/* flex - tool to generate fast lexical analyzers */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ #ifndef lint char copyright[] = "@(#) Copyright (c) 1990 The Regents of the University of California.\n\ All rights reserved.\n"; #endif /* not lint */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ //#include "flexdef.h" /* flexdef - definitions file for flex */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* @(#) $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flexdef.h,v 1.1.1.1 2003/06/02 17:12:00 expsir Exp $ (LBL) */ #include #include #if HAVE_STRING_H #include #else #include #endif #if __STDC__ #include #endif /* Always be prepared to generate an 8-bit scanner. */ #define CSIZE 256 #define Char unsigned char /* Size of input alphabet - should be size of ASCII set. */ #ifndef DEFAULT_CSIZE #define DEFAULT_CSIZE 128 #endif #ifndef PROTO #ifdef __STDC__ #define PROTO(proto) proto #else #define PROTO(proto) () #endif #endif #ifdef VMS #define unlink delete #define SHORT_FILE_NAMES #endif #ifdef MS_DOS #define SHORT_FILE_NAMES #endif /* Maximum line length we'll have to deal with. */ #define MAXLINE 2048 #ifndef MIN #define MIN(x,y) ((x) < (y) ? (x) : (y)) #endif #ifndef MAX #define MAX(x,y) ((x) > (y) ? (x) : (y)) #endif #ifndef ABS #define ABS(x) ((x) < 0 ? -(x) : (x)) #endif /* ANSI C does not guarantee that isascii() is defined */ #ifndef isascii #define isascii(c) ((c) <= 0177) #endif #define true 1 #define false 0 /* Special chk[] values marking the slots taking by end-of-buffer and action * numbers. */ #define EOB_POSITION -1 #define ACTION_POSITION -2 /* Number of data items per line for -f output. */ #define NUMDATAITEMS 10 /* Number of lines of data in -f output before inserting a blank line for * readability. */ #define NUMDATALINES 10 /* Transition_struct_out() definitions. */ #define TRANS_STRUCT_PRINT_LENGTH 15 /* Returns true if an nfa state has an epsilon out-transition slot * that can be used. This definition is currently not used. */ #define FREE_EPSILON(state) \ (transchar[state] == SYM_EPSILON && \ trans2[state] == NO_TRANSITION && \ finalst[state] != state) /* Returns true if an nfa state has an epsilon out-transition character * and both slots are free */ #define SUPER_FREE_EPSILON(state) \ (transchar[state] == SYM_EPSILON && \ trans1[state] == NO_TRANSITION) \ /* Maximum number of NFA states that can comprise a DFA state. It's real * big because if there's a lot of rules, the initial state will have a * huge epsilon closure. */ #define INITIAL_MAX_DFA_SIZE 750 #define MAX_DFA_SIZE_INCREMENT 750 /* A note on the following masks. They are used to mark accepting numbers * as being special. As such, they implicitly limit the number of accepting * numbers (i.e., rules) because if there are too many rules the rule numbers * will overload the mask bits. Fortunately, this limit is \large/ (0x2000 == * 8192) so unlikely to actually cause any problems. A check is made in * new_rule() to ensure that this limit is not reached. */ /* Mask to mark a trailing context accepting number. */ #define YY_TRAILING_MASK 0x2000 /* Mask to mark the accepting number of the "head" of a trailing context * rule. */ #define YY_TRAILING_HEAD_MASK 0x4000 /* Maximum number of rules, as outlined in the above note. */ #define MAX_RULE (YY_TRAILING_MASK - 1) /* NIL must be 0. If not, its special meaning when making equivalence classes * (it marks the representative of a given e.c.) will be unidentifiable. */ #define NIL 0 #define JAM -1 /* to mark a missing DFA transition */ #define NO_TRANSITION NIL #define UNIQUE -1 /* marks a symbol as an e.c. representative */ #define INFINITY -1 /* for x{5,} constructions */ #define INITIAL_MAX_CCLS 100 /* max number of unique character classes */ #define MAX_CCLS_INCREMENT 100 /* Size of table holding members of character classes. */ #define INITIAL_MAX_CCL_TBL_SIZE 500 #define MAX_CCL_TBL_SIZE_INCREMENT 250 #define INITIAL_MAX_RULES 100 /* default maximum number of rules */ #define MAX_RULES_INCREMENT 100 #define INITIAL_MNS 2000 /* default maximum number of nfa states */ #define MNS_INCREMENT 1000 /* amount to bump above by if it's not enough */ #define INITIAL_MAX_DFAS 1000 /* default maximum number of dfa states */ #define MAX_DFAS_INCREMENT 1000 #define JAMSTATE -32766 /* marks a reference to the state that always jams */ /* Enough so that if it's subtracted from an NFA state number, the result * is guaranteed to be negative. */ #define MARKER_DIFFERENCE 32000 #define MAXIMUM_MNS 31999 /* Maximum number of nxt/chk pairs for non-templates. */ #define INITIAL_MAX_XPAIRS 2000 #define MAX_XPAIRS_INCREMENT 2000 /* Maximum number of nxt/chk pairs needed for templates. */ #define INITIAL_MAX_TEMPLATE_XPAIRS 2500 #define MAX_TEMPLATE_XPAIRS_INCREMENT 2500 #define SYM_EPSILON (CSIZE + 1) /* to mark transitions on the symbol epsilon */ #define INITIAL_MAX_SCS 40 /* maximum number of start conditions */ #define MAX_SCS_INCREMENT 40 /* amount to bump by if it's not enough */ #define ONE_STACK_SIZE 500 /* stack of states with only one out-transition */ #define SAME_TRANS -1 /* transition is the same as "default" entry for state */ /* The following percentages are used to tune table compression: * The percentage the number of out-transitions a state must be of the * number of equivalence classes in order to be considered for table * compaction by using protos. */ #define PROTO_SIZE_PERCENTAGE 15 /* The percentage the number of homogeneous out-transitions of a state * must be of the number of total out-transitions of the state in order * that the state's transition table is first compared with a potential * template of the most common out-transition instead of with the first * proto in the proto queue. */ #define CHECK_COM_PERCENTAGE 50 /* The percentage the number of differences between a state's transition * table and the proto it was first compared with must be of the total * number of out-transitions of the state in order to keep the first * proto as a good match and not search any further. */ #define FIRST_MATCH_DIFF_PERCENTAGE 10 /* The percentage the number of differences between a state's transition * table and the most similar proto must be of the state's total number * of out-transitions to use the proto as an acceptable close match. */ #define ACCEPTABLE_DIFF_PERCENTAGE 50 /* The percentage the number of homogeneous out-transitions of a state * must be of the number of total out-transitions of the state in order * to consider making a template from the state. */ #define TEMPLATE_SAME_PERCENTAGE 60 /* The percentage the number of differences between a state's transition * table and the most similar proto must be of the state's total number * of out-transitions to create a new proto from the state. */ #define NEW_PROTO_DIFF_PERCENTAGE 20 /* The percentage the total number of out-transitions of a state must be * of the number of equivalence classes in order to consider trying to * fit the transition table into "holes" inside the nxt/chk table. */ #define INTERIOR_FIT_PERCENTAGE 15 /* Size of region set aside to cache the complete transition table of * protos on the proto queue to enable quick comparisons. */ #define PROT_SAVE_SIZE 2000 #define MSP 50 /* maximum number of saved protos (protos on the proto queue) */ /* Maximum number of out-transitions a state can have that we'll rummage * around through the interior of the internal fast table looking for a * spot for it. */ #define MAX_XTIONS_FULL_INTERIOR_FIT 4 /* Maximum number of rules which will be reported as being associated * with a DFA state. */ #define MAX_ASSOC_RULES 100 /* Number that, if used to subscript an array, has a good chance of producing * an error; should be small enough to fit into a short. */ #define BAD_SUBSCRIPT -32767 /* Absolute value of largest number that can be stored in a short, with a * bit of slop thrown in for general paranoia. */ #define MAX_SHORT 32700 /* Declarations for global variables. */ /* Variables for symbol tables: * sctbl - start-condition symbol table * ndtbl - name-definition symbol table * ccltab - character class text symbol table */ struct hash_entry { struct hash_entry *prev, *next; char *name; char *str_val; int int_val; } ; typedef struct hash_entry **hash_table; #define NAME_TABLE_HASH_SIZE 101 #define START_COND_HASH_SIZE 101 #define CCL_HASH_SIZE 101 extern struct hash_entry *ndtbl[NAME_TABLE_HASH_SIZE]; extern struct hash_entry *sctbl[START_COND_HASH_SIZE]; extern struct hash_entry *ccltab[CCL_HASH_SIZE]; /* Variables for flags: * printstats - if true (-v), dump statistics * syntaxerror - true if a syntax error has been found * eofseen - true if we've seen an eof in the input file * ddebug - if true (-d), make a "debug" scanner * trace - if true (-T), trace processing * nowarn - if true (-w), do not generate warnings * spprdflt - if true (-s), suppress the default rule * interactive - if true (-I), generate an interactive scanner * caseins - if true (-i), generate a case-insensitive scanner * lex_compat - if true (-l), maximize compatibility with AT&T lex * useecs - if true (-Ce flag), use equivalence classes * fulltbl - if true (-Cf flag), don't compress the DFA state table * usemecs - if true (-Cm flag), use meta-equivalence classes * fullspd - if true (-F flag), use Jacobson method of table representation * gen_line_dirs - if true (i.e., no -L flag), generate #line directives * performance_report - if > 0 (i.e., -p flag), generate a report relating * to scanner performance; if > 1 (-p -p), report on minor performance * problems, too * backing_up_report - if true (i.e., -b flag), generate "lex.backup" file * listing backing-up states * C_plus_plus - if true (i.e., -+ flag), generate a C++ scanner class; * otherwise, a standard C scanner * long_align - if true (-Ca flag), favor long-word alignment. * use_read - if true (-f, -F, or -Cr) then use read() for scanner input; * otherwise, use fread(). * yytext_is_array - if true (i.e., %array directive), then declare * yytext as a array instead of a character pointer. Nice and inefficient. * csize - size of character set for the scanner we're generating; * 128 for 7-bit chars and 256 for 8-bit * yymore_used - if true, yymore() is used in input rules * reject - if true, generate back-up tables for REJECT macro * real_reject - if true, scanner really uses REJECT (as opposed to just * having "reject" set for variable trailing context) * continued_action - true if this rule's action is to "fall through" to * the next rule's action (i.e., the '|' action) * yymore_really_used - has a REALLY_xxx value indicating whether a * %used or %notused was used with yymore() * reject_really_used - same for REJECT */ extern int printstats, syntaxerror, eofseen, ddebug, trace, nowarn, spprdflt; extern int interactive, caseins, lex_compat, useecs, fulltbl, usemecs; extern int fullspd, gen_line_dirs, performance_report, backing_up_report; extern int C_plus_plus, long_align, use_read, yytext_is_array, csize; extern int yymore_used, reject, real_reject, continued_action; #define REALLY_NOT_DETERMINED 0 #define REALLY_USED 1 #define REALLY_NOT_USED 2 extern int yymore_really_used, reject_really_used; /* Variables used in the flex input routines: * datapos - characters on current output line * dataline - number of contiguous lines of data in current data * statement. Used to generate readable -f output * linenum - current input line number * skelfile - the skeleton file * skel - compiled-in skeleton array * skel_ind - index into "skel" array, if skelfile is nil * yyin - input file * backing_up_file - file to summarize backing-up states to * infilename - name of input file * input_files - array holding names of input files * num_input_files - size of input_files array * program_name - name with which program was invoked * * action_array - array to hold the rule actions * action_size - size of action_array * defs1_offset - index where the user's section 1 definitions start * in action_array * prolog_offset - index where the prolog starts in action_array * action_offset - index where the non-prolog starts in action_array * action_index - index where the next action should go, with respect * to "action_array" */ extern int datapos, dataline, linenum; extern FILE *skelfile, *yyin, *backing_up_file; extern char *skel[]; extern int skel_ind; extern char *infilename; extern char **input_files; extern int num_input_files; extern char *program_name; extern char *action_array; extern int action_size; extern int defs1_offset, prolog_offset, action_offset, action_index; /* Variables for stack of states having only one out-transition: * onestate - state number * onesym - transition symbol * onenext - target state * onedef - default base entry * onesp - stack pointer */ extern int onestate[ONE_STACK_SIZE], onesym[ONE_STACK_SIZE]; extern int onenext[ONE_STACK_SIZE], onedef[ONE_STACK_SIZE], onesp; /* Variables for nfa machine data: * current_mns - current maximum on number of NFA states * num_rules - number of the last accepting state; also is number of * rules created so far * num_eof_rules - number of <> rules * default_rule - number of the default rule * current_max_rules - current maximum number of rules * lastnfa - last nfa state number created * firstst - physically the first state of a fragment * lastst - last physical state of fragment * finalst - last logical state of fragment * transchar - transition character * trans1 - transition state * trans2 - 2nd transition state for epsilons * accptnum - accepting number * assoc_rule - rule associated with this NFA state (or 0 if none) * state_type - a STATE_xxx type identifying whether the state is part * of a normal rule, the leading state in a trailing context * rule (i.e., the state which marks the transition from * recognizing the text-to-be-matched to the beginning of * the trailing context), or a subsequent state in a trailing * context rule * rule_type - a RULE_xxx type identifying whether this a ho-hum * normal rule or one which has variable head & trailing * context * rule_linenum - line number associated with rule * rule_useful - true if we've determined that the rule can be matched */ extern int current_mns, num_rules, num_eof_rules, default_rule; extern int current_max_rules, lastnfa; extern int *firstst, *lastst, *finalst, *transchar, *trans1, *trans2; extern int *accptnum, *assoc_rule, *state_type; extern int *rule_type, *rule_linenum, *rule_useful; /* Different types of states; values are useful as masks, as well, for * routines like check_trailing_context(). */ #define STATE_NORMAL 0x1 #define STATE_TRAILING_CONTEXT 0x2 /* Global holding current type of state we're making. */ extern int current_state_type; /* Different types of rules. */ #define RULE_NORMAL 0 #define RULE_VARIABLE 1 /* True if the input rules include a rule with both variable-length head * and trailing context, false otherwise. */ extern int variable_trailing_context_rules; /* Variables for protos: * numtemps - number of templates created * numprots - number of protos created * protprev - backlink to a more-recently used proto * protnext - forward link to a less-recently used proto * prottbl - base/def table entry for proto * protcomst - common state of proto * firstprot - number of the most recently used proto * lastprot - number of the least recently used proto * protsave contains the entire state array for protos */ extern int numtemps, numprots, protprev[MSP], protnext[MSP], prottbl[MSP]; extern int protcomst[MSP], firstprot, lastprot, protsave[PROT_SAVE_SIZE]; /* Variables for managing equivalence classes: * numecs - number of equivalence classes * nextecm - forward link of Equivalence Class members * ecgroup - class number or backward link of EC members * nummecs - number of meta-equivalence classes (used to compress * templates) * tecfwd - forward link of meta-equivalence classes members * tecbck - backward link of MEC's */ /* Reserve enough room in the equivalence class arrays so that we * can use the CSIZE'th element to hold equivalence class information * for the NUL character. Later we'll move this information into * the 0th element. */ extern int numecs, nextecm[CSIZE + 1], ecgroup[CSIZE + 1], nummecs; /* Meta-equivalence classes are indexed starting at 1, so it's possible * that they will require positions from 1 .. CSIZE, i.e., CSIZE + 1 * slots total (since the arrays are 0-based). nextecm[] and ecgroup[] * don't require the extra position since they're indexed from 1 .. CSIZE - 1. */ extern int tecfwd[CSIZE + 1], tecbck[CSIZE + 1]; /* Variables for start conditions: * lastsc - last start condition created * current_max_scs - current limit on number of start conditions * scset - set of rules active in start condition * scbol - set of rules active only at the beginning of line in a s.c. * scxclu - true if start condition is exclusive * sceof - true if start condition has EOF rule * scname - start condition name * actvsc - stack of active start conditions for the current rule; * a negative entry means that the start condition is *not* * active for the current rule. Start conditions may appear * multiple times on the stack; the entry for it closest * to the top of the stack (i.e., actvsc[actvp]) is the * one to use. Others are present from "{" scoping * constructs. */ extern int lastsc, current_max_scs, *scset, *scbol, *scxclu, *sceof, *actvsc; extern char **scname; /* Variables for dfa machine data: * current_max_dfa_size - current maximum number of NFA states in DFA * current_max_xpairs - current maximum number of non-template xtion pairs * current_max_template_xpairs - current maximum number of template pairs * current_max_dfas - current maximum number DFA states * lastdfa - last dfa state number created * nxt - state to enter upon reading character * chk - check value to see if "nxt" applies * tnxt - internal nxt table for templates * base - offset into "nxt" for given state * def - where to go if "chk" disallows "nxt" entry * nultrans - NUL transition for each state * NUL_ec - equivalence class of the NUL character * tblend - last "nxt/chk" table entry being used * firstfree - first empty entry in "nxt/chk" table * dss - nfa state set for each dfa * dfasiz - size of nfa state set for each dfa * dfaacc - accepting set for each dfa state (if using REJECT), or accepting * number, if not * accsiz - size of accepting set for each dfa state * dhash - dfa state hash value * numas - number of DFA accepting states created; note that this * is not necessarily the same value as num_rules, which is the analogous * value for the NFA * numsnpairs - number of state/nextstate transition pairs * jambase - position in base/def where the default jam table starts * jamstate - state number corresponding to "jam" state * end_of_buffer_state - end-of-buffer dfa state number */ extern int current_max_dfa_size, current_max_xpairs; extern int current_max_template_xpairs, current_max_dfas; extern int lastdfa, *nxt, *chk, *tnxt; extern int *base, *def, *nultrans, NUL_ec, tblend, firstfree, **dss, *dfasiz; extern union dfaacc_union { int *dfaacc_set; int dfaacc_state; } *dfaacc; extern int *accsiz, *dhash, numas; extern int numsnpairs, jambase, jamstate; extern int end_of_buffer_state; /* Variables for ccl information: * lastccl - ccl index of the last created ccl * current_maxccls - current limit on the maximum number of unique ccl's * cclmap - maps a ccl index to its set pointer * ccllen - gives the length of a ccl * cclng - true for a given ccl if the ccl is negated * cclreuse - counts how many times a ccl is re-used * current_max_ccl_tbl_size - current limit on number of characters needed * to represent the unique ccl's * ccltbl - holds the characters in each ccl - indexed by cclmap */ extern int lastccl, current_maxccls, *cclmap, *ccllen, *cclng, cclreuse; extern int current_max_ccl_tbl_size; extern Char *ccltbl; /* Variables for miscellaneous information: * nmstr - last NAME scanned by the scanner * sectnum - section number currently being parsed * nummt - number of empty nxt/chk table entries * hshcol - number of hash collisions detected by snstods * dfaeql - number of times a newly created dfa was equal to an old one * numeps - number of epsilon NFA states created * eps2 - number of epsilon states which have 2 out-transitions * num_reallocs - number of times it was necessary to realloc() a group * of arrays * tmpuses - number of DFA states that chain to templates * totnst - total number of NFA states used to make DFA states * peakpairs - peak number of transition pairs we had to store internally * numuniq - number of unique transitions * numdup - number of duplicate transitions * hshsave - number of hash collisions saved by checking number of states * num_backing_up - number of DFA states requiring backing up * bol_needed - whether scanner needs beginning-of-line recognition */ extern char nmstr[MAXLINE]; extern int sectnum, nummt, hshcol, dfaeql, numeps, eps2, num_reallocs; extern int tmpuses, totnst, peakpairs, numuniq, numdup, hshsave; extern int num_backing_up, bol_needed; void *allocate_array PROTO((int, int)); void *reallocate_array PROTO((void*, int, int)); void *flex_alloc PROTO((unsigned int)); void *flex_realloc PROTO((void*, unsigned int)); void flex_free PROTO((void*)); #define allocate_integer_array(size) \ (int *) allocate_array( size, sizeof( int ) ) #define reallocate_integer_array(array,size) \ (int *) reallocate_array( (void *) array, size, sizeof( int ) ) #define allocate_int_ptr_array(size) \ (int **) allocate_array( size, sizeof( int * ) ) #define allocate_char_ptr_array(size) \ (char **) allocate_array( size, sizeof( char * ) ) #define allocate_dfaacc_union(size) \ (union dfaacc_union *) \ allocate_array( size, sizeof( union dfaacc_union ) ) #define reallocate_int_ptr_array(array,size) \ (int **) reallocate_array( (void *) array, size, sizeof( int * ) ) #define reallocate_char_ptr_array(array,size) \ (char **) reallocate_array( (void *) array, size, sizeof( char * ) ) #define reallocate_dfaacc_union(array, size) \ (union dfaacc_union *) \ reallocate_array( (void *) array, size, sizeof( union dfaacc_union ) ) #define allocate_character_array(size) \ (char *) allocate_array( size, sizeof( char ) ) #define reallocate_character_array(array,size) \ (char *) reallocate_array( (void *) array, size, sizeof( char ) ) #define allocate_Character_array(size) \ (Char *) allocate_array( size, sizeof( Char ) ) #define reallocate_Character_array(array,size) \ (Char *) reallocate_array( (void *) array, size, sizeof( Char ) ) /* Used to communicate between scanner and parser. The type should really * be YYSTYPE, but we can't easily get our hands on it. */ extern int yylval; /* External functions that are cross-referenced among the flex source files. */ /* from file ccl.c */ extern void ccladd PROTO((int, int)); /* add a single character to a ccl */ extern int cclinit PROTO((void)); /* make an empty ccl */ extern void cclnegate PROTO((int)); /* negate a ccl */ /* List the members of a set of characters in CCL form. */ extern void list_character_set PROTO((FILE*, int[])); /* from file dfa.c */ /* Increase the maximum number of dfas. */ extern void increase_max_dfas PROTO((void)); extern void ntod PROTO((void)); /* convert a ndfa to a dfa */ /* from file ecs.c */ /* Convert character classes to set of equivalence classes. */ extern void ccl2ecl PROTO((void)); /* Associate equivalence class numbers with class members. */ extern int cre8ecs PROTO((int[], int[], int)); /* Update equivalence classes based on character class transitions. */ extern void mkeccl PROTO((Char[], int, int[], int[], int, int)); /* Create equivalence class for single character. */ extern void mkechar PROTO((int, int[], int[])); /* from file gen.c */ extern void make_tables PROTO((void)); /* generate transition tables */ /* from file main.c */ extern void flexend PROTO((int)); extern void usage PROTO((void)); /* from file misc.c */ /* Add the given text to the stored actions. */ extern void add_action PROTO(( char *new_text )); /* True if a string is all lower case. */ extern int all_lower PROTO((register char *)); /* True if a string is all upper case. */ extern int all_upper PROTO((register char *)); /* Bubble sort an integer array. */ extern void bubble PROTO((int [], int)); /* Check a character to make sure it's in the expected range. */ extern void check_char PROTO((int c)); /* Shell sort a character array. */ extern void cshell PROTO((Char [], int, int)); /* Finish up a block of data declarations. */ extern void dataend PROTO((void)); /* Report an error message and terminate. */ extern void flexerror PROTO((char[])); /* Report a fatal error message and terminate. */ extern void flexfatal PROTO((char[])); /* Report an error message formatted with one integer argument. */ extern void lerrif PROTO((char[], int)); /* Report an error message formatted with one string argument. */ extern void lerrsf PROTO((char[], char[])); /* Spit out a "# line" statement. */ extern void line_directive_out PROTO((FILE*)); /* Mark the current position in the action array as the end of the section 1 * user defs. */ extern void mark_defs1 PROTO((void)); /* Mark the current position in the action array as the end of the prolog. */ extern void mark_prolog PROTO((void)); /* Generate a data statment for a two-dimensional array. */ extern void mk2data PROTO((int)); extern void mkdata PROTO((int)); /* generate a data statement */ /* Return the integer represented by a string of digits. */ extern int myctoi PROTO((char [])); /* Return a printable version of the given character, which might be * 8-bit. */ extern char *readable_form PROTO((int)); /* Write out one section of the skeleton file. */ extern void skelout PROTO((void)); /* Output a yy_trans_info structure. */ extern void transition_struct_out PROTO((int, int)); /* Only needed when using certain broken versions of bison to build parse.c. */ extern void *yy_flex_xmalloc PROTO(( int )); /* Set a region of memory to 0. */ extern void zero_out PROTO((char *, int)); /* from file nfa.c */ /* Add an accepting state to a machine. */ extern void add_accept PROTO((int, int)); /* Make a given number of copies of a singleton machine. */ extern int copysingl PROTO((int, int)); /* Debugging routine to write out an nfa. */ extern void dumpnfa PROTO((int)); /* Finish up the processing for a rule. */ extern void finish_rule PROTO((int, int, int, int)); /* Connect two machines together. */ extern int link_machines PROTO((int, int)); /* Mark each "beginning" state in a machine as being a "normal" (i.e., * not trailing context associated) state. */ extern void mark_beginning_as_normal PROTO((register int)); /* Make a machine that branches to two machines. */ extern int mkbranch PROTO((int, int)); extern int mkclos PROTO((int)); /* convert a machine into a closure */ extern int mkopt PROTO((int)); /* make a machine optional */ /* Make a machine that matches either one of two machines. */ extern int mkor PROTO((int, int)); /* Convert a machine into a positive closure. */ extern int mkposcl PROTO((int)); extern int mkrep PROTO((int, int, int)); /* make a replicated machine */ /* Create a state with a transition on a given symbol. */ extern int mkstate PROTO((int)); extern void new_rule PROTO((void)); /* initialize for a new rule */ /* from file parse.y */ /* Write out a message formatted with one string, pinpointing its location. */ extern void format_pinpoint_message PROTO((char[], char[])); /* Write out a message, pinpointing its location. */ extern void pinpoint_message PROTO((char[])); /* Write out a warning, pinpointing it at the given line. */ void line_warning PROTO(( char[], int )); /* Write out a message, pinpointing it at the given line. */ void line_pinpoint PROTO(( char[], int )); /* Report a formatted syntax error. */ extern void format_synerr PROTO((char [], char[])); extern void synerr PROTO((char [])); /* report a syntax error */ extern void warn PROTO((char [])); /* report a warning */ extern int yyparse PROTO((void)); /* the YACC parser */ /* from file scan.l */ /* The Flex-generated scanner for flex. */ extern int flexscan PROTO((void)); /* Open the given file (if NULL, stdin) for scanning. */ extern void set_input_file PROTO((char*)); /* Wrapup a file in the lexical analyzer. */ extern int yywrap PROTO((void)); /* from file sym.c */ /* Save the text of a character class. */ extern void cclinstal PROTO ((Char [], int)); /* Lookup the number associated with character class. */ extern int ccllookup PROTO((Char [])); extern void ndinstal PROTO((char[], Char[])); /* install a name definition */ /* Increase maximum number of SC's. */ extern void scextend PROTO((void)); extern void scinstal PROTO((char[], int)); /* make a start condition */ /* Lookup the number associated with a start condition. */ extern int sclookup PROTO((char[])); /* from file tblcmp.c */ /* Build table entries for dfa state. */ extern void bldtbl PROTO((int[], int, int, int, int)); extern void cmptmps PROTO((void)); /* compress template table entries */ extern void expand_nxt_chk PROTO((void)); /* increase nxt/chk arrays */ extern void inittbl PROTO((void)); /* initialize transition tables */ /* Make the default, "jam" table entries. */ extern void mkdeftbl PROTO((void)); /* Create table entries for a state (or state fragment) which has * only one out-transition. */ extern void mk1tbl PROTO((int, int, int, int)); /* Place a state into full speed transition table. */ extern void place_state PROTO((int*, int, int)); /* Save states with only one out-transition to be processed later. */ extern void stack1 PROTO((int, int, int, int)); /* from file yylex.c */ extern int yylex PROTO((void)); //#include "version.h" #define FLEX_VERSION "2.4.7" #include //#include "parse.h" #ifndef YYSTYPE #define YYSTYPE int #endif #define CHAR 257 #define NUMBER 258 #define SECTEND 259 #define SCDECL 260 #define XSCDECL 261 #define WHITESPACE 262 #define NAME 263 #define PREVCCL 264 #define EOF_OP 265 extern YYSTYPE yylval; static char flex_version[] = FLEX_VERSION; /* declare functions that have forward references */ void flexinit PROTO((int, char**)); void readin PROTO((void)); void set_up_initial_allocations PROTO((void)); /* these globals are all defined and commented in flexdef.h */ int printstats, syntaxerror, eofseen, ddebug, trace, nowarn, spprdflt; int interactive, caseins, lex_compat, useecs, fulltbl, usemecs; int fullspd, gen_line_dirs, performance_report, backing_up_report; int C_plus_plus, long_align, use_read, yytext_is_array, csize; int yymore_used, reject, real_reject, continued_action; int yymore_really_used, reject_really_used; int datapos, dataline, linenum; FILE *skelfile = NULL; /* ### add for testing ### */ FILE *err = NULL; /* ####################### */ int skel_ind = 0; char *action_array; int action_size, defs1_offset, prolog_offset, action_offset, action_index; char *infilename = NULL; int onestate[ONE_STACK_SIZE], onesym[ONE_STACK_SIZE]; int onenext[ONE_STACK_SIZE], onedef[ONE_STACK_SIZE], onesp; int current_mns, num_rules, num_eof_rules, default_rule; int current_max_rules, lastnfa; int *firstst, *lastst, *finalst, *transchar, *trans1, *trans2; int *accptnum, *assoc_rule, *state_type; int *rule_type, *rule_linenum, *rule_useful; int current_state_type; int variable_trailing_context_rules; int numtemps, numprots, protprev[MSP], protnext[MSP], prottbl[MSP]; int protcomst[MSP], firstprot, lastprot, protsave[PROT_SAVE_SIZE]; int numecs, nextecm[CSIZE + 1], ecgroup[CSIZE + 1], nummecs, tecfwd[CSIZE + 1]; int tecbck[CSIZE + 1]; int lastsc, current_max_scs, *scset, *scbol, *scxclu, *sceof, *actvsc; char **scname; int current_max_dfa_size, current_max_xpairs; int current_max_template_xpairs, current_max_dfas; int lastdfa, *nxt, *chk, *tnxt; int *base, *def, *nultrans, NUL_ec, tblend, firstfree, **dss, *dfasiz; union dfaacc_union *dfaacc; int *accsiz, *dhash, numas; int numsnpairs, jambase, jamstate; int lastccl, current_maxccls, *cclmap, *ccllen, *cclng, cclreuse; int current_max_ccl_tbl_size; Char *ccltbl; char nmstr[MAXLINE]; int sectnum, nummt, hshcol, dfaeql, numeps, eps2, num_reallocs; int tmpuses, totnst, peakpairs, numuniq, numdup, hshsave; int num_backing_up, bol_needed; FILE *backing_up_file; int end_of_buffer_state; char **input_files; int num_input_files; char *program_name; #ifndef SHORT_FILE_NAMES static char *outfile_template = "lex.%s.%s"; #else static char *outfile_template = "lex%s.