FLEXDOC(1) FLEXDOC(1) NNAAMMEE flexdoc - documentation for flex, fast lexical analyzer generator SSYYNNOOPPSSIISS fflleexx [[--bbccddffhhiillnnppssttvvwwBBFFIILLTTVV7788++ --CC[[aaeeffFFmmrr]] --PPpprreeffiixx --SSsskkeellee-- ttoonn]] _[_f_i_l_e_n_a_m_e _._._._] DDEESSCCRRIIPPTTIIOONN _f_l_e_x is a tool for generating _s_c_a_n_n_e_r_s_: programs which recognized lexical patterns in text. _f_l_e_x reads the given input files, or its standard input if no file names are given, for a description of a scanner to generate. The description is in the form of pairs of regular expressions and C code, called _r_u_l_e_s_. _f_l_e_x generates as output a C source file, lleexx..yyyy..cc,, which defines a routine yyyylleexx(()).. This file is compiled and linked with the --llffll library to produce an executable. When the executable is run, it analyzes its input for occurrences of the regular expres- sions. Whenever it finds one, it executes the correspond- ing C code. SSOOMMEE SSIIMMPPLLEE EEXXAAMMPPLLEESS First some simple examples to get the flavor of how one uses _f_l_e_x_. The following _f_l_e_x input specifies a scanner which whenever it encounters the string "username" will replace it with the user's login name: %% username printf( "%s", getlogin() ); By default, any text not matched by a _f_l_e_x scanner is copied to the output, so the net effect of this scanner is to copy its input file to its output with each occurrence of "username" expanded. In this input, there is just one rule. "username" is the _p_a_t_t_e_r_n and the "printf" is the _a_c_t_i_o_n_. The "%%" marks the beginning of the rules. Here's another simple example: int num_lines = 0, num_chars = 0; %% \n ++num_lines; ++num_chars; . ++num_chars; %% main() { yylex(); printf( "# of lines = %d, # of chars = %d\n", num_lines, num_chars ); } Version 2.4 November 1993 1 FLEXDOC(1) FLEXDOC(1) This scanner counts the number of characters and the num- ber of lines in its input (it produces no output other than the final report on the counts). The first line declares two globals, "num_lines" and "num_chars", which are accessible both inside yyyylleexx(()) and in the mmaaiinn(()) rou- tine declared after the second "%%". There are two rules, one which matches a newline ("\n") and increments both the line count and the character count, and one which matches any character other than a newline (indicated by the "." regular expression). A somewhat more complicated example: /* scanner for a toy Pascal-like language */ %{ /* need this for the call to atof() below */ #include %} DIGIT [0-9] ID [a-z][a-z0-9]* %% {DIGIT}+ { printf( "An integer: %s (%d)\n", yytext, atoi( yytext ) ); } {DIGIT}+"."{DIGIT}* { printf( "A float: %s (%g)\n", yytext, atof( yytext ) ); } if|then|begin|end|procedure|function { printf( "A keyword: %s\n", yytext ); } {ID} printf( "An identifier: %s\n", yytext ); "+"|"-"|"*"|"/" printf( "An operator: %s\n", yytext ); "{"[^}\n]*"}" /* eat up one-line comments */ [ \t\n]+ /* eat up whitespace */ . printf( "Unrecognized character: %s\n", yytext ); %% main( argc, argv ) int argc; char **argv; Version 2.4 November 1993 2 FLEXDOC(1) FLEXDOC(1) { ++argv, --argc; /* skip over program name */ if ( argc > 0 ) yyin = fopen( argv[0], "r" ); else yyin = stdin; yylex(); } This is the beginnings of a simple scanner for a language like Pascal. It identifies different types of _t_o_k_e_n_s and reports on what it has seen. The details of this example will be explained in the fol- lowing sections. FFOORRMMAATT OOFF TTHHEE IINNPPUUTT FFIILLEE The _f_l_e_x input file consists of three sections, separated by a line with just %%%% in it: definitions %% rules %% user code The _d_e_f_i_n_i_t_i_o_n_s section contains declarations of simple _n_a_m_e definitions to simplify the scanner specification, and declarations of _s_t_a_r_t _c_o_n_d_i_t_i_o_n_s_, which are explained in a later section. Name definitions have the form: name definition The "name" is a word beginning with a letter or an under- score ('_') followed by zero or more letters, digits, '_', or '-' (dash). The definition is taken to begin at the first non-white-space character following the name and continuing to the end of the line. The definition can subsequently be referred to using "{name}", which will expand to "(definition)". For example, DIGIT [0-9] ID [a-z][a-z0-9]* defines "DIGIT" to be a regular expression which matches a single digit, and "ID" to be a regular expression which matches a letter followed by zero-or-more letters-or- digits. A subsequent reference to {DIGIT}+"."{DIGIT}* Version 2.4 November 1993 3 FLEXDOC(1) FLEXDOC(1) is identical to ([0-9])+"."([0-9])* and matches one-or-more digits followed by a '.' followed by zero-or-more digits. The _r_u_l_e_s section of the _f_l_e_x input contains a series of rules of the form: pattern action where the pattern must be unindented and the action must begin on the same line. See below for a further description of patterns and actions. Finally, the user code section is simply copied to lleexx..yyyy..cc verbatim. It is used for companion routines which call or are called by the scanner. The presence of this section is optional; if it is missing, the second %%%% in the input file may be skipped, too. In the definitions and rules sections, any _i_n_d_e_n_t_e_d text or text enclosed in %%{{ and %%}} is copied verbatim to the output (with the %{}'s removed). The %{}'s must appear unindented on lines by themselves. In the rules section, any indented or %{} text appearing before the first rule may be used to declare variables which are local to the scanning routine and (after the declarations) code which is to be executed whenever the scanning routine is entered. Other indented or %{} text in the rule section is still copied to the output, but its meaning is not well-defined and it may well cause compile- time errors (this feature is present for _P_O_S_I_X compliance; see below for other such features). In the definitions section (but not in the rules section), an unindented comment (i.e., a line beginning with "/*") is also copied verbatim to the output up to the next "*/". PPAATTTTEERRNNSS The patterns in the input are written using an extended set of regular expressions. These are: x match the character 'x' . any character except newline [xyz] a "character class"; in this case, the pattern matches either an 'x', a 'y', or a 'z' [abj-oZ] a "character class" with a range in it; matches an 'a', a 'b', any letter from 'j' through 'o', or a 'Z' Version 2.4 November 1993 4 FLEXDOC(1) FLEXDOC(1) [^A-Z] a "negated character class", i.e., any character but those in the class. In this case, any character EXCEPT an uppercase letter. [^A-Z\n] any character EXCEPT an uppercase letter or a newline r* zero or more r's, where r is any regular expression r+ one or more r's r? zero or one r's (that is, "an optional r") r{2,5} anywhere from two to five r's r{2,} two or more r's r{4} exactly 4 r's {name} the expansion of the "name" definition (see above) "[xyz]\"foo" the literal string: [xyz]"foo \X if X is an 'a', 'b', 'f', 'n', 'r', 't', or 'v', then the ANSI-C interpretation of \x. Otherwise, a literal 'X' (used to escape operators such as '*') \123 the character with octal value 123 \x2a the character with hexadecimal value 2a (r) match an r; parentheses are used to override precedence (see below) rs the regular expression r followed by the regular expression s; called "concatenation" r|s either an r or an s r/s an r but only if it is followed by an s. The s is not part of the matched text. This type of pattern is called as "trailing context". ^r an r, but only at the beginning of a line r$ an r, but only at the end of a line. Equivalent to "r/\n". r an r, but only in start condition s (see below for discussion of start conditions) r same, but in any of start conditions s1, s2, or s3 <*>r an r in any start condition, even an exclusive one. <> an end-of-file <> an end-of-file when in start condition s1 or s2 Note that inside of a character class, all regular expres- sion operators lose their special meaning except escape Version 2.4 November 1993 5 FLEXDOC(1) FLEXDOC(1) ('\') and the character class operators, '-', ']', and, at the beginning of the class, '^'. The regular expressions listed above are grouped according to precedence, from highest precedence at the top to low- est at the bottom. Those grouped together have equal precedence. For example, foo|bar* is the same as (foo)|(ba(r*)) since the '*' operator has higher precedence than concate- nation, and concatenation higher than alternation ('|'). This pattern therefore matches _e_i_t_h_e_r the string "foo" _o_r the string "ba" followed by zero-or-more r's. To match "foo" or zero-or-more "bar"'s, use: foo|(bar)* and to match zero-or-more "foo"'s-or-"bar"'s: (foo|bar)* Some notes on patterns: - A negated character class such as the example "[^A- Z]" above _w_i_l_l _m_a_t_c_h _a _n_e_w_l_i_n_e unless "\n" (or an equivalent escape sequence) is one of the charac- ters explicitly present in the negated character class (e.g., "[^A-Z\n]"). This is unlike how many other regular expression tools treat negated char- acter classes, but unfortunately the inconsistency is historically entrenched. Matching newlines means that a pattern like [^"]* can match the entire input unless there's another quote in the input. - A rule can have at most one instance of trailing context (the '/' operator or the '$' operator). The start condition, '^', and "<>" patterns can only occur at the beginning of a pattern, and, as well as with '/' and '$', cannot be grouped