0] && c <= p[1])
+ return 1;
+ p = bsearch(c, __tolower1, nelem(__tolower1)/2, 2);
+ if(p && c == p[0])
+ return 1;
+ return 0;
+}
+
+int
+isalpharune(Rune c)
+{
+ Rune *p;
+
+ if(isupperrune(c) || islowerrune(c))
+ return 1;
+ p = bsearch(c, __alpha2, nelem(__alpha2)/2, 2);
+ if(p && c >= p[0] && c <= p[1])
+ return 1;
+ p = bsearch(c, __alpha1, nelem(__alpha1), 1);
+ if(p && c == p[0])
+ return 1;
+ return 0;
+}
+
+int
+istitlerune(Rune c)
+{
+ return isupperrune(c) && islowerrune(c);
+}
+
+int
+isspacerune(Rune c)
+{
+ Rune *p;
+
+ p = bsearch(c, __space2, nelem(__space2)/2, 2);
+ if(p && c >= p[0] && c <= p[1])
+ return 1;
+ return 0;
+}
DIR diff --git a/src/libutf/utf.7 b/src/libutf/utf.7
t@@ -0,0 +1,91 @@
+.TH UTF 7
+.SH NAME
+UTF, Unicode, ASCII, rune \- character set and format
+.SH DESCRIPTION
+The Plan 9 character set and representation are
+based on the Unicode Standard and on the ISO multibyte
+.SM UTF-8
+encoding (Universal Character
+Set Transformation Format, 8 bits wide).
+The Unicode Standard represents its characters in 16
+bits;
+.SM UTF-8
+represents such
+values in an 8-bit byte stream.
+Throughout this manual,
+.SM UTF-8
+is shortened to
+.SM UTF.
+.PP
+In Plan 9, a
+.I rune
+is a 16-bit quantity representing a Unicode character.
+Internally, programs may store characters as runes.
+However, any external manifestation of textual information,
+in files or at the interface between programs, uses a
+machine-independent, byte-stream encoding called
+.SM UTF.
+.PP
+.SM UTF
+is designed so the 7-bit
+.SM ASCII
+set (values hexadecimal 00 to 7F),
+appear only as themselves
+in the encoding.
+Runes with values above 7F appear as sequences of two or more
+bytes with values only from 80 to FF.
+.PP
+The
+.SM UTF
+encoding of the Unicode Standard is backward compatible with
+.SM ASCII\c
+:
+programs presented only with
+.SM ASCII
+work on Plan 9
+even if not written to deal with
+.SM UTF,
+as do
+programs that deal with uninterpreted byte streams.
+However, programs that perform semantic processing on
+.SM ASCII
+graphic
+characters must convert from
+.SM UTF
+to runes
+in order to work properly with non-\c
+.SM ASCII
+input.
+See
+.IR rune (2).
+.PP
+Letting numbers be binary,
+a rune x is converted to a multibyte
+.SM UTF
+sequence
+as follows:
+.PP
+01. x in [00000000.0bbbbbbb] → 0bbbbbbb
+.br
+10. x in [00000bbb.bbbbbbbb] → 110bbbbb, 10bbbbbb
+.br
+11. x in [bbbbbbbb.bbbbbbbb] → 1110bbbb, 10bbbbbb, 10bbbbbb
+.br
+.PP
+Conversion 01 provides a one-byte sequence that spans the
+.SM ASCII
+character set in a compatible way.
+Conversions 10 and 11 represent higher-valued characters
+as sequences of two or three bytes with the high bit set.
+Plan 9 does not support the 4, 5, and 6 byte sequences proposed by X-Open.
+When there are multiple ways to encode a value, for example rune 0,
+the shortest encoding is used.
+.PP
+In the inverse mapping,
+any sequence except those described above
+is incorrect and is converted to rune hexadecimal 0080.
+.SH "SEE ALSO"
+.IR ascii (1),
+.IR tcs (1),
+.IR rune (3),
+.IR "The Unicode Standard" .