%s"; #endif static char outfile_path[64]; static int outfile_created = 0; static int use_stdout; static char *skelname = NULL; static char *prefix = "yy"; int main( ) { int argc; char **argv,*shuru[4]; int i,result; shuru[0]="good"; shuru[1]="luck"; shuru[2]="-++cto"; shuru[3]="you"; argv=&shuru; argc=3; /* ### add for testing ### */ if(--argc >0) { err = fopen( *++argv, "w" ); --argv; ++argc; } /* ####################### */ //fun1 flexinit( argc, argv ); return 0; /* keep compilers/lint happy */ } //fun1 /* flexinit - initialize flex */ void flexinit( argc, argv ) int argc; char **argv; { int i, sawcmpflag; int csize_given, interactive_given; char *arg, *mktemp(); printstats = syntaxerror = trace = spprdflt = caseins = false; lex_compat = false; C_plus_plus = backing_up_report = ddebug = fulltbl = fullspd = false; long_align = nowarn = yymore_used = continued_action = reject = false; yytext_is_array = yymore_really_used = reject_really_used = false; gen_line_dirs = usemecs = useecs = true; performance_report = 0; sawcmpflag = false; use_read = use_stdout = false; csize_given = false; interactive_given = false; /* Initialize dynamic array for holding the rule actions. */ action_size = 2048; /* default size of action array in bytes */ action_array = allocate_character_array( action_size ); defs1_offset = prolog_offset = action_offset = action_index = 0; action_array[0] = '\0'; /* Commented by Amit Goel on 5th September 2001 to remove nomdeterminism */ /* program_name = argv[0]; */ /* Added by Amit Goel on 5th September 2001 */ program_name = "Executable"; if ( program_name[0] != '\0' && program_name[strlen( program_name ) - 1] == '+' ) C_plus_plus = true; /* ### add for testing ### */ ++argv;--argc; /* ####################### */ /* read flags */ for ( --argc, ++argv; argc ; --argc, ++argv ) { if ( argv[0][0] != '-' || argv[0][1] == '\0' ) break; arg = argv[0]; for ( i = 1; arg[i] != '\0'; ++i ) switch ( arg[i] ) { case '+': C_plus_plus = true; break; case 'B': interactive = false; interactive_given = true; break; case 'b': backing_up_report = true; break; case 'c': fprintf( err,"%s: Assuming use of deprecated -c flag is really intended to be -C\n", program_name ); /* fall through */ case 'C': if ( i != 1 ) flexerror("-C flag must be given separately" ); if ( ! sawcmpflag ) { useecs = false; usemecs = false; fulltbl = false; sawcmpflag = true; } for ( ++i; arg[i] != '\0'; ++i ) switch ( arg[i] ) { case 'a': long_align =true; break; case 'e': useecs = true; break; case 'F': fullspd = true; break; case 'f': fulltbl = true; break; case 'm': usemecs = true; break; case 'r': use_read = true; break; default: lerrif("unknown -C option '%c'",(int) arg[i]); //fun1-1 break; } goto get_next_arg; case 'd': ddebug = true; break; case 'f': useecs = usemecs = false; use_read = fulltbl = true; break; case 'F': useecs = usemecs = false; use_read = fullspd = true; break; case 'h': usage(); exit( 0 ); case 'I': interactive = true; interactive_given = true; break; case 'i': caseins = true; break; case 'l': lex_compat = true; break; case 'L': gen_line_dirs = false; break; case 'n': /* Stupid do-nothing deprecated * option. */ break; case 'P': if ( i != 1 ) flexerror( "-P flag must be given separately" ); prefix = arg + i + 1; goto get_next_arg; case 'p': ++performance_report; break; case 'S': if ( i != 1 ) flexerror( "-S flag must be given separately" ); skelname = arg + i + 1; goto get_next_arg; case 's': spprdflt = true; break; case 't': use_stdout = true; break; case 'T': trace = true; break; case 'v': printstats = true; break; case 'V': fprintf( err, "%s version %s\n", program_name, flex_version ); exit( 0 ); case 'w': nowarn = true; break; case '7': csize = 128; csize_given = true; break; case '8': csize = CSIZE; csize_given = true; break; default: fprintf( err, "%s: unknown flag '%c'\n", program_name, (int) arg[i] ); usage(); exit( 1 ); } /* Used by -C, -S and -P flags in lieu of a "continue 2" * control. */ get_next_arg: ; } if ( ! csize_given ) { if ( (fulltbl || fullspd) && ! useecs ) csize = DEFAULT_CSIZE; else csize = CSIZE; } if ( ! interactive_given ) { if ( fulltbl || fullspd ) interactive = false; else interactive = true; } if ( lex_compat ) { if ( C_plus_plus ) flexerror( "Can't use -+ with -l option" ); if ( fulltbl || fullspd ) flexerror( "Can't use -f or -F with -l option" ); /* Don't rely on detecting use of yymore() and REJECT, * just assume they'll be used. */ yymore_really_used = reject_really_used = true; yytext_is_array = true; use_read = false; } if ( (fulltbl || fullspd) && usemecs ) flexerror( "-Cf/-CF and -Cm don't make sense together" ); if ( (fulltbl || fullspd) && interactive ) flexerror( "-Cf/-CF and -I are incompatible" ); if ( fulltbl && fullspd ) flexerror( "-Cf and -CF are mutually exclusive" ); if ( C_plus_plus && fullspd ) flexerror( "Can't use -+ with -CF option" ); if ( ! use_stdout ) { FILE *prev_stdout; char *suffix; if ( C_plus_plus ) suffix = "cc"; else suffix = "c"; sprintf( outfile_path, outfile_template, prefix, suffix ); prev_stdout = freopen( outfile_path, "w", stdout ); if ( prev_stdout == NULL ) lerrsf( "could not create %s", outfile_path ); outfile_created = 1; } num_input_files = argc; input_files = argv; set_input_file( num_input_files > 0 ? input_files[0] : NULL ); if ( backing_up_report ) { #ifndef SHORT_FILE_NAMES backing_up_file = fopen( "lex.backup", "w" ); #else backing_up_file = fopen( "lex.bck", "w" ); #endif if ( backing_up_file == NULL ) flexerror( "could not create lex.backup" ); } else backing_up_file = NULL; lastccl = 0; lastsc = 0; if ( skelname && (skelfile = fopen( skelname, "r" )) == NULL ) lerrsf( "can't open skeleton file %s", skelname ); if ( strcmp( prefix, "yy" ) ) { #define GEN_PREFIX(name) printf( "#define yy%s %s%s\n", name, prefix, name ); GEN_PREFIX( "FlexLexer" ); GEN_PREFIX( "_create_buffer" ); GEN_PREFIX( "_delete_buffer" ); GEN_PREFIX( "_flex_debug" ); GEN_PREFIX( "_init_buffer" ); GEN_PREFIX( "_load_buffer_state" ); GEN_PREFIX( "_switch_to_buffer" ); GEN_PREFIX( "in" ); GEN_PREFIX( "leng" ); GEN_PREFIX( "lex" ); GEN_PREFIX( "out" ); GEN_PREFIX( "restart" ); GEN_PREFIX( "text" ); GEN_PREFIX( "wrap" ); printf( "\n" ); } lastdfa = lastnfa = 0; num_rules = num_eof_rules = default_rule = 0; numas = numsnpairs = tmpuses = 0; numecs = numeps = eps2 = num_reallocs = hshcol = dfaeql = totnst = 0; numuniq = numdup = hshsave = eofseen = datapos = dataline = 0; num_backing_up = onesp = numprots = 0; variable_trailing_context_rules = bol_needed = false; linenum = sectnum = 1; firstprot = NIL; /* Used in mkprot() so that the first proto goes in slot 1 * of the proto queue. */ lastprot = 1; if ( useecs ) { /* Set up doubly-linked equivalence classes. */ /* We loop all the way up to csize, since ecgroup[csize] is * the position used for NUL characters. */ ecgroup[1] = NIL; for ( i = 2; i <= csize; ++i ) { ecgroup[i] = i - 1; nextecm[i - 1] = i; } nextecm[csize] = NIL; } else { /* Put everything in its own equivalence class. */ for ( i = 1; i <= csize; ++i ) { ecgroup[i] = i; nextecm[i] = BAD_SUBSCRIPT; /* to catch errors */ } } set_up_initial_allocations(); } //fun1-1 /* lerrif - report an error message formatted with one integer argument */ void lerrif( msg, arg ) char msg[]; int arg; { char errmsg[MAXLINE]; (void) sprintf( errmsg, msg, arg ); flexerror( errmsg ); } //fun1-2 void usage() { fprintf( err, "%s [-bcdfhilnpstvwBFILTV78+ -C[aefFmr] -Pprefix -Sskeleton] [file ...]\n", program_name ); fprintf( err, "\t-b generate backing-up information to lex.backup\n" ); fprintf( err, "\t-c do-nothing POSIX option\n" ); fprintf( err, "\t-d turn on debug mode in generated scanner\n" ); fprintf( err, "\t-f generate fast, large scanner\n" ); fprintf( err, "\t-h produce this help message\n" ); fprintf( err, "\t-i generate case-insensitive scanner\n" ); fprintf( err, "\t-l maximal compatibility with original lex\n" ); fprintf( err, "\t-n do-nothing POSIX option\n" ); fprintf( err, "\t-p generate performance report to stderr\n" ); fprintf( err, "\t-s suppress default rule to ECHO unmatched text\n" ); fprintf( err, "\t-t write generated scanner on stdout instead of lex.yy.c\n" ); fprintf( err, "\t-v write summary of scanner statistics to stderr\n" ); fprintf( err, "\t-w do not generate warnings\n" ); fprintf( err, "\t-B generate batch scanner (opposite of -I)\n" ); fprintf( err, "\t-F use alternative fast scanner representation\n" ); fprintf( err, "\t-I generate interactive scanner (opposite of -B)\n" ); fprintf( err, "\t-L suppress #line directives in scanner\n" ); fprintf( err, "\t-T %s should run in trace mode\n", program_name ); fprintf( err, "\t-V report %s version\n", program_name ); fprintf( err, "\t-7 generate 7-bit scanner\n" ); fprintf( err, "\t-8 generate 8-bit scanner\n" ); fprintf( err, "\t-+ generate C++ scanner class\n" ); fprintf( err, "\t-C specify degree of table compression (default is -Cem):\n" ); fprintf( err, "\t\t-Ca trade off larger tables for better memory alignment\n" ); fprintf( err, "\t\t-Ce construct equivalence classes\n" ); fprintf( err, "\t\t-Cf do not compress scanner tables; use -f representation\n" ); fprintf( err, "\t\t-CF do not compress scanner tables; use -F representation\n" ); fprintf( err, "\t\t-Cm construct meta-equivalence classes\n" ); fprintf( err, "\t\t-Cr use read() instead of stdio for scanner input\n" ); fprintf( err, "\t-P specify scanner prefix other than \"yy\"\n" ); fprintf( err, "\t-S specify skeleton file\n" ); } //fun1-3 /* lerrsf - report an error message formatted with one string argument */ void lerrsf( msg, arg ) char msg[], arg[]; { char errmsg[MAXLINE]; (void) sprintf( errmsg, msg, arg ); flexerror( errmsg ); } //fun1-4 /* set_input_file - open the given file (if NULL, stdin) for scanning */ void set_input_file( file ) char *file; { if ( file ) { infilename = file; yyin = fopen( infilename, "r" ); if ( yyin == NULL ) lerrsf( "can't open %s", file ); } else { yyin = stdin; infilename = ""; } } //fun1-5 /* flexerror - report an error message and terminate */ void flexerror( msg ) char msg[]; { fprintf( err, "%s: %s\n", program_name, msg ); //fun1-5(1) flexend( 1 ); } //fun1-5(1) /* flexend - terminate flex * * note * This routine does not return. */ void flexend( exit_status ) int exit_status; { int tblsiz; int unlink(); if ( skelfile != NULL ) { if ( ferror( skelfile ) ) flexfatal( "error occurred when reading skeleton file" ); else if ( fclose( skelfile ) ) flexfatal( "error occurred when closing skeleton file" ); } if ( exit_status != 0 && outfile_created ) { if ( ferror( stdout ) ) flexfatal( "error occurred when writing output file" ); else if ( fclose( stdout ) ) flexfatal( "error occurred when closing output file" ); else if ( unlink( outfile_path ) ) flexfatal( "error occurred when deleting output file" ); } if ( backing_up_report && backing_up_file ) { if ( num_backing_up == 0 ) fprintf( backing_up_file, "No backing up.\n" ); else if ( fullspd || fulltbl ) fprintf( backing_up_file, "%d backing up (non-accepting) states.\n", num_backing_up ); else fprintf( backing_up_file, "Compressed tables always back up.\n" ); if ( ferror( backing_up_file ) ) flexfatal( "error occurred when writing backup file" ); else if ( fclose( backing_up_file ) ) flexfatal( "error occurred when closing backup file" ); } if ( printstats ) { fprintf( err, "%s version %s usage statistics:\n", program_name, flex_version ); fprintf( err, " scanner options: -" ); if ( C_plus_plus ) putc( '+', err ); if ( backing_up_report ) putc( 'b', err ); if ( ddebug ) putc( 'd', err ); if ( caseins ) putc( 'i', err ); if ( lex_compat ) putc( 'l', err ); if ( performance_report != 0 ) //mutated statement putc( 'p', err ); if ( performance_report > 1 ) putc( 'p', err ); if ( spprdflt ) putc( 's', err ); if ( use_stdout ) putc( 't', err ); if ( printstats ) putc( 'v', err ); /* always true! */ if ( nowarn ) putc( 'w', err ); if ( ! interactive ) putc( 'B', err ); if ( interactive ) putc( 'I', err ); if ( ! gen_line_dirs ) putc( 'L', err ); if ( trace ) putc( 'T', err ); if ( csize == 128 ) putc( '7', err ); else putc( '8', err ); fprintf( err, " -C" ); if ( long_align ) putc( 'a', err ); if ( fulltbl ) putc( 'f', err ); if ( fullspd ) putc( 'F', err ); if ( useecs ) putc( 'e', err ); if ( usemecs ) putc( 'm', err ); if ( use_read ) putc( 'r', err ); if ( skelname ) fprintf( err, " -S%s", skelname ); if ( strcmp( prefix, "yy" ) ) fprintf( err, " -P%s", prefix ); putc( '\n', err ); fprintf( err, " %d/%d NFA states\n", lastnfa, current_mns ); fprintf( err, " %d/%d DFA states (%d words)\n", lastdfa, current_max_dfas, totnst ); fprintf( err, " %d rules\n", num_rules + num_eof_rules - 1 /* - 1 for def. rule */ ); if ( num_backing_up == 0 ) fprintf( err, " No backing up\n" ); else if ( fullspd || fulltbl ) fprintf( err, " %d backing-up (non-accepting) states\n", num_backing_up ); else fprintf( err, " Compressed tables always back-up\n" ); if ( bol_needed ) fprintf( err, " Beginning-of-line patterns used\n" ); fprintf( err, " %d/%d start conditions\n", lastsc, current_max_scs ); fprintf( err, " %d epsilon states, %d double epsilon states\n", numeps, eps2 ); if ( lastccl == 0 ) fprintf( err, " no character classes\n" ); else fprintf( err, " %d/%d character classes needed %d/%d words of storage, %d reused\n", lastccl, current_maxccls, cclmap[lastccl] + ccllen[lastccl], current_max_ccl_tbl_size, cclreuse ); fprintf( err, " %d state/nextstate pairs created\n", numsnpairs ); fprintf( err, " %d/%d unique/duplicate transitions\n", numuniq, numdup ); if ( fulltbl ) { tblsiz = lastdfa * numecs; fprintf( err, " %d table entries\n", tblsiz ); } else { tblsiz = 2 * (lastdfa + numtemps) + 2 * tblend; fprintf( err, " %d/%d base-def entries created\n", lastdfa + numtemps, current_max_dfas ); fprintf( err, " %d/%d (peak %d) nxt-chk entries created\n", tblend, current_max_xpairs, peakpairs ); fprintf( err, " %d/%d (peak %d) template nxt-chk entries created\n", numtemps * nummecs, current_max_template_xpairs, numtemps * numecs ); fprintf( err, " %d empty table entries\n", nummt ); fprintf( err, " %d protos created\n", numprots ); fprintf( err, " %d templates created, %d uses\n", numtemps, tmpuses ); } if ( useecs ) { tblsiz = tblsiz + csize; fprintf( err, " %d/%d equivalence classes created\n", numecs, csize ); } if ( usemecs ) { tblsiz = tblsiz + numecs; fprintf( err, " %d/%d meta-equivalence classes created\n", nummecs, csize ); } fprintf( err, " %d (%d saved) hash collisions, %d DFAs equal\n", hshcol, hshsave, dfaeql ); fprintf( err, " %d sets of reallocations needed\n", num_reallocs ); fprintf( err, " %d total table entries needed\n", tblsiz ); } #ifndef VMS exit( exit_status ); #else exit( exit_status + 1 ); #endif } //fun1-5(1)-1 /* flexfatal - report a fatal error message and terminate */ void flexfatal( msg ) char msg[]; { fprintf( err, "%s: fatal internal error, %s\n", program_name, msg ); exit( 1 ); } //fun1-6 /* set_up_initial_allocations - allocate memory for internal tables */ void set_up_initial_allocations() { current_mns = INITIAL_MNS; firstst = allocate_integer_array( current_mns ); lastst = allocate_integer_array( current_mns ); finalst = allocate_integer_array( current_mns ); transchar = allocate_integer_array( current_mns ); trans1 = allocate_integer_array( current_mns ); trans2 = allocate_integer_array( current_mns ); accptnum = allocate_integer_array( current_mns ); assoc_rule = allocate_integer_array( current_mns ); state_type = allocate_integer_array( current_mns ); current_max_rules = INITIAL_MAX_RULES; rule_type = allocate_integer_array( current_max_rules ); rule_linenum = allocate_integer_array( current_max_rules ); rule_useful = allocate_integer_array( current_max_rules ); current_max_scs = INITIAL_MAX_SCS; scset = allocate_integer_array( current_max_scs ); scbol = allocate_integer_array( current_max_scs ); scxclu = allocate_integer_array( current_max_scs ); sceof = allocate_integer_array( current_max_scs ); scname = allocate_char_ptr_array( current_max_scs ); actvsc = allocate_integer_array( current_max_scs ); current_maxccls = INITIAL_MAX_CCLS; cclmap = allocate_integer_array( current_maxccls ); ccllen = allocate_integer_array( current_maxccls ); cclng = allocate_integer_array( current_maxccls ); current_max_ccl_tbl_size = INITIAL_MAX_CCL_TBL_SIZE; ccltbl = allocate_Character_array( current_max_ccl_tbl_size ); current_max_dfa_size = INITIAL_MAX_DFA_SIZE; current_max_xpairs = INITIAL_MAX_XPAIRS; nxt = allocate_integer_array( current_max_xpairs ); chk = allocate_integer_array( current_max_xpairs ); current_max_template_xpairs = INITIAL_MAX_TEMPLATE_XPAIRS; tnxt = allocate_integer_array( current_max_template_xpairs ); current_max_dfas = INITIAL_MAX_DFAS; base = allocate_integer_array( current_max_dfas ); def = allocate_integer_array( current_max_dfas ); dfasiz = allocate_integer_array( current_max_dfas ); accsiz = allocate_integer_array( current_max_dfas ); dhash = allocate_integer_array( current_max_dfas ); dss = allocate_int_ptr_array( current_max_dfas ); dfaacc = allocate_dfaacc_union( current_max_dfas ); nultrans = (int *) 0; } /* readin - read in the rules section of the input file(s) */ void readin() { skelout(); line_directive_out( (FILE *) 0 ); if ( yyparse() ) { pinpoint_message( "fatal parse error" ); flexend( 1 ); } if ( syntaxerror ) flexend( 1 ); if ( yymore_really_used == REALLY_USED ) yymore_used = true; else if ( yymore_really_used == REALLY_NOT_USED ) yymore_used = false; if ( reject_really_used == REALLY_USED ) reject = true; else if ( reject_really_used == REALLY_NOT_USED ) reject = false; if ( performance_report > 0 ) { if ( lex_compat ) { fprintf( err, "-l AT&T lex compatibility option entails a large performance penalty\n" ); fprintf( err, " and may be the actual source of other reported performance penalties\n" ); } if ( performance_report > 1 ) { if ( interactive ) fprintf( err, "-I (interactive) entails a minor performance penalty\n" ); if ( yymore_used ) fprintf( err, "yymore() entails a minor performance penalty\n" ); } if ( reject ) fprintf( err, "REJECT entails a large performance penalty\n" ); if ( variable_trailing_context_rules ) fprintf( err, "Variable trailing context rules entail a large performance penalty\n" ); } if ( reject ) real_reject = true; if ( variable_trailing_context_rules ) reject = true; if ( (fulltbl || fullspd) && reject ) { if ( real_reject ) flexerror( "REJECT cannot be used with -f or -F" ); else flexerror( "variable trailing context rules cannot be used with -f or -F" ); } if ( csize == 256 ) puts( "typedef unsigned char YY_CHAR;" ); else puts( "typedef char YY_CHAR;" ); if ( C_plus_plus ) { puts( "#define yytext_ptr yytext" ); if ( interactive ) puts( "#define YY_INTERACTIVE" ); } if ( fullspd ) printf( "typedef const struct yy_trans_info *yy_state_type;\n" ); else if ( ! C_plus_plus ) printf( "typedef int yy_state_type;\n" ); if ( reject ) printf( "\n#define YY_USES_REJECT\n" ); if ( ddebug ) puts( "\n#define FLEX_DEBUG" ); if ( lex_compat ) { printf( "FILE *yyin = stdin, *yyout = stdout;\n" ); printf( "extern int yylineno;\n" ); printf( "int yylineno = 1;\n" ); } else if ( ! C_plus_plus ) printf( "FILE *yyin = (FILE *) 0, *yyout = (FILE *) 0;\n" ); if ( C_plus_plus ) printf( "\n#include \n" ); else { if ( yytext_is_array ) puts( "extern char yytext[];\n" ); else { puts( "extern char *yytext;" ); puts( "#define yytext_ptr yytext" ); } } if ( useecs ) numecs = cre8ecs( nextecm, ecgroup, csize ); else numecs = csize; /* Now map the equivalence class for NUL to its expected place. */ ecgroup[0] = ecgroup[csize]; NUL_ec = ABS( ecgroup[0] ); if ( useecs ) ccl2ecl(); } /* ccladd - add a single character to a ccl */ void ccladd( cclp, ch ) int cclp; int ch; { int ind, len, newpos, i; check_char( ch ); len = ccllen[cclp]; ind = cclmap[cclp]; /* check to see if the character is already in the ccl */ for ( i = 0; i < len; ++i ) if ( ccltbl[ind + i] == ch ) return; newpos = ind + len; if ( newpos >= current_max_ccl_tbl_size ) { current_max_ccl_tbl_size += MAX_CCL_TBL_SIZE_INCREMENT; ++num_reallocs; ccltbl = reallocate_Character_array( ccltbl, current_max_ccl_tbl_size ); } ccllen[cclp] = len + 1; ccltbl[newpos] = ch; } /* cclinit - return an empty ccl */ int cclinit() { if ( ++lastccl >= current_maxccls ) { current_maxccls += MAX_CCLS_INCREMENT; ++num_reallocs; cclmap = reallocate_integer_array( cclmap, current_maxccls ); ccllen = reallocate_integer_array( ccllen, current_maxccls ); cclng = reallocate_integer_array( cclng, current_maxccls ); } if ( lastccl == 1 ) /* we're making the first ccl */ cclmap[lastccl] = 0; else /* The new pointer is just past the end of the last ccl. * Since the cclmap points to the \first/ character of a * ccl, adding the length of the ccl to the cclmap pointer * will produce a cursor to the first free space. */ cclmap[lastccl] = cclmap[lastccl - 1] + ccllen[lastccl - 1]; ccllen[lastccl] = 0; cclng[lastccl] = 0; /* ccl's start out life un-negated */ return lastccl; } /* cclnegate - negate the given ccl */ void cclnegate( cclp ) int cclp; { cclng[cclp] = 1; } /* list_character_set - list the members of a set of characters in CCL form * * Writes to the given file a character-class representation of those * characters present in the given CCL. A character is present if it * has a non-zero value in the cset array. */ void list_character_set( file, cset ) FILE *file; int cset[]; { register int i; putc( '[', file ); for ( i = 0; i < csize; ++i ) { if ( cset[i] ) { register int start_char = i; putc( ' ', file ); fputs( readable_form( i ), file ); while ( ++i < csize && cset[i] ) ; if ( i - 1 > start_char ) /* this was a run */ fprintf( file, "-%s", readable_form( i - 1 ) ); putc( ' ', file ); } } putc( ']', file ); } /* dfa - DFA construction routines */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* declare functions that have forward references */ void dump_associated_rules PROTO((FILE*, int)); void dump_transitions PROTO((FILE*, int[])); void sympartition PROTO((int[], int, int[], int[])); int symfollowset PROTO((int[], int, int, int[])); /* check_for_backing_up - check a DFA state for backing up * * synopsis * void check_for_backing_up( int ds, int state[numecs] ); * * ds is the number of the state to check and state[] is its out-transitions, * indexed by equivalence class. */ void check_for_backing_up( ds, state ) int ds; int state[]; { if ( (reject && ! dfaacc[ds].dfaacc_set) || (! reject && ! dfaacc[ds].dfaacc_state) ) { /* state is non-accepting */ ++num_backing_up; if ( backing_up_report ) { fprintf( backing_up_file, "State #%d is non-accepting -\n", ds ); /* identify the state */ dump_associated_rules( backing_up_file, ds ); /* Now identify it further using the out- and * jam-transitions. */ dump_transitions( backing_up_file, state ); putc( '\n', backing_up_file ); } } } /* check_trailing_context - check to see if NFA state set constitutes * "dangerous" trailing context * * synopsis * void check_trailing_context( int nfa_states[num_states+1], int num_states, * int accset[nacc+1], int nacc ); * * NOTES * Trailing context is "dangerous" if both the head and the trailing * part are of variable size \and/ there's a DFA state which contains * both an accepting state for the head part of the rule and NFA states * which occur after the beginning of the trailing context. * * When such a rule is matched, it's impossible to tell if having been * in the DFA state indicates the beginning of the trailing context or * further-along scanning of the pattern. In these cases, a warning * message is issued. * * nfa_states[1 .. num_states] is the list of NFA states in the DFA. * accset[1 .. nacc] is the list of accepting numbers for the DFA state. */ void check_trailing_context( nfa_states, num_states, accset, nacc ) int *nfa_states, num_states; int *accset; register int nacc; { register int i, j; for ( i = 1; i <= num_states; ++i ) { int ns = nfa_states[i]; register int type = state_type[ns]; register int ar = assoc_rule[ns]; if ( type == STATE_NORMAL || rule_type[ar] != RULE_VARIABLE ) { /* do nothing */ } else if ( type == STATE_TRAILING_CONTEXT ) { /* Potential trouble. Scan set of accepting numbers * for the one marking the end of the "head". We * assume that this looping will be fairly cheap * since it's rare that an accepting number set * is large. */ for ( j = 1; j <= nacc; ++j ) if ( accset[j] & YY_TRAILING_HEAD_MASK ) { line_warning( "dangerous trailing context", rule_linenum[ar] ); return; } } } } /* dump_associated_rules - list the rules associated with a DFA state * * Goes through the set of NFA states associated with the DFA and * extracts the first MAX_ASSOC_RULES unique rules, sorts them, * and writes a report to the given file. */ void dump_associated_rules( file, ds ) FILE *file; int ds; { register int i, j; register int num_associated_rules = 0; int rule_set[MAX_ASSOC_RULES + 1]; int *dset = dss[ds]; int size = dfasiz[ds]; for ( i = 1; i <= size; ++i ) { register int rule_num = rule_linenum[assoc_rule[dset[i]]]; for ( j = 1; j <= num_associated_rules; ++j ) if ( rule_num == rule_set[j] ) break; if ( j > num_associated_rules ) { /* new rule */ if ( num_associated_rules < MAX_ASSOC_RULES ) rule_set[++num_associated_rules] = rule_num; } } bubble( rule_set, num_associated_rules ); fprintf( file, " associated rule line numbers:" ); for ( i = 1; i <= num_associated_rules; ++i ) { if ( i % 8 == 1 ) putc( '\n', file ); fprintf( file, "\t%d", rule_set[i] ); } putc( '\n', file ); } /* dump_transitions - list the transitions associated with a DFA state * * synopsis * dump_transitions( FILE *file, int state[numecs] ); * * Goes through the set of out-transitions and lists them in human-readable * form (i.e., not as equivalence classes); also lists jam transitions * (i.e., all those which are not out-transitions, plus EOF). The dump * is done to the given file. */ void dump_transitions( file, state ) FILE *file; int state[]; { register int i, ec; int out_char_set[CSIZE]; for ( i = 0; i < csize; ++i ) { ec = ABS( ecgroup[i] ); out_char_set[i] = state[ec]; } fprintf( file, " out-transitions: " ); list_character_set( file, out_char_set ); /* now invert the members of the set to get the jam transitions */ for ( i = 0; i < csize; ++i ) out_char_set[i] = ! out_char_set[i]; fprintf( file, "\n jam-transitions: EOF " ); list_character_set( file, out_char_set ); putc( '\n', file ); } /* epsclosure - construct the epsilon closure of a set of ndfa states * * synopsis * int *epsclosure( int t[num_states], int *numstates_addr, * int accset[num_rules+1], int *nacc_addr, * int *hashval_addr ); * * NOTES * The epsilon closure is the set of all states reachable by an arbitrary * number of epsilon transitions, which themselves do not have epsilon * transitions going out, unioned with the set of states which have non-null * accepting numbers. t is an array of size numstates of nfa state numbers. * Upon return, t holds the epsilon closure and *numstates_addr is updated. * accset holds a list of the accepting numbers, and the size of accset is * given by *nacc_addr. t may be subjected to reallocation if it is not * large enough to hold the epsilon closure. * * hashval is the hash value for the dfa corresponding to the state set. */ int *epsclosure( t, ns_addr, accset, nacc_addr, hv_addr ) int *t, *ns_addr, accset[], *nacc_addr, *hv_addr; { register int stkpos, ns, tsp; int numstates = *ns_addr, nacc, hashval, transsym, nfaccnum; int stkend, nstate; static int did_stk_init = false, *stk; #define MARK_STATE(state) \ trans1[state] = trans1[state] - MARKER_DIFFERENCE; #define IS_MARKED(state) (trans1[state] < 0) #define UNMARK_STATE(state) \ trans1[state] = trans1[state] + MARKER_DIFFERENCE; #define CHECK_ACCEPT(state) \ { \ nfaccnum = accptnum[state]; \ if ( nfaccnum != NIL ) \ accset[++nacc] = nfaccnum; \ } #define DO_REALLOCATION \ { \ current_max_dfa_size += MAX_DFA_SIZE_INCREMENT; \ ++num_reallocs; \ t = reallocate_integer_array( t, current_max_dfa_size ); \ stk = reallocate_integer_array( stk, current_max_dfa_size ); \ } \ #define PUT_ON_STACK(state) \ { \ if ( ++stkend >= current_max_dfa_size ) \ DO_REALLOCATION \ stk[stkend] = state; \ MARK_STATE(state) \ } #define ADD_STATE(state) \ { \ if ( ++numstates >= current_max_dfa_size ) \ DO_REALLOCATION \ t[numstates] = state; \ hashval += state; \ } #define STACK_STATE(state) \ { \ PUT_ON_STACK(state) \ CHECK_ACCEPT(state) \ if ( nfaccnum != NIL || transchar[state] != SYM_EPSILON ) \ ADD_STATE(state) \ } if ( ! did_stk_init ) { stk = allocate_integer_array( current_max_dfa_size ); did_stk_init = true; } nacc = stkend = hashval = 0; for ( nstate = 1; nstate <= numstates; ++nstate ) { ns = t[nstate]; /* The state could be marked if we've already pushed it onto * the stack. */ if ( ! IS_MARKED(ns) ) { PUT_ON_STACK(ns) CHECK_ACCEPT(ns) hashval += ns; } } for ( stkpos = 1; stkpos <= stkend; ++stkpos ) { ns = stk[stkpos]; transsym = transchar[ns]; if ( transsym == SYM_EPSILON ) { tsp = trans1[ns] + MARKER_DIFFERENCE; if ( tsp != NO_TRANSITION ) { if ( ! IS_MARKED(tsp) ) STACK_STATE(tsp) tsp = trans2[ns]; if ( tsp != NO_TRANSITION && ! IS_MARKED(tsp) ) STACK_STATE(tsp) } } } /* Clear out "visit" markers. */ for ( stkpos = 1; stkpos <= stkend; ++stkpos ) { if ( IS_MARKED(stk[stkpos]) ) UNMARK_STATE(stk[stkpos]) else flexfatal( "consistency check failed in epsclosure()" ); } *ns_addr = numstates; *hv_addr = hashval; *nacc_addr = nacc; return t; } /* increase_max_dfas - increase the maximum number of DFAs */ void increase_max_dfas() { current_max_dfas += MAX_DFAS_INCREMENT; ++num_reallocs; base = reallocate_integer_array( base, current_max_dfas ); def = reallocate_integer_array( def, current_max_dfas ); dfasiz = reallocate_integer_array( dfasiz, current_max_dfas ); accsiz = reallocate_integer_array( accsiz, current_max_dfas ); dhash = reallocate_integer_array( dhash, current_max_dfas ); dss = reallocate_int_ptr_array( dss, current_max_dfas ); dfaacc = reallocate_dfaacc_union( dfaacc, current_max_dfas ); if ( nultrans ) nultrans = reallocate_integer_array( nultrans, current_max_dfas ); } /* ntod - convert an ndfa to a dfa * * Creates the dfa corresponding to the ndfa we've constructed. The * dfa starts out in state #1. */ void ntod() { int *accset, ds, nacc, newds; int sym, hashval, numstates, dsize; int num_full_table_rows; /* used only for -f */ int *nset, *dset; int targptr, totaltrans, i, comstate, comfreq, targ; int *epsclosure(), snstods(), symlist[CSIZE + 1]; int num_start_states; int todo_head, todo_next; /* Note that the following are indexed by *equivalence classes* * and not by characters. Since equivalence classes are indexed * beginning with 1, even if the scanner accepts NUL's, this * means that (since every character is potentially in its own * equivalence class) these arrays must have room for indices * from 1 to CSIZE, so their size must be CSIZE + 1. */ int duplist[CSIZE + 1], state[CSIZE + 1]; int targfreq[CSIZE + 1], targstate[CSIZE + 1]; accset = allocate_integer_array( num_rules + 1 ); nset = allocate_integer_array( current_max_dfa_size ); /* The "todo" queue is represented by the head, which is the DFA * state currently being processed, and the "next", which is the * next DFA state number available (not in use). We depend on the * fact that snstods() returns DFA's \in increasing order/, and thus * need only know the bounds of the dfas to be processed. */ todo_head = todo_next = 0; for ( i = 0; i <= csize; ++i ) { duplist[i] = NIL; symlist[i] = false; } for ( i = 0; i <= num_rules; ++i ) accset[i] = NIL; if ( trace ) { dumpnfa( scset[1] ); fputs( "\n\nDFA Dump:\n\n", err ); } inittbl(); /* Check to see whether we should build a separate table for * transitions on NUL characters. We don't do this for full-speed * (-F) scanners, since for them we don't have a simple state * number lying around with which to index the table. We also * don't bother doing it for scanners unless (1) NUL is in its own * equivalence class (indicated by a positive value of * ecgroup[NUL]), (2) NUL's equivalence class is the last * equivalence class, and (3) the number of equivalence classes is * the same as the number of characters. This latter case comes * about when useecs is false or when it's true but every character * still manages to land in its own class (unlikely, but it's * cheap to check for). If all these things are true then the * character code needed to represent NUL's equivalence class for * indexing the tables is going to take one more bit than the * number of characters, and therefore we won't be assured of * being able to fit it into a YY_CHAR variable. This rules out * storing the transitions in a compressed table, since the code * for interpreting them uses a YY_CHAR variable (perhaps it * should just use an integer, though; this is worth pondering ... * ###). * * Finally, for full tables, we want the number of entries in the * table to be a power of two so the array references go fast (it * will just take a shift to compute the major index). If * encoding NUL's transitions in the table will spoil this, we * give it its own table (note that this will be the case if we're * not using equivalence classes). */ /* Note that the test for ecgroup[0] == numecs below accomplishes * both (1) and (2) above */ if ( ! fullspd && ecgroup[0] == numecs ) { /* NUL is alone in its equivalence class, which is the * last one. */ int use_NUL_table = (numecs == csize); if ( fulltbl && ! use_NUL_table ) { /* We still may want to use the table if numecs * is a power of 2. */ int power_of_two; for ( power_of_two = 1; power_of_two <= csize; power_of_two *= 2 ) if ( numecs == power_of_two ) { use_NUL_table = true; break; } } if ( use_NUL_table ) nultrans = allocate_integer_array( current_max_dfas ); /* From now on, nultrans != nil indicates that we're * saving null transitions for later, separate encoding. */ } if ( fullspd ) { for ( i = 0; i <= numecs; ++i ) state[i] = 0; place_state( state, 0, 0 ); dfaacc[i].dfaacc_state = 0; } else if ( fulltbl ) { if ( nultrans ) /* We won't be including NUL's transitions in the * table, so build it for entries from 0 .. numecs - 1. */ num_full_table_rows = numecs; else /* Take into account the fact that we'll be including * the NUL entries in the transition table. Build it * from 0 .. numecs. */ num_full_table_rows = numecs + 1; /* Unless -Ca, declare it "short" because it's a real * long-shot that that won't be large enough. */ printf( "static const %s yy_nxt[][%d] =\n {\n", /* '}' so vi doesn't get too confused */ long_align ? "long" : "short", num_full_table_rows ); /* Generate 0 entries for state #0. */ for ( i = 0; i < num_full_table_rows; ++i ) mk2data( 0 ); /* Force ',' and dataflush() next call to mk2data().*/ datapos = NUMDATAITEMS; /* Force extra blank line next dataflush(). */ dataline = NUMDATALINES; } /* Create the first states. */ num_start_states = lastsc * 2; for ( i = 1; i <= num_start_states; ++i ) { numstates = 1; /* For each start condition, make one state for the case when * we're at the beginning of the line (the '^' operator) and * one for the case when we're not. */ if ( i % 2 == 1 ) nset[numstates] = scset[(i / 2) + 1]; else nset[numstates] = mkbranch( scbol[i / 2], scset[i / 2] ); nset = epsclosure( nset, &numstates, accset, &nacc, &hashval ); if ( snstods( nset, numstates, accset, nacc, hashval, &ds ) ) { numas += nacc; totnst += numstates; ++todo_next; if ( variable_trailing_context_rules && nacc > 0 ) check_trailing_context( nset, numstates, accset, nacc ); } } if ( ! fullspd ) { if ( ! snstods( nset, 0, accset, 0, 0, &end_of_buffer_state ) ) flexfatal( "could not create unique end-of-buffer state" ); ++numas; ++num_start_states; ++todo_next; } while ( todo_head < todo_next ) { targptr = 0; totaltrans = 0; for ( i = 1; i <= numecs; ++i ) state[i] = 0; ds = ++todo_head; dset = dss[ds]; dsize = dfasiz[ds]; if ( trace ) fprintf( err, "state # %d:\n", ds ); sympartition( dset, dsize, symlist, duplist ); for ( sym = 1; sym <= numecs; ++sym ) { if ( symlist[sym] ) { symlist[sym] = 0; if ( duplist[sym] == NIL ) { /* Symbol has unique out-transitions. */ numstates = symfollowset( dset, dsize, sym, nset ); nset = epsclosure( nset, &numstates, accset, &nacc, &hashval ); if ( snstods( nset, numstates, accset, nacc, hashval, &newds ) ) { totnst = totnst + numstates; ++todo_next; numas += nacc; if ( variable_trailing_context_rules && nacc > 0 ) check_trailing_context( nset, numstates, accset, nacc ); } state[sym] = newds; if ( trace ) fprintf( err, "\t%d\t%d\n", sym, newds ); targfreq[++targptr] = 1; targstate[targptr] = newds; ++numuniq; } else { /* sym's equivalence class has the same * transitions as duplist(sym)'s * equivalence class. */ targ = state[duplist[sym]]; state[sym] = targ; if ( trace ) fprintf( err, "\t%d\t%d\n", sym, targ ); /* Update frequency count for * destination state. */ i = 0; while ( targstate[++i] != targ ) ; ++targfreq[i]; ++numdup; } ++totaltrans; duplist[sym] = NIL; } } numsnpairs = numsnpairs + totaltrans; if ( caseins && ! useecs ) { register int j; for ( i = 'A', j = 'a'; i <= 'Z'; ++i, ++j ) state[i] = state[j]; } if ( ds > num_start_states ) check_for_backing_up( ds, state ); if ( nultrans ) { nultrans[ds] = state[NUL_ec]; state[NUL_ec] = 0; /* remove transition */ } if ( fulltbl ) { /* Supply array's 0-element. */ if ( ds == end_of_buffer_state ) mk2data( -end_of_buffer_state ); else mk2data( end_of_buffer_state ); for ( i = 1; i < num_full_table_rows; ++i ) /* Jams are marked by negative of state * number. */ mk2data( state[i] ? state[i] : -ds ); /* Force ',' and dataflush() next call to mk2data().