inside parentheses. A '^' which does not occur at the beginning of a rule or a '$' which does not occur at the end of a rule loses its special prop- erties and is treated as a normal character. The following are illegal: foo/bar$ Version 2.4 November 1993 6 FLEXDOC(1) FLEXDOC(1) foobar Note that the first of these, can be written "foo/bar\n". The following will result in '$' or '^' being treated as a normal character: foo|(bar$) foo|^bar If what's wanted is a "foo" or a bar-followed-by-a- newline, the following could be used (the special '|' action is explained below): foo | bar$ /* action goes here */ A similar trick will work for matching a foo or a bar-at-the-beginning-of-a-line. HHOOWW TTHHEE IINNPPUUTT IISS MMAATTCCHHEEDD When the generated scanner is run, it analyzes its input looking for strings which match any of its patterns. If it finds more than one match, it takes the one matching the most text (for trailing context rules, this includes the length of the trailing part, even though it will then be returned to the input). If it finds two or more matches of the same length, the rule listed first in the _f_l_e_x input file is chosen. Once the match is determined, the text corresponding to the match (called the _t_o_k_e_n_) is made available in the global character pointer yyyytteexxtt,, and its length in the global integer yyyylleenngg.. The _a_c_t_i_o_n corresponding to the matched pattern is then executed (a more detailed descrip- tion of actions follows), and then the remaining input is scanned for another match. If no match is found, then the _d_e_f_a_u_l_t _r_u_l_e is executed: the next character in the input is considered matched and copied to the standard output. Thus, the simplest legal _f_l_e_x input is: %% which generates a scanner that simply copies its input (one character at a time) to its output. Note that yyyytteexxtt can be defined in two different ways: either as a character _p_o_i_n_t_e_r or as a character _a_r_r_a_y_. You can control which definition _f_l_e_x uses by including one of the special directives %%ppooiinntteerr or %%aarrrraayy in the first (definitions) section of your flex input. The Version 2.4 November 1993 7 FLEXDOC(1) FLEXDOC(1) default is %%ppooiinntteerr,, unless you use the --ll lex compatibil- ity option, in which case yyyytteexxtt will be an array. The advantage of using %%ppooiinntteerr is substantially faster scan- ning and no buffer overflow when matching very large tokens (unless you run out of dynamic memory). The disad- vantage is that you are restricted in how your actions can modify yyyytteexxtt (see the next section), and calls to the iinnppuutt(()) and uunnppuutt(()) functions destroy the present contents of yyyytteexxtt,, which can be a considerable porting headache when moving between different _l_e_x versions. The advantage of %%aarrrraayy is that you can then modify yyyytteexxtt to your heart's content, and calls to iinnppuutt(()) and uunnppuutt(()) do not destroy yyyytteexxtt (see below). Furthermore, existing _l_e_x programs sometimes access yyyytteexxtt externally using dec- larations of the form: extern char yytext[]; This definition is erroneous when used with %%ppooiinntteerr,, but correct for %%aarrrraayy.. %%aarrrraayy defines yyyytteexxtt to be an array of YYYYLLMMAAXX characters, which defaults to a fairly large value. You can change the size by simply #define'ing YYYYLLMMAAXX to a different value in the first section of your _f_l_e_x input. As mentioned above, with %%ppooiinntteerr yytext grows dynamically to accomo- date large tokens. While this means your %%ppooiinntteerr scanner can accomodate very large tokens (such as matching entire blocks of comments), bear in mind that each time the scan- ner must resize yyyytteexxtt it also must rescan the entire token from the beginning, so matching such tokens can prove slow. yyyytteexxtt presently does _n_o_t dynamically grow if a call to uunnppuutt(()) results in too much text being pushed back; instead, a run-time error results. Also note that you cannot use %%aarrrraayy with C++ scanner classes (the --++ option; see below). AACCTTIIOONNSS Each pattern in a rule has a corresponding action, which can be any arbitrary C statement. The pattern ends at the first non-escaped whitespace character; the remainder of the line is its action. If the action is empty, then when the pattern is matched the input token is simply dis- carded. For example, here is the specification for a pro- gram which deletes all occurrences of "zap me" from its input: %% "zap me" (It will copy all other characters in the input to the output since they will be matched by the default rule.) Here is a program which compresses multiple blanks and Version 2.4 November 1993 8 FLEXDOC(1) FLEXDOC(1) tabs down to a single blank, and throws away whitespace found at the end of a line: %% [ \t]+ putchar( ' ' ); [ \t]+$ /* ignore this token */ If the action contains a '{', then the action spans till the balancing '}' is found, and the action may cross mul- tiple lines. _f_l_e_x knows about C strings and comments and won't be fooled by braces found within them, but also allows actions to begin with %%{{ and will consider the action to be all the text up to the next %%}} (regardless of ordinary braces inside the action). An action consisting solely of a vertical bar ('|') means "same as the action for the next rule." See below for an illustration. Actions can include arbitrary C code, including rreettuurrnn statements to return a value to whatever routine called yyyylleexx(()).. Each time yyyylleexx(()) is called it continues pro- cessing tokens from where it last left off until it either reaches the end of the file or executes a return. Actions are free to modify yyyytteexxtt except for lengthening it (adding characters to its end--these will overwrite later characters in the input stream). Modifying the final character of yytext may alter whether when scanning resumes rules anchored with '^' are active. Specifically, changing the final character of yytext to a newline will activate such rules on the next scan, and changing it to anything else will deactivate the rules. Users should not rely on this behavior being present in future releases. Finally, note that none of this paragraph applies when using %%aarrrraayy (see above). Actions are free to modify yyyylleenngg except they should not do so if the action also includes use of yyyymmoorree(()) (see below). There are a number of special directives which can be included within an action: - EECCHHOO copies yytext to the scanner's output. - BBEEGGIINN followed by the name of a start condition places the scanner in the corresponding start con- dition (see below). - RREEJJEECCTT directs the scanner to proceed on to the "second best" rule which matched the input (or a prefix of the input). The rule is chosen as Version 2.4 November 1993 9 FLEXDOC(1) FLEXDOC(1) described above in "How the Input is Matched", and yyyytteexxtt and yyyylleenngg set up appropriately. It may either be one which matched as much text as the originally chosen rule but came later in the _f_l_e_x input file, or one which matched less text. For example, the following will both count the words in the input and call the routine special() whenever "frob" is seen: int word_count = 0; %% frob special(); REJECT; [^ \t\n]+ ++word_count; Without the RREEJJEECCTT,, any "frob"'s in the input would not be counted as words, since the scanner normally executes only one action per token. Multiple RREEJJEECCTT''ss are allowed, each one finding the next best choice to the currently active rule. For example, when the following scanner scans the token "abcd", it will write "abcdabcaba" to the output: %% a | ab | abc | abcd ECHO; REJECT; .|\n /* eat up any unmatched character */ (The first three rules share the fourth's action since they use the special '|' action.) RREEJJEECCTT is a particularly expensive feature in terms scanner performance; if it is used in _a_n_y of the scanner's actions it will slow down _a_l_l of the scanner's matching. Furthermore, RREEJJEECCTT cannot be used with the _-_C_f or _-_C_F options (see below). Note also that unlike the other special actions, RREEJJEECCTT is a _b_r_a_n_c_h_; code immediately following it in the action will _n_o_t be executed. - yyyymmoorree(()) tells the scanner that the next time it matches a rule, the corresponding token should be _a_p_p_e_n_d_e_d onto the current value of yyyytteexxtt rather than replacing it. For example, given the input "mega-kludge" the following will write "mega-mega- kludge" to the output: %% mega- ECHO; yymore(); kludge ECHO; First "mega-" is matched and echoed to the output. Version 2.4 November 1993 10 FLEXDOC(1) FLEXDOC(1) Then "kludge" is matched, but the previous "mega-" is still hanging around at the beginning of yyyytteexxtt so the EECCHHOO for the "kludge" rule will actually write "mega-kludge". The presence of yyyymmoorree(()) in the scanner's action entails a minor performance penalty in the scanner's matching speed. - yyyylleessss((nn)) returns all but the first _n characters of the current token back to the input stream, where they will be rescanned when the scanner looks for the next match. yyyytteexxtt and yyyylleenngg are adjusted appropriately (e.g., yyyylleenngg will now be equal to _n ). For example, on the input "foobar" the follow- ing will write out "foobarbar": %% foobar ECHO; yyless(3); [a-z]+ ECHO; An argument of 0 to yyyylleessss will cause the entire current input string to be scanned again. Unless you've changed how the scanner will subsequently