DIR diff --git a/src/libutf/utf.h b/src/libutf/utf.h
t@@ -0,0 +1,51 @@
+#ifndef _UTFH_
+#define _UTFH_ 1
+
+typedef unsigned short Rune; /* 16 bits */
+
+enum
+{
+ UTFmax = 3, /* maximum bytes per rune */
+ Runesync = 0x80, /* cannot represent part of a UTF sequence (<) */
+ Runeself = 0x80, /* rune and UTF sequences are the same (<) */
+ Runeerror = 0x80, /* decoding error in UTF */
+};
+
+/*
+ * rune routines
+ */
+extern int runetochar(char*, Rune*);
+extern int chartorune(Rune*, char*);
+extern int runelen(long);
+extern int runenlen(Rune*, int);
+extern int fullrune(char*, int);
+extern int utflen(char*);
+extern int utfnlen(char*, long);
+extern char* utfrune(char*, long);
+extern char* utfrrune(char*, long);
+extern char* utfutf(char*, char*);
+extern char* utfecpy(char*, char*, char*);
+
+extern Rune* runestrcat(Rune*, Rune*);
+extern Rune* runestrchr(Rune*, Rune);
+extern int runestrcmp(Rune*, Rune*);
+extern Rune* runestrcpy(Rune*, Rune*);
+extern Rune* runestrncpy(Rune*, Rune*, long);
+extern Rune* runestrecpy(Rune*, Rune*, Rune*);
+extern Rune* runestrdup(Rune*);
+extern Rune* runestrncat(Rune*, Rune*, long);
+extern int runestrncmp(Rune*, Rune*, long);
+extern Rune* runestrrchr(Rune*, Rune);
+extern long runestrlen(Rune*);
+extern Rune* runestrstr(Rune*, Rune*);
+
+extern Rune tolowerrune(Rune);
+extern Rune totitlerune(Rune);
+extern Rune toupperrune(Rune);
+extern int isalpharune(Rune);
+extern int islowerrune(Rune);
+extern int isspacerune(Rune);
+extern int istitlerune(Rune);
+extern int isupperrune(Rune);
+
+#endif
DIR diff --git a/src/libutf/utfdef.h b/src/libutf/utfdef.h
t@@ -0,0 +1,14 @@
+#define uchar _utfuchar
+#define ushort _utfushort
+#define uint _utfuint
+#define ulong _utfulong
+#define vlong _utfvlong
+#define uvlong _utfuvlong
+
+typedef unsigned char uchar;
+typedef unsigned short ushort;
+typedef unsigned int uint;
+typedef unsigned long ulong;
+
+#define nelem(x) (sizeof(x)/sizeof((x)[0]))
+#define nil ((void*)0)
DIR diff --git a/src/libutf/utfecpy.c b/src/libutf/utfecpy.c
t@@ -0,0 +1,36 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+char*
+utfecpy(char *to, char *e, char *from)
+{
+ char *end;
+
+ if(to >= e)
+ return to;
+ end = memccpy(to, from, '\0', e - to);
+ if(end == nil){
+ end = e-1;
+ while(end>to && (*--end&0xC0)==0x80)
+ ;
+ *end = '\0';
+ }else{
+ end--;
+ }
+ return end;
+}
DIR diff --git a/src/libutf/utflen.c b/src/libutf/utflen.c
t@@ -0,0 +1,38 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+int
+utflen(char *s)
+{
+ int c;
+ long n;
+ Rune rune;
+
+ n = 0;
+ for(;;) {
+ c = *(uchar*)s;
+ if(c < Runeself) {
+ if(c == 0)
+ return n;
+ s++;
+ } else
+ s += chartorune(&rune, s);
+ n++;
+ }
+ return 0;
+}
DIR diff --git a/src/libutf/utfnlen.c b/src/libutf/utfnlen.c
t@@ -0,0 +1,41 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+int
+utfnlen(char *s, long m)
+{
+ int c;
+ long n;
+ Rune rune;
+ char *es;
+
+ es = s + m;
+ for(n = 0; s < es; n++) {
+ c = *(uchar*)s;
+ if(c < Runeself){
+ if(c == '\0')
+ break;
+ s++;
+ continue;
+ }
+ if(!fullrune(s, es-s))
+ break;
+ s += chartorune(&rune, s);
+ }
+ return n;
+}
DIR diff --git a/src/libutf/utfrrune.c b/src/libutf/utfrrune.c
t@@ -0,0 +1,46 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+char*
+utfrrune(char *s, long c)
+{
+ long c1;
+ Rune r;
+ char *s1;
+
+ if(c < Runesync) /* not part of utf sequence */
+ return strrchr(s, c);
+
+ s1 = 0;
+ for(;;) {
+ c1 = *(uchar*)s;
+ if(c1 < Runeself) { /* one byte rune */
+ if(c1 == 0)
+ return s1;
+ if(c1 == c)
+ s1 = s;
+ s++;
+ continue;
+ }
+ c1 = chartorune(&r, s);
+ if(r == c)
+ s1 = s;
+ s += c1;
+ }
+ return 0;
+}
DIR diff --git a/src/libutf/utfrune.c b/src/libutf/utfrune.c
t@@ -0,0 +1,45 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+char*
+utfrune(char *s, long c)
+{
+ long c1;
+ Rune r;
+ int n;
+
+ if(c < Runesync) /* not part of utf sequence */
+ return strchr(s, c);
+
+ for(;;) {
+ c1 = *(uchar*)s;
+ if(c1 < Runeself) { /* one byte rune */
+ if(c1 == 0)
+ return 0;
+ if(c1 == c)
+ return s;
+ s++;
+ continue;
+ }
+ n = chartorune(&r, s);
+ if(r == c)
+ return s;
+ s += n;
+ }
+ return 0;
+}
DIR diff --git a/src/libutf/utfutf.c b/src/libutf/utfutf.c
t@@ -0,0 +1,41 @@
+/*
+ * The authors of this software are Rob Pike and Ken Thompson.
+ * Copyright (c) 2002 by Lucent Technologies.
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose without fee is hereby granted, provided that this entire notice
+ * is included in all copies of any software which is or includes a copy
+ * or modification of this software and in all copies of the supporting
+ * documentation for such software.
+ * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
+ * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+ * OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+ */
+#include <stdarg.h>
+#include <string.h>
+#include "utf.h"
+#include "utfdef.h"
+
+
+/*
+ * Return pointer to first occurrence of s2 in s1,
+ * 0 if none
+ */
+char*
+utfutf(char *s1, char *s2)
+{
+ char *p;
+ long f, n1, n2;
+ Rune r;
+
+ n1 = chartorune(&r, s2);
+ f = r;
+ if(f <= Runesync) /* represents self */
+ return strstr(s1, s2);
+
+ n2 = strlen(s2);
+ for(p=s1; p=utfrune(p, f); p+=n1)
+ if(strncmp(p, s2, n2) == 0)
+ return p;
+ return 0;
+}