*/ datapos = NUMDATAITEMS; /* Force extra blank line next dataflush(). */ dataline = NUMDATALINES; } else if ( fullspd ) place_state( state, ds, totaltrans ); else if ( ds == end_of_buffer_state ) /* Special case this state to make sure it does what * it's supposed to, i.e., jam on end-of-buffer. */ stack1( ds, 0, 0, JAMSTATE ); else /* normal, compressed state */ { /* Determine which destination state is the most * common, and how many transitions to it there are. */ comfreq = 0; comstate = 0; for ( i = 1; i <= targptr; ++i ) if ( targfreq[i] > comfreq ) { comfreq = targfreq[i]; comstate = targstate[i]; } bldtbl( state, ds, totaltrans, comstate, comfreq ); } } if ( fulltbl ) dataend(); else if ( ! fullspd ) { cmptmps(); /* create compressed template entries */ /* Create tables for all the states with only one * out-transition. */ while ( onesp > 0 ) { mk1tbl( onestate[onesp], onesym[onesp], onenext[onesp], onedef[onesp] ); --onesp; } mkdeftbl(); } flex_free( (void *) accset ); flex_free( (void *) nset ); } /* snstods - converts a set of ndfa states into a dfa state * * synopsis * is_new_state = snstods( int sns[numstates], int numstates, * int accset[num_rules+1], int nacc, * int hashval, int *newds_addr ); * * On return, the dfa state number is in newds. */ int snstods( sns, numstates, accset, nacc, hashval, newds_addr ) int sns[], numstates, accset[], nacc, hashval, *newds_addr; { int didsort = 0; register int i, j; int newds, *oldsns; for ( i = 1; i <= lastdfa; ++i ) if ( hashval == dhash[i] ) { if ( numstates == dfasiz[i] ) { oldsns = dss[i]; if ( ! didsort ) { /* We sort the states in sns so we * can compare it to oldsns quickly. * We use bubble because there probably * aren't very many states. */ bubble( sns, numstates ); didsort = 1; } for ( j = 1; j <= numstates; ++j ) if ( sns[j] != oldsns[j] ) break; if ( j > numstates ) { ++dfaeql; *newds_addr = i; return 0; } ++hshcol; } else ++hshsave; } /* Make a new dfa. */ if ( ++lastdfa >= current_max_dfas ) increase_max_dfas(); newds = lastdfa; dss[newds] = allocate_integer_array( numstates + 1 ); /* If we haven't already sorted the states in sns, we do so now, * so that future comparisons with it can be made quickly. */ if ( ! didsort ) bubble( sns, numstates ); for ( i = 1; i <= numstates; ++i ) dss[newds][i] = sns[i]; dfasiz[newds] = numstates; dhash[newds] = hashval; if ( nacc == 0 ) { if ( reject ) dfaacc[newds].dfaacc_set = (int *) 0; else dfaacc[newds].dfaacc_state = 0; accsiz[newds] = 0; } else if ( reject ) { /* We sort the accepting set in increasing order so the * disambiguating rule that the first rule listed is considered * match in the event of ties will work. We use a bubble * sort since the list is probably quite small. */ bubble( accset, nacc ); dfaacc[newds].dfaacc_set = allocate_integer_array( nacc + 1 ); /* Save the accepting set for later */ for ( i = 1; i <= nacc; ++i ) { dfaacc[newds].dfaacc_set[i] = accset[i]; if ( accset[i] <= num_rules ) /* Who knows, perhaps a REJECT can yield * this rule. */ rule_useful[accset[i]] = true; } accsiz[newds] = nacc; } else { /* Find lowest numbered rule so the disambiguating rule * will work. */ j = num_rules + 1; for ( i = 1; i <= nacc; ++i ) if ( accset[i] < j ) j = accset[i]; dfaacc[newds].dfaacc_state = j; if ( j <= num_rules ) rule_useful[j] = true; } *newds_addr = newds; return 1; } /* symfollowset - follow the symbol transitions one step * * synopsis * numstates = symfollowset( int ds[current_max_dfa_size], int dsize, * int transsym, int nset[current_max_dfa_size] ); */ int symfollowset( ds, dsize, transsym, nset ) int ds[], dsize, transsym, nset[]; { int ns, tsp, sym, i, j, lenccl, ch, numstates, ccllist; numstates = 0; for ( i = 1; i <= dsize; ++i ) { /* for each nfa state ns in the state set of ds */ ns = ds[i]; sym = transchar[ns]; tsp = trans1[ns]; if ( sym < 0 ) { /* it's a character class */ sym = -sym; ccllist = cclmap[sym]; lenccl = ccllen[sym]; if ( cclng[sym] ) { for ( j = 0; j < lenccl; ++j ) { /* Loop through negated character * class. */ ch = ccltbl[ccllist + j]; if ( ch == 0 ) ch = NUL_ec; if ( ch > transsym ) /* Transsym isn't in negated * ccl. */ break; else if ( ch == transsym ) /* next 2 */ goto bottom; } /* Didn't find transsym in ccl. */ nset[++numstates] = tsp; } else for ( j = 0; j < lenccl; ++j ) { ch = ccltbl[ccllist + j]; if ( ch == 0 ) ch = NUL_ec; if ( ch > transsym ) break; else if ( ch == transsym ) { nset[++numstates] = tsp; break; } } } else if ( sym >= 'A' && sym <= 'Z' && caseins ) flexfatal( "consistency check failed in symfollowset" ); else if ( sym == SYM_EPSILON ) { /* do nothing */ } else if ( ABS( ecgroup[sym] ) == transsym ) nset[++numstates] = tsp; bottom: ; } return numstates; } /* sympartition - partition characters with same out-transitions * * synopsis * sympartition( int ds[current_max_dfa_size], int numstates, * int symlist[numecs], int duplist[numecs] ); */ void sympartition( ds, numstates, symlist, duplist ) int ds[], numstates; int symlist[], duplist[]; { int tch, i, j, k, ns, dupfwd[CSIZE + 1], lenccl, cclp, ich; /* Partitioning is done by creating equivalence classes for those * characters which have out-transitions from the given state. Thus * we are really creating equivalence classes of equivalence classes. */ for ( i = 1; i <= numecs; ++i ) { /* initialize equivalence class list */ duplist[i] = i - 1; dupfwd[i] = i + 1; } duplist[1] = NIL; dupfwd[numecs] = NIL; for ( i = 1; i <= numstates; ++i ) { ns = ds[i]; tch = transchar[ns]; if ( tch != SYM_EPSILON ) { if ( tch < -lastccl || tch >= csize ) { flexfatal( "bad transition character detected in sympartition()" ); } if ( tch >= 0 ) { /* character transition */ int ec = ecgroup[tch]; mkechar( ec, dupfwd, duplist ); symlist[ec] = 1; } else { /* character class */ tch = -tch; lenccl = ccllen[tch]; cclp = cclmap[tch]; mkeccl( ccltbl + cclp, lenccl, dupfwd, duplist, numecs, NUL_ec ); if ( cclng[tch] ) { j = 0; for ( k = 0; k < lenccl; ++k ) { ich = ccltbl[cclp + k]; if ( ich == 0 ) ich = NUL_ec; for ( ++j; j < ich; ++j ) symlist[j] = 1; } for ( ++j; j <= numecs; ++j ) symlist[j] = 1; } else for ( k = 0; k < lenccl; ++k ) { ich = ccltbl[cclp + k]; if ( ich == 0 ) ich = NUL_ec; symlist[ich] = 1; } } } } } /* ecs - equivalence class routines */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* ccl2ecl - convert character classes to set of equivalence classes */ void ccl2ecl() { int i, ich, newlen, cclp, ccls, cclmec; for ( i = 1; i <= lastccl; ++i ) { /* We loop through each character class, and for each character * in the class, add the character's equivalence class to the * new "character" class we are creating. Thus when we are all * done, character classes will really consist of collections * of equivalence classes */ newlen = 0; cclp = cclmap[i]; for ( ccls = 0; ccls < ccllen[i]; ++ccls ) { ich = ccltbl[cclp + ccls]; cclmec = ecgroup[ich]; if ( cclmec > 0 ) { ccltbl[cclp + newlen] = cclmec; ++newlen; } } ccllen[i] = newlen; } } /* cre8ecs - associate equivalence class numbers with class members * * fwd is the forward linked-list of equivalence class members. bck * is the backward linked-list, and num is the number of class members. * * Returned is the number of classes. */ int cre8ecs( fwd, bck, num ) int fwd[], bck[], num; { int i, j, numcl; numcl = 0; /* Create equivalence class numbers. From now on, ABS( bck(x) ) * is the equivalence class number for object x. If bck(x) * is positive, then x is the representative of its equivalence * class. */ for ( i = 1; i <= num; ++i ) if ( bck[i] == NIL ) { bck[i] = ++numcl; for ( j = fwd[i]; j != NIL; j = fwd[j] ) bck[j] = -numcl; } return numcl; } /* mkeccl - update equivalence classes based on character class xtions * * synopsis * Char ccls[]; * int lenccl, fwd[llsiz], bck[llsiz], llsiz, NUL_mapping; * void mkeccl( Char ccls[], int lenccl, int fwd[llsiz], int bck[llsiz], * int llsiz, int NUL_mapping ); * * ccls contains the elements of the character class, lenccl is the * number of elements in the ccl, fwd is the forward link-list of equivalent * characters, bck is the backward link-list, and llsiz size of the link-list. * * NUL_mapping is the value which NUL (0) should be mapped to. */ void mkeccl( ccls, lenccl, fwd, bck, llsiz, NUL_mapping ) Char ccls[]; int lenccl, fwd[], bck[], llsiz, NUL_mapping; { int cclp, oldec, newec; int cclm, i, j; static unsigned char cclflags[CSIZE]; /* initialized to all '\0' */ /* Note that it doesn't matter whether or not the character class is * negated. The same results will be obtained in either case. */ cclp = 0; while ( cclp < lenccl ) { cclm = ccls[cclp]; if ( NUL_mapping && cclm == 0 ) cclm = NUL_mapping; oldec = bck[cclm]; newec = cclm; j = cclp + 1; for ( i = fwd[cclm]; i != NIL && i <= llsiz; i = fwd[i] ) { /* look for the symbol in the character class */ for ( ; j < lenccl; ++j ) { register int ccl_char; if ( NUL_mapping && ccls[j] == 0 ) ccl_char = NUL_mapping; else ccl_char = ccls[j]; if ( ccl_char > i ) break; if ( ccl_char == i && ! cclflags[j] ) { /* We found an old companion of cclm * in the ccl. Link it into the new * equivalence class and flag it as * having been processed. */ bck[i] = newec; fwd[newec] = i; newec = i; /* Set flag so we don't reprocess. */ cclflags[j] = 1; /* Get next equivalence class member. */ /* continue 2 */ goto next_pt; } } /* Symbol isn't in character class. Put it in the old * equivalence class. */ bck[i] = oldec; if ( oldec != NIL ) fwd[oldec] = i; oldec = i; next_pt: ; } if ( bck[cclm] != NIL || oldec != bck[cclm] ) { bck[cclm] = NIL; fwd[oldec] = NIL; } fwd[newec] = NIL; /* Find next ccl member to process. */ for ( ++cclp; cclflags[cclp] && cclp < lenccl; ++cclp ) { /* Reset "doesn't need processing" flag. */ cclflags[cclp] = 0; } } } /* mkechar - create equivalence class for single character */ void mkechar( tch, fwd, bck ) int tch, fwd[], bck[]; { /* If until now the character has been a proper subset of * an equivalence class, break it away to create a new ec */ if ( fwd[tch] != NIL ) bck[fwd[tch]] = bck[tch]; if ( bck[tch] != NIL ) fwd[bck[tch]] = fwd[tch]; fwd[tch] = NIL; bck[tch] = NIL; } /* gen - actual generation (writing) of flex scanners */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* declare functions that have forward references */ void gen_next_state PROTO((int)); void genecs PROTO((void)); void indent_put2s PROTO((char [], char [])); void indent_puts PROTO((char [])); static int indent_level = 0; /* each level is 8 spaces */ #define indent_up() (++indent_level) #define indent_down() (--indent_level) #define set_indent(indent_val) indent_level = indent_val /* Almost everything is done in terms of arrays starting at 1, so provide * a null entry for the zero element of all C arrays. (The exception * to this is that the fast table representation generally uses the * 0 elements of its arrays, too.) */ static char C_int_decl[] = "static const int %s[%d] =\n { 0,\n"; static char C_short_decl[] = "static const short int %s[%d] =\n { 0,\n"; static char C_long_decl[] = "static const long int %s[%d] =\n { 0,\n"; static char C_state_decl[] = "static const yy_state_type %s[%d] =\n { 0,\n"; /* Indent to the current level. */ void do_indent() { register int i = indent_level * 8; while ( i >= 8 ) { putchar( '\t' ); i -= 8; } while ( i > 0 ) { putchar( ' ' ); --i; } } /* Generate the code to keep backing-up information. */ void gen_backing_up() { if ( reject || num_backing_up == 0 ) return; if ( fullspd ) indent_puts( "if ( yy_current_state[-1].yy_nxt )" ); else indent_puts( "if ( yy_accept[yy_current_state] )" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_last_accepting_state = yy_current_state;" ); indent_puts( "yy_last_accepting_cpos = yy_cp;" ); indent_puts( "}" ); indent_down(); } /* Generate the code to perform the backing up. */ void gen_bu_action() { if ( reject || num_backing_up == 0 ) return; set_indent( 3 ); indent_puts( "case 0: /* must back up */" ); indent_puts( "/* undo the effects of YY_DO_BEFORE_ACTION */" ); indent_puts( "*yy_cp = yy_hold_char;" ); if ( fullspd || fulltbl ) indent_puts( "yy_cp = yy_last_accepting_cpos + 1;" ); else /* Backing-up info for compressed tables is taken \after/ * yy_cp has been incremented for the next state. */ indent_puts( "yy_cp = yy_last_accepting_cpos;" ); indent_puts( "yy_current_state = yy_last_accepting_state;" ); indent_puts( "goto yy_find_action;" ); putchar( '\n' ); set_indent( 0 ); } /* genctbl - generates full speed compressed transition table */ void genctbl() { register int i; int end_of_buffer_action = num_rules + 1; /* Table of verify for transition and offset to next state. */ printf( "static const struct yy_trans_info yy_transition[%d] =\n", tblend + numecs + 1 ); printf( " {\n" ); /* We want the transition to be represented as the offset to the * next state, not the actual state number, which is what it currently * is. The offset is base[nxt[i]] - (base of current state)]. That's * just the difference between the starting points of the two involved * states (to - from). * * First, though, we need to find some way to put in our end-of-buffer * flags and states. We do this by making a state with absolutely no * transitions. We put it at the end of the table. */ /* We need to have room in nxt/chk for two more slots: One for the * action and one for the end-of-buffer transition. We now *assume* * that we're guaranteed the only character we'll try to index this * nxt/chk pair with is EOB, i.e., 0, so we don't have to make sure * there's room for jam entries for other characters. */ while ( tblend + 2 >= current_max_xpairs ) expand_nxt_chk(); while ( lastdfa + 1 >= current_max_dfas ) increase_max_dfas(); base[lastdfa + 1] = tblend + 2; nxt[tblend + 1] = end_of_buffer_action; chk[tblend + 1] = numecs + 1; chk[tblend + 2] = 1; /* anything but EOB */ /* So that "make test" won't show arb. differences. */ nxt[tblend + 2] = 0; /* Make sure every state has an end-of-buffer transition and an * action #. */ for ( i = 0; i <= lastdfa; ++i ) { int anum = dfaacc[i].dfaacc_state; int offset = base[i]; chk[offset] = EOB_POSITION; chk[offset - 1] = ACTION_POSITION; nxt[offset - 1] = anum; /* action number */ } for ( i = 0; i <= tblend; ++i ) { if ( chk[i] == EOB_POSITION ) transition_struct_out( 0, base[lastdfa + 1] - i ); else if ( chk[i] == ACTION_POSITION ) transition_struct_out( 0, nxt[i] ); else if ( chk[i] > numecs || chk[i] == 0 ) transition_struct_out( 0, 0 ); /* unused slot */ else /* verify, transition */ transition_struct_out( chk[i], base[nxt[i]] - (i - chk[i]) ); } /* Here's the final, end-of-buffer state. */ transition_struct_out( chk[tblend + 1], nxt[tblend + 1] ); transition_struct_out( chk[tblend + 2], nxt[tblend + 2] ); printf( " };\n" ); printf( "\n" ); /* Table of pointers to start states. */ printf( "static const struct yy_trans_info *yy_start_state_list[%d] =\n", lastsc * 2 + 1 ); printf( " {\n" ); /* } so vi doesn't get confused */ for ( i = 0; i <= lastsc * 2; ++i ) printf( " &yy_transition[%d],\n", base[i] ); dataend(); if ( useecs ) genecs(); } /* Generate equivalence-class tables. */ void genecs() { Char clower(); register int i, j; int numrows; printf( C_int_decl, "yy_ec", csize ); for ( i = 1; i < csize; ++i ) { if ( caseins && (i >= 'A') && (i <= 'Z') ) ecgroup[i] = ecgroup[clower( i )]; ecgroup[i] = ABS( ecgroup[i] ); mkdata( ecgroup[i] ); } dataend(); if ( trace ) { fputs( "\n\nEquivalence Classes:\n\n", err ); numrows = csize / 8; for ( j = 0; j < numrows; ++j ) { for ( i = j; i < csize; i = i + numrows ) { fprintf( err, "%4s = %-2d", readable_form( i ), ecgroup[i] ); putc( ' ', err ); } putc( '\n', err ); } } } /* Generate the code to find the action number. */ void gen_find_action() { if ( fullspd ) indent_puts( "yy_act = yy_current_state[-1].yy_nxt;" ); else if ( fulltbl ) indent_puts( "yy_act = yy_accept[yy_current_state];" ); else if ( reject ) { indent_puts( "yy_current_state = *--yy_state_ptr;" ); indent_puts( "yy_lp = yy_accept[yy_current_state];" ); puts( "find_rule: /* we branch to this label when backing up */" ); indent_puts( "for ( ; ; ) /* until we find what rule we matched */" ); indent_up(); indent_puts( "{" ); indent_puts( "if ( yy_lp && yy_lp < yy_accept[yy_current_state + 1] )" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_act = yy_acclist[yy_lp];" ); if ( variable_trailing_context_rules ) { indent_puts( "if ( yy_act & YY_TRAILING_HEAD_MASK ||" ); indent_puts( " yy_looking_for_trail_begin )" ); indent_up(); indent_puts( "{" ); indent_puts( "if ( yy_act == yy_looking_for_trail_begin )" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_looking_for_trail_begin = 0;" ); indent_puts( "yy_act &= ~YY_TRAILING_HEAD_MASK;" ); indent_puts( "break;" ); indent_puts( "}" ); indent_down(); indent_puts( "}" ); indent_down(); indent_puts( "else if ( yy_act & YY_TRAILING_MASK )" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_looking_for_trail_begin = yy_act & ~YY_TRAILING_MASK;" ); indent_puts( "yy_looking_for_trail_begin |= YY_TRAILING_HEAD_MASK;" ); if ( real_reject ) { /* Remember matched text in case we back up * due to REJECT. */ indent_puts( "yy_full_match = yy_cp;" ); indent_puts( "yy_full_state = yy_state_ptr;" ); indent_puts( "yy_full_lp = yy_lp;" ); } indent_puts( "}" ); indent_down(); indent_puts( "else" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_full_match = yy_cp;" ); indent_puts( "yy_full_state = yy_state_ptr;" ); indent_puts( "yy_full_lp = yy_lp;" ); indent_puts( "break;" ); indent_puts( "}" ); indent_down(); indent_puts( "++yy_lp;" ); indent_puts( "goto find_rule;" ); } else { /* Remember matched text in case we back up due to trailing * context plus REJECT. */ indent_up(); indent_puts( "{" ); indent_puts( "yy_full_match = yy_cp;" ); indent_puts( "break;" ); indent_puts( "}" ); indent_down(); } indent_puts( "}" ); indent_down(); indent_puts( "--yy_cp;" ); /* We could consolidate the following two lines with those at * the beginning, but at the cost of complaints that we're * branching inside a loop. */ indent_puts( "yy_current_state = *--yy_state_ptr;" ); indent_puts( "yy_lp = yy_accept[yy_current_state];" ); indent_puts( "}" ); indent_down(); } else /* compressed */ indent_puts( "yy_act = yy_accept[yy_current_state];" ); } /* genftbl - generates full transition table */ void genftbl() { register int i; int end_of_buffer_action = num_rules + 1; printf( long_align ? C_long_decl : C_short_decl, "yy_accept", lastdfa + 1 ); dfaacc[end_of_buffer_state].dfaacc_state = end_of_buffer_action; for ( i = 1; i <= lastdfa; ++i ) { register int anum = dfaacc[i].dfaacc_state; mkdata( anum ); if ( trace && anum ) fprintf( err, "state # %d accepts: [%d]\n", i, anum ); } dataend(); if ( useecs ) genecs(); /* Don't have to dump the actual full table entries - they were * created on-the-fly. */ } /* Generate the code to find the next compressed-table state. */ void gen_next_compressed_state( char_map ) char *char_map; { indent_put2s( "register YY_CHAR yy_c = %s;", char_map ); /* Save the backing-up info \before/ computing the next state * because we always compute one more state than needed - we * always proceed until we reach a jam state */ gen_backing_up(); indent_puts( "while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state )" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_current_state = (int) yy_def[yy_current_state];" ); if ( usemecs ) { /* We've arrange it so that templates are never chained * to one another. This means we can afford to make a * very simple test to see if we need to convert to * yy_c's meta-equivalence class without worrying * about erroneously looking up the meta-equivalence * class twice */ do_indent(); /* lastdfa + 2 is the beginning of the templates */ printf( "if ( yy_current_state >= %d )\n", lastdfa + 2 ); indent_up(); indent_puts( "yy_c = yy_meta[(unsigned int) yy_c];" ); indent_down(); } indent_puts( "}" ); indent_down(); indent_puts( "yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c];" ); } /* Generate the code to find the next match. */ void gen_next_match() { /* NOTE - changes in here should be reflected in gen_next_state() and * gen_NUL_trans(). */ char *char_map = useecs ? "yy_ec[YY_SC_TO_UI(*yy_cp)]" : "YY_SC_TO_UI(*yy_cp)"; char *char_map_2 = useecs ? "yy_ec[YY_SC_TO_UI(*++yy_cp)]" : "YY_SC_TO_UI(*++yy_cp)"; if ( fulltbl ) { indent_put2s( "while ( (yy_current_state = yy_nxt[yy_current_state][%s]) > 0 )", char_map ); indent_up(); if ( num_backing_up > 0 ) { indent_puts( "{" ); /* } for vi */ gen_backing_up(); putchar( '\n' ); } indent_puts( "++yy_cp;" ); if ( num_backing_up > 0 ) /* { for vi */ indent_puts( "}" ); indent_down(); putchar( '\n' ); indent_puts( "yy_current_state = -yy_current_state;" ); } else if ( fullspd ) { indent_puts( "{" ); /* } for vi */ indent_puts( "register const struct yy_trans_info *yy_trans_info;\n" ); indent_puts( "register YY_CHAR yy_c;\n" ); indent_put2s( "for ( yy_c = %s;", char_map ); indent_puts( " (yy_trans_info = &yy_current_state[(unsigned int) yy_c])->" ); indent_puts( "yy_verify == yy_c;" ); indent_put2s( " yy_c = %s )", char_map_2 ); indent_up(); if ( num_backing_up > 0 ) indent_puts( "{" ); /* } for vi */ indent_puts( "yy_current_state += yy_trans_info->yy_nxt;" ); if ( num_backing_up > 0 ) { putchar( '\n' ); gen_backing_up(); /* { for vi */ indent_puts( "}" ); } indent_down(); /* { for vi */ indent_puts( "}" ); } else { /* compressed */ indent_puts( "do" ); indent_up(); indent_puts( "{" ); /* } for vi */ gen_next_state( false ); indent_puts( "++yy_cp;" ); /* { for vi */ indent_puts( "}" ); indent_down(); do_indent(); if ( interactive ) printf( "while ( yy_base[yy_current_state] != %d );\n", jambase ); else printf( "while ( yy_current_state != %d );\n", jamstate ); if ( ! reject && ! interactive ) { /* Do the guaranteed-needed backing up to figure out * the match. */ indent_puts( "yy_cp = yy_last_accepting_cpos;" ); indent_puts( "yy_current_state = yy_last_accepting_state;" ); } } } /* Generate the code to find the next state. */ void gen_next_state( worry_about_NULs ) int worry_about_NULs; { /* NOTE - changes in here should be reflected in get_next_match() */ char char_map[256]; if ( worry_about_NULs && ! nultrans ) { if ( useecs ) (void) sprintf( char_map, "(*yy_cp ? yy_ec[YY_SC_TO_UI(*yy_cp)] : %d)", NUL_ec ); else (void) sprintf( char_map, "(*yy_cp ? YY_SC_TO_UI(*yy_cp) : %d)", NUL_ec ); } else strcpy( char_map, useecs ? "yy_ec[YY_SC_TO_UI(*yy_cp)]" : "YY_SC_TO_UI(*yy_cp)" ); if ( worry_about_NULs && nultrans ) { if ( ! fulltbl && ! fullspd ) /* Compressed tables back up *before* they match. */ gen_backing_up(); indent_puts( "if ( *yy_cp )" ); indent_up(); indent_puts( "{" ); /* } for vi */ } if ( fulltbl ) indent_put2s( "yy_current_state = yy_nxt[yy_current_state][%s];", char_map ); else if ( fullspd ) indent_put2s( "yy_current_state += yy_current_state[%s].yy_nxt;", char_map ); else gen_next_compressed_state( char_map ); if ( worry_about_NULs && nultrans ) { /* { for vi */ indent_puts( "}" ); indent_down(); indent_puts( "else" ); indent_up(); indent_puts( "yy_current_state = yy_NUL_trans[yy_current_state];" ); indent_down(); } if ( fullspd || fulltbl ) gen_backing_up(); if ( reject ) indent_puts( "*yy_state_ptr++ = yy_current_state;" ); } /* Generate the code to make a NUL transition. */ void gen_NUL_trans() { /* NOTE - changes in here should be reflected in get_next_match() */ int need_backing_up = (num_backing_up > 0 && ! reject); if ( need_backing_up ) /* We'll need yy_cp lying around for the gen_backing_up(). */ indent_puts( "register char *yy_cp = yy_c_buf_p;" ); putchar( '\n' ); if ( nultrans ) { indent_puts( "yy_current_state = yy_NUL_trans[yy_current_state];" ); indent_puts( "yy_is_jam = (yy_current_state == 0);" ); } else if ( fulltbl ) { do_indent(); printf( "yy_current_state = yy_nxt[yy_current_state][%d];\n", NUL_ec ); indent_puts( "yy_is_jam = (yy_current_state <= 0);" ); } else if ( fullspd ) { do_indent(); printf( "register int yy_c = %d;\n", NUL_ec ); indent_puts( "register const struct yy_trans_info *yy_trans_info;\n" ); indent_puts( "yy_trans_info = &yy_current_state[(unsigned int) yy_c];" ); indent_puts( "yy_current_state += yy_trans_info->yy_nxt;" ); indent_puts( "yy_is_jam = (yy_trans_info->yy_verify != yy_c);" ); } else { char NUL_ec_str[20]; (void) sprintf( NUL_ec_str, "%d", NUL_ec ); gen_next_compressed_state( NUL_ec_str ); if ( reject ) indent_puts( "*yy_state_ptr++ = yy_current_state;" ); do_indent(); printf( "yy_is_jam = (yy_current_state == %d);\n", jamstate ); } /* If we've entered an accepting state, back up; note that * compressed tables have *already* done such backing up, so * we needn't bother with it again. */ if ( need_backing_up && (fullspd || fulltbl) ) { putchar( '\n' ); indent_puts( "if ( ! yy_is_jam )" ); indent_up(); indent_puts( "{" ); gen_backing_up(); indent_puts( "}" ); indent_down(); } } /* Generate the code to find the start state. */ void gen_start_state() { if ( fullspd ) indent_put2s( "yy_current_state = yy_start_state_list[yy_start%s];", bol_needed ? " + (yy_bp[-1] == '\\n' ? 1 : 0)" : "" ); else { indent_puts( "yy_current_state = yy_start;" ); if ( bol_needed ) { indent_puts( "if ( yy_bp[-1] == '\\n' )" ); indent_up(); indent_puts( "++yy_current_state;" ); indent_down(); } if ( reject ) { /* Set up for storing up states. */ indent_puts( "yy_state_ptr = yy_state_buf;" ); indent_puts( "*yy_state_ptr++ = yy_current_state;" ); } } } /* gentabs - generate data statements for the transition tables */ void gentabs() { int i, j, k, *accset, nacc, *acc_array, total_states; int end_of_buffer_action = num_rules + 1; /* *Everything* is done in terms of arrays starting at 1, so provide * a null entry for the zero element of all C arrays. */ static char C_char_decl[] = "static const YY_CHAR %s[%d] =\n { 0,\n"; /* } for vi */ acc_array = allocate_integer_array( current_max_dfas ); nummt = 0; /* The compressed table format jams by entering the "jam state", * losing information about the previous state in the process. * In order to recover the previous state, we effectively need * to keep backing-up information. */ ++num_backing_up; if ( reject ) { /* Write out accepting list and pointer list. * * First we generate the "yy_acclist" array. In the process, * we compute the indices that will go into the "yy_accept" * array, and save the indices in the dfaacc array. */ int EOB_accepting_list[2]; /* Set up accepting structures for the End Of Buffer state. */ EOB_accepting_list[0] = 0; EOB_accepting_list[1] = end_of_buffer_action; accsiz[end_of_buffer_state] = 1; dfaacc[end_of_buffer_state].dfaacc_set = EOB_accepting_list; printf( long_align ? C_long_decl : C_short_decl, "yy_acclist", MAX( numas, 1 ) + 1 ); j = 1; /* index into "yy_acclist" array */ for ( i = 1; i <= lastdfa; ++i ) { acc_array[i] = j; if ( accsiz[i] != 0 ) { accset = dfaacc[i].dfaacc_set; nacc = accsiz[i]; if ( trace ) fprintf( err, "state # %d accepts: ", i ); for ( k = 1; k <= nacc; ++k ) { int accnum = accset[k]; ++j; if ( variable_trailing_context_rules && ! (accnum & YY_TRAILING_HEAD_MASK) && accnum > 0 && accnum <= num_rules && rule_type[accnum] == RULE_VARIABLE ) { /* Special hack to flag * accepting number as part * of trailing context rule. */ accnum |= YY_TRAILING_MASK; } mkdata( accnum ); if ( trace ) { fprintf( err, "[%d]", accset[k] ); if ( k < nacc ) fputs( ", ", err ); else putc( '\n', err ); } } } } /* add accepting number for the "jam" state */ acc_array[i] = j; dataend(); } else { dfaacc[end_of_buffer_state].dfaacc_state = end_of_buffer_action; for ( i = 1; i <= lastdfa; ++i ) acc_array[i] = dfaacc[i].dfaacc_state; /* add accepting number for jam state */ acc_array[i] = 0; } /* Spit out "yy_accept" array. If we're doing "reject", it'll be * pointers into the "yy_acclist" array. Otherwise it's actual * accepting numbers. In either case, we just dump the numbers. */ /* "lastdfa + 2" is the size of "yy_accept"; includes room for C arrays * beginning at 0 and for "jam" state. */ k = lastdfa + 2; if ( reject ) /* We put a "cap" on the table associating lists of accepting * numbers with state numbers. This is needed because we tell * where the end of an accepting list is by looking at where * the list for the next state starts. */ ++k; printf( long_align ? C_long_decl : C_short_decl, "yy_accept", k ); for ( i = 1; i <= lastdfa; ++i ) { mkdata( acc_array[i] ); if ( ! reject && trace && acc_array[i] ) fprintf( err, "state # %d accepts: [%d]\n", i, acc_array[i] ); } /* Add entry for "jam" state. */ mkdata( acc_array[i] ); if ( reject ) /* Add "cap" for the list. */ mkdata( acc_array[i] ); dataend(); if ( useecs ) genecs(); if ( usemecs ) { /* Write out meta-equivalence classes (used to index * templates with). */ if ( trace ) fputs( "\n\nMeta-Equivalence Classes:\n", err ); printf( C_int_decl, "yy_meta", numecs + 1 ); for ( i = 1; i <= numecs; ++i ) { if ( trace ) fprintf( err, "%d = %d\n", i, ABS( tecbck[i] ) ); mkdata( ABS( tecbck[i] ) ); } dataend(); } total_states = lastdfa + numtemps; printf( (tblend >= MAX_SHORT || long_align) ? C_long_decl : C_short_decl, "yy_base", total_states + 1 ); for ( i = 1; i <= lastdfa; ++i ) { register int d = def[i]; if ( base[i] == JAMSTATE ) base[i] = jambase; if ( d == JAMSTATE ) def[i] = jamstate; else if ( d < 0 ) { /* Template reference. */ ++tmpuses; def[i] = lastdfa - d + 1; } mkdata( base[i] ); } /* Generate jam state's base index. */ mkdata( base[i] ); for ( ++i /* skip jam state */; i <= total_states; ++i ) { mkdata( base[i] ); def[i] = jamstate; } dataend(); printf( (total_states >= MAX_SHORT || long_align) ? C_long_decl : C_short_decl, "yy_def", total_states + 1 ); for ( i = 1; i <= total_states; ++i ) mkdata( def[i] ); dataend(); printf( (total_states >= MAX_SHORT || long_align) ? C_long_decl : C_short_decl, "yy_nxt", tblend + 1 ); for ( i = 1; i <= tblend; ++i ) { if ( nxt[i] == 0 || chk[i] == 0 ) nxt[i] = jamstate; /* new state is the JAM state */ mkdata( nxt[i] ); } dataend(); printf( (total_states >= MAX_SHORT || long_align) ? C_long_decl : C_short_decl, "yy_chk", tblend + 1 ); for ( i = 1; i <= tblend; ++i ) { if ( chk[i] == 0 ) ++nummt; mkdata( chk[i] ); } dataend(); } /* Write out a formatted string (with a secondary string argument) at the * current indentation level, adding a final newline. */ void indent_put2s( fmt, arg ) char fmt[], arg[]; { do_indent(); printf( fmt, arg ); putchar( '\n' ); } /* Write out a string at the current indentation level, adding a final * newline. */ void indent_puts( str ) char str[]; { do_indent(); puts( str ); } /* make_tables - generate transition tables and finishes generating output file */ void make_tables() { register int i; int did_eof_rule = false; skelout(); /* First, take care of YY_DO_BEFORE_ACTION depending on yymore * being used. */ set_indent( 1 ); if ( yymore_used ) { indent_puts( "yytext_ptr -= yy_more_len; \\" ); indent_puts( "yyleng = yy_cp - yytext_ptr; \\" ); } else indent_puts( "yyleng = yy_cp - yy_bp; \\" ); /* Now also deal with copying yytext_ptr to yytext if needed. */ skelout(); if ( yytext_is_array ) { indent_puts( "if ( yyleng >= YYLMAX ) \\" ); indent_up(); indent_puts( "YY_FATAL_ERROR( \"token too large, exceeds YYLMAX\" ); \\" ); indent_down(); indent_puts( "yy_flex_strncpy( yytext, yytext_ptr, yyleng + 1 ); \\" ); } set_indent( 0 ); skelout(); printf( "#define YY_END_OF_BUFFER %d\n", num_rules + 1 ); if ( fullspd ) { /* Need to define the transet type as a size large * enough to hold the biggest offset. */ int total_table_size = tblend + numecs + 1; char *trans_offset_type = (total_table_size >= MAX_SHORT || long_align) ? "long" : "short"; set_indent( 0 ); indent_puts( "struct yy_trans_info" ); indent_up(); indent_puts( "{" ); /* } for vi */ if ( long_align ) indent_puts( "long yy_verify;" ); else indent_puts( "short yy_verify;" ); /* In cases where its sister yy_verify *is* a "yes, there is * a transition", yy_nxt is the offset (in records) to the * next state. In most cases where there is no transition, * the value of yy_nxt is irrelevant. If yy_nxt is the -1th * record of a state, though, then yy_nxt is the action number * for that state. */ indent_put2s( "%s yy_nxt;", trans_offset_type ); indent_puts( "};" ); indent_down(); } if ( fullspd ) genctbl(); else if ( fulltbl ) genftbl(); else gentabs(); /* Definitions for backing up. We don't need them if REJECT * is being used because then we use an alternative backin-up * technique instead. */ if ( num_backing_up > 0 && ! reject ) { if ( ! C_plus_plus ) { indent_puts( "static yy_state_type yy_last_accepting_state;" ); indent_puts( "static char *yy_last_accepting_cpos;\n" ); } } if ( nultrans ) { printf( C_state_decl, "yy_NUL_trans", lastdfa + 1 ); for ( i = 1; i <= lastdfa; ++i ) { if ( fullspd ) printf( " &yy_transition[%d],\n", base[i] ); else mkdata( nultrans[i] ); } dataend(); } if ( ddebug ) { /* Spit out table mapping rules to line numbers. */ indent_puts( "extern int yy_flex_debug;" ); indent_puts( "int yy_flex_debug = 1;\n" ); printf( long_align ? C_long_decl : C_short_decl, "yy_rule_linenum", num_rules ); for ( i = 1; i < num_rules; ++i ) mkdata( rule_linenum[i] ); dataend(); } if ( reject ) { /* Declare state buffer variables. */ if ( ! C_plus_plus ) { puts( "static yy_state_type yy_state_buf[YY_BUF_SIZE + 2], *yy_state_ptr;" ); puts( "static char *yy_full_match;" ); puts( "static int yy_lp;" ); } if ( variable_trailing_context_rules ) { if ( ! C_plus_plus ) { puts( "static int yy_looking_for_trail_begin = 0;" ); puts( "static int yy_full_lp;" ); puts( "static int *yy_full_state;" ); } printf( "#define YY_TRAILING_MASK 0x%x\n", (unsigned int) YY_TRAILING_MASK ); printf( "#define YY_TRAILING_HEAD_MASK 0x%x\n", (unsigned int) YY_TRAILING_HEAD_MASK ); } puts( "#define REJECT \\" ); puts( "{ \\" ); /* } for vi */ puts( "*yy_cp = yy_hold_char; /* undo effects of setting up yytext */ \\" ); puts( "yy_cp = yy_full_match; /* restore poss. backed-over text */ \\" ); if ( variable_trailing_context_rules ) { puts( "yy_lp = yy_full_lp; /* restore orig. accepting pos. */ \\" ); puts( "yy_state_ptr = yy_full_state; /* restore orig. state */ \\" ); puts( "yy_current_state = *yy_state_ptr; /* restore curr. state */ \\" ); } puts( "++yy_lp; \\" ); puts( "goto find_rule; \\" ); /* { for vi */ puts( "}" ); } else { puts( "/* The intent behind this definition is that it'll catch" ); puts( " * any uses of REJECT which flex missed." ); puts( " */" ); puts( "#define REJECT reject_used_but_not_detected" ); } if ( yymore_used ) { if ( ! C_plus_plus ) { indent_puts( "static int yy_more_flag = 0;" ); indent_puts( "static int yy_more_len = 0;" ); } indent_puts( "#define yymore() (yy_more_flag = 1)" ); indent_puts( "#define YY_MORE_ADJ yy_more_len" ); } else { indent_puts( "#define yymore() yymore_used_but_not_detected" ); indent_puts( "#define YY_MORE_ADJ 0" ); } if ( ! C_plus_plus ) { if ( yytext_is_array ) { puts( "#ifndef YYLMAX" ); puts( "#define YYLMAX 8192" ); puts( "#endif\n" ); puts( "char yytext[YYLMAX];" ); puts( "char *yytext_ptr;" ); } else puts( "char *yytext;" ); } fputs( &action_array[defs1_offset], stdout ); skelout(); if ( ! C_plus_plus ) { if ( use_read ) { printf( "\tif ( (result = read( fileno(yyin), (char *) buf, max_size )) < 0 ) \\\n" ); printf( "\t\tYY_FATAL_ERROR( \"input in flex scanner failed\" );\n" ); } else { printf( "\tif ( yy_current_buffer->yy_is_interactive ) \\\n" ); printf( "\t\t{ \\\n" ); printf( "\t\tint c = getc( yyin ); \\\n" ); printf( "\t\tresult = c == EOF ? 