process its input (using BBEEGGIINN,, for example), this will result in an endless loop. Note that yyyylleessss is a macro and can only be used in the flex input file, not from other source files. - uunnppuutt((cc)) puts the character _c back onto the input stream. It will be the next character scanned. The following action will take the current token and cause it to be rescanned enclosed in parenthe- ses. { int i; unput( ')' ); for ( i = yyleng - 1; i >= 0; --i ) unput( yytext[i] ); unput( '(' ); } Note that since each uunnppuutt(()) puts the given charac- ter back at the _b_e_g_i_n_n_i_n_g of the input stream, pushing back strings must be done back-to-front. Also note that you cannot put back EEOOFF to attempt to mark the input stream with an end-of-file. - iinnppuutt(()) reads the next character from the input stream. For example, the following is one way to eat up C comments: %% "/*" { Version 2.4 November 1993 11 FLEXDOC(1) FLEXDOC(1) register int c; for ( ; ; ) { while ( (c = input()) != '*' && c != EOF ) ; /* eat up text of comment */ if ( c == '*' ) { while ( (c = input()) == '*' ) ; if ( c == '/' ) break; /* found the end */ } if ( c == EOF ) { error( "EOF in comment" ); break; } } } (Note that if the scanner is compiled using CC++++,, then iinnppuutt(()) is instead referred to as yyyyiinnppuutt(()),, in order to avoid a name clash with the CC++++ stream by the name of _i_n_p_u_t_._) - yyyytteerrmmiinnaattee(()) can be used in lieu of a return statement in an action. It terminates the scanner and returns a 0 to the scanner's caller, indicating "all done". By default, yyyytteerrmmiinnaattee(()) is also called when an end-of-file is encountered. It is a macro and may be redefined. TTHHEE GGEENNEERRAATTEEDD SSCCAANNNNEERR The output of _f_l_e_x is the file lleexx..yyyy..cc,, which contains the scanning routine yyyylleexx(()),, a number of tables used by it for matching tokens, and a number of auxiliary routines and macros. By default, yyyylleexx(()) is declared as follows: int yylex() { ... various definitions and the actions in here ... } (If your environment supports function prototypes, then it will be "int yylex( void )".) This definition may be changed by defining the "YY_DECL" macro. For example, you could use: #define YY_DECL float lexscan( a, b ) float a, b; Version 2.4 November 1993 12 FLEXDOC(1) FLEXDOC(1) to give the scanning routine the name _l_e_x_s_c_a_n_, returning a float, and taking two floats as arguments. Note that if you give arguments to the scanning routine using a K&R- style/non-prototyped function declaration, you must termi- nate the definition with a semi-colon (;). Whenever yyyylleexx(()) is called, it scans tokens from the global input file _y_y_i_n (which defaults to stdin). It con- tinues until it either reaches an end-of-file (at which point it returns the value 0) or one of its actions exe- cutes a _r_e_t_u_r_n statement. If the scanner reaches an end-of-file, subsequent calls are undefined unless either _y_y_i_n is pointed at a new input file (in which case scanning continues from that file), or yyyyrreessttaarrtt(()) is called. yyyyrreessttaarrtt(()) takes one argument, a FFIILLEE ** pointer, and initializes _y_y_i_n for scanning from that file. Essentially there is no difference between just assigning _y_y_i_n to a new input file or using yyyyrreessttaarrtt(()) to do so; the latter is available for compati- bility with previous versions of _f_l_e_x_, and because it can be used to switch input files in the middle of scanning. It can also be used to throw away the current input buffer, by calling it with an argument of _y_y_i_n_. If yyyylleexx(()) stops scanning due to executing a _r_e_t_u_r_n state- ment in one of the actions, the scanner may then be called again and it will resume scanning where it left off. By default (and for purposes of efficiency), the scanner uses block-reads rather than simple _g_e_t_c_(_) calls to read characters from _y_y_i_n_. The nature of how it gets its input can be controlled by defining the YYYY__IINNPPUUTT macro. YY_INPUT's calling sequence is "YY_INPUT(buf,result,max_size)". Its action is to place up to _m_a_x___s_i_z_e characters in the character array _b_u_f and return in the integer variable _r_e_s_u_l_t either the number of characters read or the constant YY_NULL (0 on Unix sys- tems) to indicate EOF. The default YY_INPUT reads from the global file-pointer "yyin". A sample definition of YY_INPUT (in the definitions sec- tion of the input file): %{ #define YY_INPUT(buf,result,max_size) \ { \ int c = getchar(); \ result = (c == EOF) ? YY_NULL : (buf[0] = c, 1); \ } %} This definition will change the input processing to occur one character at a time. Version 2.4 November 1993 13 FLEXDOC(1) FLEXDOC(1) You also can add in things like keeping track of the input line number this way; but don't expect your scanner to go very fast. When the scanner receives an end-of-file indication from YY_INPUT, it then checks the yyyywwrraapp(()) function. If yyyywwrraapp(()) returns false (zero), then it is assumed that the function has gone ahead and set up _y_y_i_n to point to another input file, and scanning continues. If it returns true (non-zero), then the scanner terminates, returning 0 to its caller. The default yyyywwrraapp(()) always returns 1. The scanner writes its EECCHHOO output to the _y_y_o_u_t global (default, stdout), which may be redefined by the user sim- ply by assigning it to some other FFIILLEE pointer. SSTTAARRTT CCOONNDDIITTIIOONNSS _f_l_e_x provides a mechanism for conditionally activating rules. Any rule whose pattern is prefixed with "" will only be active when the scanner is in the start con- dition named "sc". For example, [^"]* { /* eat up the string body ... */ ... } will be active only when the scanner is in the "STRING" start condition, and \. { /* handle an escape ... */ ... } will be active only when the current start condition is either "INITIAL", "STRING", or "QUOTE". Start conditions are declared in the definitions (first) section of the input using unindented lines beginning with either %%ss or %%xx followed by a list of names. The former declares _i_n_c_l_u_s_i_v_e start conditions, the latter _e_x_c_l_u_s_i_v_e start conditions. A start condition is activated using the BBEEGGIINN action. Until the next BBEEGGIINN action is exe- cuted, rules with the given start condition will be active and rules with other start conditions will be inactive. If the start condition is _i_n_c_l_u_s_i_v_e_, then rules with no start conditions at all will also be active. If it is _e_x_c_l_u_s_i_v_e_, then _o_n_l_y rules qualified with the start condi- tion will be active. A set of rules contingent on the same exclusive start condition describe a scanner which is independent of any of the other rules in the _f_l_e_x input. Because of this, exclusive start conditions make it easy to specify "mini-scanners" which scan portions of the Version 2.4 November 1993 14 FLEXDOC(1) FLEXDOC(1) input that are syntactically different from the rest (e.g., comments). If the distinction between inclusive and exclusive start conditions is still a little vague, here's a simple exam- ple illustrating the connection between the two. The set of rules: %s example %% foo /* do something */ is equivalent to %x example %% foo /* do something */ Also note that the special start-condition specifier <<**>> matches every start condition. Thus, the above example could also have been written; %x example %% <*>foo /* do something */ The default rule (to EECCHHOO any unmatched character) remains active in start conditions. BBEEGGIINN((00)) returns to the original state where only the rules with no start conditions are active. This state can also be referred to as the start-condition "INITIAL", so BBEEGGIINN((IINNIITTIIAALL)) is equivalent to BBEEGGIINN((00)).. (The parenthe- ses around the start condition name are not required but are considered good style.) BBEEGGIINN actions can also be given as indented code at the beginning of the rules section. For example, the follow- ing will cause the scanner to enter the "SPECIAL" start condition whenever _y_y_l_e_x_(_) is called and the global vari- able _e_n_t_e_r___s_p_e_c_i_a_l is true: int enter_special; %x SPECIAL %% if ( enter_special ) BEGIN(SPECIAL); blahblahblah ...more rules follow... Version 2.4 November 1993 15 FLEXDOC(1) FLEXDOC(1) To illustrate the uses of start conditions, here is a scanner which provides two different interpretations of a string like "123.456". By default it will treat it as as three tokens, the integer "123", a dot ('.'), and the integer "456". But if the string is preceded earlier in the line by the string "expect-floats" it will treat it as a single token, the floating-point number 123.456: %{ #include %} %s expect %% expect-floats BEGIN(expect); [0-9]+"."