0 : 1; \\\n" ); printf( "\t\tbuf[0] = (char) c; \\\n" ); printf( "\t\t} \\\n" ); printf( "\telse if ( ((result = fread( buf, 1, max_size, yyin )) == 0) \\\n" ); printf( "\t\t && ferror( yyin ) ) \\\n" ); printf( "\t\tYY_FATAL_ERROR( \"input in flex scanner failed\" );\n" ); } } skelout(); /* Copy prolog to output file. */ fputs( &action_array[prolog_offset], stdout ); skelout(); set_indent( 2 ); if ( yymore_used ) { indent_puts( "yy_more_len = 0;" ); indent_puts( "if ( yy_more_flag )" ); indent_up(); indent_puts( "{" ); indent_puts( "yy_more_len = yyleng;" ); indent_puts( "yy_more_flag = 0;" ); indent_puts( "}" ); indent_down(); } skelout(); gen_start_state(); /* Note, don't use any indentation. */ puts( "yy_match:" ); gen_next_match(); skelout(); set_indent( 2 ); gen_find_action(); skelout(); if ( lex_compat ) { indent_puts( "if ( yy_act != YY_END_OF_BUFFER )" ); indent_up(); indent_puts( "{" ); indent_puts( "int yyl;" ); indent_puts( "for ( yyl = 0; yyl < yyleng; ++yyl )" ); indent_up(); indent_puts( "if ( yytext[yyl] == '\\n' )" ); indent_up(); indent_puts( "++yylineno;" ); indent_down(); indent_down(); indent_puts( "}" ); indent_down(); } skelout(); if ( ddebug ) { indent_puts( "if ( yy_flex_debug )" ); indent_up(); indent_puts( "{" ); indent_puts( "if ( yy_act == 0 )" ); indent_up(); indent_puts( "fprintf( stderr, \"--scanner backing up\\n\" );" ); indent_down(); do_indent(); printf( "else if ( yy_act < %d )\n", num_rules ); indent_up(); indent_puts( "fprintf( stderr, \"--accepting rule at line %d (\\\"%s\\\")\\n\"," ); indent_puts( " yy_rule_linenum[yy_act], yytext );" ); indent_down(); do_indent(); printf( "else if ( yy_act == %d )\n", num_rules ); indent_up(); indent_puts( "fprintf( stderr, \"--accepting default rule (\\\"%s\\\")\\n\"," ); indent_puts( " yytext );" ); indent_down(); do_indent(); printf( "else if ( yy_act == %d )\n", num_rules + 1 ); indent_up(); indent_puts( "fprintf( stderr, \"--(end of buffer or a NUL)\\n\" );" ); indent_down(); do_indent(); printf( "else\n" ); indent_up(); indent_puts( "fprintf( stderr, \"--EOF (start condition %d)\\n\", YY_START );" ); indent_down(); indent_puts( "}" ); indent_down(); } /* Copy actions to output file. */ skelout(); indent_up(); gen_bu_action(); fputs( &action_array[action_offset], stdout ); /* generate cases for any missing EOF rules */ for ( i = 1; i <= lastsc; ++i ) if ( ! sceof[i] ) { do_indent(); printf( "case YY_STATE_EOF(%s):\n", scname[i] ); did_eof_rule = true; } if ( did_eof_rule ) { indent_up(); indent_puts( "yyterminate();" ); indent_down(); } /* Generate code for handling NUL's, if needed. */ /* First, deal with backing up and setting up yy_cp if the scanner * finds that it should JAM on the NUL. */ skelout(); set_indent( 7 ); if ( fullspd || fulltbl ) indent_puts( "yy_cp = yy_c_buf_p;" ); else { /* compressed table */ if ( ! reject && ! interactive ) { /* Do the guaranteed-needed backing up to figure * out the match. */ indent_puts( "yy_cp = yy_last_accepting_cpos;" ); indent_puts( "yy_current_state = yy_last_accepting_state;" ); } else /* Still need to initialize yy_cp, though * yy_current_state was set up by * yy_get_previous_state(). */ indent_puts( "yy_cp = yy_c_buf_p;" ); } /* Generate code for yy_get_previous_state(). */ set_indent( 1 ); skelout(); if ( bol_needed ) indent_puts( "register char *yy_bp = yytext_ptr;\n" ); gen_start_state(); set_indent( 2 ); skelout(); gen_next_state( true ); set_indent( 1 ); skelout(); gen_NUL_trans(); skelout(); if ( lex_compat ) { /* update yylineno inside of unput() */ indent_puts( "if ( c == '\\n' )" ); indent_up(); indent_puts( "--yylineno;" ); indent_down(); } skelout(); /* Copy remainder of input to output. */ line_directive_out( stdout ); if ( sectnum == 3 ) (void) flexscan(); /* copy remainder of input to output */ } /* misc - miscellaneous flex routines */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* declare functions that have forward references */ void dataflush PROTO((void)); int otoi PROTO((Char [])); void add_action( new_text ) char *new_text; { int len = strlen( new_text ); while ( len + action_index >= action_size - 10 /* slop */ ) { action_size *= 2; action_array = reallocate_character_array( action_array, action_size ); } strcpy( &action_array[action_index], new_text ); action_index += len; } /* allocate_array - allocate memory for an integer array of the given size */ void *allocate_array( size, element_size ) int size, element_size; { register void *mem; /* On 16-bit int machines (e.g., 80286) we might be trying to * allocate more than a signed int can hold, and that won't * work. Cheap test: */ if ( element_size * size <= 0 ) flexfatal( "request for < 1 byte in allocate_array()" ); mem = flex_alloc( element_size * size); ""; if ( mem == NULL ) flexfatal( "memory allocation failed in allocate_array()" ); return mem; } /* all_lower - true if a string is all lower-case */ int all_lower( str ) register char *str; { while ( *str ) { if ( ! isascii( (Char) *str ) || ! islower( *str ) ) return 0; ++str; } return 1; } /* all_upper - true if a string is all upper-case */ int all_upper( str ) register char *str; { while ( *str ) { if ( ! isascii( (Char) *str ) || ! isupper( *str ) ) return 0; ++str; } return 1; } /* bubble - bubble sort an integer array in increasing order * * synopsis * int v[n], n; * void bubble( v, n ); * * description * sorts the first n elements of array v and replaces them in * increasing order. * * passed * v - the array to be sorted * n - the number of elements of 'v' to be sorted */ void bubble( v, n ) int v[], n; { register int i, j, k; for ( i = n; i > 1; --i ) for ( j = 1; j < i; ++j ) if ( v[j] > v[j + 1] ) /* compare */ { k = v[j]; /* exchange */ v[j] = v[j + 1]; v[j + 1] = k; } } /* check_char - checks a character to make sure it's within the range * we're expecting. If not, generates fatal error message * and exits. */ void check_char( c ) int c; { if ( c >= CSIZE ) lerrsf( "bad character '%s' detected in check_char()", readable_form( c ) ); if ( c >= csize ) lerrsf( "scanner requires -8 flag to use the character '%s'", readable_form( c ) ); } /* clower - replace upper-case letter to lower-case */ Char clower( c ) register int c; { return (Char) ((isascii( c ) && isupper( c )) ? tolower( c ) : c); } /* copy_string - returns a dynamically allocated copy of a string */ char *copy_string( str ) register char *str; { register char *c; char *copy; /* find length */ for ( c = str; *c; ++c ) ; copy = (char *) flex_alloc( (c - str + 1) * sizeof( char ) ); if ( copy == NULL ) flexfatal( "dynamic memory failure in copy_string()" ); for ( c = copy; (*c++ = *str++); ) ; return copy; } /* copy_unsigned_string - * returns a dynamically allocated copy of a (potentially) unsigned string */ Char *copy_unsigned_string( str ) register Char *str; { register Char *c; Char *copy; /* find length */ for ( c = str; *c; ++c ) ; copy = allocate_Character_array( c - str + 1 ); for ( c = copy; (*c++ = *str++); ) ; return copy; } /* cshell - shell sort a character array in increasing order * * synopsis * * Char v[n]; * int n, special_case_0; * cshell( v, n, special_case_0 ); * * description * Does a shell sort of the first n elements of array v. * If special_case_0 is true, then any element equal to 0 * is instead assumed to have infinite weight. * * passed * v - array to be sorted * n - number of elements of v to be sorted */ void cshell( v, n, special_case_0 ) Char v[]; int n, special_case_0; { int gap, i, j, jg; Char k; for ( gap = n / 2; gap > 0; gap = gap / 2 ) for ( i = gap; i < n; ++i ) for ( j = i - gap; j >= 0; j = j - gap ) { jg = j + gap; if ( special_case_0 ) { if ( v[jg] == 0 ) break; else if ( v[j] != 0 && v[j] <= v[jg] ) break; } else if ( v[j] <= v[jg] ) break; k = v[j]; v[j] = v[jg]; v[jg] = k; } } /* dataend - finish up a block of data declarations */ void dataend() { if ( datapos > 0 ) dataflush(); /* add terminator for initialization; { for vi */ puts( " } ;\n" ); dataline = 0; datapos = 0; } /* dataflush - flush generated data statements */ void dataflush() { putchar( '\n' ); if ( ++dataline >= NUMDATALINES ) { /* Put out a blank line so that the table is grouped into * large blocks that enable the user to find elements easily. */ putchar( '\n' ); dataline = 0; } /* Reset the number of characters written on the current line. */ datapos = 0; } /* htoi - convert a hexadecimal digit string to an integer value */ int htoi( str ) Char str[]; { unsigned int result; (void) sscanf( (char *) str, "%x", &result ); return result; } /* is_hex_digit - returns true if a character is a valid hex digit, false * otherwise */ int is_hex_digit( ch ) int ch; { if ( isdigit( ch ) ) return 1; switch ( clower( ch ) ) { case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': return 1; default: return 0; } } /* line_directive_out - spit out a "# line" statement */ void line_directive_out( output_file ) FILE *output_file; { if ( infilename && gen_line_dirs ) { char directive[MAXLINE]; sprintf( directive, "# line %d \"%s\"\n", linenum, infilename ); /* If output_file is nil then we should put the directive in * the accumulated actions. */ if ( output_file ) fputs( directive, output_file ); else add_action( directive ); } } /* mark_defs1 - mark the current position in the action array as * representing where the user's section 1 definitions end * and the prolog begins */ void mark_defs1() { defs1_offset = 0; action_array[action_index++] = '\0'; action_offset = prolog_offset = action_index; action_array[action_index] = '\0'; } /* mark_prolog - mark the current position in the action array as * representing the end of the action prolog */ void mark_prolog() { action_array[action_index++] = '\0'; action_offset = action_index; action_array[action_index] = '\0'; } /* mk2data - generate a data statement for a two-dimensional array * * Generates a data statement initializing the current 2-D array to "value". */ void mk2data( value ) int value; { if ( datapos >= NUMDATAITEMS ) { putchar( ',' ); dataflush(); } if ( datapos == 0 ) /* Indent. */ fputs( " ", stdout ); else putchar( ',' ); ++datapos; printf( "%5d", value ); } /* mkdata - generate a data statement * * Generates a data statement initializing the current array element to * "value". */ void mkdata( value ) int value; { if ( datapos >= NUMDATAITEMS ) { putchar( ',' ); dataflush(); } if ( datapos == 0 ) /* Indent. */ fputs( " ", stdout ); else putchar( ',' ); ++datapos; printf( "%5d", value ); } /* myctoi - return the integer represented by a string of digits */ int myctoi( array ) char array[]; { int val = 0; (void) sscanf( array, "%d", &val ); return val; } /* myesc - return character corresponding to escape sequence */ Char myesc( array ) Char array[]; { Char c, esc_char; switch ( array[1] ) { case 'b': return '\b'; case 'f': return '\f'; case 'n': return '\n'; case 'r': return '\r'; case 't': return '\t'; #ifdef __STDC__ case 'a': return '\a'; case 'v': return '\v'; #else case 'a': return '\007'; case 'v': return '\013'; #endif case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': case '8': case '9': { /* \ */ int sptr = 1; while ( isascii( array[sptr] ) && isdigit( array[sptr] ) ) /* Don't increment inside loop control * because if isdigit() is a macro it might * expand into multiple increments ... */ ++sptr; c = array[sptr]; array[sptr] = '\0'; esc_char = otoi( array + 1 ); array[sptr] = c; return esc_char; } case 'x': { /* \x */ int sptr = 2; while ( isascii( array[sptr] ) && is_hex_digit( (char) array[sptr] ) ) /* Don't increment inside loop control * because if isdigit() is a macro it might * expand into multiple increments ... */ ++sptr; c = array[sptr]; array[sptr] = '\0'; esc_char = htoi( array + 2 ); array[sptr] = c; return esc_char; } default: return array[1]; } } /* otoi - convert an octal digit string to an integer value */ int otoi( str ) Char str[]; { unsigned int result; (void) sscanf( (char *) str, "%o", &result ); return result; } /* readable_form - return the the human-readable form of a character * * The returned string is in static storage. */ char *readable_form( c ) register int c; { static char rform[10]; if ( (c >= 0 && c < 32) || c >= 127 ) { switch ( c ) { case '\b': return "\\b"; case '\f': return "\\f"; case '\n': return "\\n"; case '\r': return "\\r"; case '\t': return "\\t"; #ifdef __STDC__ case '\a': return "\\a"; case '\v': return "\\v"; #endif default: (void) sprintf( rform, "\\%.3o", (unsigned int) c ); return rform; } } else if ( c == ' ' ) return "' '"; else { rform[0] = c; rform[1] = '\0'; return rform; } } /* reallocate_array - increase the size of a dynamic array */ void *reallocate_array( array, size, element_size ) void *array; int size, element_size; { register void *new_array; /* Same worry as in allocate_array(): */ if ( size * element_size <= 0 ) flexfatal( "attempt to increase array size by less than 1 byte" ); new_array = flex_realloc( array, size * element_size ); if ( new_array == NULL ) flexfatal( "attempt to increase array size failed" ); return new_array; } /* skelout - write out one section of the skeleton file * * Description * Copies skelfile or skel array to stdout until a line beginning with * "%%" or EOF is found. */ void skelout() { char buf_storage[MAXLINE]; char *buf = buf_storage; int do_copy = 1; /* Loop pulling lines either from the skelfile, if we're using * one, or from the skel[] array. */ while ( skelfile ? (fgets( buf, MAXLINE, skelfile ) != NULL) : ((buf = skel[skel_ind++]) != 0) ) { /* copy from skel array */ if ( buf[0] == '%' ) { /* control line */ switch ( buf[1] ) { case '%': return; case '+': do_copy = C_plus_plus; break; case '-': do_copy = ! C_plus_plus; break; case '*': do_copy = 1; break; default: flexfatal( "bad line in skeleton file" ); } } else if ( do_copy ) { if ( skelfile ) /* Skeleton file reads include final * newline, skel[] array does not. */ fputs( buf, stdout ); else printf( "%s\n", buf ); } } } /* transition_struct_out - output a yy_trans_info structure * * outputs the yy_trans_info structure with the two elements, element_v and * element_n. Formats the output with spaces and carriage returns. */ void transition_struct_out( element_v, element_n ) int element_v, element_n; { printf( "%7d, %5d,", element_v, element_n ); datapos += TRANS_STRUCT_PRINT_LENGTH; if ( datapos >= 75 ) { putchar( '\n' ); if ( ++dataline % 10 == 0 ) putchar( '\n' ); datapos = 0; } } /* The following is only needed when building flex's parser using certain * broken versions of bison. */ void *yy_flex_xmalloc( size ) int size; { void *result = flex_alloc( size ); if ( ! result ) flexfatal( "memory allocation failed in yy_flex_xmalloc()" ); return result; } /* zero_out - set a region of memory to 0 * * Sets region_ptr[0] through region_ptr[size_in_bytes - 1] to zero. */ void zero_out( region_ptr, size_in_bytes ) char *region_ptr; int size_in_bytes; { register char *rp, *rp_end; rp = region_ptr; rp_end = region_ptr + size_in_bytes; while ( rp < rp_end ) *rp++ = 0; } /* nfa - NFA construction routines */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* declare functions that have forward references */ int dupmachine PROTO((int)); void mkxtion PROTO((int, int)); /* add_accept - add an accepting state to a machine * * accepting_number becomes mach's accepting number. */ void add_accept( mach, accepting_number ) int mach, accepting_number; { /* Hang the accepting number off an epsilon state. if it is associated * with a state that has a non-epsilon out-transition, then the state * will accept BEFORE it makes that transition, i.e., one character * too soon. */ if ( transchar[finalst[mach]] == SYM_EPSILON ) accptnum[finalst[mach]] = accepting_number; else { int astate = mkstate( SYM_EPSILON ); accptnum[astate] = accepting_number; (void) link_machines( mach, astate ); } } /* copysingl - make a given number of copies of a singleton machine * * synopsis * * newsng = copysingl( singl, num ); * * newsng - a new singleton composed of num copies of singl * singl - a singleton machine * num - the number of copies of singl to be present in newsng */ int copysingl( singl, num ) int singl, num; { int copy, i; copy = mkstate( SYM_EPSILON ); for ( i = 1; i <= num; ++i ) copy = link_machines( copy, dupmachine( singl ) ); return copy; } /* dumpnfa - debugging routine to write out an nfa */ void dumpnfa( state1 ) int state1; { int sym, tsp1, tsp2, anum, ns; fprintf( err, "\n\n********** beginning dump of nfa with start state %d\n", state1 ); /* We probably should loop starting at firstst[state1] and going to * lastst[state1], but they're not maintained properly when we "or" * all of the rules together. So we use our knowledge that the machine * starts at state 1 and ends at lastnfa. */ /* for ( ns = firstst[state1]; ns <= lastst[state1]; ++ns ) */ for ( ns = 1; ns <= lastnfa; ++ns ) { fprintf( err, "state # %4d\t", ns ); sym = transchar[ns]; tsp1 = trans1[ns]; tsp2 = trans2[ns]; anum = accptnum[ns]; fprintf( err, "%3d: %4d, %4d", sym, tsp1, tsp2 ); if ( anum != NIL ) fprintf( err, " [%d]", anum ); fprintf( err, "\n" ); } fprintf( err, "********** end of dump\n" ); } /* dupmachine - make a duplicate of a given machine * * synopsis * * copy = dupmachine( mach ); * * copy - holds duplicate of mach * mach - machine to be duplicated * * note that the copy of mach is NOT an exact duplicate; rather, all the * transition states values are adjusted so that the copy is self-contained, * as the original should have been. * * also note that the original MUST be contiguous, with its low and high * states accessible by the arrays firstst and lastst */ int dupmachine( mach ) int mach; { int i, init, state_offset; int state = 0; int last = lastst[mach]; for ( i = firstst[mach]; i <= last; ++i ) { state = mkstate( transchar[i] ); if ( trans1[i] != NO_TRANSITION ) { mkxtion( finalst[state], trans1[i] + state - i ); if ( transchar[i] == SYM_EPSILON && trans2[i] != NO_TRANSITION ) mkxtion( finalst[state], trans2[i] + state - i ); } accptnum[state] = accptnum[i]; } if ( state == 0 ) flexfatal( "empty machine in dupmachine()" ); state_offset = state - i + 1; init = mach + state_offset; firstst[init] = firstst[mach] + state_offset; finalst[init] = finalst[mach] + state_offset; lastst[init] = lastst[mach] + state_offset; return init; } /* finish_rule - finish up the processing for a rule * * An accepting number is added to the given machine. If variable_trail_rule * is true then the rule has trailing context and both the head and trail * are variable size. Otherwise if headcnt or trailcnt is non-zero then * the machine recognizes a pattern with trailing context and headcnt is * the number of characters in the matched part of the pattern, or zero * if the matched part has variable length. trailcnt is the number of * trailing context characters in the pattern, or zero if the trailing * context has variable length. */ void finish_rule( mach, variable_trail_rule, headcnt, trailcnt ) int mach, variable_trail_rule, headcnt, trailcnt; { char action_text[MAXLINE]; add_accept( mach, num_rules ); /* We did this in new_rule(), but it often gets the wrong * number because we do it before we start parsing the current rule. */ rule_linenum[num_rules] = linenum; /* If this is a continued action, then the line-number has already * been updated, giving us the wrong number. */ if ( continued_action ) --rule_linenum[num_rules]; sprintf( action_text, "case %d:\n", num_rules ); add_action( action_text ); if ( variable_trail_rule ) { rule_type[num_rules] = RULE_VARIABLE; if ( performance_report > 0 ) fprintf( err, "Variable trailing context rule at line %d\n", rule_linenum[num_rules] ); variable_trailing_context_rules = true; } else { rule_type[num_rules] = RULE_NORMAL; if ( headcnt > 0 || trailcnt > 0 ) { /* Do trailing context magic to not match the trailing * characters. */ char *scanner_cp = "yy_c_buf_p = yy_cp"; char *scanner_bp = "yy_bp"; add_action( "*yy_cp = yy_hold_char; /* undo effects of setting up yytext */\n" ); if ( headcnt > 0 ) { sprintf( action_text, "%s = %s + %d;\n", scanner_cp, scanner_bp, headcnt ); add_action( action_text ); } else { sprintf( action_text, "%s -= %d;\n", scanner_cp, trailcnt ); add_action( action_text ); } add_action( "YY_DO_BEFORE_ACTION; /* set up yytext again */\n" ); } } /* Okay, in the action code at this point yytext and yyleng have * their proper final values for this rule, so here's the point * to do any user action. But don't do it for continued actions, * as that'll result in multiple YY_USER_ACTION's. */ if ( ! continued_action ) add_action( "YY_USER_ACTION\n" ); line_directive_out( (FILE *) 0 ); } /* link_machines - connect two machines together * * synopsis * * new = link_machines( first, last ); * * new - a machine constructed by connecting first to last * first - the machine whose successor is to be last * last - the machine whose predecessor is to be first * * note: this routine concatenates the machine first with the machine * last to produce a machine new which will pattern-match first first * and then last, and will fail if either of the sub-patterns fails. * FIRST is set to new by the operation. last is unmolested. */ int link_machines( first, last ) int first, last; { if ( first == NIL ) return last; else if ( last == NIL ) return first; else { mkxtion( finalst[first], last ); finalst[first] = finalst[last]; lastst[first] = MAX( lastst[first], lastst[last] ); firstst[first] = MIN( firstst[first], firstst[last] ); return first; } } /* mark_beginning_as_normal - mark each "beginning" state in a machine * as being a "normal" (i.e., not trailing context- * associated) states * * The "beginning" states are the epsilon closure of the first state */ void mark_beginning_as_normal( mach ) register int mach; { switch ( state_type[mach] ) { case STATE_NORMAL: /* Oh, we've already visited here. */ return; case STATE_TRAILING_CONTEXT: state_type[mach] = STATE_NORMAL; if ( transchar[mach] == SYM_EPSILON ) { if ( trans1[mach] != NO_TRANSITION ) mark_beginning_as_normal( trans1[mach] ); if ( trans2[mach] != NO_TRANSITION ) mark_beginning_as_normal( trans2[mach] ); } break; default: flexerror( "bad state type in mark_beginning_as_normal()" ); break; } } /* mkbranch - make a machine that branches to two machines * * synopsis * * branch = mkbranch( first, second ); * * branch - a machine which matches either first's pattern or second's * first, second - machines whose patterns are to be or'ed (the | operator) * * Note that first and second are NEITHER destroyed by the operation. Also, * the resulting machine CANNOT be used with any other "mk" operation except * more mkbranch's. Compare with mkor() */ int mkbranch( first, second ) int first, second; { int eps; if ( first == NO_TRANSITION ) return second; else if ( second == NO_TRANSITION ) return first; eps = mkstate( SYM_EPSILON ); mkxtion( eps, first ); mkxtion( eps, second ); return eps; } /* mkclos - convert a machine into a closure * * synopsis * new = mkclos( state ); * * new - a new state which matches the closure of "state" */ int mkclos( state ) int state; { return mkopt( mkposcl( state ) ); } /* mkopt - make a machine optional * * synopsis * * new = mkopt( mach ); * * new - a machine which optionally matches whatever mach matched * mach - the machine to make optional * * notes: * 1. mach must be the last machine created * 2. mach is destroyed by the call */ int mkopt( mach ) int mach; { int eps; if ( ! SUPER_FREE_EPSILON(finalst[mach]) ) { eps = mkstate( SYM_EPSILON ); mach = link_machines( mach, eps ); } /* Can't skimp on the following if FREE_EPSILON(mach) is true because * some state interior to "mach" might point back to the beginning * for a closure. */ eps = mkstate( SYM_EPSILON ); mach = link_machines( eps, mach ); mkxtion( mach, finalst[mach] ); return mach; } /* mkor - make a machine that matches either one of two machines * * synopsis * * new = mkor( first, second ); * * new - a machine which matches either first's pattern or second's * first, second - machines whose patterns are to be or'ed (the | operator) * * note that first and second are both destroyed by the operation * the code is rather convoluted because an attempt is made to minimize * the number of epsilon states needed */ int mkor( first, second ) int first, second; { int eps, orend; if ( first == NIL ) return second; else if ( second == NIL ) return first; else { /* See comment in mkopt() about why we can't use the first * state of "first" or "second" if they satisfy "FREE_EPSILON". */ eps = mkstate( SYM_EPSILON ); first = link_machines( eps, first ); mkxtion( first, second ); if ( SUPER_FREE_EPSILON(finalst[first]) && accptnum[finalst[first]] == NIL ) { orend = finalst[first]; mkxtion( finalst[second], orend ); } else if ( SUPER_FREE_EPSILON(finalst[second]) && accptnum[finalst[second]] == NIL ) { orend = finalst[second]; mkxtion( finalst[first], orend ); } else { eps = mkstate( SYM_EPSILON ); first = link_machines( first, eps ); orend = finalst[first]; mkxtion( finalst[second], orend ); } } finalst[first] = orend; return first; } /* mkposcl - convert a machine into a positive closure * * synopsis * new = mkposcl( state ); * * new - a machine matching the positive closure of "state" */ int mkposcl( state ) int state; { int eps; if ( SUPER_FREE_EPSILON(finalst[state]) ) { mkxtion( finalst[state], state ); return state; } else { eps = mkstate( SYM_EPSILON ); mkxtion( eps, state ); return link_machines( state, eps ); } } /* mkrep - make a replicated machine * * synopsis * new = mkrep( mach, lb, ub ); * * new - a machine that matches whatever "mach" matched from "lb" * number of times to "ub" number of times * * note * if "ub" is INFINITY then "new" matches "lb" or more occurrences of "mach" */ int mkrep( mach, lb, ub ) int mach, lb, ub; { int base_mach, tail, copy, i; base_mach = copysingl( mach, lb - 1 ); if ( ub == INFINITY ) { copy = dupmachine( mach ); mach = link_machines( mach, link_machines( base_mach, mkclos( copy ) ) ); } else { tail = mkstate( SYM_EPSILON ); for ( i = lb; i < ub; ++i ) { copy = dupmachine( mach ); tail = mkopt( link_machines( copy, tail ) ); } mach = link_machines( mach, link_machines( base_mach, tail ) ); } return mach; } /* mkstate - create a state with a transition on a given symbol * * synopsis * * state = mkstate( sym ); * * state - a new state matching sym * sym - the symbol the new state is to have an out-transition on * * note that this routine makes new states in ascending order through the * state array (and increments LASTNFA accordingly). The routine DUPMACHINE * relies on machines being made in ascending order and that they are * CONTIGUOUS. Change it and you will have to rewrite DUPMACHINE (kludge * that it admittedly is) */ int mkstate( sym ) int sym; { if ( ++lastnfa >= current_mns ) { if ( (current_mns += MNS_INCREMENT) >= MAXIMUM_MNS ) lerrif( "input rules are too complicated (>= %d NFA states)", current_mns ); ++num_reallocs; firstst = reallocate_integer_array( firstst, current_mns ); lastst = reallocate_integer_array( lastst, current_mns ); finalst = reallocate_integer_array( finalst, current_mns ); transchar = reallocate_integer_array( transchar, current_mns ); trans1 = reallocate_integer_array( trans1, current_mns ); trans2 = reallocate_integer_array( trans2, current_mns ); accptnum = reallocate_integer_array( accptnum, current_mns ); assoc_rule = reallocate_integer_array( assoc_rule, current_mns ); state_type = reallocate_integer_array( state_type, current_mns ); } firstst[lastnfa] = lastnfa; finalst[lastnfa] = lastnfa; lastst[lastnfa] = lastnfa; transchar[lastnfa] = sym; trans1[lastnfa] = NO_TRANSITION; trans2[lastnfa] = NO_TRANSITION; accptnum[lastnfa] = NIL; assoc_rule[lastnfa] = num_rules; state_type[lastnfa] = current_state_type; /* Fix up equivalence classes base on this transition. Note that any * character which has its own transition gets its own equivalence * class. Thus only characters which are only in character classes * have a chance at being in the same equivalence class. E.g. "a|b" * puts 'a' and 'b' into two different equivalence classes. "[ab]" * puts them in the same equivalence class (barring other differences * elsewhere in the input). */ if ( sym < 0 ) { /* We don't have to update the equivalence classes since * that was already done when the ccl was created for the * first time. */ } else if ( sym == SYM_EPSILON ) ++numeps; else { check_char( sym ); if ( useecs ) /* Map NUL's to csize. */ mkechar( sym ? sym : csize, nextecm, ecgroup ); } return lastnfa; } /* mkxtion - make a transition from one state to another * * synopsis * * mkxtion( statefrom, stateto ); * * statefrom - the state from which the transition is to be made * stateto - the state to which the transition is to be made */ void mkxtion( statefrom, stateto ) int statefrom, stateto; { if ( trans1[statefrom] == NO_TRANSITION ) trans1[statefrom] = stateto; else if ( (transchar[statefrom] != SYM_EPSILON) || (trans2[statefrom] != NO_TRANSITION) ) flexfatal( "found too many transitions in mkxtion()" ); else { /* second out-transition for an epsilon state */ ++eps2; trans2[statefrom] = stateto; } } /* new_rule - initialize for a new rule */ void new_rule() { if ( ++num_rules >= current_max_rules ) { ++num_reallocs; current_max_rules += MAX_RULES_INCREMENT; rule_type = reallocate_integer_array( rule_type, current_max_rules ); rule_linenum = reallocate_integer_array( rule_linenum, current_max_rules ); rule_useful = reallocate_integer_array( rule_useful, current_max_rules ); } if ( num_rules > MAX_RULE ) lerrif( "too many rules (> %d)!", MAX_RULE ); rule_linenum[num_rules] = linenum; rule_useful[num_rules] = false; } /* File created from flex.skel via mkskel.sh */ char *skel[] = { "/* A lexical scanner generated by flex */", "", "/* Scanner skeleton version:", " * $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $", " */", "", "#define FLEX_SCANNER", "", "%-", "#include ", "%*", "", "", "/* cfront 1.2 defines \"c_plusplus\" instead of \"__cplusplus\" */", "#ifdef c_plusplus", "#ifndef __cplusplus", "#define __cplusplus", "#endif", "#endif", "", "", "#ifdef __cplusplus", "", "#include ", "%+", "class istream;", "%*", "#include ", "", "/* Use prototypes in function declarations. */", "#define YY_USE_PROTOS", "", "/* The \"const\" storage-class-modifier is valid. */", "#define YY_USE_CONST", "", "#else /* ! __cplusplus */", "", "#ifdef __STDC__", "", "#define YY_USE_PROTOS", "#define YY_USE_CONST", "", "#endif /* __STDC__ */", "#endif /* ! __cplusplus */", "", "", "#ifdef __TURBOC__", "#define YY_USE_CONST", "#endif", "", "", "#ifndef YY_USE_CONST", "#ifndef const", "#define const", "#endif", "#endif", "", "", "#ifdef YY_USE_PROTOS", "#define YY_PROTO(proto) proto", "#else", "#define YY_PROTO(proto) ()", "#endif", "", "/* Returned upon end-of-file. */", "#define YY_NULL 0", "", "/* Promotes a possibly negative, possibly signed char to an unsigned", " * integer for use as an array index. If the signed char is negative,", " * we want to instead treat it as an 8-bit unsigned char, hence the", " * double cast.", " */", "#define YY_SC_TO_UI(c) ((unsigned int) (unsigned char) c)", "", "/* Enter a start condition. This macro really ought to take a parameter,", " * but we do it the disgusting crufty way forced on us by the ()-less", " * definition of BEGIN.", " */", "#define BEGIN yy_start = 1 + 2 *", "", "/* Translate the current start state into a value that can be later handed", " * to BEGIN to return to the state.", " */", "#define YY_START ((yy_start - 1) / 2)", "", "/* Action number for EOF rule of a given start state. */", "#define YY_STATE_EOF(state) (YY_END_OF_BUFFER + state + 1)", "", "/* Special action meaning \"start processing a new file\". Now included", " * only for backward compatibility with previous versions of flex.", " */", "#define YY_NEW_FILE yyrestart( yyin )", "", "#define YY_END_OF_BUFFER_CHAR 0", "", "/* Size of default input buffer. */", "#define YY_BUF_SIZE 16384", "", "typedef struct yy_buffer_state *YY_BUFFER_STATE;", "", "extern int yyleng;", "%-", "extern FILE *yyin, *yyout;", "%*", "", "#ifdef __cplusplus", "extern \"C\" {", "#endif", " extern int yywrap YY_PROTO(( void ));", "#ifdef __cplusplus", " }", "#endif", "", "#define EOB_ACT_CONTINUE_SCAN 0", "#define EOB_ACT_END_OF_FILE 1", "#define EOB_ACT_LAST_MATCH 2", "", "/* The funky do-while in the following #define is used to turn the definition", " * int a single C statement (which needs a semi-colon terminator). This", " * avoids problems with code like:", " *", " * if ( condition_holds )", " * yyless( 5 );", " * else", " * do_something_else();", " *", " * Prior to using the do-while the compiler would get upset at the", " * \"else\" because it interpreted the \"if\" statement as being all", " * done when it reached the ';' after the yyless() call.", " */", "", "/* Return all but the first 'n' matched characters back to the input stream. */", "", "#define yyless(n) \\", " do \\", " { \\", " /* Undo effects of setting up yytext. */ \\", " *yy_cp = yy_hold_char; \\", " yy_c_buf_p = yy_cp = yy_bp + n - YY_MORE_ADJ; \\", " YY_DO_BEFORE_ACTION; /* set up yytext again */ \\", " } \\", " while ( 0 )", "", "#define unput(c) yyunput( c, yytext_ptr )", "", "", "struct yy_buffer_state", " {", "%-", " FILE *yy_input_file;", "%+", " istream* yy_input_file;", "%*", "", " char *yy_ch_buf; /* input buffer */", " char *yy_buf_pos; /* current position in input buffer */", "", " /* Size of input buffer in bytes, not including room for EOB", " * characters.", " */", " int yy_buf_size;", "", " /* Number of characters read into yy_ch_buf, not including EOB", " * characters.", " */", " int yy_n_chars;", "", " /* Whether this is an \"interactive\" input source; if so, and", " * if we're using stdio for input, then we want to use getc()", " * instead of fread(), to make sure we stop fetching input after", " * each newline.", " */", " int yy_is_interactive;", "", " /* Whether to try to fill the input buffer when we reach the", " * end of it.", " */", " int yy_fill_buffer;", "", " int yy_buffer_status;", "#define YY_BUFFER_NEW 0", "#define YY_BUFFER_NORMAL 1", " /* When an EOF's been seen but there's still some text to process", " * then we mark the buffer as YY_EOF_PENDING, to indicate that we", " * shouldn't try reading from the input source any more. We might", " * still have a bunch of tokens to match, though, because of", " * possible backing-up.", " *", " * When we actually see the EOF, we change the status to \"new\"", " * (via yyrestart()), so that the user can continue scanning by", " * just pointing yyin at a new input file.", " */", "#define YY_BUFFER_EOF_PENDING 2", " };", "", "%- Standard (non-C++) definition", "static YY_BUFFER_STATE yy_current_buffer = 0;", "%*", "", "/* We provide macros for accessing buffer states in case in the", " * future we want to put the buffer states in a more general", " * \"scanner state\".", " */", "#define YY_CURRENT_BUFFER yy_current_buffer", "", "", "%- Standard (non-C++) definition", "/* yy_hold_char holds the character lost when yytext is formed. */", "static char yy_hold_char;", "", "static int yy_n_chars; /* number of characters read into yy_ch_buf */", "", "", "int yyleng;", "", "/* Points to current character in buffer. */", "static char *yy_c_buf_p = (char *) 0;", "static int yy_init = 1; /* whether we need to initialize */", "static int yy_start = 0; /* start state number */", "", "/* Flag which is used to allow yywrap()'s to do buffer switches", " * instead of setting up a fresh yyin. A bit of a hack ...", " */", "static int yy_did_buffer_switch_on_eof;", "", "static void yyunput YY_PROTO(( int c, char *buf_ptr ));", "void yyrestart YY_PROTO(( FILE *input_file ));", "void yy_switch_to_buffer YY_PROTO(( YY_BUFFER_STATE new_buffer ));", "void yy_load_buffer_state YY_PROTO(( void ));", "YY_BUFFER_STATE yy_create_buffer YY_PROTO(( FILE *file, int size ));", "void yy_delete_buffer YY_PROTO(( YY_BUFFER_STATE b ));", "void yy_init_buffer YY_PROTO(( YY_BUFFER_STATE b, FILE *file ));", "", "static int yy_start_stack_ptr = 0;", "static int yy_start_stack_depth = 0;", "static int *yy_start_stack = 0;", "static void yy_push_state YY_PROTO(( int new_state ));", "static void yy_pop_state YY_PROTO(( void ));", "static int yy_top_state YY_PROTO(( void ));", "%*", "", "static void *yy_flex_alloc YY_PROTO(( unsigned int ));", "static void *yy_flex_realloc YY_PROTO(( void *, unsigned int ));", "static void yy_flex_free YY_PROTO(( void * ));", "", "#define yy_new_buffer yy_create_buffer", "", "%% yytext/yyin/yyout/yy_state_type/yylineno etc. def's & init go here", "", "#ifndef yytext_ptr", "static void yy_flex_strncpy YY_PROTO(( char *, const char *, int ));", "#endif", "", "%- Standard (non-C++) definition", "#ifdef __cplusplus", "static int yyinput YY_PROTO(( void ));", "#else", "static int input YY_PROTO(( void ));", "#endif", "%*", "", "%- Standard (non-C++) definition", "static yy_state_type yy_get_previous_state YY_PROTO(( void ));", "static yy_state_type yy_try_NUL_trans YY_PROTO(( yy_state_type current_state ));", "static int yy_get_next_buffer YY_PROTO(( void ));", "static void yy_fatal_error YY_PROTO(( const char msg[] ));", "%*", "", "/* Done after the current pattern has been matched and before the", " * corresponding action - sets up yytext.", " */", "#define YY_DO_BEFORE_ACTION \\", " yytext_ptr = yy_bp; \\", "%% code to fiddle yytext and yyleng for yymore() goes here", " yy_hold_char = *yy_cp; \\", " *yy_cp = '\\0'; \\", "%% code to copy yytext_ptr to yytext[] goes here, if %array", " yy_c_buf_p = yy_cp;", "", "%% data tables for the DFA and the user's section 1 definitions go here", "", "/* Macros after this point can all be overridden by user definitions in", " * section 1.", " */", "", "#ifdef YY_MALLOC_DECL", "YY_MALLOC_DECL", "#else", "#if __STDC__", "#ifndef __cplusplus", "#include ", "#endif", "#else", "/* Just try to get by without declaring the routines. This will fail", " * miserably on non-ANSI systems for which sizeof(size_t) != sizeof(int)", " * or sizeof(void*) != sizeof(int).", " */", "#endif", "#endif", "", "/* Amount of stuff to slurp up with each read. */", "#ifndef YY_READ_BUF_SIZE", "#define YY_READ_BUF_SIZE 8192", "#endif", "", "/* Copy whatever the last rule matched to the standard output. */", "", "#ifndef ECHO", "%- Standard (non-C++) definition", "/* This used to be an fputs(), but since the string might contain NUL's,", " * we now use fwrite().", " */", "#define ECHO (void) fwrite( yytext, yyleng, 1, yyout )", "%+ C++ definition", "#define ECHO LexerOutput( yytext, yyleng )", "%*", "#endif", "", "/* Gets input and stuffs it into \"buf\". number of characters read, or YY_NULL,", " * is returned in \"result\".", " */", "#ifndef YY_INPUT", "#define YY_INPUT(buf,result,max_size) \\", "%% fread()/read() definition of YY_INPUT goes here unless we're doing C++", "%+ C++ definition", " if ( (result = LexerInput( (char *) buf, max_size )) < 0 ) \\", " YY_FATAL_ERROR( \"input in flex scanner failed\" );", "%*", "#endif", "", "/* No semi-colon after return; correct usage is to write \"yyterminate();\" -", " * we don't want an extra ';' after the \"return\" because that will cause", " * some compilers to complain about unreachable statements.", " */", "#ifndef yyterminate", "#define yyterminate() return YY_NULL", "#endif", "", "/* Number of entries by which start-condition stack grows. */", "#ifndef YY_START_STACK_INCR", "#define YY_START_STACK_INCR 25", "#endif", "", "/* Report a fatal error. */", "#ifndef YY_FATAL_ERROR", "%-", "#define YY_FATAL_ERROR(msg) yy_fatal_error( msg )", "%+", "#define YY_FATAL_ERROR(msg) LexerError( msg )", "%*", "#endif", "", "/* Default declaration of generated scanner - a define so the user can", " * easily add parameters.", " */", "#ifndef YY_DECL", "%- Standard (non-C++) definition", "#define YY_DECL int yylex YY_PROTO(( void ))", "%+ C++ definition", "#define YY_DECL int yyFlexLexer::yylex()", "%*", "#endif", "", "/* Code executed at the beginning of each rule, after yytext and yyleng", " * have been set up.", " */", "#ifndef YY_USER_ACTION", "#define YY_USER_ACTION", "#endif", "", "/* Code executed at the end of each rule. */", "#ifndef YY_BREAK", "#define YY_BREAK break;", "#endif", "", "YY_DECL", " {", " register yy_state_type yy_current_state;", " register char *yy_cp, *yy_bp;", " register int yy_act;", "", "%% user's declarations go here", "", " if ( yy_init )", " {", "#ifdef YY_USER_INIT", " YY_USER_INIT;", "#endif", "", " if ( ! yy_start )", " yy_start = 1; /* first start state */", "", " if ( ! yyin )", "%-", " yyin = stdin;", "%+", " yyin = &cin;", "%*", "", " if ( ! yyout )", "%-", " yyout = stdout;", "%+", " yyout = &cout;", "%*", "", " if ( yy_current_buffer )", " yy_init_buffer( yy_current_buffer, yyin );", " else", " yy_current_buffer =", " yy_create_buffer( yyin, YY_BUF_SIZE );", "", " yy_load_buffer_state();", "", " yy_init = 0;", " }", "", " while ( 1 ) /* loops until end-of-file is reached */", " {", "%% yymore()-related code goes here", " yy_cp = yy_c_buf_p;", "", " /* Support of yytext. */", " *yy_cp = yy_hold_char;", "", " /* yy_bp points to the position in yy_ch_buf of the start of", " * the current run.", " */", " yy_bp = yy_cp;", "", "%% code to set up and find next match goes here", "", "yy_find_action:", "%% code to find the action number goes here", "", " YY_DO_BEFORE_ACTION;", "", "%% code for yylineno update goes here, if -l option", "", "do_action: /* This label is used only to access EOF actions. */", "", "%% debug code goes here", "", " switch ( yy_act )", " { /* beginning of action switch */", "%% actions go here", "", " case YY_END_OF_BUFFER:", " {", " /* Amount of text matched not including the EOB char. */", " int yy_amount_of_matched_text = yy_cp - yytext_ptr - 1;", "", " /* Undo the effects of YY_DO_BEFORE_ACTION. */", " *yy_cp = yy_hold_char;", "", " if ( yy_current_buffer->yy_buffer_status == YY_BUFFER_NEW )", " {", " /* We're scanning a new file or input source. It's", " * possible that this happened because the user", " * just pointed yyin at a new source and called", " * yylex(). If so, then we have to assure", " * consistency between yy_current_buffer and our", " * globals. Here is the right place to do so, because", " * this is the first action (other than possibly a", " * back-up) that will match for the new input source.", " */", " yy_n_chars = yy_current_buffer->yy_n_chars;", " yy_current_buffer->yy_input_file = yyin;", " yy_current_buffer->yy_buffer_status = YY_BUFFER_NORMAL;", " }", "", " /* Note that here we test for yy_c_buf_p \"<=\" to the position", " * of the first EOB in the buffer, since yy_c_buf_p will", " * already have been incremented past the NUL character", " * (since all states make transitions on EOB to the", " * end-of-buffer state). Contrast this with the test", " * in input().", " */", " if ( yy_c_buf_p <= &yy_current_buffer->yy_ch_buf[yy_n_chars] )", " { /* This was really a NUL. */", " yy_state_type yy_next_state;", "", " yy_c_buf_p = yytext_ptr + yy_amount_of_matched_text;", "", " yy_current_state = yy_get_previous_state();", "", " /* Okay, we're now positioned to make the NUL", " * transition. We couldn't have", " * yy_get_previous_state() go ahead and do it", " * for us because it doesn't know how to deal", " * with the possibility of jamming (and we don't", " * want to build jamming into it because then it", " * will run more slowly).", " */", "", " yy_next_state = yy_try_NUL_trans( yy_current_state );", "", " yy_bp = yytext_ptr + YY_MORE_ADJ;", "", " if ( yy_next_state )", " {", " /* Consume the NUL. */", " yy_cp = ++yy_c_buf_p;", " yy_current_state = yy_next_state;", " goto yy_match;", " }", "", " else", " {", "%% code to do back-up for compressed tables and set up yy_cp goes here", " goto yy_find_action;", " }", " }", "", " else switch ( yy_get_next_buffer() )", " {", " case EOB_ACT_END_OF_FILE:", " {", " yy_did_buffer_switch_on_eof = 0;", "", " if ( yywrap() )", " {", " /* Note: because we've taken care in", " * yy_get_next_buffer() to have set up", " * yytext, we can now set up", " * yy_c_buf_p so that if some total", " * hoser (like flex itself) wants to", " * call the scanner after we return the", " * YY_NULL, it'll still work - another", " * YY_NULL will get returned.", " */", " yy_c_buf_p = yytext_ptr + YY_MORE_ADJ;", "", " yy_act = YY_STATE_EOF(YY_START);", " goto do_action;", " }", "", " else", " {", " if ( ! yy_did_buffer_switch_on_eof )", " YY_NEW_FILE;", " }", " break;", " }", "", " case EOB_ACT_CONTINUE_SCAN:", " yy_c_buf_p =", " yytext_ptr + yy_amount_of_matched_text;", "", " yy_current_state = yy_get_previous_state();", "", " yy_cp = yy_c_buf_p;", " yy_bp = yytext_ptr + YY_MORE_ADJ;", " goto yy_match;", "", " case EOB_ACT_LAST_MATCH:", " yy_c_buf_p =", " &yy_current_buffer->yy_ch_buf[yy_n_chars];", "", " yy_current_state = yy_get_previous_state();", "", " yy_cp = yy_c_buf_p;", " yy_bp = yytext_ptr + YY_MORE_ADJ;", " goto yy_find_action;", " }", " break;", " }", "", " default:", " YY_FATAL_ERROR(", " \"fatal flex scanner internal error--no action found\" );", " } /* end of action switch */", " } /* end of scanning one token */", " } /* end of yylex */", "", "%+", "#ifdef YY_INTERACTIVE", "int yyFlexLexer::LexerInput( char* buf, int /* max_size */ )", "#else", "int yyFlexLexer::LexerInput( char* buf, int max_size )", "#endif", " {", " if ( yyin->eof() || yyin->fail() )", " return 0;", "", "#ifdef YY_INTERACTIVE", " yyin->get( buf[0] );", "", " if ( yyin->eof() )", " return 0;", "", " if ( yyin->bad() )", " return -1;", "", " return 1;", "", "#else", " (void) yyin->read( buf, max_size );", "", " if ( yyin->bad() )", " return -1;", " else", " return yyin->gcount();", "#endif", " }", "", "void yyFlexLexer::LexerOutput( const char* buf, int size )", " {", " (void) yyout->write( buf, size );", " }", "%*", "", "/* yy_get_next_buffer - try to read in a new buffer", " *", " * Returns a code representing an action:", " * EOB_ACT_LAST_MATCH -", " * EOB_ACT_CONTINUE_SCAN - continue scanning from current position", " * EOB_ACT_END_OF_FILE - end of file", " */", "", "%-", "static int yy_get_next_buffer()", "%+", "int yyFlexLexer::yy_get_next_buffer()", "%*", " {", " register char *dest = yy_current_buffer->yy_ch_buf;", " register char *source = yytext_ptr - 1; /* copy prev. char, too */", " register int number_to_move, i;", " int ret_val;", "", " if ( yy_c_buf_p > &yy_current_buffer->yy_ch_buf[yy_n_chars + 1] )", " YY_FATAL_ERROR(", " \"fatal flex scanner internal error--end of buffer missed\" );", "", " if ( yy_current_buffer->yy_fill_buffer == 0 )", " { /* Don't try to fill the buffer, so this is an EOF. */", " if ( yy_c_buf_p - yytext_ptr - YY_MORE_ADJ == 1 )", " {", " /* We matched a singled characater, the EOB, so", " * treat this as a final EOF.", " */", " return EOB_ACT_END_OF_FILE;", " }", "", " else", " {", " /* We matched some text prior to the EOB, first", " * process it.", " */", " return EOB_ACT_LAST_MATCH;", " }", " }", "", " /* Try to read more data. */", "", " /* First move last chars to start of buffer. */", " number_to_move = yy_c_buf_p - yytext_ptr;", "", " for ( i = 0; i < number_to_move; ++i )", " *(dest++) = *(source++);", "", " if ( yy_current_buffer->yy_buffer_status == YY_BUFFER_EOF_PENDING )", " /* don't do the read, it's not guaranteed to return an EOF,", " * just force an EOF", " */", " yy_n_chars = 0;", "", " else", " {", " int num_to_read =", " yy_current_buffer->yy_buf_size - number_to_move - 1;", "", " while ( num_to_read <= 0 )", " { /* Not enough room in the buffer - grow it. */", "#ifdef YY_USES_REJECT", " YY_FATAL_ERROR(", "\"input buffer overflow, can't enlarge buffer because scanner uses REJECT\" );", "#else", "", " /* just a shorter name for the current buffer */", " YY_BUFFER_STATE b = yy_current_buffer;", "", " int yy_c_buf_p_offset = yy_c_buf_p - b->yy_ch_buf;", "", " b->yy_buf_size *= 2;", " b->yy_ch_buf = (char *)", " yy_flex_realloc( (void *) b->yy_ch_buf,", " b->yy_buf_size );", "", " if ( ! b->yy_ch_buf )", " YY_FATAL_ERROR(", " \"fatal error - scanner input buffer overflow\" );", "", " yy_c_buf_p = &b->yy_ch_buf[yy_c_buf_p_offset];", "", " num_to_read = yy_current_buffer->yy_buf_size -", " number_to_move - 1;", "#endif", " }", "", " if ( num_to_read > YY_READ_BUF_SIZE )", " num_to_read = YY_READ_BUF_SIZE;", "", " /* Read in more data. */", " YY_INPUT( (&yy_current_buffer->yy_ch_buf[number_to_move]),", " yy_n_chars, num_to_read );", " }", "", " if ( yy_n_chars == 0 )", " {", " if ( number_to_move - YY_MORE_ADJ == 1 )", " {", " ret_val = EOB_ACT_END_OF_FILE;", " yyrestart( yyin );", " }", "", " else", " {", " ret_val = EOB_ACT_LAST_MATCH;", " yy_current_buffer->yy_buffer_status =", " YY_BUFFER_EOF_PENDING;", " }", " }", "", " else", " ret_val = EOB_ACT_CONTINUE_SCAN;", "", " yy_n_chars += number_to_move;", " yy_current_buffer->yy_ch_buf[yy_n_chars] = YY_END_OF_BUFFER_CHAR;", " yy_current_buffer->yy_ch_buf[yy_n_chars + 1] = YY_END_OF_BUFFER_CHAR;", "", " /* yytext begins at the second character in yy_ch_buf; the first", " * character is the one which preceded it before reading in the latest", " * buffer; it needs to be kept around in case it's a newline, so", " * yy_get_previous_state() will have with '^' rules active.", " */", "", " yytext_ptr = &yy_current_buffer->yy_ch_buf[1];", "", " return ret_val;", " }", "", "", "/* yy_get_previous_state - get the state just before the EOB char was reached */", "", "%-", "static yy_state_type yy_get_previous_state()", "%+", "yy_state_type yyFlexLexer::yy_get_previous_state()", "%*", " {", " register yy_state_type yy_current_state;", " register char *yy_cp;", "", "%% code to get the start state into yy_current_state goes here", "", " for ( yy_cp = yytext_ptr + YY_MORE_ADJ; yy_cp < yy_c_buf_p; ++yy_cp )", " {", "%% code to find the next state goes here", " }", "", " return yy_current_state;", " }", "", "", "/* yy_try_NUL_trans - try to make a transition on the NUL character", " *", " * synopsis", " * next_state = yy_try_NUL_trans( current_state );", " */", "", "%-", "#ifdef YY_USE_PROTOS", "static yy_state_type yy_try_NUL_trans( yy_state_type yy_current_state )", "#else", "static yy_state_type yy_try_NUL_trans( yy_current_state )", "yy_state_type yy_current_state;", "#endif", "%+", "yy_state_type yyFlexLexer::yy_try_NUL_trans( yy_state_type yy_current_state )", "%*", " {", " register int yy_is_jam;", "%% code to find the next state, and perhaps do backing up, goes here", "", " return yy_is_jam ? 0 : yy_current_state;", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "static void yyunput( int c, register char *yy_bp )", "#else", "static void yyunput( c, yy_bp )", "int c;", "register char *yy_bp;", "#endif", "%+", "void yyFlexLexer::yyunput( int c, register char* yy_bp )", "%*", " {", " register char *yy_cp = yy_c_buf_p;", "", " /* undo effects of setting up yytext */", " *yy_cp = yy_hold_char;", "", " if ( yy_cp < yy_current_buffer->yy_ch_buf + 2 )", " { /* need to shift things up to make room */", " /* +2 for EOB chars. */", " register int number_to_move = yy_n_chars + 2;", " register char *dest = &yy_current_buffer->yy_ch_buf[", " yy_current_buffer->yy_buf_size + 2];", " register char *source =", " &yy_current_buffer->yy_ch_buf[number_to_move];", "", " while ( source > yy_current_buffer->yy_ch_buf )", " *--dest = *--source;", "", " yy_cp += dest - source;", " yy_bp += dest - source;", " yy_n_chars = yy_current_buffer->yy_buf_size;", "", " if ( yy_cp < yy_current_buffer->yy_ch_buf + 2 )", " YY_FATAL_ERROR( \"flex scanner push-back overflow\" );", " }", "", " if ( yy_cp > yy_bp && yy_cp[-1] == '\\n' )", " yy_cp[-2] = '\\n';", "", " *--yy_cp = (char) c;", "", "%% update yylineno here, if doing -l", "", " /* Note: the formal parameter *must* be called \"yy_bp\" for this", " * macro to now work correctly.", " */", " YY_DO_BEFORE_ACTION; /* set up yytext again */", " }", "", "", "%-", "#ifdef __cplusplus", "static int yyinput()", "#else", "static int input()", "#endif", "%+", "int yyFlexLexer::yyinput()", "%*", " {", " int c;", "", " *yy_c_buf_p = yy_hold_char;", "", " if ( *yy_c_buf_p == YY_END_OF_BUFFER_CHAR )", " {", " /* yy_c_buf_p now points to the character we want to return.", " * If this occurs *before* the EOB characters, then it's a", " * valid NUL; if not, then we've hit the end of the buffer.", " */", " if ( yy_c_buf_p < &yy_current_buffer->yy_ch_buf[yy_n_chars] )", " /* This was really a NUL. */", " *yy_c_buf_p = '\\0';", "", " else", " { /* need more input */", " yytext_ptr = yy_c_buf_p;", " ++yy_c_buf_p;", "", " switch ( yy_get_next_buffer() )", " {", " case EOB_ACT_END_OF_FILE:", " {", " if ( yywrap() )", " {", " yy_c_buf_p =", " yytext_ptr + YY_MORE_ADJ;", " return EOF;", " }", "", " YY_NEW_FILE;", "#ifdef __cplusplus", " return yyinput();", "#else", " return input();", "#endif", " }", "", " case EOB_ACT_CONTINUE_SCAN:", " yy_c_buf_p = yytext_ptr + YY_MORE_ADJ;", " break;", "", " case EOB_ACT_LAST_MATCH:", "#ifdef __cplusplus", " YY_FATAL_ERROR(", " \"unexpected last match in yyinput()\" );", "#else", " YY_FATAL_ERROR(", " \"unexpected last match in input()\" );", "#endif", " }", " }", " }", "", " c = *(unsigned char *) yy_c_buf_p; /* cast for 8-bit char's */", " *yy_c_buf_p = '\\0'; /* preserve yytext */", " yy_hold_char = *++yy_c_buf_p;", "", " return c;", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "void yyrestart( FILE *input_file )", "#else", "void yyrestart( input_file )", "FILE *input_file;", "#endif", "%+", "void yyFlexLexer::yyrestart( istream* input_file )", "%*", " {", " if ( ! yy_current_buffer )", " yy_current_buffer = yy_create_buffer( yyin, YY_BUF_SIZE );", "", " yy_init_buffer( yy_current_buffer, input_file );", " yy_load_buffer_state();", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "void yy_switch_to_buffer( YY_BUFFER_STATE new_buffer )", "#else", "void yy_switch_to_buffer( new_buffer )", "YY_BUFFER_STATE new_buffer;", "#endif", "%+", "void yyFlexLexer::yy_switch_to_buffer( YY_BUFFER_STATE new_buffer )", "%*", " {", " if ( yy_current_buffer == new_buffer )", " return;", "", " if ( yy_current_buffer )", " {", " /* Flush out information for old buffer. */", " *yy_c_buf_p = yy_hold_char;", " yy_current_buffer->yy_buf_pos = yy_c_buf_p;", " yy_current_buffer->yy_n_chars = yy_n_chars;", " }", "", " yy_current_buffer = new_buffer;", " yy_load_buffer_state();", "", " /* We don't actually know whether we did this switch during", " * EOF (yywrap()) processing, but the only time this flag", " * is looked at is after yywrap() is called, so it's safe", " * to go ahead and always set it.", " */", " yy_did_buffer_switch_on_eof = 1;", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "void yy_load_buffer_state( void )", "#else", "void yy_load_buffer_state()", "#endif", "%+", "void yyFlexLexer::yy_load_buffer_state()", "%*", " {", " yy_n_chars = yy_current_buffer->yy_n_chars;", " yytext_ptr = yy_c_buf_p = yy_current_buffer->yy_buf_pos;", " yyin = yy_current_buffer->yy_input_file;", " yy_hold_char = *yy_c_buf_p;", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "YY_BUFFER_STATE yy_create_buffer( FILE *file, int size )", "#else", "YY_BUFFER_STATE yy_create_buffer( file, size )", "FILE *file;", "int size;", "#endif", "%+", "YY_BUFFER_STATE yyFlexLexer::yy_create_buffer( istream* file, int size )", "%*", " {", " YY_BUFFER_STATE b;", "", " b = (YY_BUFFER_STATE) yy_flex_alloc( sizeof( struct yy_buffer_state ) );", "", " if ( ! b )", " YY_FATAL_ERROR( \"out of dynamic memory in yy_create_buffer()\" );", "", " b->yy_buf_size = size;", "", " /* yy_ch_buf has to be 2 characters longer than the size given because", " * we need to put in 2 end-of-buffer characters.", " */", " b->yy_ch_buf = (char *) yy_flex_alloc( b->yy_buf_size + 2 );", "", " if ( ! b->yy_ch_buf )", " YY_FATAL_ERROR( \"out of dynamic memory in yy_create_buffer()\" );", "", " yy_init_buffer( b, file );", "", " return b;", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "void yy_delete_buffer( YY_BUFFER_STATE b )", "#else", "void yy_delete_buffer( b )", "YY_BUFFER_STATE b;", "#endif", "%+", "void yyFlexLexer::yy_delete_buffer( YY_BUFFER_STATE b )", "%*", " {", " if ( b == yy_current_buffer )", " yy_current_buffer = (YY_BUFFER_STATE) 0;", "", " yy_flex_free( (void *) b->yy_ch_buf );", " yy_flex_free( (void *) b );", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "void yy_init_buffer( YY_BUFFER_STATE b, FILE *file )", "#else", "void yy_init_buffer( b, file )", "YY_BUFFER_STATE b;", "FILE *file;", "#endif", "%+", "void yyFlexLexer::yy_init_buffer( YY_BUFFER_STATE b, istream* file )", "%*", " {", " b->yy_input_file = file;", "", " /* We put in the '\\n' and start reading from [1] so that an", " * initial match-at-newline will be true.", " */", "", " b->yy_ch_buf[0] = '\\n';", " b->yy_n_chars = 1;", "", " /* We always need two end-of-buffer characters. The first causes", " * a transition to the end-of-buffer state. The second causes", " * a jam in that state.", " */", " b->yy_ch_buf[1] = YY_END_OF_BUFFER_CHAR;", " b->yy_ch_buf[2] = YY_END_OF_BUFFER_CHAR;", "", " b->yy_buf_pos = &b->yy_ch_buf[1];", "", "%-", " b->yy_is_interactive = file ? isatty( fileno(file) ) : 0;", "%+", " b->yy_is_interactive = 0;", "%*", "", " b->yy_fill_buffer = 1;", "", " b->yy_buffer_status = YY_BUFFER_NEW;", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "static void yy_push_state( int new_state )", "#else", "static void yy_push_state( new_state )", "int new_state;", "#endif", "%+", "void yyFlexLexer::yy_push_state( int new_state )", "%*", " {", " if ( yy_start_stack_ptr >= yy_start_stack_depth )", " {", " int new_size;", "", " yy_start_stack_depth += YY_START_STACK_INCR;", " new_size = yy_start_stack_depth * sizeof( int );", "", " if ( ! yy_start_stack )", " yy_start_stack = (int *) yy_flex_alloc( new_size );", "", " else", " yy_start_stack = (int *) yy_flex_realloc(", " (void *) yy_start_stack, new_size );", "", " if ( ! yy_start_stack )", " YY_FATAL_ERROR(", " \"out of memory expanding start-condition stack\" );", " }", "", " yy_start_stack[yy_start_stack_ptr++] = YY_START;", "", " BEGIN(new_state);", " }", "", "", "%-", "static void yy_pop_state()", "%+", "void yyFlexLexer::yy_pop_state()", "%*", " {", " if ( --yy_start_stack_ptr < 0 )", " YY_FATAL_ERROR( \"start-condition stack underflow\" );", "", " BEGIN(yy_start_stack[yy_start_stack_ptr]);", " }", "", "", "%-", "static int yy_top_state()", "%+", "int yyFlexLexer::yy_top_state()", "%*", " {", " return yy_start_stack[yy_start_stack_ptr - 1];", " }", "", "", "%-", "#ifdef YY_USE_PROTOS", "static void yy_fatal_error( const char msg[] )", "#else", "static void yy_fatal_error( msg )", "char msg[];", "#endif", " {", "FILE *err;", /* ### modified(Jan.24.2001) ### */ " (void) fprintf( err, \"%s\\n\", msg );", " exit( 1 );", " }", "", "%+", "", "void yyFlexLexer::LexerError( const char msg[] )", " {", " cerr << msg << '\\n';", " exit( 1 );", " }", "%*", "", "", "/* Redefine yyless() so it works in section 3 code. */", "", "#undef yyless", "#define yyless(n) \\", " do \\", " { \\", " /* Undo effects of setting up yytext. */ \\", " yytext[yyleng] = yy_hold_char; \\", " yy_c_buf_p = yytext + n - YY_MORE_ADJ; \\", " yy_hold_char = *yy_c_buf_p; \\", " *yy_c_buf_p = '\\0'; \\", " yyleng = n; \\", " } \\", " while ( 0 )", "", "", "/* Internal utility routines. */", "", "#ifndef yytext_ptr", "#ifdef YY_USE_PROTOS", "static void yy_flex_strncpy( char *s1, const char *s2, int n )", "#else", "static void yy_flex_strncpy( s1, s2, n )", "char *s1;", "const char *s2;", "int n;", "#endif", " {", " register int i;", " for ( i = 0; i < n; ++i )", " s1[i] = s2[i];", " }", "#endif", "", "", "#ifdef YY_USE_PROTOS", "static void *yy_flex_alloc( unsigned int size )", "#else", "static void *yy_flex_alloc( size )", "unsigned int size;", "#endif", " {", " return (void *) malloc( size );", " }", "", "#ifdef YY_USE_PROTOS", "static void *yy_flex_realloc( void *ptr, unsigned int size )", "#else", "static void *yy_flex_realloc( ptr, size )", "void *ptr;", "unsigned int size;", "#endif", " {", " return (void *) realloc( ptr, size );", " }", "", "#ifdef YY_USE_PROTOS", "static void yy_flex_free( void *ptr )", "#else", "static void yy_flex_free( ptr )", "void *ptr;", "#endif", " {", " free( ptr );", " }", 0 }; /* sym - symbol table routines */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* declare functions that have forward references */ int hashfunct PROTO((register char[], int)); struct hash_entry *ndtbl[NAME_TABLE_HASH_SIZE]; struct hash_entry *sctbl[START_COND_HASH_SIZE]; struct hash_entry *ccltab[CCL_HASH_SIZE]; struct hash_entry *findsym(); /* addsym - add symbol and definitions to symbol table * * -1 is returned if the symbol already exists, and the change not made. */ int addsym( sym, str_def, int_def, table, table_size ) register char sym[]; char *str_def; int int_def; hash_table table; int table_size; { int hash_val = hashfunct( sym, table_size ); register struct hash_entry *sym_entry = table[hash_val]; register struct hash_entry *new_entry; register struct hash_entry *successor; while ( sym_entry ) { if ( ! strcmp( sym, sym_entry->name ) ) { /* entry already exists */ return -1; } sym_entry = sym_entry->next; } /* create new entry */ new_entry = (struct hash_entry *) flex_alloc( sizeof( struct hash_entry ) ); if ( new_entry == NULL ) flexfatal( "symbol table memory allocation failed" ); if ( (successor = table[hash_val]) ) { new_entry->next = successor; successor->prev = new_entry; } else new_entry->next = NULL; new_entry->prev = NULL; new_entry->name = sym; new_entry->str_val = str_def; new_entry->int_val = int_def; table[hash_val] = new_entry; return 0; } /* cclinstal - save the text of a character class */ void cclinstal( ccltxt, cclnum ) Char ccltxt[]; int cclnum; { /* We don't bother checking the return status because we are not * called unless the symbol is new. */ Char *copy_unsigned_string(); (void) addsym( (char *) copy_unsigned_string( ccltxt ), (char *) 0, cclnum, ccltab, CCL_HASH_SIZE ); } /* ccllookup - lookup the number associated with character class text * * Returns 0 if there's no CCL associated with the text. */ int ccllookup( ccltxt ) Char ccltxt[]; { return findsym( (char *) ccltxt, ccltab, CCL_HASH_SIZE )->int_val; } /* findsym - find symbol in symbol table */ struct hash_entry *findsym( sym, table, table_size ) register char sym[]; hash_table table; int table_size; { static struct hash_entry empty_entry = { (struct hash_entry *) 0, (struct hash_entry *) 0, (char *) 0, (char *) 0, 0, } ; register struct hash_entry *sym_entry = table[hashfunct( sym, table_size )]; while ( sym_entry ) { if ( ! strcmp( sym, sym_entry->name ) ) return sym_entry; sym_entry = sym_entry->next; } return &empty_entry; } /* hashfunct - compute the hash value for "str" and hash size "hash_size" */ int hashfunct( str, hash_size ) register char str[]; int hash_size; { register int hashval; register int locstr; hashval = 0; locstr = 0; while ( str[locstr] ) { hashval = (hashval << 1) + (unsigned char) str[locstr++]; hashval %= hash_size; } return hashval; } /* ndinstal - install a name definition */ void ndinstal( name, definition ) char name[]; Char definition[]; { char *copy_string(); Char *copy_unsigned_string(); if ( addsym( copy_string( name ), (char *) copy_unsigned_string( definition ), 0, ndtbl, NAME_TABLE_HASH_SIZE ) ) synerr( "name defined twice" ); } /* ndlookup - lookup a name definition * * Returns a nil pointer if the name definition does not exist. */ Char *ndlookup( nd ) char nd[]; { return (Char *) findsym( nd, ndtbl, NAME_TABLE_HASH_SIZE )->str_val; } /* scextend - increase the maximum number of start conditions */ void scextend() { current_max_scs += MAX_SCS_INCREMENT; ++num_reallocs; scset = reallocate_integer_array( scset, current_max_scs ); scbol = reallocate_integer_array( scbol, current_max_scs ); scxclu = reallocate_integer_array( scxclu, current_max_scs ); sceof = reallocate_integer_array( sceof, current_max_scs ); scname = reallocate_char_ptr_array( scname, current_max_scs ); actvsc = reallocate_integer_array( actvsc, current_max_scs ); } /* scinstal - make a start condition * * NOTE * The start condition is "exclusive" if xcluflg is true. */ void scinstal( str, xcluflg ) char str[]; int xcluflg; { char *copy_string(); /* Generate start condition definition, for use in BEGIN et al. */ printf( "#define %s %d\n", str, lastsc ); if ( ++lastsc >= current_max_scs ) scextend(); scname[lastsc] = copy_string( str ); if ( addsym( scname[lastsc], (char *) 0, lastsc, sctbl, START_COND_HASH_SIZE ) ) format_pinpoint_message( "start condition %s declared twice", str ); scset[lastsc] = mkstate( SYM_EPSILON ); scbol[lastsc] = mkstate( SYM_EPSILON ); scxclu[lastsc] = xcluflg; sceof[lastsc] = false; } /* sclookup - lookup the number associated with a start condition * * Returns 0 if no such start condition. */ int sclookup( str ) char str[]; { return findsym( str, sctbl, START_COND_HASH_SIZE )->int_val; } /* tblcmp - table compression routines */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* declarations for functions that have forward references */ void mkentry PROTO((register int*, int, int, int, int)); void mkprot PROTO((int[], int, int)); void mktemplate PROTO((int[], int, int)); void mv2front PROTO((int)); int tbldiff PROTO((int[], int, int[])); /* bldtbl - build table entries for dfa state * * synopsis * int state[numecs], statenum, totaltrans, comstate, comfreq; * bldtbl( state, statenum, totaltrans, comstate, comfreq ); * * State is the statenum'th dfa state. It is indexed by equivalence class and * gives the number of the state to enter for a given equivalence class. * totaltrans is the total number of transitions out of the state. Comstate * is that state which is the destination of the most transitions out of State. * Comfreq is how many transitions there are out of State to Comstate. * * A note on terminology: * "protos" are transition tables which have a high probability of * either being redundant (a state processed later will have an identical * transition table) or nearly redundant (a state processed later will have * many of the same out-transitions). A "most recently used" queue of * protos is kept around with the hope that most states will find a proto * which is similar enough to be usable, and therefore compacting the * output tables. * "templates" are a special type of proto. If a transition table is * homogeneous or nearly homogeneous (all transitions go to the same * destination) then the odds are good that future states will also go * to the same destination state on basically the same character set. * These homogeneous states are so common when dealing with large rule * sets that they merit special attention. If the transition table were * simply made into a proto, then (typically) each subsequent, similar * state will differ from the proto for two out-transitions. One of these * out-transitions will be that character on which the proto does not go * to the common destination, and one will be that character on which the * state does not go to the common destination. Templates, on the other * hand, go to the common state on EVERY transition character, and therefore * cost only one difference. */ void bldtbl( state, statenum, totaltrans, comstate, comfreq ) int state[], statenum, totaltrans, comstate, comfreq; { int extptr, extrct[2][CSIZE + 1]; int mindiff, minprot, i, d; /* If extptr is 0 then the first array of extrct holds the result * of the "best difference" to date, which is those transitions * which occur in "state" but not in the proto which, to date, * has the fewest differences between itself and "state". If * extptr is 1 then the second array of extrct hold the best * difference. The two arrays are toggled between so that the * best difference to date can be kept around and also a difference * just created by checking against a candidate "best" proto. */ extptr = 0; /* If the state has too few out-transitions, don't bother trying to * compact its tables. */ if ( (totaltrans * 100) < (numecs * PROTO_SIZE_PERCENTAGE) ) mkentry( state, numecs, statenum, JAMSTATE, totaltrans ); else { /* "checkcom" is true if we should only check "state" against * protos which have the same "comstate" value. */ int checkcom = comfreq * 100 > totaltrans * CHECK_COM_PERCENTAGE; minprot = firstprot; mindiff = totaltrans; if ( checkcom ) { /* Find first proto which has the same "comstate". */ for ( i = firstprot; i != NIL; i = protnext[i] ) if ( protcomst[i] == comstate ) { minprot = i; mindiff = tbldiff( state, minprot, extrct[extptr] ); break; } } else { /* Since we've decided that the most common destination * out of "state" does not occur with a high enough * frequency, we set the "comstate" to zero, assuring * that if this state is entered into the proto list, * it will not be considered a template. */ comstate = 0; if ( firstprot != NIL ) { minprot = firstprot; mindiff = tbldiff( state, minprot, extrct[extptr] ); } } /* We now have the first interesting proto in "minprot". If * it matches within the tolerances set for the first proto, * we don't want to bother scanning the rest of the proto list * to see if we have any other reasonable matches. */ if ( mindiff * 100 > totaltrans * FIRST_MATCH_DIFF_PERCENTAGE ) { /* Not a good enough match. Scan the rest of the * protos. */ for ( i = minprot; i != NIL; i = protnext[i] ) { d = tbldiff( state, i, extrct[1 - extptr] ); if ( d < mindiff ) { extptr = 1 - extptr; mindiff = d; minprot = i; } } } /* Check if the proto we've decided on as our best bet is close * enough to the state we want to match to be usable. */ if ( mindiff * 100 > totaltrans * ACCEPTABLE_DIFF_PERCENTAGE ) { /* No good. If the state is homogeneous enough, * we make a template out of it. Otherwise, we * make a proto. */ if ( comfreq * 100 >= totaltrans * TEMPLATE_SAME_PERCENTAGE ) mktemplate( state, statenum, comstate ); else { mkprot( state, statenum, comstate ); mkentry( state, numecs, statenum, JAMSTATE, totaltrans ); } } else { /* use the proto */ mkentry( extrct[extptr], numecs, statenum, prottbl[minprot], mindiff ); /* If this state was sufficiently different from the * proto we built it from, make it, too, a proto. */ if ( mindiff * 100 >= totaltrans * NEW_PROTO_DIFF_PERCENTAGE ) mkprot( state, statenum, comstate ); /* Since mkprot added a new proto to the proto queue, * it's possible that "minprot" is no longer on the * proto queue (if it happened to have been the last * entry, it would have been bumped off). If it's * not there, then the new proto took its physical * place (though logically the new proto is at the * beginning of the queue), so in that case the * following call will do nothing. */ mv2front( minprot ); } } } /* cmptmps - compress template table entries * * Template tables are compressed by using the 'template equivalence * classes', which are collections of transition character equivalence * classes which always appear together in templates - really meta-equivalence * classes. */ void cmptmps() { int tmpstorage[CSIZE + 1]; register int *tmp = tmpstorage, i, j; int totaltrans, trans; peakpairs = numtemps * numecs + tblend; if ( usemecs ) { /* Create equivalence classes based on data gathered on * template transitions. */ nummecs = cre8ecs( tecfwd, tecbck, numecs ); } else nummecs = numecs; while ( lastdfa + numtemps + 1 >= current_max_dfas ) increase_max_dfas(); /* Loop through each template. */ for ( i = 1; i <= numtemps; ++i ) { /* Number of non-jam transitions out of this template. */ totaltrans = 0; for ( j = 1; j <= numecs; ++j ) { trans = tnxt[numecs * i + j]; if ( usemecs ) { /* The absolute value of tecbck is the * meta-equivalence class of a given * equivalence class, as set up by cre8ecs(). */ if ( tecbck[j] > 0 ) { tmp[tecbck[j]] = trans; if ( trans > 0 ) ++totaltrans; } } else { tmp[j] = trans; if ( trans > 0 ) ++totaltrans; } } /* It is assumed (in a rather subtle way) in the skeleton * that if we're using meta-equivalence classes, the def[] * entry for all templates is the jam template, i.e., * templates never default to other non-jam table entries * (e.g., another template) */ /* Leave room for the jam-state after the last real state. */ mkentry( tmp, nummecs, lastdfa + i + 1, JAMSTATE, totaltrans ); } } /* expand_nxt_chk - expand the next check arrays */ void expand_nxt_chk() { register int old_max = current_max_xpairs; current_max_xpairs += MAX_XPAIRS_INCREMENT; ++num_reallocs; nxt = reallocate_integer_array( nxt, current_max_xpairs ); chk = reallocate_integer_array( chk, current_max_xpairs ); zero_out( (char *) (chk + old_max), MAX_XPAIRS_INCREMENT * sizeof( int ) / sizeof( char ) ); } /* find_table_space - finds a space in the table for a state to be placed * * synopsis * int *state, numtrans, block_start; * int find_table_space(); * * block_start = find_table_space( state, numtrans ); * * State is the state to be added to the full speed transition table. * Numtrans is the number of out-transitions for the state. * * find_table_space() returns the position of the start of the first block (in * chk) able to accommodate the state * * In determining if a state will or will not fit, find_table_space() must take * into account the fact that an end-of-buffer state will be added at [0], * and an action number will be added in [-1]. */ int find_table_space( state, numtrans ) int *state, numtrans; { /* Firstfree is the position of the first possible occurrence of two * consecutive unused records in the chk and nxt arrays. */ register int i; register int *state_ptr, *chk_ptr; register int *ptr_to_last_entry_in_state; /* If there are too many out-transitions, put the state at the end of * nxt and chk. */ if ( numtrans > MAX_XTIONS_FULL_INTERIOR_FIT ) { /* If table is empty, return the first available spot in * chk/nxt, which should be 1. */ if ( tblend < 2 ) return 1; /* Start searching for table space near the end of * chk/nxt arrays. */ i = tblend - numecs; } else /* Start searching for table space from the beginning * (skipping only the elements which will definitely not * hold the new state). */ i = firstfree; while ( 1 ) /* loops until a space is found */ { while ( i + numecs >= current_max_xpairs ) expand_nxt_chk(); /* Loops until space for end-of-buffer and action number * are found. */ while ( 1 ) { /* Check for action number space. */ if ( chk[i - 1] == 0 ) { /* Check for end-of-buffer space. */ if ( chk[i] == 0 ) break; else /* Since i != 0, there is no use * checking to see if (++i) - 1 == 0, * because that's the same as i == 0, * so we skip a space. */ i += 2; } else ++i; while ( i + numecs >= current_max_xpairs ) expand_nxt_chk(); } /* If we started