[0-9]+ { printf( "found a float, = %f\n", atof( yytext ) ); } \n { /* that's the end of the line, so * we need another "expect-number" * before we'll recognize any more * numbers */ BEGIN(INITIAL); } [0-9]+ { printf( "found an integer, = %d\n", atoi( yytext ) ); } "." printf( "found a dot\n" ); Here is a scanner which recognizes (and discards) C com- ments while maintaining a count of the current input line. %x comment %% int line_num = 1; "/*" BEGIN(comment); [^*\n]* /* eat anything that's not a '*' */ "*"+[^*/\n]* /* eat up '*'s not followed by '/'s */ \n ++line_num; "*"+"/" BEGIN(INITIAL); This scanner goes to a bit of trouble to match as much text as possible with each rule. In general, when attempting to write a high-speed scanner try to match as much possible in each rule, as it's a big win. Version 2.4 November 1993 16 FLEXDOC(1) FLEXDOC(1) Note that start-conditions names are really integer values and can be stored as such. Thus, the above could be extended in the following fashion: %x comment foo %% int line_num = 1; int comment_caller; "/*" { comment_caller = INITIAL; BEGIN(comment); } ... "/*" { comment_caller = foo; BEGIN(comment); } [^*\n]* /* eat anything that's not a '*' */ "*"+[^*/\n]* /* eat up '*'s not followed by '/'s */ \n ++line_num; "*"+"/" BEGIN(comment_caller); Furthermore, you can access the current start condition using the integer-valued YYYY__SSTTAARRTT macro. For example, the above assignments to _c_o_m_m_e_n_t___c_a_l_l_e_r could instead be writ- ten comment_caller = YY_START; Note that start conditions do not have their own name- space; %s's and %x's declare names in the same fashion as #define's. Finally, here's an example of how to match C-style quoted strings using exclusive start conditions, including expanded escape sequences (but not including checking for a string that's too long): %x str %% char string_buf[MAX_STR_CONST]; char *string_buf_ptr; \" string_buf_ptr = string_buf; BEGIN(str); \" { /* saw closing quote - all done */ BEGIN(INITIAL); *string_buf_ptr = '\0'; Version 2.4 November 1993 17 FLEXDOC(1) FLEXDOC(1) /* return string constant token type and * value to parser */ } \n { /* error - unterminated string constant */ /* generate error message */ } \\[0-7]{1,3} { /* octal escape sequence */ int result; (void) sscanf( yytext + 1, "%o", &result ); if ( result > 0xff ) /* error, constant is out-of-bounds */ *string_buf_ptr++ = result; } \\[0-9]+ { /* generate error - bad escape sequence; something * like '\48' or '\0777777' */ } \\n *string_buf_ptr++ = '\n'; \\t *string_buf_ptr++ = '\t'; \\r *string_buf_ptr++ = '\r'; \\b *string_buf_ptr++ = '\b'; \\f *string_buf_ptr++ = '\f'; \\(.|\n) *string_buf_ptr++ = yytext[1]; [^\\\n\"]+ { char *yytext_ptr = yytext; while ( *yytext_ptr ) *string_buf_ptr++ = *yytext_ptr++; } MMUULLTTIIPPLLEE IINNPPUUTT BBUUFFFFEERRSS Some scanners (such as those which support "include" files) require reading from several input streams. As _f_l_e_x scanners do a large amount of buffering, one cannot control where the next input will be read from by simply writing a YYYY__IINNPPUUTT which is sensitive to the scanning con- text. YYYY__IINNPPUUTT is only called when the scanner reaches the end of its buffer, which may be a long time after scanning a statement such as an "include" which requires switching the input source. Version 2.4 November 1993 18 FLEXDOC(1) FLEXDOC(1) To negotiate these sorts of problems, _f_l_e_x provides a mechanism for creating and switching between multiple input buffers. An input buffer is created by using: YY_BUFFER_STATE yy_create_buffer( FILE *file, int size ) which takes a _F_I_L_E pointer and a size and creates a buffer associated with the given file and large enough to hold _s_i_z_e characters (when in doubt, use YYYY__BBUUFF__SSIIZZEE for the size). It returns a YYYY__BBUUFFFFEERR__SSTTAATTEE handle, which may then be passed to other routines: void yy_switch_to_buffer( YY_BUFFER_STATE new_buffer ) switches the scanner's input buffer so subsequent tokens will come from _n_e_w___b_u_f_f_e_r_. Note that yyyy__sswwiittcchh__ttoo__bbuuffffeerr(()) may be used by yywrap() to set things up for continued scanning, instead of opening a new file and pointing _y_y_i_n at it. void yy_delete_buffer( YY_BUFFER_STATE buffer ) is used to reclaim the storage associated with a buffer. yyyy__nneeww__bbuuffffeerr(()) is an alias for yyyy__ccrreeaattee__bbuuffffeerr(()),, pro- vided for compatibility with the C++ use of _n_e_w and _d_e_l_e_t_e for creating and destroying dynamic objects. Finally, the YYYY__CCUURRRREENNTT__BBUUFFFFEERR macro returns a YYYY__BBUUFFFFEERR__SSTTAATTEE handle to the current buffer. Here is an example of using these features for writing a scanner which expands include files (the <<<>>> feature is discussed below): /* the "incl" state is used for picking up the name * of an include file */ %x incl %{ #define MAX_INCLUDE_DEPTH 10 YY_BUFFER_STATE include_stack[MAX_INCLUDE_DEPTH]; int include_stack_ptr = 0; %} %% include BEGIN(incl); [a-z]+ ECHO; [^a-z\n]*\n? ECHO; [ \t]* /* eat the whitespace */ [^ \t\n]+ { /* got the include file name */ Version 2.4 November 1993 19 FLEXDOC(1) FLEXDOC(1) if ( include_stack_ptr >= MAX_INCLUDE_DEPTH ) { fprintf( stderr, "Includes nested too deeply" ); exit( 1 ); } include_stack[include_stack_ptr++] = YY_CURRENT_BUFFER; yyin = fopen( yytext, "r" ); if ( ! yyin ) error( ... ); yy_switch_to_buffer( yy_create_buffer( yyin, YY_BUF_SIZE ) ); BEGIN(INITIAL); } <> { if ( --include_stack_ptr < 0 ) { yyterminate(); } else { yy_delete_buffer( YY_CURRENT_BUFFER ); yy_switch_to_buffer( include_stack[include_stack_ptr] ); } } EENNDD--OOFF--FFIILLEE RRUULLEESS The special rule "<>" indicates actions which are to be taken when an end-of-file is encountered and yywrap() returns non-zero (i.e., indicates no further files to pro- cess). The action must finish by doing one of four things: - assigning _y_y_i_n to a new input file (in previous versions of flex, after doing the assignment you had to call the special action YYYY__NNEEWW__FFIILLEE;; this is no longer necessary); - executing a _r_e_t_u_r_n statement; - executing the special yyyytteerrmmiinnaattee(()) action; - or, switching to a new buffer using yyyy__sswwiittcchh__ttoo__bbuuffffeerr(()) as shown in the example above. Version 2.4 November 1993 20 FLEXDOC(1) FLEXDOC(1) <> rules may not be used with other patterns; they may only be qualified with a list of start conditions. If an unqualified <> rule is given, it applies to _a_l_l start conditions which do not already have <> actions. To specify an <> rule for only the initial start condition, use <> These rules are useful for catching things like unclosed comments. An example: %x quote %% ...other rules for dealing with quotes... <> { error( "unterminated quote" ); yyterminate(); } <> { if ( *++filelist ) yyin = fopen( *filelist, "r" ); else yyterminate(); } MMIISSCCEELLLLAANNEEOOUUSS MMAACCRROOSS The macro YY_USER_ACTION can be defined to provide an action which is always executed prior to the matched rule's action. For example, it could be #define'd to call a routine to convert yytext to lower-case. The macro YYYY__UUSSEERR__IINNIITT may be defined to provide an action which is always executed before the first scan (and before the scanner's internal initializations are done). For example, it could be used to call a routine to read in a data table or open a logging file. In the generated scanner, the actions are all gathered in one large switch statement and separated using YYYY__BBRREEAAKK,, which may be redefined. By default, it is simply a "break", to separate each rule's action from the following rule's. Redefining YYYY__BBRREEAAKK allows, for example, C++ users to #define YY_BREAK to do nothing (while being very careful that every rule ends with a "break" or a "return"!) to avoid suffering from unreachable statement warnings where because a rule's action ends with "return", the YYYY__BBRREEAAKK is inaccessible. Version 2.4 November 1993 21 FLEXDOC(1) FLEXDOC(1) IINNTTEERRFFAACCIINNGG WWIITTHH YYAACCCC One of the main uses of _f_l_e_x is as a companion to the _y_a_c_c parser-generator. _y_a_c_c parsers expect to call a routine named yyyylleexx(()) to find the next input token. The routine is supposed to return the type of the next token as well as putting any associated value in the global yyyyllvvaall.. To use _f_l_e_x with _y_a_c_c_, one specifies the --dd option to _y_a_c_c to instruct it to generate the file yy..ttaabb..hh containing defi- nitions of all the %%ttookkeennss appearing in the _y_a_c_c input. This file is then included in the _f_l_e_x scanner. For exam- ple, if one of the tokens is "TOK_NUMBER", part of the scanner might look like: %{ #include "y.tab.h" %} %% [0-9]+ yylval = atoi( yytext ); return TOK_NUMBER; OOPPTTIIOONNSS _f_l_e_x has the following options: --bb Generate backing-up information to _l_e_x_._b_a_c_k_u_p_. This is a list of scanner states which require backing up and the input characters on which they do so. By adding rules one can remove backing-up states. If all backing-up states are eliminated and --CCff or --CCFF is used, the generated scanner will run faster (see the --pp flag). Only users who wish to squeeze every last cycle out of their scanners need worry about this option. (See the section on Performance Considerations below.) --cc is a do-nothing, deprecated option included for POSIX compliance. NNOOTTEE:: in previous releases of _f_l_e_x --cc specified table-compression options. This functionality is now given by the --CC flag. To ease the the impact of this change, when _f_l_e_x encounters --cc,, it cur- rently issues a warning message and assumes that --CC was desired instead. In the future this "promo- tion" of --cc to --CC will go away in the name of full POSIX compliance (unless the POSIX meaning is removed first). --dd makes the generated scanner run in _d_e_b_u_g mode. Whenever a pattern is recognized and the global yyyy__fflleexx__ddeebbuugg is non-zero (which is the default), the scanner will write to _s_t_d_e_r_r a line of the form: Version 2.4 November 1993 22 FLEXDOC(1) FLEXDOC(1) --accepting rule at line 53 ("the matched text") The line number refers to the location of the rule in the file defining the scanner (i.e., the file that was fed to flex). Messages are also generated when the scanner