search from the beginning, store the new * firstfree for the next call of find_table_space(). */ if ( numtrans <= MAX_XTIONS_FULL_INTERIOR_FIT ) firstfree = i + 1; /* Check to see if all elements in chk (and therefore nxt) * that are needed for the new state have not yet been taken. */ state_ptr = &state[1]; ptr_to_last_entry_in_state = &chk[i + numecs + 1]; for ( chk_ptr = &chk[i + 1]; chk_ptr != ptr_to_last_entry_in_state; ++chk_ptr ) if ( *(state_ptr++) != 0 && *chk_ptr != 0 ) break; if ( chk_ptr == ptr_to_last_entry_in_state ) return i; else ++i; } } /* inittbl - initialize transition tables * * Initializes "firstfree" to be one beyond the end of the table. Initializes * all "chk" entries to be zero. */ void inittbl() { register int i; zero_out( (char *) chk, current_max_xpairs * sizeof( int ) / sizeof( char ) ); tblend = 0; firstfree = tblend + 1; numtemps = 0; if ( usemecs ) { /* Set up doubly-linked meta-equivalence classes; these * are sets of equivalence classes which all have identical * transitions out of TEMPLATES. */ tecbck[1] = NIL; for ( i = 2; i <= numecs; ++i ) { tecbck[i] = i - 1; tecfwd[i - 1] = i; } tecfwd[numecs] = NIL; } } /* mkdeftbl - make the default, "jam" table entries */ void mkdeftbl() { int i; jamstate = lastdfa + 1; ++tblend; /* room for transition on end-of-buffer character */ while ( tblend + numecs >= current_max_xpairs ) expand_nxt_chk(); /* Add in default end-of-buffer transition. */ nxt[tblend] = end_of_buffer_state; chk[tblend] = jamstate; for ( i = 1; i <= numecs; ++i ) { nxt[tblend + i] = 0; chk[tblend + i] = jamstate; } jambase = tblend; base[jamstate] = jambase; def[jamstate] = 0; tblend += numecs; ++numtemps; } /* mkentry - create base/def and nxt/chk entries for transition array * * synopsis * int state[numchars + 1], numchars, statenum, deflink, totaltrans; * mkentry( state, numchars, statenum, deflink, totaltrans ); * * "state" is a transition array "numchars" characters in size, "statenum" * is the offset to be used into the base/def tables, and "deflink" is the * entry to put in the "def" table entry. If "deflink" is equal to * "JAMSTATE", then no attempt will be made to fit zero entries of "state" * (i.e., jam entries) into the table. It is assumed that by linking to * "JAMSTATE" they will be taken care of. In any case, entries in "state" * marking transitions to "SAME_TRANS" are treated as though they will be * taken care of by whereever "deflink" points. "totaltrans" is the total * number of transitions out of the state. If it is below a certain threshold, * the tables are searched for an interior spot that will accommodate the * state array. */ void mkentry( state, numchars, statenum, deflink, totaltrans ) register int *state; int numchars, statenum, deflink, totaltrans; { register int minec, maxec, i, baseaddr; int tblbase, tbllast; if ( totaltrans == 0 ) { /* there are no out-transitions */ if ( deflink == JAMSTATE ) base[statenum] = JAMSTATE; else base[statenum] = 0; def[statenum] = deflink; return; } for ( minec = 1; minec <= numchars; ++minec ) { if ( state[minec] != SAME_TRANS ) if ( state[minec] != 0 || deflink != JAMSTATE ) break; } if ( totaltrans == 1 ) { /* There's only one out-transition. Save it for later to fill * in holes in the tables. */ stack1( statenum, minec, state[minec], deflink ); return; } for ( maxec = numchars; maxec > 0; --maxec ) { if ( state[maxec] != SAME_TRANS ) if ( state[maxec] != 0 || deflink != JAMSTATE ) break; } /* Whether we try to fit the state table in the middle of the table * entries we have already generated, or if we just take the state * table at the end of the nxt/chk tables, we must make sure that we * have a valid base address (i.e., non-negative). Note that * negative base addresses dangerous at run-time (because indexing * the nxt array with one and a low-valued character will access * memory before the start of the array. */ /* Find the first transition of state that we need to worry about. */ if ( totaltrans * 100 <= numchars * INTERIOR_FIT_PERCENTAGE ) { /* Attempt to squeeze it into the middle of the tables. */ baseaddr = firstfree; while ( baseaddr < minec ) { /* Using baseaddr would result in a negative base * address below; find the next free slot. */ for ( ++baseaddr; chk[baseaddr] != 0; ++baseaddr ) ; } while ( baseaddr + maxec - minec + 1 >= current_max_xpairs ) expand_nxt_chk(); for ( i = minec; i <= maxec; ++i ) if ( state[i] != SAME_TRANS && (state[i] != 0 || deflink != JAMSTATE) && chk[baseaddr + i - minec] != 0 ) { /* baseaddr unsuitable - find another */ for ( ++baseaddr; baseaddr < current_max_xpairs && chk[baseaddr] != 0; ++baseaddr ) ; while ( baseaddr + maxec - minec + 1 >= current_max_xpairs ) expand_nxt_chk(); /* Reset the loop counter so we'll start all * over again next time it's incremented. */ i = minec - 1; } } else { /* Ensure that the base address we eventually generate is * non-negative. */ baseaddr = MAX( tblend + 1, minec ); } tblbase = baseaddr - minec; tbllast = tblbase + maxec; while ( tbllast + 1 >= current_max_xpairs ) expand_nxt_chk(); base[statenum] = tblbase; def[statenum] = deflink; for ( i = minec; i <= maxec; ++i ) if ( state[i] != SAME_TRANS ) if ( state[i] != 0 || deflink != JAMSTATE ) { nxt[tblbase + i] = state[i]; chk[tblbase + i] = statenum; } if ( baseaddr == firstfree ) /* Find next free slot in tables. */ for ( ++firstfree; chk[firstfree] != 0; ++firstfree ) ; tblend = MAX( tblend, tbllast ); } /* mk1tbl - create table entries for a state (or state fragment) which * has only one out-transition */ void mk1tbl( state, sym, onenxt, onedef ) int state, sym, onenxt, onedef; { if ( firstfree < sym ) firstfree = sym; while ( chk[firstfree] != 0 ) if ( ++firstfree >= current_max_xpairs ) expand_nxt_chk(); base[state] = firstfree - sym; def[state] = onedef; chk[firstfree] = state; nxt[firstfree] = onenxt; if ( firstfree > tblend ) { tblend = firstfree++; if ( firstfree >= current_max_xpairs ) expand_nxt_chk(); } } /* mkprot - create new proto entry */ void mkprot( state, statenum, comstate ) int state[], statenum, comstate; { int i, slot, tblbase; if ( ++numprots >= MSP || numecs * numprots >= PROT_SAVE_SIZE ) { /* Gotta make room for the new proto by dropping last entry in * the queue. */ slot = lastprot; lastprot = protprev[lastprot]; protnext[lastprot] = NIL; } else slot = numprots; protnext[slot] = firstprot; if ( firstprot != NIL ) protprev[firstprot] = slot; firstprot = slot; prottbl[slot] = statenum; protcomst[slot] = comstate; /* Copy state into save area so it can be compared with rapidly. */ tblbase = numecs * (slot - 1); for ( i = 1; i <= numecs; ++i ) protsave[tblbase + i] = state[i]; } /* mktemplate - create a template entry based on a state, and connect the state * to it */ void mktemplate( state, statenum, comstate ) int state[], statenum, comstate; { int i, numdiff, tmpbase, tmp[CSIZE + 1]; Char transset[CSIZE + 1]; int tsptr; ++numtemps; tsptr = 0; /* Calculate where we will temporarily store the transition table * of the template in the tnxt[] array. The final transition table * gets created by cmptmps(). */ tmpbase = numtemps * numecs; if ( tmpbase + numecs >= current_max_template_xpairs ) { current_max_template_xpairs += MAX_TEMPLATE_XPAIRS_INCREMENT; ++num_reallocs; tnxt = reallocate_integer_array( tnxt, current_max_template_xpairs ); } for ( i = 1; i <= numecs; ++i ) if ( state[i] == 0 ) tnxt[tmpbase + i] = 0; else { transset[tsptr++] = i; tnxt[tmpbase + i] = comstate; } if ( usemecs ) mkeccl( transset, tsptr, tecfwd, tecbck, numecs, 0 ); mkprot( tnxt + tmpbase, -numtemps, comstate ); /* We rely on the fact that mkprot adds things to the beginning * of the proto queue. */ numdiff = tbldiff( state, firstprot, tmp ); mkentry( tmp, numecs, statenum, -numtemps, numdiff ); } /* mv2front - move proto queue element to front of queue */ void mv2front( qelm ) int qelm; { if ( firstprot != qelm ) { if ( qelm == lastprot ) lastprot = protprev[lastprot]; protnext[protprev[qelm]] = protnext[qelm]; if ( protnext[qelm] != NIL ) protprev[protnext[qelm]] = protprev[qelm]; protprev[qelm] = NIL; protnext[qelm] = firstprot; protprev[firstprot] = qelm; firstprot = qelm; } } /* place_state - place a state into full speed transition table * * State is the statenum'th state. It is indexed by equivalence class and * gives the number of the state to enter for a given equivalence class. * Transnum is the number of out-transitions for the state. */ void place_state( state, statenum, transnum ) int *state, statenum, transnum; { register int i; register int *state_ptr; int position = find_table_space( state, transnum ); /* "base" is the table of start positions. */ base[statenum] = position; /* Put in action number marker; this non-zero number makes sure that * find_table_space() knows that this position in chk/nxt is taken * and should not be used for another accepting number in another * state. */ chk[position - 1] = 1; /* Put in end-of-buffer marker; this is for the same purposes as * above. */ chk[position] = 1; /* Place the state into chk and nxt. */ state_ptr = &state[1]; for ( i = 1; i <= numecs; ++i, ++state_ptr ) if ( *state_ptr != 0 ) { chk[position + i] = i; nxt[position + i] = *state_ptr; } if ( position + numecs > tblend ) tblend = position + numecs; } /* stack1 - save states with only one out-transition to be processed later * * If there's room for another state on the "one-transition" stack, the * state is pushed onto it, to be processed later by mk1tbl. If there's * no room, we process the sucker right now. */ void stack1( statenum, sym, nextstate, deflink ) int statenum, sym, nextstate, deflink; { if ( onesp >= ONE_STACK_SIZE - 1 ) mk1tbl( statenum, sym, nextstate, deflink ); else { ++onesp; onestate[onesp] = statenum; onesym[onesp] = sym; onenext[onesp] = nextstate; onedef[onesp] = deflink; } } /* tbldiff - compute differences between two state tables * * "state" is the state array which is to be extracted from the pr'th * proto. "pr" is both the number of the proto we are extracting from * and an index into the save area where we can find the proto's complete * state table. Each entry in "state" which differs from the corresponding * entry of "pr" will appear in "ext". * * Entries which are the same in both "state" and "pr" will be marked * as transitions to "SAME_TRANS" in "ext". The total number of differences * between "state" and "pr" is returned as function value. Note that this * number is "numecs" minus the number of "SAME_TRANS" entries in "ext". */ int tbldiff( state, pr, ext ) int state[], pr, ext[]; { register int i, *sp = state, *ep = ext, *protp; register int numdiff = 0; protp = &protsave[numecs * (pr - 1)]; for ( i = numecs; i > 0; --i ) { if ( *++protp == *++sp ) *++ep = SAME_TRANS; else { *++ep = *sp; ++numdiff; } } return numdiff; } /* yylex - scanner front-end for flex */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* yylex - scan for a regular expression token */ int yylex() { int toktype; static int beglin = false; if ( eofseen ) toktype = EOF; else toktype = flexscan(); if ( toktype == EOF || toktype == 0 ) { eofseen = 1; if ( sectnum == 1 ) { synerr( "premature EOF" ); sectnum = 2; toktype = SECTEND; } else toktype = 0; } if ( trace ) { if ( beglin ) { fprintf( err, "%d\t", num_rules + 1 ); beglin = 0; } switch ( toktype ) { case '<': case '>': case '^': case '$': case '"': case '[': case ']': case '{': case '}': case '|': case '(': case ')': case '-': case '/': case '\\': case '?': case '.': case '*': case '+': case ',': (void) putc( toktype, err ); break; case '\n': (void) putc( '\n', err ); if ( sectnum == 2 ) beglin = 1; break; case SCDECL: fputs( "%s", err ); break; case XSCDECL: fputs( "%x", err ); break; case WHITESPACE: (void) putc( ' ', err ); break; case SECTEND: fputs( "%%\n", err ); /* We set beglin to be true so we'll start * writing out numbers as we echo rules. * flexscan() has already assigned sectnum. */ if ( sectnum == 2 ) beglin = 1; break; case NAME: fprintf( err, "'%s'", nmstr ); break; case CHAR: switch ( yylval ) { case '<': case '>': case '^': case '$': case '"': case '[': case ']': case '{': case '}': case '|': case '(': case ')': case '-': case '/': case '\\': case '?': case '.': case '*': case '+': case ',': fprintf( err, "\\%c", yylval ); break; default: if ( ! isascii( yylval ) || ! isprint( yylval ) ) fprintf( err, "\\%.3o", (unsigned int) yylval ); else (void) putc( yylval, err ); break; } break; case NUMBER: fprintf( err, "%d", yylval ); break; case PREVCCL: fprintf( err, "[%d]", yylval ); break; case EOF_OP: fprintf( err, "<>" ); break; case 0: fprintf( err, "End Marker" ); break; default: fprintf( err, "*Something Weird* - tok: %d val: %d\n", toktype, yylval ); break; } } return toktype; } /* A Bison parser, made from ./parse.y by GNU Bison version 1.28 */ #define YYBISON 1 /* Identify Bison output. */ #define CHAR 257 #define NUMBER 258 #define SECTEND 259 #define SCDECL 260 #define XSCDECL 261 #define WHITESPACE 262 #define NAME 263 #define PREVCCL 264 #define EOF_OP 265 /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ /* Some versions of bison are broken in that they use alloca() but don't * declare it properly. The following is the patented (just kidding!) * #ifdef chud to fix the problem, courtesy of Francois Pinard. */ #ifdef YYBISON /* AIX requires this to be the first thing in the file. */ #ifdef __GNUC__ #define alloca __builtin_alloca #else /* not __GNUC__ */ #if HAVE_ALLOCA_H #include #else /* not HAVE_ALLOCA_H */ #ifdef _AIX #pragma alloca #else /* not _AIX */ char *alloca (); #endif /* not _AIX */ #endif /* not HAVE_ALLOCA_H */ #endif /* not __GNUC__ */ #endif /* YYBISON */ /* Bletch, ^^^^ that was ugly! */ int pat, scnum, eps, headcnt, trailcnt, anyccl, lastchar, i, actvp, rulelen; int trlcontxt, xcluflg, cclsorted, varlength, variable_trail_rule; int *active_ss; Char clower(); void build_eof_action(); void yyerror(); static int madeany = false; /* whether we've made the '.' character class */ int previous_continued_action; /* whether the previous rule's action was '|' */ /* On some over-ambitious machines, such as DEC Alpha's, the default * token type is "long" instead of "int"; this leads to problems with * declaring yylval in flexdef.h. But so far, all the yacc's I've seen * wrap their definitions of YYSTYPE with "#ifndef YYSTYPE"'s, so the * following should ensure that the default token type is "int". */ #define YYSTYPE int #ifndef YYSTYPE #define YYSTYPE int #endif #include #ifndef __cplusplus #ifndef __STDC__ #define const #endif #endif #define YYFINAL 85 #define YYFLAG -32768 #define YYNTBASE 32 #define YYTRANSLATE(x) ((unsigned)(x) <= 265 ? yytranslate[x] : 53) static const char yytranslate[] = { 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 12, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 26, 2, 18, 2, 2, 2, 27, 28, 16, 21, 17, 31, 25, 20, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 14, 2, 15, 22, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 29, 2, 30, 13, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 23, 19, 24, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 3, 4, 5, 6, 7, 8, 9, 10, 11 }; #if YYDEBUG != 0 static const short yyprhs[] = { 0, 0, 6, 7, 13, 14, 17, 19, 21, 23, 27, 29, 31, 36, 37, 38, 42, 45, 48, 50, 53, 55, 57, 61, 65, 69, 70, 73, 75, 77, 80, 84, 87, 89, 93, 95, 98, 101, 103, 106, 109, 112, 119, 125, 130, 132, 134, 136, 140, 144, 146, 150, 155, 160, 163, 164, 167 }; static const short yyrhs[] = { 33, 34, 35, 38, 39, 0, 0, 34, 36, 8, 37, 12, 0, 0, 1, 12, 0, 5, 0, 6, 0, 7, 0, 37, 8, 9, 0, 9, 0, 1, 0, 38, 39, 40, 12, 0, 0, 0, 41, 13, 45, 0, 41, 45, 0, 13, 45, 0, 45, 0, 41, 11, 0, 11, 0, 1, 0, 14, 42, 15, 0, 14, 16, 15, 0, 42, 17, 44, 0, 0, 43, 44, 0, 1, 0, 9, 0, 47, 46, 0, 47, 46, 18, 0, 46, 18, 0, 46, 0, 46, 19, 48, 0, 48, 0, 46, 20, 0, 48, 49, 0, 49, 0, 49, 16, 0, 49, 21, 0, 49, 22, 0, 49, 23, 4, 17, 4, 24, 0, 49, 23, 4, 17, 24, 0, 49, 23, 4, 24, 0, 25, 0, 50, 0, 10, 0, 26, 52, 26, 0, 27, 46, 28, 0, 3, 0, 29, 51, 30, 0, 29, 13, 51, 30, 0, 51, 3, 31, 3, 0, 51, 3, 0, 0, 52, 3, 0, 0 }; #endif #if YYDEBUG != 0 static const short yyrline[] = { 0, 82, 111, 124, 125, 126, 130, 143, 146, 150, 153, 156, 160, 161, 164, 175, 195, 206, 230, 241, 244, 263, 267, 269, 278, 280, 282, 282, 286, 307, 362, 365, 408, 426, 432, 437, 464, 472, 476, 483, 489, 495, 523, 537, 556, 578, 596, 603, 606, 609, 620, 623, 630, 658, 669, 677, 687 }; #endif #if YYDEBUG != 0 || defined (YYERROR_VERBOSE) static const char * const yytname[] = { "$","error","$undefined.","CHAR","NUMBER", "SECTEND","SCDECL","XSCDECL","WHITESPACE","NAME","PREVCCL","EOF_OP","'\\n'", "'^'","'<'","'>'","'*'","','","'$'","'|'","'/'","'+'","'?'","'{'","'}'","'.'", "'\\\"'","'('","')'","'['","']'","'-'","goal","initlex","sect1","sect1end","startconddecl", "namelist1","sect2","initforrule","flexrule","scon","namelist2","@1","sconname", "rule","re","re2","series","singleton","fullccl","ccl","string", NULL }; #endif static const short yyr1[] = { 0, 32, 33, 34, 34, 34, 35, 36, 36, 37, 37, 37, 38, 38, 39, 40, 40, 40, 40, 40, 40, 40, 41, 41, 42, 43, 42, 42, 44, 45, 45, 45, 45, 46, 46, 47, 48, 48, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 49, 50, 50, 51, 51, 51, 52, 52 }; static const short yyr2[] = { 0, 5, 0, 5, 0, 2, 1, 1, 1, 3, 1, 1, 4, 0, 0, 3, 2, 2, 1, 2, 1, 1, 3, 3, 3, 0, 2, 1, 1, 2, 3, 2, 1, 3, 1, 2, 2, 1, 2, 2, 2, 6, 5, 4, 1, 1, 1, 3, 3, 1, 3, 4, 4, 2, 0, 2, 0 }; static const short yydefact[] = { 2, 0, 0, 0, 5, 6, 7, 8, 13, 0, 14, 0, 0, 11, 10, 0, 21, 49, 46, 20, 0, 0, 44, 56, 0, 54, 0, 0, 18, 32, 0, 34, 37, 45, 0, 3, 17, 27, 0, 0, 0, 0, 0, 54, 0, 12, 19, 0, 16, 31, 0, 35, 29, 36, 38, 39, 40, 0, 9, 23, 22, 0, 28, 26, 55, 47, 48, 0, 53, 50, 15, 33, 30, 0, 24, 51, 0, 0, 43, 52, 0, 42, 41, 0, 0, 0 }; static const short yydefgoto[] = { 83, 1, 3, 8, 9, 15, 10, 12, 26, 27, 39, 40, 63, 28, 29, 30, 31, 32, 33, 44, 41 }; static const short yypact[] = {-32768, 61, -2, 64,-32768,-32768,-32768,-32768,-32768, 13,-32768, 34, 1,-32768,-32768, 30,-32768,-32768,-32768,-32768, 21, 48,-32768,-32768, 21, 31, 33, 26,-32768, 54, 21, 21, 38,-32768, 47,-32768,-32768,-32768, 43, 5, 56, 15, -11,-32768, 2,-32768,-32768, 21,-32768,-32768, 21, -32768, 57, 38,-32768,-32768,-32768, 59,-32768,-32768,-32768, 56,-32768,-32768,-32768,-32768,-32768, 4, 46,-32768,-32768, 21,-32768, -8,-32768,-32768, 75, -1,-32768,-32768, 55, -32768,-32768, 80, 81,-32768 }; static const short yypgoto[] = {-32768, -32768,-32768,-32768,-32768,-32768,-32768,-32768,-32768,-32768,-32768, -32768, 22, -14, -5,-32768, 32, -31,-32768, 41,-32768 }; #define YYLAST 84 static const short yytable[] = { 53, -1, 16, 80, 17, 68, 36, 68, 50, 77, 4, 18, 19, 48, 20, 21, 78, 66, 64, 42, 60, 11, 61, 81, 17, 52, 22, 23, 24, 17, 25, 18, 69, 70, 75, 13, 18, 46, 34, 47, 53, 65, 35, 14, 43, 45, 22, 23, 24, 37, 25, 22, 23, 24, 54, 25, 58, -25, 59, 55, 56, 57, 2, 73, 38, 62, -4, -4, -4, 5, 6, 7, 49, 50, 51, 72, 50, 76, 79, 82, 84, 85, 71, 74, 67 }; static const short yycheck[] = { 31, 0, 1, 4, 3, 3, 20, 3, 19, 17, 12, 10, 11, 27, 13, 14, 24, 28, 3, 24, 15, 8, 17, 24, 3, 30, 25, 26, 27, 3, 29, 10, 30, 47, 30, 1, 10, 11, 8, 13, 71, 26, 12, 9, 13, 12, 25, 26, 27, 1, 29, 25, 26, 27, 16, 29, 9, 9, 15, 21, 22, 23, 1, 4, 16, 9, 5, 6, 7, 5, 6, 7, 18, 19, 20, 18, 19, 31, 3, 24, 0, 0, 50, 61, 43 }; /* -*-C-*- Note some compilers choke on comments on `#line' lines. */ /* This file comes from bison-1.28. */ /* Skeleton output parser for bison, Copyright (C) 1984, 1989, 1990 Free Software Foundation, Inc. This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version. This program 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. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ /* As a special exception, when this file is copied by Bison into a Bison output file, you may use that output file without restriction. This special exception was added by the Free Software Foundation in version 1.24 of Bison. */ /* This is the parser code that is written into each bison parser when the %semantic_parser declaration is not specified in the grammar. It was written by Richard Stallman by simplifying the hairy parser used when %semantic_parser is specified. */ #ifndef YYSTACK_USE_ALLOCA #ifdef alloca #define YYSTACK_USE_ALLOCA #else /* alloca not defined */ #ifdef __GNUC__ #define YYSTACK_USE_ALLOCA #define alloca __builtin_alloca #else /* not GNU C. */ #if (!defined (__STDC__) && defined (sparc)) || defined (__sparc__) || defined (__sparc) || defined (__sgi) || (defined (__sun) && defined (__i386)) #define YYSTACK_USE_ALLOCA #include #else /* not sparc */ /* We think this test detects Watcom and Microsoft C. */ /* This used to test MSDOS, but that is a bad idea since that symbol is in the user namespace. */ #if (defined (_MSDOS) || defined (_MSDOS_)) && !defined (__TURBOC__) #if 0 /* No need for malloc.h, which pollutes the namespace; instead, just don't use alloca. */ #include #endif #else /* not MSDOS, or __TURBOC__ */ #if defined(_AIX) /* I don't know what this was needed for, but it pollutes the namespace. So I turned it off. rms, 2 May 1997. */ /* #include */ #pragma alloca #define YYSTACK_USE_ALLOCA #else /* not MSDOS, or __TURBOC__, or _AIX */ #if 0 #ifdef __hpux /* haible@ilog.fr says this works for HPUX 9.05 and up, and on HPUX 10. Eventually we can turn this on. */ #define YYSTACK_USE_ALLOCA #define alloca __builtin_alloca #endif /* __hpux */ #endif #endif /* not _AIX */ #endif /* not MSDOS, or __TURBOC__ */ #endif /* not sparc */ #endif /* not GNU C */ #endif /* alloca not defined */ #endif /* YYSTACK_USE_ALLOCA not defined */ #ifdef YYSTACK_USE_ALLOCA #define YYSTACK_ALLOC alloca #else #define YYSTACK_ALLOC malloc #endif /* Note: there must be only one dollar sign in this file. It is replaced by the list of actions, each action as one case of the switch. */ #define yyerrok (yyerrstatus = 0) #define yyclearin (yychar = YYEMPTY) #define YYEMPTY -2 #define YYEOF 0 #define YYACCEPT goto yyacceptlab #define YYABORT goto yyabortlab #define YYERROR goto yyerrlab1 /* Like YYERROR except do call yyerror. This remains here temporarily to ease the transition to the new meaning of YYERROR, for GCC. Once GCC version 2 has supplanted version 1, this can go. */ #define YYFAIL goto yyerrlab #define YYRECOVERING() (!!yyerrstatus) #define YYBACKUP(token, value) \ do \ if (yychar == YYEMPTY && yylen == 1) \ { yychar = (token), yylval = (value); \ yychar1 = YYTRANSLATE (yychar); \ YYPOPSTACK; \ goto yybackup; \ } \ else \ { yyerror ("syntax error: cannot back up"); YYERROR; } \ while (0) #define YYTERROR 1 #define YYERRCODE 256 #ifndef YYPURE #define YYLEX yylex() #endif #ifdef YYPURE #ifdef YYLSP_NEEDED #ifdef YYLEX_PARAM #define YYLEX yylex(&yylval, &yylloc, YYLEX_PARAM) #else #define YYLEX yylex(&yylval, &yylloc) #endif #else /* not YYLSP_NEEDED */ #ifdef YYLEX_PARAM #define YYLEX yylex(&yylval, YYLEX_PARAM) #else #define YYLEX yylex(&yylval) #endif #endif /* not YYLSP_NEEDED */ #endif /* If nonreentrant, generate the variables here */ #ifndef YYPURE int yychar; /* the lookahead symbol */ YYSTYPE yylval; /* the semantic value of the */ /* lookahead symbol */ #ifdef YYLSP_NEEDED YYLTYPE yylloc; /* location data for the lookahead */ /* symbol */ #endif int yynerrs; /* number of parse errors so far */ #endif /* not YYPURE */ #if YYDEBUG != 0 int yydebug; /* nonzero means print parse trace */ /* Since this is uninitialized, it does not stop multiple parsers from coexisting. */ #endif /* YYINITDEPTH indicates the initial size of the parser's stacks */ #ifndef YYINITDEPTH #define YYINITDEPTH 200 #endif /* YYMAXDEPTH is the maximum size the stacks can grow to (effective only if the built-in stack extension method is used). */ #if YYMAXDEPTH == 0 #undef YYMAXDEPTH #endif #ifndef YYMAXDEPTH #define YYMAXDEPTH 10000 #endif /* Define __yy_memcpy. Note that the size argument should be passed with type unsigned int, because that is what the non-GCC definitions require. With GCC, __builtin_memcpy takes an arg of type size_t, but it can handle unsigned int. */ #if __GNUC__ > 1 /* GNU C and GNU C++ define this. */ #define __yy_memcpy(TO,FROM,COUNT) __builtin_memcpy(TO,FROM,COUNT) #else /* not GNU C or C++ */ #ifndef __cplusplus /* This is the most reliable way to avoid incompatibilities in available built-in functions on various systems. */ static void __yy_memcpy (to, from, count) char *to; char *from; unsigned int count; { register char *f = from; register char *t = to; register int i = count; while (i-- > 0) *t++ = *f++; } #else /* __cplusplus */ /* This is the most reliable way to avoid incompatibilities in available built-in functions on various systems. */ static void __yy_memcpy (char *to, char *from, unsigned int count) { register char *t = to; register char *f = from; register int i = count; while (i-- > 0) *t++ = *f++; } #endif #endif /* The user can define YYPARSE_PARAM as the name of an argument to be passed into yyparse. The argument should have type void *. It should actually point to an object. Grammar actions can access the variable by casting it to the proper pointer type. */ #ifdef YYPARSE_PARAM #ifdef __cplusplus #define YYPARSE_PARAM_ARG void *YYPARSE_PARAM #define YYPARSE_PARAM_DECL #else /* not __cplusplus */ #define YYPARSE_PARAM_ARG YYPARSE_PARAM #define YYPARSE_PARAM_DECL void *YYPARSE_PARAM; #endif /* not __cplusplus */ #else /* not YYPARSE_PARAM */ #define YYPARSE_PARAM_ARG #define YYPARSE_PARAM_DECL #endif /* not YYPARSE_PARAM */ /* Prevent warning if -Wstrict-prototypes. */ #ifdef __GNUC__ #ifdef YYPARSE_PARAM int yyparse (void *); #else int yyparse (void); #endif #endif int yyparse(YYPARSE_PARAM_ARG) YYPARSE_PARAM_DECL { register int yystate; register int yyn; register short *yyssp; register YYSTYPE *yyvsp; int yyerrstatus; /* number of tokens to shift before error messages enabled */ int yychar1 = 0; /* lookahead token as an internal (translated) token number */ short yyssa[YYINITDEPTH]; /* the state stack */ YYSTYPE yyvsa[YYINITDEPTH]; /* the semantic value stack */ short *yyss = yyssa; /* refer to the stacks thru separate pointers */ YYSTYPE *yyvs = yyvsa; /* to allow yyoverflow to reallocate them elsewhere */ #ifdef YYLSP_NEEDED YYLTYPE yylsa[YYINITDEPTH]; /* the location stack */ YYLTYPE *yyls = yylsa; YYLTYPE *yylsp; #define YYPOPSTACK (yyvsp--, yyssp--, yylsp--) #else #define YYPOPSTACK (yyvsp--, yyssp--) #endif int yystacksize = YYINITDEPTH; int yyfree_stacks = 0; #ifdef YYPURE int yychar; YYSTYPE yylval; int yynerrs; #ifdef YYLSP_NEEDED YYLTYPE yylloc; #endif #endif YYSTYPE yyval; /* the variable used to return */ /* semantic values from the action */ /* routines */ int yylen; #if YYDEBUG != 0 if (yydebug) fprintf(err, "Starting parse\n"); #endif yystate = 0; yyerrstatus = 0; yynerrs = 0; yychar = YYEMPTY; /* Cause a token to be read. */ /* Initialize stack pointers. Waste one element of value and location stack so that they stay on the same level as the state stack. The wasted elements are never initialized. */ yyssp = yyss - 1; yyvsp = yyvs; #ifdef YYLSP_NEEDED yylsp = yyls; #endif /* Push a new state, which is found in yystate . */ /* In all cases, when you get here, the value and location stacks have just been pushed. so pushing a state here evens the stacks. */ yynewstate: *++yyssp = yystate; if (yyssp >= yyss + yystacksize - 1) { /* Give user a chance to reallocate the stack */ /* Use copies of these so that the &'s don't force the real ones into memory. */ YYSTYPE *yyvs1 = yyvs; short *yyss1 = yyss; #ifdef YYLSP_NEEDED YYLTYPE *yyls1 = yyls; #endif /* Get the current used size of the three stacks, in elements. */ int size = yyssp - yyss + 1; #ifdef yyoverflow /* Each stack pointer address is followed by the size of the data in use in that stack, in bytes. */ #ifdef YYLSP_NEEDED /* This used to be a conditional around just the two extra args, but that might be undefined if yyoverflow is a macro. */ yyoverflow("parser stack overflow", &yyss1, size * sizeof (*yyssp), &yyvs1, size * sizeof (*yyvsp), &yyls1, size * sizeof (*yylsp), &yystacksize); #else yyoverflow("parser stack overflow", &yyss1, size * sizeof (*yyssp), &yyvs1, size * sizeof (*yyvsp), &yystacksize); #endif yyss = yyss1; yyvs = yyvs1; #ifdef YYLSP_NEEDED yyls = yyls1; #endif #else /* no yyoverflow */ /* Extend the stack our own way. */ if (yystacksize >= YYMAXDEPTH) { yyerror("parser stack overflow"); if (yyfree_stacks) { free (yyss); free (yyvs); #ifdef YYLSP_NEEDED free (yyls); #endif } return 2; } yystacksize *= 2; if (yystacksize > YYMAXDEPTH) yystacksize = YYMAXDEPTH; #ifndef YYSTACK_USE_ALLOCA yyfree_stacks = 1; #endif yyss = (short *) YYSTACK_ALLOC (yystacksize * sizeof (*yyssp)); __yy_memcpy ((char *)yyss, (char *)yyss1, size * (unsigned int) sizeof (*yyssp)); yyvs = (YYSTYPE *) YYSTACK_ALLOC (yystacksize * sizeof (*yyvsp)); __yy_memcpy ((char *)yyvs, (char *)yyvs1, size * (unsigned int) sizeof (*yyvsp)); #ifdef YYLSP_NEEDED yyls = (YYLTYPE *) YYSTACK_ALLOC (yystacksize * sizeof (*yylsp)); __yy_memcpy ((char *)yyls, (char *)yyls1, size * (unsigned int) sizeof (*yylsp)); #endif #endif /* no yyoverflow */ yyssp = yyss + size - 1; yyvsp = yyvs + size - 1; #ifdef YYLSP_NEEDED yylsp = yyls + size - 1; #endif #if YYDEBUG != 0 if (yydebug) fprintf(err, "Stack size increased to %d\n", yystacksize); #endif if (yyssp >= yyss + yystacksize - 1) YYABORT; } #if YYDEBUG != 0 if (yydebug) fprintf(err, "Entering state %d\n", yystate); #endif goto yybackup; yybackup: /* Do appropriate processing given the current state. */ /* Read a lookahead token if we need one and don't already have one. */ /* yyresume: */ /* First try to decide what to do without reference to lookahead token. */ yyn = yypact[yystate]; if (yyn == YYFLAG) goto yydefault; /* Not known => get a lookahead token if don't already have one. */ /* yychar is either YYEMPTY or YYEOF or a valid token in external form. */ if (yychar == YYEMPTY) { #if YYDEBUG != 0 if (yydebug) fprintf(err, "Reading a token: "); #endif yychar = YYLEX; } /* Convert token to internal form (in yychar1) for indexing tables with */ if (yychar <= 0) /* This means end of input. */ { yychar1 = 0; yychar = YYEOF; /* Don't call YYLEX any more */ #if YYDEBUG != 0 if (yydebug) fprintf(err, "Now at end of input.\n"); #endif } else { yychar1 = YYTRANSLATE(yychar); #if YYDEBUG != 0 if (yydebug) { fprintf (err, "Next token is %d (%s", yychar, yytname[yychar1]); /* Give the individual parser a way to print the precise meaning of a token, for further debugging info. */ #ifdef YYPRINT YYPRINT (err, yychar, yylval); #endif fprintf (err, ")\n"); } #endif } yyn += yychar1; if (yyn < 0 || yyn > YYLAST || yycheck[yyn] != yychar1) goto yydefault; yyn = yytable[yyn]; /* yyn is what to do for this token type in this state. Negative => reduce, -yyn is rule number. Positive => shift, yyn is new state. New state is final state => don't bother to shift, just return success. 