backs up, accepts the default rule, reaches the end of its input buffer (or encounters a NUL; at this point, the two look the same as far as the scanner's concerned), or reaches an end-of-file. --ff specifies _f_a_s_t _s_c_a_n_n_e_r_. No table compression is done and stdio is bypassed. The result is large but fast. This option is equivalent to --CCffrr (see below). --hh generates a "help" summary of _f_l_e_x_'_s options to _s_t_d_e_r_r and then exits. --ii instructs _f_l_e_x to generate a _c_a_s_e_-_i_n_s_e_n_s_i_t_i_v_e scan- ner. The case of letters given in the _f_l_e_x input patterns will be ignored, and tokens in the input will be matched regardless of case. The matched text given in _y_y_t_e_x_t will have the preserved case (i.e., it will not be folded). --ll turns on maximum compatibility with the original AT&T _l_e_x implementation. Note that this does not mean _f_u_l_l compatibility. Use of this option costs a considerable amount of performance, and it cannot be used with the --++,, --ff,, --FF,, --CCff,, or --CCFF options. For details on the compatibilities it provides, see the section "Incompatibilities With Lex And POSIX" below. --nn is another do-nothing, deprecated option included only for POSIX compliance. --pp generates a performance report to stderr. The report consists of comments regarding features of the _f_l_e_x input file which will cause a serious loss of performance in the resulting scanner. If you give the flag twice, you will also get comments regarding features that lead to minor performance losses. Note that the use of RREEJJEECCTT and variable trailing context (see the Bugs section in flex(1)) entails a substantial performance penalty; use of _y_y_m_o_r_e_(_)_, the ^^ operator, and the --II flag entail minor per- formance penalties. --ss causes the _d_e_f_a_u_l_t _r_u_l_e (that unmatched scanner input is echoed to _s_t_d_o_u_t_) to be suppressed. If Version 2.4 November 1993 23 FLEXDOC(1) FLEXDOC(1) the scanner encounters input that does not match any of its rules, it aborts with an error. This option is useful for finding holes in a scanner's rule set. --tt instructs _f_l_e_x to write the scanner it generates to standard output instead of lleexx..yyyy..cc.. --vv specifies that _f_l_e_x should write to _s_t_d_e_r_r a sum- mary of statistics regarding the scanner it gener- ates. Most of the statistics are meaningless to the casual _f_l_e_x user, but the first line identifies the version of _f_l_e_x (same as reported by --VV)),, and the next line the flags used when generating the scanner, including those that are on by default. --ww suppresses warning messages. --BB instructs _f_l_e_x to generate a _b_a_t_c_h scanner, the opposite of _i_n_t_e_r_a_c_t_i_v_e scanners generated by --II (see below). In general, you use --BB when you are _c_e_r_t_a_i_n that your scanner will never be used inter- actively, and you want to squeeze a _l_i_t_t_l_e more performance out of it. If your goal is instead to squeeze out a _l_o_t more performance, you should be using the --CCff or --CCFF options (discussed below), which turn on --BB automatically anyway. --FF specifies that the _f_a_s_t scanner table representa- tion should be used (and stdio bypassed). This representation is about as fast as the full table representation ((--ff)),, and for some sets of patterns will be considerably smaller (and for others, larger). In general, if the pattern set contains both "keywords" and a catch-all, "identifier" rule, such as in the set: "case" return TOK_CASE; "switch" return TOK_SWITCH; ... "default" return TOK_DEFAULT; [a-z]+ return TOK_ID; then you're better off using the full table repre- sentation. If only the "identifier" rule is pre- sent and you then use a hash table or some such to detect the keywords, you're better off using --FF.. This option is equivalent to --CCFFrr (see below). It cannot be used with --++.. --II instructs _f_l_e_x to generate an _i_n_t_e_r_a_c_t_i_v_e scanner. An interactive scanner is one that only looks ahead to decide what token has been matched if it Version 2.4 November 1993 24 FLEXDOC(1) FLEXDOC(1) absolutely must. It turns out that always looking one extra character ahead, even if the scanner has already seen enough text to disambiguate the cur- rent token, is a bit faster than only looking ahead when necessary. But scanners that always look ahead give dreadful interactive performance; for example, when a user types a newline, it is not recognized as a newline token until they enter _a_n_o_t_h_e_r token, which often means typing in another whole line. _F_l_e_x scanners default to _i_n_t_e_r_a_c_t_i_v_e unless you use the --CCff or --CCFF table-compression options (see below). That's because if you're looking for high- performance you should be using one of these options, so if you didn't, _f_l_e_x assumes you'd rather trade off a bit of run-time performance for intuitive interactive behavior. Note also that you _c_a_n_n_o_t use --II in conjunction with --CCff or --CCFF.. Thus, this option is not really needed; it is on by default for all those cases in which it is allowed. You can force a scanner to _n_o_t be interactive by using --BB (see above). --LL instructs _f_l_e_x not to generate ##lliinnee directives. Without this option, _f_l_e_x peppers the generated scanner with #line directives so error messages in the actions will be correctly located with respect to the original _f_l_e_x input file, and not to the fairly meaningless line numbers of lleexx..yyyy..cc.. (Unfortunately _f_l_e_x does not presently generate the necessary directives to "retarget" the line numbers for those parts of lleexx..yyyy..cc which it generated. So if there is an error in the generated code, a mean- ingless line number is reported.) --TT makes _f_l_e_x run in _t_r_a_c_e mode. It will generate a lot of messages to _s_t_d_e_r_r concerning the form of the input and the resultant non-deterministic and deterministic finite automata. This option is mostly for use in maintaining _f_l_e_x_. --VV prints the version number to _s_t_d_e_r_r and exits. --77 instructs _f_l_e_x to generate a 7-bit scanner, i.e., one which can only recognized 7-bit characters in its input. The advantage of using --77 is that the scanner's tables can be up to half the size of those generated using the --88 option (see below). The disadvantage is that such scanners often hang or crash if their input contains an 8-bit charac- ter. Version 2.4 November 1993 25 FLEXDOC(1) FLEXDOC(1) Note, however, that unless you generate your scan- ner using the --CCff or --CCFF table compression options, use of --77 will save only a small amount of table space, and make your scanner considerably less portable. _F_l_e_x_'_s default behavior is to generate an 8-bit scanner unless you use the --CCff or --CCFF,, in which case _f_l_e_x defaults to generating 7-bit scan- ners unless your site was always configured to gen- erate 8-bit scanners (as will often be the case with non-USA sites). You can tell whether flex generated a 7-bit or an 8-bit scanner by inspecting the flag summary in the --vv output as described above. Note that if you use --CCffee or --CCFFee (those table com- pression options, but also using equivalence classes as discussed see below), flex still defaults to generating an 8-bit scanner, since usu- ally with these compression options full 8-bit tables are not much more expensive than 7-bit tables. --88 instructs _f_l_e_x to generate an 8-bit scanner, i.e., one which can recognize 8-bit characters. This flag is only needed for scanners generated using --CCff or --CCFF,, as otherwise flex defaults to generat- ing an 8-bit scanner anyway. See the discussion of --77 above for flex's default behavior and the tradeoffs between 7-bit and 8-bit scanners. --++ specifies that you want flex to generate a C++ scanner class. See the section on Generating C++ Scanners below for details. --CC[[aaeeffFFmmrr]] controls the degree of table compression and, more generally, trade-offs between small scanners and fast scanners. --CCaa ("align") instructs flex to trade off larger tables in the generated scanner for faster perfor- mance because the elements of the tables are better aligned for memory access and computation. On some RISC architectures, fetching and manipulating long- words is more efficient than with smaller-sized datums such as shortwords. This option can double the size of the tables used by your scanner. --CCee directs _f_l_e_x to construct _e_q_u_i_v_a_l_e_n_c_e _c_l_a_s_s_e_s_, i.e., sets of characters which have identical lexi- cal properties (for example, if the only appearance of digits in the _f_l_e_x input is in the character Version 2.4 November 1993 26 FLEXDOC(1) FLEXDOC(1) class "[0-9]" then the digits '0', '1', ..., '9' will all be put in the same equivalence class). Equivalence classes usually give dramatic reduc- tions in the final table/object file sizes (typi- cally a factor of 2-5) and are pretty cheap perfor- mance-wise (one array look-up per character scanned). --CCff specifies that the _f_u_l_l scanner tables should be generated - _f_l_e_x should not compress the tables by taking advantages of similar transition func- tions for different states. --CCFF specifies that the alternate fast scanner rep- resentation (described above under the --FF flag) should be used. This option cannot be used with --++.. --CCmm directs _f_l_e_x to construct _m_e_t_a_-_e_q_u_i_v_a_l_e_n_c_e _c_l_a_s_s_e_s_, which are sets of equivalence classes (or characters, if equivalence classes are not being used) that are commonly used together. Meta- equivalence