0, or most negative number => error. */ if (yyn < 0) { if (yyn == YYFLAG) goto yyerrlab; yyn = -yyn; goto yyreduce; } else if (yyn == 0) goto yyerrlab; if (yyn == YYFINAL) YYACCEPT; /* Shift the lookahead token. */ #if YYDEBUG != 0 if (yydebug) fprintf(err, "Shifting token %d (%s), ", yychar, yytname[yychar1]); #endif /* Discard the token being shifted unless it is eof. */ if (yychar != YYEOF) yychar = YYEMPTY; *++yyvsp = yylval; #ifdef YYLSP_NEEDED *++yylsp = yylloc; #endif /* count tokens shifted since error; after three, turn off error status. */ if (yyerrstatus) yyerrstatus--; yystate = yyn; goto yynewstate; /* Do the default action for the current state. */ yydefault: yyn = yydefact[yystate]; if (yyn == 0) goto yyerrlab; /* Do a reduction. yyn is the number of a rule to reduce with. */ yyreduce: yylen = yyr2[yyn]; if (yylen > 0) yyval = yyvsp[1-yylen]; /* implement default value of the action */ #if YYDEBUG != 0 if (yydebug) { int i; fprintf (err, "Reducing via rule %d (line %d), ", yyn, yyrline[yyn]); /* Print the symbols being reduced, and their result. */ for (i = yyprhs[yyn]; yyrhs[i] > 0; i++) fprintf (err, "%s ", yytname[yyrhs[i]]); fprintf (err, " -> %s\n", yytname[yyr1[yyn]]); } #endif switch (yyn) { case 1: { /* add default rule */ int def_rule; pat = cclinit(); cclnegate( pat ); def_rule = mkstate( -pat ); /* Remember the number of the default rule so we * don't generate "can't match" warnings for it. */ default_rule = num_rules; finish_rule( def_rule, false, 0, 0 ); for ( i = 1; i <= lastsc; ++i ) scset[i] = mkbranch( scset[i], def_rule ); if ( spprdflt ) add_action( "YY_FATAL_ERROR( \"flex scanner jammed\" )" ); else add_action( "ECHO" ); add_action( ";\n\tYY_BREAK\n" ); ; break;} case 2: { /* initialize for processing rules */ /* Create default DFA start condition. */ scinstal( "INITIAL", false ); /* Initially, the start condition scoping is * "no start conditions active". */ actvp = 0; ; break;} case 5: { synerr( "unknown error processing section 1" ); ; break;} case 6: { /* We now know how many start conditions there * are, so create the "activity" map indicating * which conditions are active. */ active_ss = allocate_integer_array( lastsc + 1 ); for ( i = 1; i <= lastsc; ++i ) active_ss[i] = 0; ; break;} case 7: { xcluflg = false; ; break;} case 8: { xcluflg = true; ; break;} case 9: { scinstal( nmstr, xcluflg ); ; break;} case 10: { scinstal( nmstr, xcluflg ); ; break;} case 11: { synerr( "bad start condition list" ); ; break;} case 14: { /* Initialize for a parse of one rule. */ trlcontxt = variable_trail_rule = varlength = false; trailcnt = headcnt = rulelen = 0; current_state_type = STATE_NORMAL; previous_continued_action = continued_action; new_rule(); ; break;} case 15: { pat = yyvsp[0]; finish_rule( pat, variable_trail_rule, headcnt, trailcnt ); for ( i = 1; i <= actvp; ++i ) scbol[actvsc[i]] = mkbranch( scbol[actvsc[i]], pat ); if ( ! bol_needed ) { bol_needed = true; if ( performance_report > 1 ) pinpoint_message( "'^' operator results in sub-optimal performance" ); } ; break;} case 16: { pat = yyvsp[0]; finish_rule( pat, variable_trail_rule, headcnt, trailcnt ); for ( i = 1; i <= actvp; ++i ) scset[actvsc[i]] = mkbranch( scset[actvsc[i]], pat ); ; break;} case 17: { pat = yyvsp[0]; finish_rule( pat, variable_trail_rule, headcnt, trailcnt ); /* Add to all non-exclusive start conditions, * including the default (0) start condition. */ for ( i = 1; i <= lastsc; ++i ) if ( ! scxclu[i] ) scbol[i] = mkbranch( scbol[i], pat ); if ( ! bol_needed ) { bol_needed = true; if ( performance_report > 1 ) pinpoint_message( "'^' operator results in sub-optimal performance" ); } ; break;} case 18: { pat = yyvsp[0]; finish_rule( pat, variable_trail_rule, headcnt, trailcnt ); for ( i = 1; i <= lastsc; ++i ) if ( ! scxclu[i] ) scset[i] = mkbranch( scset[i], pat ); ; break;} case 19: { build_eof_action(); ; break;} case 20: { /* This EOF applies to all start conditions * which don't already have EOF actions. */ actvp = 0; for ( i = 1; i <= lastsc; ++i ) if ( ! sceof[i] ) actvsc[++actvp] = i; if ( actvp == 0 ) warn( "all start conditions already have <> rules" ); else build_eof_action(); ; break;} case 21: { synerr( "unrecognized rule" ); ; break;} case 23: { actvp = 0; for ( i = 1; i <= lastsc; ++i ) actvsc[++actvp] = i; ; break;} case 25: { actvp = 0; ; break;} case 27: { synerr( "bad start condition list" ); ; break;} case 28: { if ( (scnum = sclookup( nmstr )) == 0 ) format_pinpoint_message( "undeclared start condition %s", nmstr ); else { if ( ++actvp >= current_max_scs ) /* Some bozo has included multiple * instances of start condition names. */ pinpoint_message( "too many start conditions in <> construct!" ); else actvsc[actvp] = scnum; } ; break;} case 29: { if ( transchar[lastst[yyvsp[0]]] != SYM_EPSILON ) /* Provide final transition \now/ so it * will be marked as a trailing context * state. */ yyvsp[0] = link_machines( yyvsp[0], mkstate( SYM_EPSILON ) ); mark_beginning_as_normal( yyvsp[0] ); current_state_type = STATE_NORMAL; if ( previous_continued_action ) { /* We need to treat this as variable trailing * context so that the backup does not happen * in the action but before the action switch * statement. If the backup happens in the * action, then the rules "falling into" this * one's action will *also* do the backup, * erroneously. */ if ( ! varlength || headcnt != 0 ) warn( "trailing context made variable due to preceding '|' action" ); /* Mark as variable. */ varlength = true; headcnt = 0; } if ( lex_compat || (varlength && headcnt == 0) ) { /* variable trailing context rule */ /* Mark the first part of the rule as the * accepting "head" part of a trailing * context rule. * * By the way, we didn't do this at the * beginning of this production because back * then current_state_type was set up for a * trail rule, and add_accept() can create * a new state ... */ add_accept( yyvsp[-1], num_rules | YY_TRAILING_HEAD_MASK ); variable_trail_rule = true; } else trailcnt = rulelen; yyval = link_machines( yyvsp[-1], yyvsp[0] ); ; break;} case 30: { synerr( "trailing context used twice" ); ; break;} case 31: { headcnt = 0; trailcnt = 1; rulelen = 1; varlength = false; current_state_type = STATE_TRAILING_CONTEXT; if ( trlcontxt ) { synerr( "trailing context used twice" ); yyval = mkstate( SYM_EPSILON ); } else if ( previous_continued_action ) { /* See the comment in the rule for "re2 re" * above. */ warn( "trailing context made variable due to preceding '|' action" ); varlength = true; } if ( lex_compat || varlength ) { /* Again, see the comment in the rule for * "re2 re" above. */ add_accept( yyvsp[-1], num_rules | YY_TRAILING_HEAD_MASK ); variable_trail_rule = true; } trlcontxt = true; eps = mkstate( SYM_EPSILON ); yyval = link_machines( yyvsp[-1], link_machines( eps, mkstate( '\n' ) ) ); ; break;} case 32: { yyval = yyvsp[0]; if ( trlcontxt ) { if ( lex_compat || (varlength && headcnt == 0) ) /* Both head and trail are * variable-length. */ variable_trail_rule = true; else trailcnt = rulelen; } ; break;} case 33: { varlength = true; yyval = mkor( yyvsp[-2], yyvsp[0] ); ; break;} case 34: { yyval = yyvsp[0]; ; break;} case 35: { /* This rule is written separately so the * reduction will occur before the trailing * series is parsed. */ if ( trlcontxt ) synerr( "trailing context used twice" ); else trlcontxt = true; if ( varlength ) /* We hope the trailing context is * fixed-length. */ varlength = false; else headcnt = rulelen; rulelen = 0; current_state_type = STATE_TRAILING_CONTEXT; yyval = yyvsp[-1]; ; break;} case 36: { /* This is where concatenation of adjacent patterns * gets done. */ yyval = link_machines( yyvsp[-1], yyvsp[0] ); ; break;} case 37: { yyval = yyvsp[0]; ; break;} case 38: { varlength = true; yyval = mkclos( yyvsp[-1] ); ; break;} case 39: { varlength = true; yyval = mkposcl( yyvsp[-1] ); ; break;} case 40: { varlength = true; yyval = mkopt( yyvsp[-1] ); ; break;} case 41: { varlength = true; if ( yyvsp[-3] > yyvsp[-1] || yyvsp[-3] < 0 ) { synerr( "bad iteration values" ); yyval = yyvsp[-5]; } else { if ( yyvsp[-3] == 0 ) { if ( yyvsp[-1] <= 0 ) { synerr( "bad iteration values" ); yyval = yyvsp[-5]; } else yyval = mkopt( mkrep( yyvsp[-5], 1, yyvsp[-1] ) ); } else yyval = mkrep( yyvsp[-5], yyvsp[-3], yyvsp[-1] ); } ; break;} case 42: { varlength = true; if ( yyvsp[-2] <= 0 ) { synerr( "iteration value must be positive" ); yyval = yyvsp[-4]; } else yyval = mkrep( yyvsp[-4], yyvsp[-2], INFINITY ); ; break;} case 43: { /* The singleton could be something like "(foo)", * in which case we have no idea what its length * is, so we punt here. */ varlength = true; if ( yyvsp[-1] <= 0 ) { synerr( "iteration value must be positive" ); yyval = yyvsp[-3]; } else yyval = link_machines( yyvsp[-3], copysingl( yyvsp[-3], yyvsp[-1] - 1 ) ); ; break;} case 44: { if ( ! madeany ) { /* Create the '.' character class. */ anyccl = cclinit(); ccladd( anyccl, '\n' ); cclnegate( anyccl ); if ( useecs ) mkeccl( ccltbl + cclmap[anyccl], ccllen[anyccl], nextecm, ecgroup, csize, csize ); madeany = true; } ++rulelen; yyval = mkstate( -anyccl ); ; break;} case 45: { if ( ! cclsorted ) /* Sort characters for fast searching. We * use a shell sort since this list could * be large. */ cshell( ccltbl + cclmap[yyvsp[0]], ccllen[yyvsp[0]], true ); if ( useecs ) mkeccl( ccltbl + cclmap[yyvsp[0]], ccllen[yyvsp[0]], nextecm, ecgroup, csize, csize ); ++rulelen; yyval = mkstate( -yyvsp[0] ); ; break;} case 46: { ++rulelen; yyval = mkstate( -yyvsp[0] ); ; break;} case 47: { yyval = yyvsp[-1]; ; break;} case 48: { yyval = yyvsp[-1]; ; break;} case 49: { ++rulelen; if ( caseins && yyvsp[0] >= 'A' && yyvsp[0] <= 'Z' ) yyvsp[0] = clower( yyvsp[0] ); yyval = mkstate( yyvsp[0] ); ; break;} case 50: { yyval = yyvsp[-1]; ; break;} case 51: { cclnegate( yyvsp[-1] ); yyval = yyvsp[-1]; ; break;} case 52: { if ( caseins ) { if ( yyvsp[-2] >= 'A' && yyvsp[-2] <= 'Z' ) yyvsp[-2] = clower( yyvsp[-2] ); if ( yyvsp[0] >= 'A' && yyvsp[0] <= 'Z' ) yyvsp[0] = clower( yyvsp[0] ); } if ( yyvsp[-2] > yyvsp[0] ) synerr( "negative range in character class" ); else { for ( i = yyvsp[-2]; i <= yyvsp[0]; ++i ) ccladd( yyvsp[-3], i ); /* Keep track if this ccl is staying in * alphabetical order. */ cclsorted = cclsorted && (yyvsp[-2] > lastchar); lastchar = yyvsp[0]; } yyval = yyvsp[-3]; ; break;} case 53: { if ( caseins && yyvsp[0] >= 'A' && yyvsp[0] <= 'Z' ) yyvsp[0] = clower( yyvsp[0] ); ccladd( yyvsp[-1], yyvsp[0] ); cclsorted = cclsorted && (yyvsp[0] > lastchar); lastchar = yyvsp[0]; yyval = yyvsp[-1]; ; break;} case 54: { cclsorted = true; lastchar = 0; yyval = cclinit(); ; break;} case 55: { if ( caseins && yyvsp[0] >= 'A' && yyvsp[0] <= 'Z' ) yyvsp[0] = clower( yyvsp[0] ); ++rulelen; yyval = link_machines( yyvsp[-1], mkstate( yyvsp[0] ) ); ; break;} case 56: { yyval = mkstate( SYM_EPSILON ); ; break;} } /* the action file gets copied in in place of this dollarsign */ yyvsp -= yylen; yyssp -= yylen; #ifdef YYLSP_NEEDED yylsp -= yylen; #endif #if YYDEBUG != 0 if (yydebug) { short *ssp1 = yyss - 1; fprintf (err, "state stack now"); while (ssp1 != yyssp) fprintf (err, " %d", *++ssp1); fprintf (err, "\n"); } #endif *++yyvsp = yyval; #ifdef YYLSP_NEEDED yylsp++; if (yylen == 0) { yylsp->first_line = yylloc.first_line; yylsp->first_column = yylloc.first_column; yylsp->last_line = (yylsp-1)->last_line; yylsp->last_column = (yylsp-1)->last_column; yylsp->text = 0; } else { yylsp->last_line = (yylsp+yylen-1)->last_line; yylsp->last_column = (yylsp+yylen-1)->last_column; } #endif /* Now "shift" the result of the reduction. Determine what state that goes to, based on the state we popped back to and the rule number reduced by. */ yyn = yyr1[yyn]; yystate = yypgoto[yyn - YYNTBASE] + *yyssp; if (yystate >= 0 && yystate <= YYLAST && yycheck[yystate] == *yyssp) yystate = yytable[yystate]; else yystate = yydefgoto[yyn - YYNTBASE]; goto yynewstate; yyerrlab: /* here on detecting error */ if (! yyerrstatus) /* If not already recovering from an error, report this error. */ { ++yynerrs; #ifdef YYERROR_VERBOSE yyn = yypact[yystate]; if (yyn > YYFLAG && yyn < YYLAST) { int size = 0; char *msg; int x, count; count = 0; /* Start X at -yyn if nec to avoid negative indexes in yycheck. */ for (x = (yyn < 0 ? -yyn : 0); x < (sizeof(yytname) / sizeof(char *)); x++) if (yycheck[x + yyn] == x) size += strlen(yytname[x]) + 15, count++; msg = (char *) malloc(size + 15); if (msg != 0) { strcpy(msg, "parse error"); if (count < 5) { count = 0; for (x = (yyn < 0 ? -yyn : 0); x < (sizeof(yytname) / sizeof(char *)); x++) if (yycheck[x + yyn] == x) { strcat(msg, count == 0 ? ", expecting `" : " or `"); strcat(msg, yytname[x]); strcat(msg, "'"); count++; } } yyerror(msg); free(msg); } else yyerror ("parse error; also virtual memory exceeded"); } else #endif /* YYERROR_VERBOSE */ yyerror("parse error"); } goto yyerrlab1; yyerrlab1: /* here on error raised explicitly by an action */ if (yyerrstatus == 3) { /* if just tried and failed to reuse lookahead token after an error, discard it. */ /* return failure if at end of input */ if (yychar == YYEOF) YYABORT; #if YYDEBUG != 0 if (yydebug) fprintf(err, "Discarding token %d (%s).\n", yychar, yytname[yychar1]); #endif yychar = YYEMPTY; } /* Else will try to reuse lookahead token after shifting the error token. */ yyerrstatus = 3; /* Each real token shifted decrements this */ goto yyerrhandle; yyerrdefault: /* current state does not do anything special for the error token. */ #if 0 /* This is wrong; only states that explicitly want error tokens should shift them. */ yyn = yydefact[yystate]; /* If its default is to accept any token, ok. Otherwise pop it.*/ if (yyn) goto yydefault; #endif yyerrpop: /* pop the current state because it cannot handle the error token */ if (yyssp == yyss) YYABORT; yyvsp--; yystate = *--yyssp; #ifdef YYLSP_NEEDED yylsp--; #endif #if YYDEBUG != 0 if (yydebug) { short *ssp1 = yyss - 1; fprintf (err, "Error: state stack now"); while (ssp1 != yyssp) fprintf (err, " %d", *++ssp1); fprintf (err, "\n"); } #endif yyerrhandle: yyn = yypact[yystate]; if (yyn == YYFLAG) goto yyerrdefault; yyn += YYTERROR; if (yyn < 0 || yyn > YYLAST || yycheck[yyn] != YYTERROR) goto yyerrdefault; yyn = yytable[yyn]; if (yyn < 0) { if (yyn == YYFLAG) goto yyerrpop; yyn = -yyn; goto yyreduce; } else if (yyn == 0) goto yyerrpop; if (yyn == YYFINAL) YYACCEPT; #if YYDEBUG != 0 if (yydebug) fprintf(err, "Shifting error token, "); #endif *++yyvsp = yylval; #ifdef YYLSP_NEEDED *++yylsp = yylloc; #endif yystate = yyn; goto yynewstate; yyacceptlab: /* YYACCEPT comes here. */ if (yyfree_stacks) { free (yyss); free (yyvs); #ifdef YYLSP_NEEDED free (yyls); #endif } return 0; yyabortlab: /* YYABORT comes here. */ if (yyfree_stacks) { free (yyss); free (yyvs); #ifdef YYLSP_NEEDED free (yyls); #endif } return 1; } /* build_eof_action - build the "<>" action for the active start * conditions */ void build_eof_action() { register int i; char action_text[MAXLINE]; for ( i = 1; i <= actvp; ++i ) { if ( sceof[actvsc[i]] ) format_pinpoint_message( "multiple <> rules for start condition %s", scname[actvsc[i]] ); else { sceof[actvsc[i]] = true; sprintf( action_text, "case YY_STATE_EOF(%s):\n", scname[actvsc[i]] ); add_action( action_text ); } } line_directive_out( (FILE *) 0 ); /* This isn't a normal rule after all - don't count it as * such, so we don't have any holes in the rule numbering * (which make generating "rule can never match" warnings * more difficult. */ --num_rules; ++num_eof_rules; } /* format_synerr - write out formatted syntax error */ void format_synerr( msg, arg ) char msg[], arg[]; { char errmsg[MAXLINE]; (void) sprintf( errmsg, msg, arg ); synerr( errmsg ); } /* synerr - report a syntax error */ void synerr( str ) char str[]; { syntaxerror = true; pinpoint_message( str ); } /* warn - report a warning, unless -w was given */ void warn( str ) char str[]; { line_warning( str, linenum ); } /* format_pinpoint_message - write out a message formatted with one string, * pinpointing its location */ void format_pinpoint_message( msg, arg ) char msg[], arg[]; { char errmsg[MAXLINE]; (void) sprintf( errmsg, msg, arg ); pinpoint_message( errmsg ); } /* pinpoint_message - write out a message, pinpointing its location */ void pinpoint_message( str ) char str[]; { line_pinpoint( str, linenum ); } /* line_warning - report a warning at a given line, unless -w was given */ void line_warning( str, line ) char str[]; int line; { char warning[MAXLINE]; if ( ! nowarn ) { sprintf( warning, "warning, %s", str ); line_pinpoint( warning, line ); } } /* line_pinpoint - write out a message, pinpointing it at the given line */ void line_pinpoint( str, line ) char str[]; int line; { fprintf( err, "\"%s\", line %d: %s\n", infilename, line, str ); } /* yyerror - eat up an error message from the parser; * currently, messages are ignore */ void yyerror( msg ) char msg[]; { } /* A lexical scanner generated by flex */ /* Scanner skeleton version: * $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ #define FLEX_SCANNER #include /* cfront 1.2 defines "c_plusplus" instead of "__cplusplus" */ #ifdef c_plusplus #ifndef __cplusplus #define __cplusplus #endif #endif #ifdef __cplusplus #include #include /* Use prototypes in function declarations. */ #define YY_USE_PROTOS /* The "const" storage-class-modifier is valid. */ #define YY_USE_CONST #else /* ! __cplusplus */ #ifdef __STDC__ #define YY_USE_PROTOS #define YY_USE_CONST #endif /* __STDC__ */ #endif /* ! __cplusplus */ #ifdef __TURBOC__ #define YY_USE_CONST #endif #ifndef YY_USE_CONST #ifndef const #define const #endif #endif #ifdef YY_USE_PROTOS #define YY_PROTO(proto) proto #else #define YY_PROTO(proto) () #endif /* Returned upon end-of-file. */ #define YY_NULL 0 /* Promotes a possibly negative, possibly signed char to an unsigned * integer for use as an array index. If the signed char is negative, * we want to instead treat it as an 8-bit unsigned char, hence the * double cast. */ #define YY_SC_TO_UI(c) ((unsigned int) (unsigned char) c) /* Enter a start condition. This macro really ought to take a parameter, * but we do it the disgusting crufty way forced on us by the ()-less * definition of BEGIN. */ #define BEGIN yy_start = 1 + 2 * /* Translate the current start state into a value that can be later handed * to BEGIN to return to the state. */ #define YY_START ((yy_start - 1) / 2) /* Action number for EOF rule of a given start state. */ #define YY_STATE_EOF(state) (YY_END_OF_BUFFER + state + 1) /* Special action meaning "start processing a new file". Now included * only for backward compatibility with previous versions of flex. */ #define YY_NEW_FILE yyrestart( yyin ) #define YY_END_OF_BUFFER_CHAR 0 /* Size of default input buffer. */ #define YY_BUF_SIZE 16384 typedef struct yy_buffer_state *YY_BUFFER_STATE; extern int yyleng; extern FILE *yyin, *yyout; #ifdef __cplusplus extern "C" { #endif extern int yywrap YY_PROTO(( void )); #ifdef __cplusplus } #endif #define EOB_ACT_CONTINUE_SCAN 0 #define EOB_ACT_END_OF_FILE 1 #define EOB_ACT_LAST_MATCH 2 /* The funky do-while in the following #define is used to turn the definition * int a single C statement (which needs a semi-colon terminator). This * avoids problems with code like: * * if ( condition_holds ) * yyless( 5 ); * else * do_something_else(); * * Prior to using the do-while the compiler would get upset at the * "else" because it interpreted the "if" statement as being all * done when it reached the ';' after the yyless() call. */ /* Return all but the first 'n' matched characters back to the input stream. */ #define yyless(n) \ do \ { \ /* Undo effects of setting up yytext. */ \ *yy_cp = yy_hold_char; \ yy_c_buf_p = yy_cp = yy_bp + n - YY_MORE_ADJ; \ YY_DO_BEFORE_ACTION; /* set up yytext again */ \ } \ while ( 0 ) #define unput(c) yyunput( c, yytext_ptr ) struct yy_buffer_state { FILE *yy_input_file; char *yy_ch_buf; /* input buffer */ char *yy_buf_pos; /* current position in input buffer */ /* Size of input buffer in bytes, not including room for EOB * characters. */ int yy_buf_size; /* Number of characters read into yy_ch_buf, not including EOB * characters. */ int yy_n_chars; /* Whether this is an "interactive" input source; if so, and * if we're using stdio for input, then we want to use getc() * instead of fread(), to make sure we stop fetching input after * each newline. */ int yy_is_interactive; /* Whether to try to fill the input buffer when we reach the * end of it. */ int yy_fill_buffer; int yy_buffer_status; #define YY_BUFFER_NEW 0 #define YY_BUFFER_NORMAL 1 /* When an EOF's been seen but there's still some text to process * then we mark the buffer as YY_EOF_PENDING, to indicate that we * shouldn't try reading from the input source any more. We might * still have a bunch of tokens to match, though, because of * possible backing-up. * * When we actually see the EOF, we change the status to "new" * (via yyrestart()), so that the user can continue scanning by * just pointing yyin at a new input file. */ #define YY_BUFFER_EOF_PENDING 2 }; static YY_BUFFER_STATE yy_current_buffer = 0; /* We provide macros for accessing buffer states in case in the * future we want to put the buffer states in a more general * "scanner state". */ #define YY_CURRENT_BUFFER yy_current_buffer /* yy_hold_char holds the character lost when yytext is formed. */ static char yy_hold_char; static int yy_n_chars; /* number of characters read into yy_ch_buf */ int yyleng; /* Points to current character in buffer. */ static char *yy_c_buf_p = (char *) 0; static int yy_init = 1; /* whether we need to initialize */ static int yy_start = 0; /* start state number */ /* Flag which is used to allow yywrap()'s to do buffer switches * instead of setting up a fresh yyin. A bit of a hack ... */ static int yy_did_buffer_switch_on_eof; static void yyunput YY_PROTO(( int c, char *buf_ptr )); void yyrestart YY_PROTO(( FILE *input_file )); void yy_switch_to_buffer YY_PROTO(( YY_BUFFER_STATE new_buffer )); void yy_load_buffer_state YY_PROTO(( void )); YY_BUFFER_STATE yy_create_buffer YY_PROTO(( FILE *file, int size )); void yy_delete_buffer YY_PROTO(( YY_BUFFER_STATE b )); void yy_init_buffer YY_PROTO(( YY_BUFFER_STATE b, FILE *file )); static int yy_start_stack_ptr = 0; static int yy_start_stack_depth = 0; static int *yy_start_stack = 0; static void yy_push_state YY_PROTO(( int new_state )); static void yy_pop_state YY_PROTO(( void )); static int yy_top_state YY_PROTO(( void )); static void *yy_flex_alloc YY_PROTO(( unsigned int )); static void *yy_flex_realloc YY_PROTO(( void *, unsigned int )); static void yy_flex_free YY_PROTO(( void * )); #define yy_new_buffer yy_create_buffer #define INITIAL 0 #define SECT2 1 #define SECT2PROLOG 2 #define SECT3 3 #define CODEBLOCK 4 #define PICKUPDEF 5 #define SC 6 #define CARETISBOL 7 #define NUM 8 #define QUOTE 9 #define FIRSTCCL 10 #define CCL 11 #define ACTION 12 #define RECOVER 13 #define BRACEERROR 14 #define C_COMMENT 15 #define ACTION_COMMENT 16 #define ACTION_STRING 17 #define PERCENT_BRACE_ACTION 18 #define USED_LIST 19 #define CODEBLOCK_2 20 typedef unsigned char YY_CHAR; typedef int yy_state_type; FILE *yyin = (FILE *) 0, *yyout = (FILE *) 0; extern char *yytext; #define yytext_ptr yytext #ifndef yytext_ptr static void yy_flex_strncpy YY_PROTO(( char *, const char *, int )); #endif #ifdef __cplusplus static int yyinput YY_PROTO(( void )); #else static int input YY_PROTO(( void )); #endif static yy_state_type yy_get_previous_state YY_PROTO(( void )); static yy_state_type yy_try_NUL_trans YY_PROTO(( yy_state_type current_state )); static int yy_get_next_buffer YY_PROTO(( void )); static void yy_fatal_error YY_PROTO(( const char msg[] )); /* Done after the current pattern has been matched and before the * corresponding action - sets up yytext. */ #define YY_DO_BEFORE_ACTION \ yytext_ptr = yy_bp; \ yyleng = yy_cp - yy_bp; \ yy_hold_char = *yy_cp; \ *yy_cp = '\0'; \ yy_c_buf_p = yy_cp; #define YY_END_OF_BUFFER 113 static const short int yy_accept[408] = { 0, 0, 0, 0, 0, 41, 41, 110, 110, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 113, 111, 6, 17, 111, 15, 1, 16, 111, 111, 111, 14, 60, 53, 54, 60, 47, 60, 59, 60, 60, 60, 60, 44, 43, 60, 60, 45, 46, 41, 42, 41, 40, 39, 40, 40, 110, 110, 26, 27, 26, 26, 26, 26, 26, 26, 29, 28, 30, 29, 65, 61, 62, 64, 66, 80, 81, 80, 78, 77, 79, 67, 69, 67, 68, 67, 72, 72, 72, 74, 76, 74, 74, 74, 75, 92, 97, 92, 96, 98, 98, 93, 93, 93, 90, 91, 111, 31, 111, 83, 111, 82, 20, 22, 20, 21, 101, 102, 101, 100, 103, 105, 103, 106, 107, 88, 88, 89, 88, 88, 88, 88, 88, 88, 36, 33, 32, 36, 36, 36, 88, 6, 17, 0, 17, 15, 1, 16, 0, 16, 13, 7, 0, 0, 0, 3, 0, 4, 0, 2, 14, 53, 54, 0, 0, 0, 54, 50, 50, 0, 0, 57, 0, 108, 108, 108, 49, 48, 49, 44, 43, 0, 43, 56, 44, 41, 42, 40, 39, 39, 37, 38, 110, 110, 26, 27, 26, 26, 26, 26, 29, 28, 30, 63, 64, 81, 77, 69, 109, 109, 109, 70, 71, 76, 73, 92, 97, 0, 95, 0, 94, 93, 93, 93, 0, 31, 0, 31, 31, 83, 20, 22, 18, 101, 102, 101, 102, 102, 99, 103, 105, 104, 88, 88, 88, 89, 85, 88, 88, 88, 36, 33, 32, 36, 36, 84, 13, 7, 0, 12, 0, 0, 0, 0, 3, 0, 0, 4, 0, 5, 0, 51, 0, 52, 0, 0, 57, 0, 57, 57, 108, 108, 49, 49, 58, 56, 37, 38, 26, 26, 26, 23, 26, 0, 109, 109, 93, 93, 0, 19, 0, 85, 85, 88, 88, 36, 36, 12, 0, 0, 0, 3, 0, 0, 4, 5, 5, 52, 52, 0, 57, 57, 57, 57, 108, 26, 26, 23, 23, 0, 109, 93, 93, 19, 19, 88, 88, 36, 36, 0, 0, 0, 10, 0, 57, 57, 57, 57, 26, 26, 93, 93, 88, 88, 36, 36, 0, 0, 0, 0, 57, 57, 24, 25, 86, 87, 86, 87, 34, 35, 0, 9, 0, 0, 11, 55, 9, 9, 0, 0, 8, 0, 8, 8, 0 } ; static const int yy_ec[256] = { 0, 1, 1, 1, 1, 1, 1, 1, 1, 2, 3, 1, 1, 4, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 5, 1, 6, 7, 1, 8, 9, 9, 10, 9, 11, 12, 9, 13, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 1, 1, 15, 1, 16, 9, 1, 22, 23, 24, 25, 26, 27, 21, 21, 28, 29, 30, 21, 31, 32, 33, 34, 21, 35, 36, 37, 38, 21, 21, 39, 40, 21, 17, 18, 19, 20, 21, 1, 22, 23, 24, 25, 26, 27, 21, 21, 28, 29, 30, 21, 31, 32, 33, 34, 21, 35, 36, 37, 38, 21, 21, 39, 40, 21, 41, 42, 43, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 } ; static const int yy_meta[44] = { 0, 1, 2, 3, 1, 4, 1, 1, 5, 1, 6, 1, 7, 5, 8, 1, 1, 1, 9, 10, 1, 11, 12, 12, 12, 12, 12, 12, 11, 11, 11, 11, 11, 11, 11, 11, 11, 13, 11, 11, 11, 5, 1, 14 } ; static const short int yy_base[470] = { 0, 0, 43, 85, 126, 89, 102, 1611, 1610, 168, 1605, 108, 111, 211, 0, 1591, 1590, 252, 254, 116, 119, 98, 122, 144, 146, 297, 0, 93, 104, 338, 340, 149, 151, 257, 266, 268, 274, 383, 0, 425, 428, 1596, 1595, 1607, 1615, 278, 1602, 1602, 0, 281, 1600, 1600, 462, 1592, 0, 1615, 431, 1597, 1597, 1615, 285, 1615, 1584, 1580, 331, 503, 437, 1593, 1593, 110, 1580, 1615, 0, 1590, 1590, 0, 1590, 1588, 221, 1587, 1615, 0, 1585, 1585, 1615, 0, 1561, 1546, 1511, 0, 1551, 1543, 1543, 1615, 1615, 1498, 0, 1615, 1615, 1500, 1487, 1615, 1463, 1615, 1615, 1466, 1460, 1615, 332, 1615, 333, 126, 1615, 1411, 1398, 0, 334, 1615, 0, 1383, 1383, 1615, 341, 1371, 0, 1354, 1336, 1615, 1615, 271, 1371, 287, 1370, 1366, 1615, 0, 1362, 1349, 1331, 290, 1335, 347, 1325, 0, 1323, 1310, 1615, 0, 0, 350, 1306, 1287, 1246, 1615, 0, 1249, 1227, 0, 1264, 1261, 1255, 1225, 1197, 1213, 351, 1213, 1213, 1615, 0, 358, 1198, 1193, 1615, 0, 0, 443, 361, 447, 0, 342, 0, 363, 1615, 0, 451, 1191, 1188, 1149, 365, 1615, 1615, 1185, 1181, 1155, 1135, 423, 1615, 1125, 0, 0, 1615, 546, 588, 1121, 1108, 1615, 0, 1615, 0, 1615, 0, 0, 1095, 0, 0, 1088, 1615, 0, 1615, 0, 1061, 1041, 630, 0, 1069, 1615, 1615, 0, 1615, 838, 1615, 1615, 837, 0, 1615, 1615, 1615, 1615, 0, 1615, 434, 1615, 0, 1615, 0, 821, 817, 373, 843, 376, 1615, 842, 1615, 0, 1615, 463, 467, 834, 471, 1615, 833, 1615, 0, 1615, 1615, 0, 441, 793, 1615, 673, 0, 805, 802, 0, 830, 1615, 795, 792, 1615, 0, 0, 594, 818, 817, 597, 784, 790, 0, 775, 786, 0, 473, 807, 478, 1615, 486, 806, 590, 776, 790, 588, 469, 703, 793, 0, 0, 0, 1615, 0, 0, 0, 780, 772, 0, 800, 800, 730, 788, 0, 775, 767, 600, 795, 602, 0, 773, 772, 764, 770, 762, 1615, 610, 772, 751, 0, 740, 745, 0, 1615, 765, 1615, 764, 740, 488, 803, 601, 817, 1615, 742, 730, 1615, 760, 760, 1615, 738, 726, 1615, 756, 735, 723, 733, 721, 714, 716, 726, 1615, 724, 602, 831, 715, 658, 512, 473, 454, 459, 435, 438, 422, 430, 606, 410, 357, 342, 338, 278, 0, 0, 0, 0, 0, 0, 0, 0, 614, 255, 618, 131, 1615, 1615, 1615, 156, 620, 622, 153, 625, 1615, 95, 1615, 858, 872, 886, 900, 914, 928, 942, 956, 970, 984, 998, 1012, 1026, 1040, 1054, 1062, 1075, 1081, 1094, 1108, 1122, 1136, 1150, 1164, 1178, 1186, 1199, 1207, 1220, 1234, 1248, 1262, 1272, 1280, 1293, 1307, 1321, 1335, 1349, 1363, 1371, 1384, 1398, 1412, 1416, 1419, 1432, 1446, 1460, 710, 1474, 1487, 1501, 1515, 711, 1529, 1537, 1544, 712, 743, 1557, 1571 } ; static const short int yy_def[470] = { 0, 407, 407, 408, 408, 409, 410, 411, 411, 407, 9, 412, 412, 407, 13, 413, 413, 414, 414, 415, 415, 416, 416, 417, 417, 407, 25, 418, 418, 413, 413, 419, 419, 420, 420, 421, 421, 407, 37, 422, 422, 37, 37, 407, 407, 407, 407, 407, 423, 407, 407, 407, 424, 407, 425, 407, 407, 407, 407, 407, 407, 407, 407, 426, 427, 407, 407, 407, 407, 407, 407, 407, 428, 407, 428, 429, 430, 429, 429, 431, 407, 432, 407, 432, 407, 433, 433, 433, 432, 434, 407, 407, 434, 407, 407, 407, 435, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 427, 407, 436, 437, 407, 407, 407, 438, 427, 407, 439, 407, 439, 407, 440, 407, 441, 441, 441, 407, 407, 442, 407, 442, 407, 407, 407, 443, 407, 443, 407, 444, 407, 444, 407, 445, 407, 445, 407, 446, 447, 447, 407, 447, 447, 407, 448, 448, 448, 449, 407, 407, 449, 449, 449, 447, 407, 407, 407, 407, 423, 407, 407, 407, 407, 450, 451, 407, 407, 407, 452, 407, 453, 454, 407, 425, 407, 407, 407, 407, 455, 407, 407, 407, 407, 407, 456, 426, 407, 407, 457, 458, 407, 407, 407, 407, 407, 407, 459, 407, 428, 407, 429, 430, 430, 460, 461, 431, 407, 432, 407, 433, 433, 433, 407, 434, 407, 407, 407, 435, 407, 407, 407, 407, 407, 462, 407, 407, 407, 407, 439, 407, 440, 407, 440, 407, 441, 441, 441, 442, 407, 442, 407, 407, 407, 443, 407, 463, 444, 407, 444, 407, 407, 407, 445, 407, 407, 447, 447, 447, 407, 407, 448, 448, 448, 449, 407, 407, 449, 449, 407, 450, 451, 407, 407, 407, 407, 407, 407, 464, 407, 407, 465, 454, 407, 454, 407, 455, 407, 455, 407, 456, 456, 456, 456, 407, 466, 458, 201, 407, 459, 460, 461, 433, 433, 222, 407, 222, 222, 407, 467, 441, 441, 463, 407, 463, 269, 269, 448, 448, 449, 449, 407, 407, 407, 407, 464, 407, 407, 465, 407, 407, 407, 407, 407, 456, 456, 456, 456, 407, 433, 433, 407, 407, 316, 407, 441, 441, 407, 407, 448, 448, 449, 449, 407, 407, 407, 407, 407, 456, 456, 456, 456, 433, 433, 441, 441, 448, 448, 449, 449, 468, 407, 407, 407, 456, 456, 433, 433, 441, 441, 448, 448, 449, 449, 468, 407, 468, 407, 407, 407, 407, 407, 469, 469, 407, 469, 407, 407, 0, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407 } ; static const short int yy_nxt[1659] = { 0, 44, 45, 46, 47, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 44, 44, 44, 44, 49, 50, 51, 44, 44, 52, 44, 44, 44, 44, 44, 53, 44, 44, 44, 44, 44, 44, 44, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 54, 44, 44, 44, 56, 57, 58, 59, 60, 73, 74, 61, 61, 130, 131, 61, 405, 62, 44, 63, 64, 76, 73, 77, 130, 131, 78, 90, 91, 92, 90, 91, 92, 110, 206, 111, 105, 106, 107, 105, 106, 107, 44, 65, 61, 66, 67, 68, 59, 60, 69, 108, 61, 61, 108, 235, 61, 110, 70, 111, 63, 64, 235, 71, 113, 114, 113, 114, 207, 136, 137, 136, 137, 115, 405, 115, 138, 399, 138, 116, 117, 116, 117, 401, 65, 61, 81, 81, 82, 83, 81, 81, 81, 81, 81, 84, 81, 81, 81, 81, 81, 81, 81, 81, 81, 81, 85, 85, 85, 85, 85, 85, 85, 85, 85, 85, 85, 85, 85, 85, 86, 85, 85, 85, 85, 87, 81, 81, 81, 93, 93, 44, 93, 93, 93, 93, 93, 93, 94, 94, 93, 93, 93, 93, 95, 93, 93, 93, 93, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 96, 93, 93, 93, 99, 100, 99, 100, 399, 140, 141, 213, 101, 214, 101, 102, 142, 102, 140, 141, 144, 145, 146, 248, 249, 142, 144, 145, 146, 164, 165, 166, 169, 170, 171, 147, 190, 191, 192, 251, 249, 147, 257, 258, 103, 300, 103, 118, 118, 119, 120, 121, 118, 118, 122, 118, 118, 118, 118, 123, 118, 118, 118, 118, 118, 118, 118, 124, 124, 124, 124, 124, 124, 124, 124, 124, 124, 124, 124, 124, 124, 125, 124, 124, 124, 124, 126, 127, 118, 128, 132, 133, 132, 133, 197, 197, 232, 197, 241, 260, 258, 266, 164, 165, 166, 300, 267, 398, 242, 169, 170, 171, 281, 282, 283, 292, 293, 296, 297, 198, 198, 233, 198, 288, 284, 248, 249, 289, 251, 249, 134, 397, 134, 148, 149, 150, 151, 148, 148, 152, 148, 148, 153, 148, 148, 148, 148, 148, 148, 148, 148, 148, 148, 154, 154, 154, 154, 154, 154, 154, 154, 154, 154, 154, 154, 154, 154, 155, 154, 154, 154, 154, 156, 148, 148, 148, 158, 159, 160, 158, 159, 160, 184, 185, 186, 396, 301, 187, 202, 203, 204, 241, 266, 187, 281, 282, 283, 267, 281, 282, 283, 242, 184, 185, 186, 392, 284, 187, 391, 161, 284, 302, 161, 390, 162, 322, 323, 162, 174, 257, 258, 389, 188, 260, 258, 292, 293, 285, 188, 286, 339, 293, 345, 175, 388, 176, 300, 176, 296, 297, 387, 176, 188, 176, 176, 177, 176, 178, 386, 179, 180, 367, 181, 199, 199, 300, 199, 199, 199, 199, 199, 199, 199, 199, 199, 199, 200, 199, 199, 199, 199, 199, 199, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 199, 199, 199, 305, 305, 385, 305, 305, 305, 305, 305, 305, 305, 305, 306, 305, 306, 305, 305, 305, 305, 305, 305, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 306, 305, 305, 307, 202, 203, 204, 341, 297, 187, 281, 282, 283, 331, 282, 283, 343, 322, 323, 357, 323, 299, 284, 394, 395, 284, 331, 282, 283, 369, 383, 394, 395, 300, 300, 400, 395, 403, 404, 403, 404, 344, 406, 404, 188, 313, 313, 314, 315, 313, 313, 313, 313, 313, 316, 313, 313, 313, 313, 313, 313, 313, 313, 313, 313, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 313, 313, 313, 324, 324, 300, 324, 324, 324, 324, 324, 324, 325, 324, 324, 324, 324, 324, 324, 324, 324, 324, 324, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 324, 324, 324, 346, 304, 318, 347, 300, 304, 318, 347, 346, 346, 346, 346, 346, 346, 316, 316, 300, 352, 316, 316, 316, 316, 316, 382, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 353, 381, 380, 379, 353, 378, 377, 376, 375, 356, 374, 373, 351, 350, 372, 371, 366, 340, 338, 365, 316, 316, 316, 325, 325, 364, 325, 325, 325, 325, 325, 325, 363, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 362, 361, 360, 359, 358, 356, 355, 354, 231, 351, 350, 349, 348, 196, 300, 342, 340, 338, 336, 335, 325, 325, 325, 368, 333, 332, 330, 300, 330, 329, 328, 368, 368, 368, 368, 368, 368, 370, 274, 327, 326, 300, 269, 259, 259, 370, 370, 370, 370, 370, 370, 384, 250, 250, 320, 300, 319, 317, 229, 384, 384, 384, 384, 384, 384, 55, 55, 55, 55, 55, 55, 55, 55, 55, 55, 55, 55, 55, 55, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 72, 75, 75, 75, 75, 75, 75, 75, 75, 75, 75, 75, 75, 75, 75, 79, 79, 79, 79, 79, 79, 79, 79, 79, 79, 79, 79, 79, 79, 89, 89, 89, 89, 89, 89, 89, 89, 89, 89, 89, 89, 89, 89, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 98, 98, 98, 98, 98, 98, 98, 98, 98, 98, 98, 98, 98, 98, 104, 104, 104, 104, 104, 104, 104, 104, 104, 104, 104, 104, 104, 104, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 112, 112, 112, 112, 112, 112, 112, 112, 112, 112, 112, 112, 112, 112, 129, 129, 129, 129, 129, 129, 129, 129, 129, 129, 129, 129, 129, 129, 135, 135, 135, 135, 135, 135, 135, 135, 135, 135, 135, 135, 135, 135, 139, 139, 139, 139, 139, 139, 139, 139, 139, 139, 139, 139, 139, 139, 143, 143, 143, 143, 143, 143, 143, 143, 143, 143, 143, 143, 143, 143, 157, 157, 157, 157, 157, 157, 157, 157, 157, 157, 157, 157, 157, 157, 168, 168, 224, 312, 168, 168, 168, 173, 173, 173, 173, 173, 173, 173, 173, 173, 173, 173, 173, 183, 183, 311, 216, 183, 183, 183, 194, 194, 212, 194, 194, 194, 194, 194, 194, 194, 194, 194, 194, 194, 196, 196, 205, 196, 196, 196, 196, 196, 196, 196, 196, 196, 196, 196, 208, 208, 205, 208, 208, 208, 208, 208, 208, 208, 208, 208, 208, 208, 210, 210, 303, 210, 210, 210, 210, 210, 210, 210, 210, 210, 210, 210, 211, 211, 300, 211, 211, 211, 211, 211, 211, 211, 211, 211, 211, 211, 215, 215, 215, 215, 215, 215, 215, 215, 215, 215, 215, 215, 215, 215, 217, 217, 298, 217, 217, 190, 217, 217, 217, 217, 190, 294, 189, 217, 219, 219, 189, 172, 219, 219, 219, 223, 223, 172, 223, 223, 223, 223, 223, 223, 223, 223, 223, 223, 223, 227, 227, 167, 167, 227, 227, 227, 231, 231, 278, 231, 231, 231, 231, 231, 231, 231, 231, 231, 231, 231, 234, 234, 277, 234, 234, 234, 234, 234, 234, 234, 234, 234, 234, 234, 237, 237, 276, 237, 237, 237, 237, 237, 237, 275, 237, 237, 237, 237, 238, 238, 275, 274, 272, 238, 238, 238, 238, 238, 240, 240, 271, 240, 240, 240, 240, 240, 240, 240, 240, 240, 240, 240, 244, 244, 269, 268, 244, 244, 244, 247, 247, 247, 247, 247, 247, 247, 247, 247, 247, 247, 247, 247, 247, 253, 253, 268, 253, 253, 263, 253, 253, 253, 253, 253, 253, 253, 253, 256, 256, 256, 256, 256, 263, 256, 256, 256, 256, 256, 256, 256, 256, 262, 262, 261, 259, 262, 262, 262, 262, 255, 262, 262, 262, 262, 262, 264, 264, 254, 264, 264, 264, 264, 264, 264, 264, 264, 264, 264, 264, 265, 265, 254, 265, 265, 252, 265, 265, 265, 265, 252, 250, 246, 265, 270, 270, 245, 243, 270, 270, 270, 273, 239, 239, 273, 273, 273, 273, 273, 273, 273, 273, 273, 273, 273, 279, 279, 236, 279, 279, 279, 279, 279, 279, 279, 279, 279, 279, 279, 280, 280, 236, 280, 280, 280, 280, 280, 280, 280, 280, 280, 280, 280, 287, 287, 287, 290, 290, 290, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 295, 295, 295, 295, 295, 295, 295, 295, 295, 295, 295, 295, 295, 295, 299, 299, 230, 299, 299, 299, 299, 299, 299, 230, 299, 299, 299, 299, 305, 305, 229, 305, 305, 305, 305, 305, 305, 305, 305, 305, 305, 308, 308, 228, 308, 308, 308, 308, 308, 308, 308, 308, 308, 308, 308, 309, 309, 228, 309, 309, 309, 309, 309, 309, 309, 309, 309, 309, 309, 310, 310, 226, 310, 310, 310, 310, 310, 310, 310, 310, 310, 310, 310, 321, 321, 321, 321, 321, 321, 321, 321, 321, 321, 321, 321, 321, 321, 334, 334, 225, 225, 334, 334, 334, 337, 337, 224, 222, 337, 337, 337, 393, 393, 393, 393, 393, 393, 393, 393, 393, 393, 393, 393, 393, 393, 402, 402, 402, 402, 402, 402, 402, 402, 402, 402, 402, 402, 402, 402, 221, 220, 218, 218, 216, 209, 212, 209, 209, 193, 205, 205, 195, 193, 189, 189, 182, 172, 172, 167, 167, 407, 163, 163, 97, 97, 88, 80, 80, 43, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407 } ; static const short int yy_chk[1659] = { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 5, 5, 3, 3, 27, 27, 3, 406, 3, 21, 3, 3, 6, 6, 6, 28, 28, 6, 11, 11, 11, 12, 12, 12, 21, 69, 21, 19, 19, 19, 20, 20, 20, 22, 3, 3, 4, 4, 4, 4, 4, 4, 19, 4, 4, 20, 111, 4, 22, 4, 22, 4, 4, 111, 4, 23, 23, 24, 24, 69, 31, 31, 32, 32, 23, 403, 24, 31, 400, 32, 23, 23, 24, 24, 396, 4, 4, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 17, 17, 18, 18, 394, 33, 33, 78, 17, 78, 18, 17, 33, 18, 34, 34, 35, 35, 35, 129, 129, 34, 36, 36, 36, 45, 45, 45, 49, 49, 49, 35, 60, 60, 60, 131, 131, 36, 139, 139, 17, 384, 18, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 29, 29, 30, 30, 64, 108, 110, 116, 122, 141, 141, 149, 164, 164, 164, 383, 149, 382, 122, 169, 169, 169, 176, 176, 176, 181, 181, 188, 188, 64, 108, 110, 116, 179, 176, 247, 247, 179, 249, 249, 29, 381, 30, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 39, 39, 39, 40, 40, 40, 56, 56, 56, 380, 195, 56, 66, 66, 66, 240, 266, 66, 175, 175, 175, 266, 177, 177, 177, 240, 184, 184, 184, 378, 175, 184, 377, 39, 177, 195, 40, 376, 39, 255, 255, 40, 52, 256, 256, 375, 56, 258, 258, 291, 291, 175, 66, 177, 293, 293, 301, 52, 374, 52, 301, 52, 295, 295, 373, 52, 184, 52, 52, 52, 52, 52, 372, 52, 52, 343, 52, 65, 65, 343, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 65, 201, 201, 371, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 201, 202, 202, 202, 297, 297, 202, 281, 281, 281, 284, 284, 284, 300, 321, 321, 323, 323, 300, 281, 379, 379, 284, 331, 331, 331, 345, 367, 393, 393, 345, 367, 395, 395, 401, 401, 402, 402, 300, 404, 404, 202, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 222, 269, 269, 370, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 269, 302, 457, 462, 466, 302, 457, 462, 466, 302, 302, 302, 302, 302, 302, 316, 316, 369, 316, 316, 316, 316, 316, 316, 366, 316, 316, 316, 316, 316, 316, 316, 316, 316, 316, 467, 364, 363, 362, 467, 361, 360, 359, 358, 357, 355, 354, 352, 351, 349, 348, 342, 341, 339, 336, 316, 316, 316, 325, 325, 335, 325, 325, 325, 325, 325, 325, 333, 325, 325, 325, 325, 325, 325, 325, 325, 325, 325, 332, 329, 328, 327, 326, 322, 320, 319, 317, 315, 314, 312, 311, 303, 299, 298, 296, 292, 289, 288, 325, 325, 325, 344, 286, 285, 283, 344, 282, 277, 276, 344, 344, 344, 344, 344, 344, 346, 274, 272, 271, 346, 267, 260, 257, 346, 346, 346, 346, 346, 346, 368, 251, 248, 246, 368, 245, 232, 229, 368, 368, 368, 368, 368, 368, 408, 408, 408, 408, 408, 408, 408, 408, 408, 408, 408, 408, 408, 408, 409, 409, 409, 409, 409, 409, 409, 409, 409, 409, 409, 409, 409, 409, 410, 410, 410, 410, 410, 410, 410, 410, 410, 410, 410, 410, 410, 410, 411, 411, 411, 411, 411, 411, 411, 411, 411, 411, 411, 411, 411, 411, 412, 412, 412, 412, 412, 412, 412, 412, 412, 412, 412, 412, 412, 412, 413, 413, 413, 413, 413, 413, 413, 413, 