classes are often a big win when using compressed tables, but they have a moderate perfor- mance impact (one or two "if" tests and one array look-up per character scanned). --CCrr causes the generated scanner to _b_y_p_a_s_s use of the standard I/O library (stdio) for input. Instead of calling ffrreeaadd(()) or ggeettcc(()),, the scanner will use the rreeaadd(()) system call, resulting in a performance gain which varies from system to sys- tem, but in general is probably negligible unless you are also using --CCff or --CCFF.. Using --CCrr can cause strange behavior if, for example, you read from _y_y_i_n using stdio prior to calling the scanner (because the scanner will miss whatever text your previous reads left in the stdio input buffer). --CCrr has no effect if you define YYYY__IINNPPUUTT (see The Generated Scanner above). A lone --CC specifies that the scanner tables should be compressed but neither equivalence classes nor meta-equivalence classes should be used. The options --CCff or --CCFF and --CCmm do not make sense together - there is no opportunity for meta- equivalence classes if the table is not being com- pressed. Otherwise the options may be freely mixed, and are cumulative. The default setting is --CCeemm,, which specifies that _f_l_e_x should generate equivalence classes and meta- Version 2.4 November 1993 27 FLEXDOC(1) FLEXDOC(1) equivalence classes. This setting provides the highest degree of table compression. You can trade off faster-executing scanners at the cost of larger tables with the following generally being true: slowest & smallest -Cem -Cm -Ce -C -C{f,F}e -C{f,F} -C{f,F}a fastest & largest Note that scanners with the smallest tables are usually generated and compiled the quickest, so during development you will usually want to use the default, maximal compression. --CCffee is often a good compromise between speed and size for production scanners. --PPpprreeffiixx changes the default _y_y prefix used by _f_l_e_x for all globally-visible variable and function names to instead be _p_r_e_f_i_x_. For example, --PPffoooo changes the name of yyyytteexxtt to ffooootteexxtt.. It also changes the name of the default output file from lleexx..yyyy..cc to lleexx..ffoooo..cc.. Here are all of the names affected: yyFlexLexer yy_create_buffer yy_delete_buffer yy_flex_debug yy_init_buffer yy_load_buffer_state yy_switch_to_buffer yyin yyleng yylex yyout yyrestart yytext yywrap Within your scanner itself, you can still refer to the global variables and functions using either version of their name; but eternally, they have the modified name. This option lets you easily link together multiple _f_l_e_x programs into the same executable. Note, though, that using this option also renames Version 2.4 November 1993 28 FLEXDOC(1) FLEXDOC(1) yyyywwrraapp(()),, so you now _m_u_s_t provide your own (appro- priately-named) version of the routine for your scanner, as linking with --llffll no longer provides one for you by default. --SSsskkeelleettoonn__ffiillee overrides the default skeleton file from which _f_l_e_x constructs its scanners. You'll never need this option unless you are doing _f_l_e_x maintenance or development. PPEERRFFOORRMMAANNCCEE CCOONNSSIIDDEERRAATTIIOONNSS The main design goal of _f_l_e_x is that it generate high- performance scanners. It has been optimized for dealing well with large sets of rules. Aside from the effects on scanner speed of the table compression --CC options outlined above, there are a number of options/actions which degrade performance. These are, from most expensive to least: REJECT pattern sets that require backing up arbitrary trailing context yymore() '^' beginning-of-line operator with the first three all being quite expensive and the last two being quite cheap. Note also that uunnppuutt(()) is implemented as a routine call that potentially does quite a bit of work, while yyyylleessss(()) is a quite-cheap macro; so if just putting back some excess text you scanned, use yyyylleessss(()).. RREEJJEECCTT should be avoided at all costs when performance is important. It is a particularly expensive option. Getting rid of backing up is messy and often may be an enormous amount of work for a complicated scanner. In principal, one begins by using the --bb flag to generate a _l_e_x_._b_a_c_k_u_p file. For example, on the input %% foo return TOK_KEYWORD; foobar return TOK_KEYWORD; the file looks like: State #6 is non-accepting - associated rule line numbers: 2 3 out-transitions: [ o ] jam-transitions: EOF [ \001-n p-\177 ] Version 2.4 November 1993 29 FLEXDOC(1) FLEXDOC(1) State #8 is non-accepting - associated rule line numbers: 3 out-transitions: [ a ] jam-transitions: EOF [ \001-` b-\177 ] State #9 is non-accepting - associated rule line numbers: 3 out-transitions: [ r ] jam-transitions: EOF [ \001-q s-\177 ] Compressed tables always back up. The first few lines tell us that there's a scanner state in which it can make a transition on an 'o' but not on any other character, and that in that state the currently scanned text does not match any rule. The state occurs when trying to match the rules found at lines 2 and 3 in the input file. If the scanner is in that state and then reads something other than an 'o', it will have to back up to find a rule which is matched. With a bit of head- scratching one can see that this must be the state it's in when it has seen "fo". When this has happened, if any- thing other than another 'o' is seen, the scanner will have to back up to simply match the 'f' (by the default rule). The comment regarding State #8 indicates there's a problem when "foob" has been scanned. Indeed, on any character other than an 'a', the scanner will have to back up to accept "foo". Similarly, the comment for State #9 con- cerns when "fooba" has been scanned and an 'r' does not follow. The final comment reminds us that there's no point going to all the trouble of removing backing up from the rules unless we're using --CCff or --CCFF,, since there's no perfor- mance gain doing so with compressed scanners. The way to remove the backing up is to add "error" rules: %% foo return TOK_KEYWORD; foobar return TOK_KEYWORD; fooba | foob | fo { /* false alarm, not really a keyword */ return TOK_ID; } Version 2.4 November 1993 30 FLEXDOC(1) FLEXDOC(1) Eliminating backing up among a list of keywords can also be done using a "catch-all" rule: %% foo return TOK_KEYWORD; foobar return TOK_KEYWORD; [a-z]+ return TOK_ID; This is usually the best solution when appropriate. Backing up messages tend to cascade. With a complicated set of rules it's not uncommon to get hundreds of mes- sages. If one can decipher them, though, it often only takes a dozen or so rules to eliminate the backing up (though it's easy to make a mistake and have an error rule accidentally match a valid token. A possible future _f_l_e_x feature will be to automatically add rules to eliminate backing up). _V_a_r_i_a_b_l_e trailing context (where both the leading and trailing parts do not have a fixed length) entails almost the same performance loss as RREEJJEECCTT (i.e., substantial). So when possible a rule like: %% mouse|rat/(cat|dog) run(); is better written: %% mouse/cat|dog run(); rat/cat|dog run(); or as %% mouse|rat/cat run(); mouse|rat/dog run(); Note that here the special '|' action does _n_o_t provide any savings, and can even make things worse (see A final note regarding performance: as mentioned above in the section How the Input is Matched, dynamically resizing yyyytteexxtt to accomodate huge tokens is a slow process because it presently requires that the (huge) token be rescanned from the beginning. Thus if performance is vital, you should attempt to match "large" quantities of text but not "huge" quantities, where the cutoff between the two is at about 8K characters/token. Another area where the user can increase a scanner's per- formance (and one that's easier to implement) arises from Version 2.4 November 1993 31 FLEXDOC(1) FLEXDOC(1) the fact that the longer the tokens matched, the faster the scanner will run. This is because with long tokens the processing of most input characters takes place in the (short) inner scanning loop, and does not often have to go through the additional work of setting up the scanning environment (e.g., yyyytteexxtt)) for the action. Recall the scanner for C comments: %x comment %% int line_num = 1; "/*" BEGIN(comment); [^*\n]* "*"+[^*/\n]* \n ++line_num; "*"+"/" BEGIN(INITIAL); This could be sped up by writing it as: %x comment %% int line_num = 1; "/*" BEGIN(comment); [^*\n]* [^*\n]*\n ++line_num; "*"+[^*/\n]* "*"+[^*/\n]*\n ++line_num; "*"+"/" BEGIN(INITIAL); Now instead of each newline requiring the processing of another action, recognizing the newlines is "distributed" over the other rules to keep the matched text as long as possible. Note that _a_d_d_i_n_g rules does _n_o_t slow down the scanner! The speed of the scanner is independent of the number of rules or (modulo the considerations given at the beginning of this section) how complicated the rules are with regard to operators such as '*' and '|'. A final example in speeding up a scanner: suppose you want to scan through a file containing identifiers and key- words, one per line and with no other extraneous charac- ters, and recognize all the keywords. A natural first approach is: %% asm | auto | break | ... etc ... volatile | Version 2.4 November 1993 32 FLEXDOC(1) FLEXDOC(1) while /* it's a keyword */ .|\n /* it's not a keyword */ To eliminate the back-tracking, introduce a catch-all rule: %% asm | auto | break | ... etc ... volatile | while /* it's a keyword */ [a-z]+ | .