413, 413, 413, 413, 413, 413, 414, 414, 414, 414, 414, 414, 414, 414, 414, 414, 414, 414, 414, 414, 415, 415, 415, 415, 415, 415, 415, 415, 415, 415, 415, 415, 415, 415, 416, 416, 416, 416, 416, 416, 416, 416, 416, 416, 416, 416, 416, 416, 417, 417, 417, 417, 417, 417, 417, 417, 417, 417, 417, 417, 417, 417, 418, 418, 418, 418, 418, 418, 418, 418, 418, 418, 418, 418, 418, 418, 419, 419, 419, 419, 419, 419, 419, 419, 419, 419, 419, 419, 419, 419, 420, 420, 420, 420, 420, 420, 420, 420, 420, 420, 420, 420, 420, 420, 421, 421, 421, 421, 421, 421, 421, 421, 421, 421, 421, 421, 421, 421, 422, 422, 422, 422, 422, 422, 422, 422, 422, 422, 422, 422, 422, 422, 423, 423, 224, 221, 423, 423, 423, 424, 424, 424, 424, 424, 424, 424, 424, 424, 424, 424, 424, 425, 425, 220, 215, 425, 425, 425, 426, 426, 212, 426, 426, 426, 426, 426, 426, 426, 426, 426, 426, 426, 427, 427, 204, 427, 427, 427, 427, 427, 427, 427, 427, 427, 427, 427, 428, 428, 203, 428, 428, 428, 428, 428, 428, 428, 428, 428, 428, 428, 429, 429, 197, 429, 429, 429, 429, 429, 429, 429, 429, 429, 429, 429, 430, 430, 194, 430, 430, 430, 430, 430, 430, 430, 430, 430, 430, 430, 431, 431, 431, 431, 431, 431, 431, 431, 431, 431, 431, 431, 431, 431, 432, 432, 193, 432, 432, 192, 432, 432, 432, 432, 191, 187, 186, 432, 433, 433, 185, 171, 433, 433, 433, 434, 434, 170, 434, 434, 434, 434, 434, 434, 434, 434, 434, 434, 434, 435, 435, 166, 165, 435, 435, 435, 436, 436, 163, 436, 436, 436, 436, 436, 436, 436, 436, 436, 436, 436, 437, 437, 162, 437, 437, 437, 437, 437, 437, 437, 437, 437, 437, 437, 438, 438, 161, 438, 438, 438, 438, 438, 438, 160, 438, 438, 438, 438, 439, 439, 159, 158, 156, 439, 439, 439, 439, 439, 440, 440, 155, 440, 440, 440, 440, 440, 440, 440, 440, 440, 440, 440, 441, 441, 152, 151, 441, 441, 441, 442, 442, 442, 442, 442, 442, 442, 442, 442, 442, 442, 442, 442, 442, 443, 443, 150, 443, 443, 145, 443, 443, 443, 443, 443, 443, 443, 443, 444, 444, 444, 444, 444, 144, 444, 444, 444, 444, 444, 444, 444, 444, 445, 445, 142, 140, 445, 445, 445, 445, 138, 445, 445, 445, 445, 445, 446, 446, 137, 446, 446, 446, 446, 446, 446, 446, 446, 446, 446, 446, 447, 447, 136, 447, 447, 133, 447, 447, 447, 447, 132, 130, 126, 447, 448, 448, 125, 123, 448, 448, 448, 449, 120, 119, 449, 449, 449, 449, 449, 449, 449, 449, 449, 449, 449, 450, 450, 114, 450, 450, 450, 450, 450, 450, 450, 450, 450, 450, 450, 451, 451, 113, 451, 451, 451, 451, 451, 451, 451, 451, 451, 451, 451, 452, 452, 452, 453, 453, 453, 454, 454, 454, 454, 454, 454, 454, 454, 454, 454, 454, 454, 454, 454, 455, 455, 455, 455, 455, 455, 455, 455, 455, 455, 455, 455, 455, 455, 456, 456, 106, 456, 456, 456, 456, 456, 456, 105, 456, 456, 456, 456, 458, 458, 102, 458, 458, 458, 458, 458, 458, 458, 458, 458, 458, 459, 459, 100, 459, 459, 459, 459, 459, 459, 459, 459, 459, 459, 459, 460, 460, 99, 460, 460, 460, 460, 460, 460, 460, 460, 460, 460, 460, 461, 461, 95, 461, 461, 461, 461, 461, 461, 461, 461, 461, 461, 461, 463, 463, 463, 463, 463, 463, 463, 463, 463, 463, 463, 463, 463, 463, 464, 464, 92, 91, 464, 464, 464, 465, 465, 90, 88, 465, 465, 465, 468, 468, 468, 468, 468, 468, 468, 468, 468, 468, 468, 468, 468, 468, 469, 469, 469, 469, 469, 469, 469, 469, 469, 469, 469, 469, 469, 469, 87, 86, 83, 82, 79, 77, 76, 74, 73, 70, 68, 67, 63, 62, 58, 57, 53, 51, 50, 47, 46, 43, 42, 41, 16, 15, 10, 8, 7, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407, 407 } ; static yy_state_type yy_last_accepting_state; static char *yy_last_accepting_cpos; /* The intent behind this definition is that it'll catch * any uses of REJECT which flex missed. */ #define REJECT reject_used_but_not_detected #define yymore() yymore_used_but_not_detected #define YY_MORE_ADJ 0 char *yytext; /* scan.l - scanner for flex input */ /*- * Copyright (c) 1990 The Regents of the University of California. * All rights reserved. * * This code is derived from software contributed to Berkeley by * Vern Paxson. * * The United States Government has rights in this work pursuant * to contract no. DE-AC03-76SF00098 between the United States * Department of Energy and the University of California. * * Redistribution and use in source and binary forms are permitted provided * that: (1) source distributions retain this entire copyright notice and * comment, and (2) distributions including binaries display the following * acknowledgement: ``This product includes software developed by the * University of California, Berkeley and its contributors'' in the * documentation or other materials provided with the distribution and in * all advertising materials mentioning features or use of this software. * Neither the name of the University nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. */ /* $Header: /a1/cvsadm/cvsroot/expsir/csubjects/flex/versions.alt/versions.orig/v1/flex.c,v 1.1.1.2 2003/09/02 20:21:58 expsir Exp $ */ #define ACTION_ECHO add_action( yytext ) #define MARK_END_OF_PROLOG mark_prolog(); #define YY_DECL \ int flexscan() #define RETURNCHAR \ yylval = (unsigned char) yytext[0]; \ return CHAR; #define RETURNNAME \ strcpy( nmstr, yytext ); \ return NAME; #define PUT_BACK_STRING(str, start) \ for ( i = strlen( str ) - 1; i >= start; --i ) \ unput((str)[i]) #define CHECK_REJECT(str) \ if ( all_upper( str ) ) \ reject = true; #define CHECK_YYMORE(str) \ if ( all_lower( str ) ) \ yymore_used = true; /* Macros after this point can all be overridden by user definitions in * section 1. */ #ifdef YY_MALLOC_DECL YY_MALLOC_DECL #else #if __STDC__ #ifndef __cplusplus #include #endif #else /* Just try to get by without declaring the routines. This will fail * miserably on non-ANSI systems for which sizeof(size_t) != sizeof(int) * or sizeof(void*) != sizeof(int). */ #endif #endif /* Amount of stuff to slurp up with each read. */ #ifndef YY_READ_BUF_SIZE #define YY_READ_BUF_SIZE 8192 #endif /* Copy whatever the last rule matched to the standard output. */ #ifndef ECHO /* This used to be an fputs(), but since the string might contain NUL's, * we now use fwrite(). */ #define ECHO (void) fwrite( yytext, yyleng, 1, yyout ) #endif /* Gets input and stuffs it into "buf". number of characters read, or YY_NULL, * is returned in "result". */ #ifndef YY_INPUT #define YY_INPUT(buf,result,max_size) \ if ( yy_current_buffer->yy_is_interactive ) \ { \ int c = getc( yyin ); \ result = c == EOF ? 0 : 1; \ buf[0] = (char) c; \ } \ else if ( ((result = fread( buf, 1, max_size, yyin )) == 0) \ && ferror( yyin ) ) \ YY_FATAL_ERROR( "input in flex scanner failed" ); #endif /* No semi-colon after return; correct usage is to write "yyterminate();" - * we don't want an extra ';' after the "return" because that will cause * some compilers to complain about unreachable statements. */ #ifndef yyterminate #define yyterminate() return YY_NULL #endif /* Number of entries by which start-condition stack grows. */ #ifndef YY_START_STACK_INCR #define YY_START_STACK_INCR 25 #endif /* Report a fatal error. */ #ifndef YY_FATAL_ERROR #define YY_FATAL_ERROR(msg) yy_fatal_error( msg ) #endif /* Default declaration of generated scanner - a define so the user can * easily add parameters. */ #ifndef YY_DECL #define YY_DECL int yylex YY_PROTO(( void )) #endif /* Code executed at the beginning of each rule, after yytext and yyleng * have been set up. */ #ifndef YY_USER_ACTION #define YY_USER_ACTION #endif /* Code executed at the end of each rule. */ #ifndef YY_BREAK #define YY_BREAK break; #endif YY_DECL { register yy_state_type yy_current_state; register char *yy_cp, *yy_bp; register int yy_act; static int bracelevel, didadef, indented_code, checking_used; int doing_codeblock = false; int i; Char nmdef[MAXLINE], myesc(); if ( yy_init ) { #ifdef YY_USER_INIT YY_USER_INIT; #endif if ( ! yy_start ) yy_start = 1; /* first start state */ if ( ! yyin ) yyin = stdin; if ( ! yyout ) yyout = stdout; if ( yy_current_buffer ) yy_init_buffer( yy_current_buffer, yyin ); else yy_current_buffer = yy_create_buffer( yyin, YY_BUF_SIZE ); yy_load_buffer_state(); yy_init = 0; } while ( 1 ) /* loops until end-of-file is reached */ { yy_cp = yy_c_buf_p; /* Support of yytext. */ *yy_cp = yy_hold_char; /* yy_bp points to the position in yy_ch_buf of the start of * the current run. */ yy_bp = yy_cp; yy_current_state = yy_start; if ( yy_bp[-1] == '\n' ) ++yy_current_state; yy_match: do { register YY_CHAR yy_c = yy_ec[YY_SC_TO_UI(*yy_cp)]; if ( yy_accept[yy_current_state] ) { yy_last_accepting_state = yy_current_state; yy_last_accepting_cpos = yy_cp; } while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state ) { yy_current_state = (int) yy_def[yy_current_state]; if ( yy_current_state >= 408 ) yy_c = yy_meta[(unsigned int) yy_c]; } yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c]; ++yy_cp; } while ( yy_base[yy_current_state] != 1615 ); yy_find_action: yy_act = yy_accept[yy_current_state]; YY_DO_BEFORE_ACTION; do_action: /* This label is used only to access EOF actions. */ switch ( yy_act ) { /* beginning of action switch */ case 0: /* must back up */ /* undo the effects of YY_DO_BEFORE_ACTION */ *yy_cp = yy_hold_char; yy_cp = yy_last_accepting_cpos; yy_current_state = yy_last_accepting_state; goto yy_find_action; case 1: YY_USER_ACTION indented_code = true; BEGIN(CODEBLOCK); YY_BREAK case 2: YY_USER_ACTION ACTION_ECHO; BEGIN(C_COMMENT); YY_BREAK case 3: YY_USER_ACTION return SCDECL; YY_BREAK case 4: YY_USER_ACTION return XSCDECL; YY_BREAK case 5: YY_USER_ACTION { ++linenum; line_directive_out( (FILE *) 0 ); indented_code = false; BEGIN(CODEBLOCK); } YY_BREAK case 6: YY_USER_ACTION return WHITESPACE; YY_BREAK case 7: YY_USER_ACTION { sectnum = 2; bracelevel = 0; mark_defs1(); line_directive_out( (FILE *) 0 ); BEGIN(SECT2PROLOG); return SECTEND; } YY_BREAK case 8: YY_USER_ACTION { if ( lex_compat ) warn( "%pointer incompatible with -l option" ); else yytext_is_array = false; ++linenum; } YY_BREAK case 9: YY_USER_ACTION { if ( C_plus_plus ) warn( "%array incompatible with -+ option" ); else yytext_is_array = true; ++linenum; } YY_BREAK case 10: YY_USER_ACTION { warn( "%used/%unused have been deprecated" ); checking_used = REALLY_USED; BEGIN(USED_LIST); } YY_BREAK case 11: YY_USER_ACTION { warn( "%used/%unused have been deprecated" ); checking_used = REALLY_NOT_USED; BEGIN(USED_LIST); } YY_BREAK case 12: YY_USER_ACTION ++linenum; /* ignore */ YY_BREAK case 13: YY_USER_ACTION synerr( "unrecognized '%' directive" ); YY_BREAK case 14: YY_USER_ACTION { strcpy( nmstr, yytext ); didadef = false; BEGIN(PICKUPDEF); } YY_BREAK case 15: YY_USER_ACTION RETURNNAME; YY_BREAK case 16: YY_USER_ACTION ++linenum; /* allows blank lines in section 1 */ YY_BREAK case 17: YY_USER_ACTION ++linenum; return '\n'; YY_BREAK case 18: YY_USER_ACTION ACTION_ECHO; BEGIN(INITIAL); YY_BREAK case 19: YY_USER_ACTION ++linenum; ACTION_ECHO; BEGIN(INITIAL); YY_BREAK case 20: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 21: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 22: YY_USER_ACTION ++linenum; ACTION_ECHO; YY_BREAK case 23: YY_USER_ACTION ++linenum; BEGIN(INITIAL); YY_BREAK case 24: YY_USER_ACTION ACTION_ECHO; CHECK_REJECT(yytext); YY_BREAK case 25: YY_USER_ACTION ACTION_ECHO; CHECK_YYMORE(yytext); YY_BREAK case 26: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 27: YY_USER_ACTION { ++linenum; ACTION_ECHO; if ( indented_code ) BEGIN(INITIAL); } YY_BREAK case 28: YY_USER_ACTION /* separates name and definition */ YY_BREAK case 29: YY_USER_ACTION { strcpy( (char *) nmdef, yytext ); /* Skip trailing whitespace. */ for ( i = strlen( (char *) nmdef ) - 1; i >= 0 && (nmdef[i] == ' ' || nmdef[i] == '\t'); --i ) ; nmdef[i + 1] = '\0'; ndinstal( nmstr, nmdef ); didadef = true; } YY_BREAK case 30: YY_USER_ACTION { if ( ! didadef ) synerr( "incomplete name definition" ); BEGIN(INITIAL); ++linenum; } YY_BREAK case 31: YY_USER_ACTION ++linenum; BEGIN(INITIAL); RETURNNAME; YY_BREAK case 32: YY_USER_ACTION ++linenum; BEGIN(INITIAL); YY_BREAK case 33: YY_USER_ACTION YY_BREAK case 34: YY_USER_ACTION { if ( all_upper( yytext ) ) reject_really_used = checking_used; else synerr( "unrecognized %used/%unused construct" ); } YY_BREAK case 35: YY_USER_ACTION { if ( all_lower( yytext ) ) yymore_really_used = checking_used; else synerr( "unrecognized %used/%unused construct" ); } YY_BREAK case 36: YY_USER_ACTION synerr( "unrecognized %used/%unused construct" ); YY_BREAK case 37: YY_USER_ACTION ++bracelevel; yyless( 2 ); /* eat only %{ */ YY_BREAK case 38: YY_USER_ACTION --bracelevel; yyless( 2 ); /* eat only %} */ YY_BREAK case 39: YY_USER_ACTION ACTION_ECHO; /* indented code in prolog */ YY_BREAK case 40: YY_USER_ACTION { /* non-indented code */ if ( bracelevel <= 0 ) { /* not in %{ ... %} */ yyless( 0 ); /* put it all back */ mark_prolog(); BEGIN(SECT2); } else ACTION_ECHO; } YY_BREAK case 41: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 42: YY_USER_ACTION ++linenum; ACTION_ECHO; YY_BREAK case YY_STATE_EOF(SECT2PROLOG): { mark_prolog(); sectnum = 0; yyterminate(); /* to stop the parser */ } YY_BREAK case 43: YY_USER_ACTION ++linenum; /* allow blank lines in section 2 */ YY_BREAK case 44: YY_USER_ACTION { indented_code = (yytext[0] != '%'); doing_codeblock = true; bracelevel = 1; if ( indented_code ) ACTION_ECHO; BEGIN(CODEBLOCK_2); } YY_BREAK case 45: YY_USER_ACTION BEGIN(SC); return '<'; YY_BREAK case 46: YY_USER_ACTION return '^'; YY_BREAK case 47: YY_USER_ACTION BEGIN(QUOTE); return '"'; YY_BREAK case 48: *yy_cp = yy_hold_char; /* undo effects of setting up yytext */ yy_c_buf_p = yy_cp = yy_bp + 1; YY_DO_BEFORE_ACTION; /* set up yytext again */ YY_USER_ACTION BEGIN(NUM); return '{'; YY_BREAK case 49: YY_USER_ACTION BEGIN(BRACEERROR); YY_BREAK case 50: *yy_cp = yy_hold_char; /* undo effects of setting up yytext */ yy_c_buf_p = yy_cp = yy_bp + 1; YY_DO_BEFORE_ACTION; /* set up yytext again */ YY_USER_ACTION return '$'; YY_BREAK case 51: YY_USER_ACTION { bracelevel = 1; BEGIN(PERCENT_BRACE_ACTION); return '\n'; } YY_BREAK case 52: YY_USER_ACTION continued_action = true; ++linenum; return '\n'; YY_BREAK case 53: YY_USER_ACTION { /* This rule is separate from the one below because * otherwise we get variable trailing context, so * we can't build the scanner using -{f,F}. */ bracelevel = 0; continued_action = false; BEGIN(ACTION); return '\n'; } YY_BREAK case 54: YY_USER_ACTION { bracelevel = 0; continued_action = false; BEGIN(ACTION); unput( '\n' ); /* so sees it */ return '\n'; } YY_BREAK case 55: YY_USER_ACTION return EOF_OP; YY_BREAK case 56: YY_USER_ACTION { sectnum = 3; BEGIN(SECT3); yyterminate(); /* to stop the parser */ } YY_BREAK case 57: YY_USER_ACTION { int cclval; strcpy( nmstr, yytext ); /* Check to see if we've already encountered this * ccl. */ if ( (cclval = ccllookup( (Char *) nmstr )) ) { if ( input() != ']' ) synerr( "bad character class" ); yylval = cclval; ++cclreuse; return PREVCCL; } else { /* We fudge a bit. We know that this ccl will * soon be numbered as lastccl + 1 by cclinit. */ cclinstal( (Char *) nmstr, lastccl + 1 ); /* Push back everything but the leading bracket * so the ccl can be rescanned. */ yyless( 1 ); BEGIN(FIRSTCCL); return '['; } } YY_BREAK case 58: YY_USER_ACTION { register Char *nmdefptr; Char *ndlookup(); strcpy( nmstr, yytext + 1 ); nmstr[yyleng - 2] = '\0'; /* chop trailing brace */ if ( ! (nmdefptr = ndlookup( nmstr )) ) format_synerr( "undefined definition {%s}", nmstr ); else { /* push back name surrounded by ()'s */ int len = strlen( (char *) nmdefptr ); if ( lex_compat || nmdefptr[0] == '^' || (len > 0 && nmdefptr[len - 1] == '$') ) { /* don't use ()'s after all */ PUT_BACK_STRING((char *) nmdefptr, 0); if ( nmdefptr[0] == '^' ) BEGIN(CARETISBOL); } else { unput(')'); PUT_BACK_STRING((char *) nmdefptr, 0); unput('('); } } } YY_BREAK case 59: YY_USER_ACTION return (unsigned char) yytext[0]; YY_BREAK case 60: YY_USER_ACTION RETURNCHAR; YY_BREAK case 61: YY_USER_ACTION return (unsigned char) yytext[0]; YY_BREAK case 62: YY_USER_ACTION BEGIN(SECT2); return '>'; YY_BREAK case 63: *yy_cp = yy_hold_char; /* undo effects of setting up yytext */ yy_c_buf_p = yy_cp = yy_bp + 1; YY_DO_BEFORE_ACTION; /* set up yytext again */ YY_USER_ACTION BEGIN(CARETISBOL); return '>'; YY_BREAK case 64: YY_USER_ACTION RETURNNAME; YY_BREAK case 65: YY_USER_ACTION { format_synerr( "bad : %s", yytext ); } YY_BREAK case 66: YY_USER_ACTION BEGIN(SECT2); return '^'; YY_BREAK case 67: YY_USER_ACTION RETURNCHAR; YY_BREAK case 68: YY_USER_ACTION BEGIN(SECT2); return '"'; YY_BREAK case 69: YY_USER_ACTION { synerr( "missing quote" ); BEGIN(SECT2); ++linenum; return '"'; } YY_BREAK case 70: *yy_cp = yy_hold_char; /* undo effects of setting up yytext */ yy_c_buf_p = yy_cp = yy_bp + 1; YY_DO_BEFORE_ACTION; /* set up yytext again */ YY_USER_ACTION BEGIN(CCL); return '^'; YY_BREAK case 71: *yy_cp = yy_hold_char; /* undo effects of setting up yytext */ yy_c_buf_p = yy_cp = yy_bp + 1; YY_DO_BEFORE_ACTION; /* set up yytext again */ YY_USER_ACTION return '^'; YY_BREAK case 72: YY_USER_ACTION BEGIN(CCL); RETURNCHAR; YY_BREAK case 73: *yy_cp = yy_hold_char; /* undo effects of setting up yytext */ yy_c_buf_p = yy_cp = yy_bp + 1; YY_DO_BEFORE_ACTION; /* set up yytext again */ YY_USER_ACTION return '-'; YY_BREAK case 74: YY_USER_ACTION RETURNCHAR; YY_BREAK case 75: YY_USER_ACTION BEGIN(SECT2); return ']'; YY_BREAK case 76: YY_USER_ACTION { synerr( "bad character class" ); BEGIN(SECT2); return ']'; } YY_BREAK case 77: YY_USER_ACTION { yylval = myctoi( yytext ); return NUMBER; } YY_BREAK case 78: YY_USER_ACTION return ','; YY_BREAK case 79: YY_USER_ACTION BEGIN(SECT2); return '}'; YY_BREAK case 80: YY_USER_ACTION { synerr( "bad character inside {}'s" ); BEGIN(SECT2); return '}'; } YY_BREAK case 81: YY_USER_ACTION { synerr( "missing }" ); BEGIN(SECT2); ++linenum; return '}'; } YY_BREAK case 82: YY_USER_ACTION synerr( "bad name in {}'s" ); BEGIN(SECT2); YY_BREAK case 83: YY_USER_ACTION synerr( "missing }" ); ++linenum; BEGIN(SECT2); YY_BREAK case 84: YY_USER_ACTION ACTION_ECHO; BEGIN(ACTION_COMMENT); YY_BREAK case 85: YY_USER_ACTION bracelevel = 0; YY_BREAK case 86: YY_USER_ACTION { ACTION_ECHO; CHECK_REJECT(yytext); } YY_BREAK case 87: YY_USER_ACTION { ACTION_ECHO; CHECK_YYMORE(yytext); } YY_BREAK case 88: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 89: YY_USER_ACTION { ++linenum; ACTION_ECHO; if ( bracelevel == 0 || (doing_codeblock && indented_code) ) { if ( ! doing_codeblock ) add_action( "\tYY_BREAK\n" ); doing_codeblock = false; BEGIN(SECT2); } } YY_BREAK /* Reject and YYmore() are checked for above, in PERCENT_BRACE_ACTION */ case 90: YY_USER_ACTION ACTION_ECHO; ++bracelevel; YY_BREAK case 91: YY_USER_ACTION ACTION_ECHO; --bracelevel; YY_BREAK case 92: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 93: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 94: YY_USER_ACTION ACTION_ECHO; BEGIN(ACTION_COMMENT); YY_BREAK case 95: YY_USER_ACTION ACTION_ECHO; /* character constant */ YY_BREAK case 96: YY_USER_ACTION ACTION_ECHO; BEGIN(ACTION_STRING); YY_BREAK case 97: YY_USER_ACTION { ++linenum; ACTION_ECHO; if ( bracelevel == 0 ) { add_action( "\tYY_BREAK\n" ); BEGIN(SECT2); } } YY_BREAK case 98: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 99: YY_USER_ACTION { ACTION_ECHO; if ( doing_codeblock ) BEGIN(CODEBLOCK_2); else BEGIN(ACTION); } YY_BREAK case 100: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 101: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 102: YY_USER_ACTION ++linenum; ACTION_ECHO; YY_BREAK case 103: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 104: YY_USER_ACTION ACTION_ECHO; YY_BREAK case 105: YY_USER_ACTION ++linenum; ACTION_ECHO; YY_BREAK case 106: YY_USER_ACTION ACTION_ECHO; BEGIN(ACTION); YY_BREAK case 107: YY_USER_ACTION ACTION_ECHO; YY_BREAK case YY_STATE_EOF(ACTION): case YY_STATE_EOF(ACTION_COMMENT): case YY_STATE_EOF(ACTION_STRING): { synerr( "EOF encountered inside an action" ); yyterminate(); } YY_BREAK case 108: YY_USER_ACTION { yylval = myesc( (Char *) yytext ); return CHAR; } YY_BREAK case 109: YY_USER_ACTION { yylval = myesc( (Char *) yytext ); BEGIN(CCL); return CHAR; } YY_BREAK case 110: YY_USER_ACTION ECHO; YY_BREAK case YY_STATE_EOF(SECT3): sectnum = 0; yyterminate(); YY_BREAK case 111: YY_USER_ACTION format_synerr( "bad character: %s", yytext ); YY_BREAK case 112: YY_USER_ACTION YY_FATAL_ERROR( "flex scanner jammed" ); YY_BREAK case YY_STATE_EOF(INITIAL): case YY_STATE_EOF(SECT2): case YY_STATE_EOF(CODEBLOCK): case YY_STATE_EOF(PICKUPDEF): case YY_STATE_EOF(SC): case YY_STATE_EOF(CARETISBOL): case YY_STATE_EOF(NUM): case YY_STATE_EOF(QUOTE): case YY_STATE_EOF(FIRSTCCL): case YY_STATE_EOF(CCL): case YY_STATE_EOF(RECOVER): case YY_STATE_EOF(BRACEERROR): case YY_STATE_EOF(C_COMMENT): case YY_STATE_EOF(PERCENT_BRACE_ACTION): case YY_STATE_EOF(USED_LIST): case YY_STATE_EOF(CODEBLOCK_2): yyterminate(); case YY_END_OF_BUFFER: { /* Amount of text matched not including the EOB char. */ int yy_amount_of_matched_text = yy_cp - yytext_ptr - 1; /* Undo the effects of YY_DO_BEFORE_ACTION. */ *yy_cp = yy_hold_char; if ( yy_current_buffer->yy_buffer_status == YY_BUFFER_NEW ) { /* We're scanning a new file or input source. It's * possible that this happened because the user * just pointed yyin at a new source and called * yylex(). If so, then we have to assure * consistency between yy_current_buffer and our * globals. Here is the right place to do so, because * this is the first action (other than possibly a * back-up) that will match for the new input source. */ yy_n_chars = yy_current_buffer->yy_n_chars; yy_current_buffer->yy_input_file = yyin; yy_current_buffer->yy_buffer_status = YY_BUFFER_NORMAL; } /* Note that here we test for yy_c_buf_p "<=" to the position * of the first EOB in the buffer, since yy_c_buf_p will * already have been incremented past the NUL character * (since all states make transitions on EOB to the * end-of-buffer state). Contrast this with the test * in input(). */ if ( yy_c_buf_p <= &yy_current_buffer->yy_ch_buf[yy_n_chars] ) { /* This was really a NUL. */ yy_state_type yy_next_state; yy_c_buf_p = yytext_ptr + yy_amount_of_matched_text; yy_current_state = yy_get_previous_state(); /* Okay, we're now positioned to make the NUL * transition. We couldn't have * yy_get_previous_state() go ahead and do it * for us because it doesn't know how to deal * with the possibility of jamming (and we don't * want to build jamming into it because then it * will run more slowly). */ yy_next_state = yy_try_NUL_trans( yy_current_state ); yy_bp = yytext_ptr + YY_MORE_ADJ; if ( yy_next_state ) { /* Consume the NUL. */ yy_cp = ++yy_c_buf_p; yy_current_state = yy_next_state; goto yy_match; } else { yy_cp = yy_c_buf_p; goto yy_find_action; } } else switch ( yy_get_next_buffer() ) { case EOB_ACT_END_OF_FILE: { yy_did_buffer_switch_on_eof = 0; if ( yywrap() ) { /* Note: because we've taken care in * yy_get_next_buffer() to have set up * yytext, we can now set up * yy_c_buf_p so that if some total * hoser (like flex itself) wants to * call the scanner after we return the * YY_NULL, it'll still work - another * YY_NULL will get returned. */ yy_c_buf_p = yytext_ptr + YY_MORE_ADJ; yy_act = YY_STATE_EOF(YY_START); goto do_action; } else { if ( ! yy_did_buffer_switch_on_eof ) YY_NEW_FILE; } break; } case EOB_ACT_CONTINUE_SCAN: yy_c_buf_p = yytext_ptr + yy_amount_of_matched_text; yy_current_state = yy_get_previous_state(); yy_cp = yy_c_buf_p; yy_bp = yytext_ptr + YY_MORE_ADJ; goto yy_match; case EOB_ACT_LAST_MATCH: yy_c_buf_p = &yy_current_buffer->yy_ch_buf[yy_n_chars]; yy_current_state = yy_get_previous_state(); yy_cp = yy_c_buf_p; yy_bp = yytext_ptr + YY_MORE_ADJ; goto yy_find_action; } break; } default: YY_FATAL_ERROR( "fatal flex scanner internal error--no action found" ); } /* end of action switch */ } /* end of scanning one token */ } /* end of yylex */ /* yy_get_next_buffer - try to read in a new buffer * * Returns a code representing an action: * EOB_ACT_LAST_MATCH - * EOB_ACT_CONTINUE_SCAN - continue scanning from current position * EOB_ACT_END_OF_FILE - end of file */ static int yy_get_next_buffer() { register char *dest = yy_current_buffer->yy_ch_buf; register char *source = yytext_ptr - 1; /* copy prev. char, too */ register int number_to_move, i; int ret_val; if ( yy_c_buf_p > &yy_current_buffer->yy_ch_buf[yy_n_chars + 1] ) YY_FATAL_ERROR( "fatal flex scanner internal error--end of buffer missed" ); if ( yy_current_buffer->yy_fill_buffer == 0 ) { /* Don't try to fill the buffer, so this is an EOF. */ if ( yy_c_buf_p - yytext_ptr - YY_MORE_ADJ == 1 ) { /* We matched a singled characater, the EOB, so * treat this as a final EOF. */ return EOB_ACT_END_OF_FILE; } else { /* We matched some text prior to the EOB, first * process it. */ return EOB_ACT_LAST_MATCH; } } /* Try to read more data. */ /* First move last chars to start of buffer. */ number_to_move = yy_c_buf_p - yytext_ptr; for ( i = 0; i < number_to_move; ++i ) *(dest++) = *(source++); if ( yy_current_buffer->yy_buffer_status == YY_BUFFER_EOF_PENDING ) /* don't do the read, it's not guaranteed to return an EOF, * just force an EOF */ yy_n_chars = 0; else { int num_to_read = yy_current_buffer->yy_buf_size - number_to_move - 1; while ( num_to_read <= 0 ) { /* Not enough room in the buffer - grow it. */ #ifdef YY_USES_REJECT YY_FATAL_ERROR( "input buffer overflow, can't enlarge buffer because scanner uses REJECT" ); #else /* just a shorter name for the current buffer */ YY_BUFFER_STATE b = yy_current_buffer; int yy_c_buf_p_offset = yy_c_buf_p - b->yy_ch_buf; b->yy_buf_size *= 2; b->yy_ch_buf = (char *) yy_flex_realloc( (void *) b->yy_ch_buf, b->yy_buf_size ); if ( ! b->yy_ch_buf ) YY_FATAL_ERROR( "fatal error - scanner input buffer overflow" ); yy_c_buf_p = &b->yy_ch_buf[yy_c_buf_p_offset]; num_to_read = yy_current_buffer->yy_buf_size - number_to_move - 1; #endif } if ( num_to_read > YY_READ_BUF_SIZE ) num_to_read = YY_READ_BUF_SIZE; /* Read in more data. */ YY_INPUT( (&yy_current_buffer->yy_ch_buf[number_to_move]), yy_n_chars, num_to_read ); } if ( yy_n_chars == 0 ) { if ( number_to_move - YY_MORE_ADJ == 1 ) { ret_val = EOB_ACT_END_OF_FILE; yyrestart( yyin ); } else { ret_val = EOB_ACT_LAST_MATCH; yy_current_buffer->yy_buffer_status = YY_BUFFER_EOF_PENDING; } } else ret_val = EOB_ACT_CONTINUE_SCAN; yy_n_chars += number_to_move; yy_current_buffer->yy_ch_buf[yy_n_chars] = YY_END_OF_BUFFER_CHAR; yy_current_buffer->yy_ch_buf[yy_n_chars + 1] = YY_END_OF_BUFFER_CHAR; /* yytext begins at the second character in yy_ch_buf; the first * character is the one which preceded it before reading in the latest * buffer; it needs to be kept around in case it's a newline, so * yy_get_previous_state() will have with '^' rules active. */ yytext_ptr = &yy_current_buffer->yy_ch_buf[1]; return ret_val; } /* yy_get_previous_state - get the state just before the EOB char was reached */ static yy_state_type yy_get_previous_state() { register yy_state_type yy_current_state; register char *yy_cp; register char *yy_bp = yytext_ptr; yy_current_state = yy_start; if ( yy_bp[-1] == '\n' ) ++yy_current_state; for ( yy_cp = yytext_ptr + YY_MORE_ADJ; yy_cp < yy_c_buf_p; ++yy_cp ) { register YY_CHAR yy_c = (*yy_cp ? yy_ec[YY_SC_TO_UI(*yy_cp)] : 1); if ( yy_accept[yy_current_state] ) { yy_last_accepting_state = yy_current_state; yy_last_accepting_cpos = yy_cp; } while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state ) { yy_current_state = (int) yy_def[yy_current_state]; if ( yy_current_state >= 408 ) yy_c = yy_meta[(unsigned int) yy_c]; } yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c]; } return yy_current_state; } /* yy_try_NUL_trans - try to make a transition on the NUL character * * synopsis * next_state = yy_try_NUL_trans( current_state ); */ #ifdef YY_USE_PROTOS static yy_state_type yy_try_NUL_trans( yy_state_type yy_current_state ) #else static yy_state_type yy_try_NUL_trans( yy_current_state ) yy_state_type yy_current_state; #endif { register int yy_is_jam; register char *yy_cp = yy_c_buf_p; register YY_CHAR yy_c = 1; if ( yy_accept[yy_current_state] ) { yy_last_accepting_state = yy_current_state; yy_last_accepting_cpos = yy_cp; } while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state ) { yy_current_state = (int) yy_def[yy_current_state]; if ( yy_current_state >= 408 ) yy_c = yy_meta[(unsigned int) yy_c]; } yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c]; yy_is_jam = (yy_current_state == 407); return yy_is_jam ? 0 : yy_current_state; } #ifdef YY_USE_PROTOS static void yyunput( int c, register char *yy_bp ) #else static void yyunput( c, yy_bp ) int c; register char *yy_bp; #endif { register char *yy_cp = yy_c_buf_p; /* undo effects of setting up yytext */ *yy_cp = yy_hold_char; if ( yy_cp < yy_current_buffer->yy_ch_buf + 2 ) { /* need to shift things up to make room */ /* +2 for EOB chars. */ register int number_to_move = yy_n_chars + 2; register char *dest = &yy_current_buffer->yy_ch_buf[ yy_current_buffer->yy_buf_size + 2]; register char *source = &yy_current_buffer->yy_ch_buf[number_to_move]; while ( source > yy_current_buffer->yy_ch_buf ) *--dest = *--source; yy_cp += dest - source; yy_bp += dest - source; yy_n_chars = yy_current_buffer->yy_buf_size; if ( yy_cp < yy_current_buffer->yy_ch_buf + 2 ) YY_FATAL_ERROR( "flex scanner push-back overflow" ); } if ( yy_cp > yy_bp && yy_cp[-1] == '\n' ) yy_cp[-2] = '\n'; *--yy_cp = (char) c; /* Note: the formal parameter *must* be called "yy_bp" for this * macro to now work correctly. */ YY_DO_BEFORE_ACTION; /* set up yytext again */ } #ifdef __cplusplus static int yyinput() #else static int input() #endif { int c; *yy_c_buf_p = yy_hold_char; if ( *yy_c_buf_p == YY_END_OF_BUFFER_CHAR ) { /* yy_c_buf_p now points to the character we want to return. * If this occurs *before* the EOB characters, then it's a * valid NUL; if not, then we've hit the end of the buffer. */ if ( yy_c_buf_p < &yy_current_buffer->yy_ch_buf[yy_n_chars] ) /* This was really a NUL. */ *yy_c_buf_p = '\0'; else { /* need more input */ yytext_ptr = yy_c_buf_p; ++yy_c_buf_p; switch ( yy_get_next_buffer() ) { case EOB_ACT_END_OF_FILE: { if ( yywrap() ) { yy_c_buf_p = yytext_ptr + YY_MORE_ADJ; return EOF; } YY_NEW_FILE; #ifdef __cplusplus return yyinput(); #else return input(); #endif } case EOB_ACT_CONTINUE_SCAN: yy_c_buf_p = yytext_ptr + YY_MORE_ADJ; break; case EOB_ACT_LAST_MATCH: #ifdef __cplusplus YY_FATAL_ERROR( "unexpected last match in yyinput()" ); #else YY_FATAL_ERROR( "unexpected last match in input()" ); #endif } } } c = *(unsigned char *) yy_c_buf_p; /* cast for 8-bit char's */ *yy_c_buf_p = '\0'; /* preserve yytext */ yy_hold_char = *++yy_c_buf_p; return c; } #ifdef YY_USE_PROTOS void yyrestart( FILE *input_file ) #else void yyrestart( input_file ) FILE *input_file; #endif { if ( ! yy_current_buffer ) yy_current_buffer = yy_create_buffer( yyin, YY_BUF_SIZE ); yy_init_buffer( yy_current_buffer, input_file ); yy_load_buffer_state(); } #ifdef YY_USE_PROTOS void yy_switch_to_buffer( YY_BUFFER_STATE new_buffer ) #else void yy_switch_to_buffer( new_buffer ) YY_BUFFER_STATE new_buffer; #endif { if ( yy_current_buffer == new_buffer ) return; if ( yy_current_buffer ) { /* Flush out information for old buffer. */ *yy_c_buf_p = yy_hold_char; yy_current_buffer->yy_buf_pos = yy_c_buf_p; yy_current_buffer->yy_n_chars = yy_n_chars; } yy_current_buffer = new_buffer; yy_load_buffer_state(); /* We don't actually know whether we did this switch during * EOF (yywrap()) processing, but the only time this flag * is looked at is after yywrap() is called, so it's safe * to go ahead and always set it. */ yy_did_buffer_switch_on_eof = 1; } #ifdef YY_USE_PROTOS void yy_load_buffer_state( void ) #else void yy_load_buffer_state() #endif { yy_n_chars = yy_current_buffer->yy_n_chars; yytext_ptr = yy_c_buf_p = yy_current_buffer->yy_buf_pos; yyin = yy_current_buffer->yy_input_file; yy_hold_char = *yy_c_buf_p; } #ifdef YY_USE_PROTOS YY_BUFFER_STATE yy_create_buffer( FILE *file, int size ) #else YY_BUFFER_STATE yy_create_buffer( file, size ) FILE *file; int size; #endif { YY_BUFFER_STATE b; b = (YY_BUFFER_STATE) yy_flex_alloc( sizeof( struct yy_buffer_state ) ); if ( ! b ) YY_FATAL_ERROR( "out of dynamic memory in yy_create_buffer()" ); b->yy_buf_size = size; /* yy_ch_buf has to be 2 characters longer than the size given because * we need to put in 2 end-of-buffer characters. */ b->yy_ch_buf = (char *) yy_flex_alloc( b->yy_buf_size + 2 ); if ( ! b->yy_ch_buf ) YY_FATAL_ERROR( "out of dynamic memory in yy_create_buffer()" ); yy_init_buffer( b, file ); return b; } #ifdef YY_USE_PROTOS void yy_delete_buffer( YY_BUFFER_STATE b ) #else void yy_delete_buffer( b ) YY_BUFFER_STATE b; #endif { if ( b == yy_current_buffer ) yy_current_buffer = (YY_BUFFER_STATE) 0; yy_flex_free( (void *) b->yy_ch_buf ); yy_flex_free( (void *) b ); } #ifdef YY_USE_PROTOS void yy_init_buffer( YY_BUFFER_STATE b, FILE *file ) #else void yy_init_buffer( b, file ) YY_BUFFER_STATE b; FILE *file; #endif { b->yy_input_file = file; /* We put in the '\n' and start reading from [1] so that an * initial match-at-newline will be true. */ b->yy_ch_buf[0] = '\n'; b->yy_n_chars = 1; /* We always need two end-of-buffer characters. The first causes * a transition to the end-of-buffer state. The second causes * a jam in that state. */ b->yy_ch_buf[1] = YY_END_OF_BUFFER_CHAR; b->yy_ch_buf[2] = YY_END_OF_BUFFER_CHAR; b->yy_buf_pos = &b->yy_ch_buf[1]; b->yy_is_interactive = file ? isatty( fileno(file) ) : 0; b->yy_fill_buffer = 1; b->yy_buffer_status = YY_BUFFER_NEW; } #ifdef YY_USE_PROTOS static void yy_push_state( int new_state ) #else static void yy_push_state( new_state ) int new_state; #endif { if ( yy_start_stack_ptr >= yy_start_stack_depth ) { int new_size; yy_start_stack_depth += YY_START_STACK_INCR; new_size = yy_start_stack_depth * sizeof( int ); if ( ! yy_start_stack ) yy_start_stack = (int *) yy_flex_alloc( new_size ); else yy_start_stack = (int *) yy_flex_realloc( (void *) yy_start_stack, new_size ); if ( ! yy_start_stack ) YY_FATAL_ERROR( "out of memory expanding start-condition stack" ); } yy_start_stack[yy_start_stack_ptr++] = YY_START; BEGIN(new_state); } static void yy_pop_state() { if ( --yy_start_stack_ptr < 0 ) YY_FATAL_ERROR( "start-condition stack underflow" ); BEGIN(yy_start_stack[yy_start_stack_ptr]); } static int yy_top_state() { return yy_start_stack[yy_start_stack_ptr - 1]; } #ifdef YY_USE_PROTOS static void yy_fatal_error( const char msg[] ) #else static void yy_fatal_error( msg ) char msg[]; #endif { (void) fprintf( err, "%s\n", msg ); exit( 1 ); } /* Redefine yyless() so it works in section 3 code. */ #undef yyless #define yyless(n) \ do \ { \ /* Undo effects of setting up yytext. */ \ yytext[yyleng] = yy_hold_char; \ yy_c_buf_p = yytext + n - YY_MORE_ADJ; \ yy_hold_char = *yy_c_buf_p; \ *yy_c_buf_p = '\0'; \ yyleng = n; \ } \ while ( 0 ) /* Internal utility routines. */ #ifndef yytext_ptr #ifdef YY_USE_PROTOS static void yy_flex_strncpy( char *s1, const char *s2, int n ) #else static void yy_flex_strncpy( s1, s2, n ) char *s1; const char *s2; int n; #endif { register int i; for ( i = 0; i < n; ++i ) s1[i] = s2[i]; } #endif #ifdef YY_USE_PROTOS static void *yy_flex_alloc( unsigned int size ) #else static void *yy_flex_alloc( size ) unsigned int size; #endif { return (void *) malloc( size ); } #ifdef YY_USE_PROTOS static void *yy_flex_realloc( void *ptr, unsigned int size ) #else static void *yy_flex_realloc( ptr, size ) void *ptr; unsigned int size; #endif { return (void *) realloc( ptr, size ); } #ifdef YY_USE_PROTOS static void yy_flex_free( void *ptr ) #else static void yy_flex_free( ptr ) void *ptr; #endif { free( ptr ); } int yywrap() { if ( --num_input_files > 0 ) { set_input_file( *++input_files ); return 0; } else return 1; } /* Wrapper routines for accessing the scanner's malloc routines. */ void *flex_alloc( size ) unsigned int size; { return yy_flex_alloc( size ); } void *flex_realloc( ptr, size ) void *ptr; unsigned int size; { return yy_flex_realloc( ptr, size ); } void flex_free( ptr ) void *ptr; { yy_flex_free( ptr ); }