|\n /* it's not a keyword */ Now, if it's guaranteed that there's exactly one word per line, then we can reduce the total number of matches by a half by merging in the recognition of newlines with that of the other tokens: %% asm\n | auto\n | break\n | ... etc ... volatile\n | while\n /* it's a keyword */ [a-z]+\n | .|\n /* it's not a keyword */ One has to be careful here, as we have now reintroduced backing up into the scanner. In particular, while _w_e know that there will never be any characters in the input stream other than letters or newlines, _f_l_e_x can't figure this out, and it will plan for possibly needing to back up when it has scanned a token like "auto" and then the next character is something other than a newline or a letter. Previously it would then just match the "auto" rule and be done, but now it has no "auto" rule, only a "auto\n" rule. To eliminate the possibility of backing up, we could either duplicate all rules but without final newlines, or, since we never expect to encounter such an input and therefore don't how it's classified, we can introduce one more catch-all rule, this one which doesn't include a new- line: %% asm\n | auto\n | break\n | Version 2.4 November 1993 33 FLEXDOC(1) FLEXDOC(1) ... etc ... volatile\n | while\n /* it's a keyword */ [a-z]+\n | [a-z]+ | .|\n /* it's not a keyword */ Compiled with --CCff,, this is about as fast as one can get a _f_l_e_x scanner to go for this particular problem. A final note: _f_l_e_x is slow when matching NUL's, particu- larly when a token contains multiple NUL's. It's best to write rules which match _s_h_o_r_t amounts of text if it's anticipated that the text will often include NUL's. GGEENNEERRAATTIINNGG CC++++ SSCCAANNNNEERRSS _f_l_e_x provides two different ways to generate scanners for use with C++. The first way is to simply compile a scan- ner generated by _f_l_e_x using a C++ compiler instead of a C compiler. You should not encounter any compilations errors (please report any you find to the email address given in the Author section below). You can then use C++ code in your rule actions instead of C code. Note that the default input source for your scanner remains _y_y_i_n_, and default echoing is still done to _y_y_o_u_t_. Both of these remain _F_I_L_E _* variables and not C++ _s_t_r_e_a_m_s_. You can also use _f_l_e_x to generate a C++ scanner class, using the --++ option, which is automatically specified if the name of the flex executable ends in a '+', such as _f_l_e_x_+_+_. When using this option, flex defaults to generat- ing the scanner to the file lleexx..yyyy..cccc instead of lleexx..yyyy..cc.. The generated scanner includes the header file _F_l_e_x_L_e_x_e_r_._h_, which defines the interface to two C++ classes. The first class, FFlleexxLLeexxeerr,, provides an abstract base class defining the general scanner class interface. It provides the following member functions: ccoonnsstt cchhaarr** YYYYTTeexxtt(()) returns the text of the most recently matched token, the equivalent of yyyytteexxtt.. iinntt YYYYLLeenngg(()) returns the length of the most recently matched token, the equivalent of yyyylleenngg.. Also provided are member functions equivalent to yyyy__sswwiittcchh__ttoo__bbuuffffeerr(()),, yyyy__ccrreeaattee__bbuuffffeerr(()) (though the first argument is an iissttrreeaamm** object pointer and not a FFIILLEE**)),, yyyy__ddeelleettee__bbuuffffeerr(()),, and yyyyrreessttaarrtt(()) (again, the first argument is a iissttrreeaamm** object pointer). Version 2.4 November 1993 34 FLEXDOC(1) FLEXDOC(1) The second class defined in _F_l_e_x_L_e_x_e_r_._h is yyyyFFlleexxLLeexxeerr,, which is derived from FFlleexxLLeexxeerr.. It defines the following additional member functions: yyyyFFlleexxLLeexxeerr(( iissttrreeaamm** aarrgg__yyyyiinn == 00,, oossttrreeaamm** aarrgg__yyyyoouutt == 00 )) constructs a yyyyFFlleexxLLeexxeerr object using the given streams for input and output. If not specified, the streams default to cciinn and ccoouutt,, respectively. vviirrttuuaall iinntt yyyylleexx(()) performs the same role is yyyylleexx(()) does for ordinary flex scanners: it scans the input stream, consuming tokens, until a rule's action returns a value. In addition, yyyyFFlleexxLLeexxeerr defines the following protected virtual functions which you can redefine in derived classes to tailor the scanner's input and output: vviirrttuuaall iinntt LLeexxeerrIInnppuutt(( cchhaarr** bbuuff,, iinntt mmaaxx__ssiizzee )) reads up to mmaaxx__ssiizzee characters into bbuuff and returns the number of characters read. To indicate end-of-input, return 0 characters. Note that "interactive" scanners (see the --BB and --II flags) define the macro YYYY__IINNTTEERRAACCTTIIVVEE.. If you redefine LLeexxeerrIInnppuutt(()) and need to take different actions depending on whether or not the scanner might be scanning an interactive input source, you can test for the presence of this name via ##iiffddeeff.. vviirrttuuaall vvooiidd LLeexxeerrOOuuttppuutt(( ccoonnsstt cchhaarr** bbuuff,, iinntt ssiizzee )) writes out ssiizzee characters from the buffer bbuuff,, which, while NUL-terminated, may also contain "internal" NUL's if the scanner's rules can match text with NUL's in them. Note that a yyyyFFlleexxLLeexxeerr object contains its _e_n_t_i_r_e scan- ning state. Thus you can use such objects to create reen- trant scanners. You can instantiate multiple instances of the same yyyyFFlleexxLLeexxeerr class, and you can also combine mul- tiple C++ scanner classes together in the same program using the --PP option discussed above. Finally, note that the %%aarrrraayy feature is not available to C++ scanner classes; you must use %%ppooiinntteerr (the default). Here is an example of a simple C++ scanner: // An example of using the flex C++ scanner class. %{ int mylineno = 0; %} Version 2.4 November 1993 35 FLEXDOC(1) FLEXDOC(1) string \"[^\n"]+\" ws [ \t]+ alpha [A-Za-z] dig [0-9] name ({alpha}|{dig}|\$)({alpha}|{dig}|[_.\-/$])* num1 [-+]?{dig}+\.?([eE][-+]?{dig}+)? num2 [-+]?{dig}*\.{dig}+([eE][-+]?{dig}+)? number {num1}|{num2} %% {ws} /* skip blanks and tabs */ "/*" { int c; while((c = yyinput()) != 0) { if(c == '\n') ++mylineno; else if(c == '*') { if((c = yyinput()) == '/') break; else unput(c); } } } {number} cout << "number " << YYText() << '\n'; \n mylineno++; {name} cout << "name " << YYText() << '\n'; {string} cout << "string " << YYText() << '\n'; %% int main( int /* argc */, char** /* argv */ ) { FlexLexer* lexer = new yyFlexLexer; while(lexer->yylex() != 0) ; return 0; } IINNCCOOMMPPAATTIIBBIILLIITTIIEESS WWIITTHH LLEEXX AANNDD PPOOSSIIXX _f_l_e_x is a rewrite of the AT&T Unix _l_e_x tool (the two implementations do not share any code, though), with some Version 2.4 November 1993 36 FLEXDOC(1) FLEXDOC(1) extensions and incompatibilities, both of which are of concern to those who wish to write scanners acceptable to either implementation. The POSIX _l_e_x specification is closer to _f_l_e_x_'_s behavior than that of the original _l_e_x implementation, but there also remain some incompatibili- ties between _f_l_e_x and POSIX. The intent is that ulti- mately _f_l_e_x will be fully POSIX-conformant. In this sec- tion we discuss all of the known areas of incompatibility. _f_l_e_x_'_s --ll option turns on maximum compatibility with the original AT&T _l_e_x implementation, at the cost of a major loss in the generated scanner's performance. We note below which incompatibilities can be overcome using the --ll option. _f_l_e_x is fully compatible with _l_e_x with the following exceptions: - The undocumented _l_e_x scanner internal variable yyyylliinneennoo is not supported unless --ll is used. yylineno is not part of the POSIX specification. - The iinnppuutt(()) routine is not redefinable, though it may be called to read characters following whatever has been matched by a rule. If iinnppuutt(()) encounters an end-of-file the normal yyyywwrraapp(()) processing is done. A ``real'' end-of-file is returned by iinnppuutt(()) as _E_O_F_. Input is instead controlled by defining the YYYY__IINNPPUUTT macro. The _f_l_e_x restriction that iinnppuutt(()) cannot be rede- fined is in accordance with the POSIX specifica- tion, which simply does not specify any way of con- trolling the scanner's input other than by making an initial assignment to _y_y_i_n_. - _f_l_e_x scanners are not as reentrant as _l_e_x scanners. In particular, if you have an interactive scanner and an interrupt handler which long-jumps out of the scanner, and the scanner is subsequently called again, you may get the following message: fatal flex scanner internal error--end of buffer missed To reenter the scanner, first use yyrestart( yyin ); Note that this call will throw away any buffered input; usually this isn't a problem with an inter- active scanner. Version 2.4 November 1993 37 FLEXDOC(1) FLEXDOC(1) Also note that flex C++ scanner classes _a_r_e reen- trant, so if using C++ is an option for you, you should use them instead. See "Generating C++ Scan- ners" above for details. - oouuttppuutt(()) is not supported. Output from the EECCHHOO macro is done to the file-pointer _y_y_o_u_t (default _s_t_d_o_u_t_)_. oouuttppuutt(()) is not part of the POSIX specification. - _l_e_x does not support exclusive start conditions (%x), though they are in the POSIX specification. - When definitions are expanded, _f_l_e_x encloses them in parentheses. With lex, the following: NAME [A-Z][A-Z0-9]* %% foo{NAME}? printf( "Found it\n" ); %% will not match the string "foo" because when the macro is expanded the rule is equivalent to "foo[A- Z][A-Z0-9]*?" and the precedence is such that the '?' is associated with "[A-Z0-9]*". With _f_l_e_x_, the rule will be expanded to "foo([A-Z][A-Z0-9]*)?" and so the string "foo" will match. Note that if the definition begins with ^^ or ends with $$ then it is _n_o_t expanded with parentheses, to allow these operators to appear in definitions without losing their special meanings. But the <>,, //,, and <<<>>> operators cannot be used in a _f_l_e_x definition. Using --ll results in the _l_e_x behavior of no paren- theses around the definition. The POSIX specification is that the definition be enclosed in parentheses. - The _l_e_x %%rr (generate a Ratfor scanner) option is not supported. It is not part of the POSIX speci- fication. - After a call to uunnppuutt(()),, _y_y_t_e_x_t and _y_y_l_e_n_g are undefined until the next token is matched, unless the scanner was built using %%aarrrraayy.. This is not the case with _l_e_x or the POSIX specification. The --ll option does away with this incompatibility. - The precedence of the {{}} (numeric range) operator is different. _l_e_x interprets "abc{1,3}" as "match Version 2.4 November 1993 38 FLEXDOC(1) FLEXDOC(1) one, two, or three occurrences of 'abc'", whereas _f_l_e_x interprets it as "match 'ab' followed by one, two, or three occurrences of 'c'". The latter is in agreement with the POSIX specification. - The precedence of the ^^ operator is different. _l_e_x interprets "^foo|bar" as "match either 'foo' at the beginning of a line, or 'bar' anywhere", whereas _f_l_e_x interprets it as "match either 'foo' or 'bar' if they come at the beginning of a line". The lat- ter is in agreement with the POSIX specification. - _y_y_i_n is _i_n_i_t_i_a_l_i_z_e_d by _l_e_x to be _s_t_d_i_n_; _f_l_e_x_, on the other hand, initializes _y_y_i_n to NULL and then _a_s_s_i_g_n_s it to _s_t_d_i_n the first time the scanner is called, providing _y_y_i_n has not already been assigned to a non-NULL value. The difference is subtle, but the net effect is that with _f_l_e_x scan- ners, _y_y_i_n does not have a valid value until the scanner has been called. The --ll option does away with this incompatibility. - The special table-size declarations such as %%aa sup- ported by _l_e_x are not required by _f_l_e_x scanners; _f_l_e_x ignores them. - The name FLEX_SCANNER is #define'd so scanners may be written for use with either _f_l_e_x or _l_e_x_. The following _f_l_e_x features are not included in _l_e_x or the POSIX specification: yyterminate() <> <*> YY_DECL YY_START YY_USER_ACTION #line directives %{}'s around actions multiple actions on a line plus almost all of the flex flags. The last feature in the list refers to the fact that with _f_l_e_x you can put multiple actions on the same line, separated with semi- colons, while with _l_e_x_, the following foo handle_foo(); ++num_foos_seen; is (rather surprisingly) truncated to foo handle_foo(); Version 2.4 November 1993 39 FLEXDOC(1) FLEXDOC(1) _f_l_e_x does not truncate the action. Actions that are not enclosed in braces are simply terminated at the end of the line. DDIIAAGGNNOOSSTTIICCSS If you receive errors when linking a _f_l_e_x scanner com- plaining about the following missing routines: yywrap yy_flex_alloc ... (and various others) then you forgot to link your program with --llffll.. This run-time library is _r_e_q_u_i_r_e_d for all _f_l_e_x scanners. _w_a_r_n_i_n_g_, _r_u_l_e _c_a_n_n_o_t _b_e _m_a_t_c_h_e_d indicates that the given rule cannot be matched because it follows other rules that will always match the same text as it. For example, in the following "foo" cannot be matched because it comes after an identifier "catch-all" rule: [a-z]+ got_identifier(); foo got_foo(); Using RREEJJEECCTT in a scanner suppresses this warning. _w_a_r_n_i_n_g_, --ss _o_p_t_i_o_n _g_i_v_e_n _b_u_t _d_e_f_a_u_l_t _r_u_l_e _c_a_n _b_e _m_a_t_c_h_e_d means that it is possible (perhaps only in a particular start condition) that the default rule (match any single character) is the only one that will match a particular input. Since --ss was given, presumably this is not intended. _r_e_j_e_c_t___u_s_e_d___b_u_t___n_o_t___d_e_t_e_c_t_e_d _u_n_d_e_f_i_n_e_d or _y_y_m_o_r_e___u_s_e_d___b_u_t___n_o_t___d_e_t_e_c_t_e_d _u_n_d_e_f_i_n_e_d _- These errors can occur at compile time. They indicate that the scanner uses RREEJJEECCTT or yyyymmoorree(()) but that _f_l_e_x failed to notice the fact, meaning that _f_l_e_x scanned the first two sections looking for occurrences of these actions and failed to find any, but somehow you snuck some in (via a #include file, for example). Make an explicit reference to the action in your _f_l_e_x input file. (Note that previously _f_l_e_x supported a %%uusseedd//%%uunnuusseedd mechanism for dealing with this problem; this feature is still supported but now dep- recated, and will go away soon unless the author hears from people who can argue compellingly that they need it.) _f_l_e_x _s_c_a_n_n_e_r _j_a_m_m_e_d _- a scanner compiled with --ss has encountered an input string which wasn't matched by any of its rules. This error can also occur due to internal problems. _t_o_k_e_n _t_o_o _l_a_r_g_e_, _e_x_c_e_e_d_s _Y_Y_L_M_A_X _- your scanner uses %%aarrrraayy and one of its rules matched a string longer than the YYYYLL-- MMAAXX constant (8K bytes by default). You can increase the value by #define'ing YYYYLLMMAAXX in the definitions section of Version 2.4 November 1993 40 FLEXDOC(1) FLEXDOC(1) your _f_l_e_x input. _s_c_a_n_n_e_r _r_e_q_u_i_r_e_s _-_8 _f_l_a_g _t_o _u_s_e _t_h_e _c_h_a_r_a_c_t_e_r _'_x_' _- Your scanner specification includes recognizing the 8-bit char- acter _'_x_' and you did not specify the -8 flag, and your scanner defaulted to 7-bit because you used the --CCff or --CCFF table compression options. See the discussion of the --77 flag for details. _f_l_e_x _s_c_a_n_n_e_r _p_u_s_h_-_b_a_c_k _o_v_e_r_f_l_o_w _- you used uunnppuutt(()) to push back so much text that the scanner's buffer could not hold both the pushed-back text and the current token in yyyytteexxtt.. Ideally the scanner should dynamically resize the buffer in this case, but at present it does not. _i_n_p_u_t _b_u_f_f_e_r _o_v_e_r_f_l_o_w_, _c_a_n_'_t _e_n_l_a_r_g_e _b_u_f_f_e_r _b_e_c_a_u_s_e _s_c_a_n_- _n_e_r _u_s_e_s _R_E_J_E_C_T _- the scanner was working on matching an extremely large token and needed to expand the input buffer. This doesn't work with scanners that use RREEJJEECCTT.. _f_a_t_a_l _f_l_e_x _s_c_a_n_n_e_r _i_n_t_e_r_n_a_l _e_r_r_o_r_-_-_e_n_d _o_f _b_u_f_f_e_r _m_i_s_s_e_d _- This can occur in an scanner which is reentered after a long-jump has jumped out (or over) the scanner's activa- tion frame. Before reentering the scanner, use: yyrestart( yyin ); or, as noted above, switch to using the C++ scanner class. _t_o_o _m_a_n_y _s_t_a_r_t _c_o_n_d_i_t_i_o_n_s _i_n _<_> _c_o_n_s_t_r_u_c_t_! _- you listed more start conditions in a <> construct than exist (so you must have listed at least one of them twice). FFIILLEESS See flex(1). DDEEFFIICCIIEENNCCIIEESS // BBUUGGSS Again, see flex(1). SSEEEE AALLSSOO flex(1), lex(1), yacc(1), sed(1), awk(1). M. E. Lesk and E. Schmidt, _L_E_X _- _L_e_x_i_c_a_l _A_n_a_l_y_z_e_r _G_e_n_e_r_a_- _t_o_r AAUUTTHHOORR Vern Paxson, with the help of many ideas and much inspira- tion from Van Jacobson. Original version by Jef Poskanzer. The fast table representation is a partial implementation of a design done by Van Jacobson. The implementation was done by Kevin Gong and Vern Paxson. Thanks to the many _f_l_e_x beta-testers, feedbackers, and contributors, especially Francois Pinard, Casey Leedom, Version 2.4 November 1993 41 FLEXDOC(1) FLEXDOC(1) Nelson H.F. Beebe, benson@odi.com, Peter A. Bigot, Keith Bostic, Frederic Brehm, Nick Christopher, Jason Coughlin, Bill Cox, Dave Curtis, Scott David Daniels, Chris G. Demetriou, Mike Donahue, Chuck Doucette, Tom Epperly, Leo Eskin, Chris Faylor, Jon Forrest, Kaveh R. Ghazi, Eric Goldman, Ulrich Grepel, Jan Hajic, Jarkko Hietaniemi, Eric Hughes, John Interrante, Ceriel Jacobs, Jeffrey R. Jones, Henry Juengst, Amir Katz, ken@ken.hilco.com, Kevin B. Kenny, Marq Kole, Ronald Lamprecht, Greg Lee, Craig Leres, John Levine, Mohamed el Lozy, Chris Metcalf, Luke Mewburn, Jim Meyering, G.T. Nicol, Landon Noll, Marc Nozell, Richard Ohnemus, Sven Panne, Roland Pesch, Walter Pelis- sero, Gaumond Pierre, Esmond Pitt, Jef Poskanzer, Joe Rah- meh, Kevin Rodgers, Jim Roskind, Doug Schmidt, Alex Siegel, Paul Stuart, Dave Tallman, Paul Tuinenga, Gary Weik, Frank Whaley, Gerhard Wilhelms, Kent Williams, Ken Yap, Nathan Zelle, David Zuhn, and those whose names have slipped my marginal mail-archiving skills but whose con- tributions are appreciated all the same. Thanks to Keith Bostic, Noah Friedman, John Gilmore, Craig Leres, Bob Mulcahy, G.T. Nicol, Francois Pinard, Rich Salz, and Richard Stallman for help with various distribu- tion headaches. Thanks to Esmond Pitt and Earle Horton for 8-bit character support; to Benson Margulies and Fred Burke for C++ sup- port; to Kent Williams and Tom Epperly for C++ class sup- port; to Ove Ewerlid for support of NUL's; and to Eric Hughes for support of multiple buffers. This work was primarily done when I was with the Real Time Systems Group at the Lawrence Berkeley Laboratory in Berkeley, CA. Many thanks to all there for the support I received. Send comments to: Vern Paxson Systems Engineering Bldg. 46A, Room 1123 Lawrence Berkeley Laboratory University of California Berkeley, CA 94720 vern@ee.lbl.gov Version 2.4 November 1993 42 .