flexdef.h

C++版 词法分析、语法分析器

 * 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 32766


/* 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
 * 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
 * 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 true (i.e., -p flag), generate a report relating
 *   to scanner performance
 * backtrack_report - if true (i.e., -b flag), generate "lex.backtrack" file
 *   listing backtracking states
 * 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 backtracking 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, spprdflt;
extern int interactive, caseins, useecs, fulltbl, usemecs;
extern int fullspd, gen_line_dirs, performance_report, backtrack_report, 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
 * yyin - input file
 * temp_action_file - temporary file to hold actions
 * backtrack_file - file to summarize backtracking states to
 * infilename - name of input file
 * headerfilename - name of output header file
 * includefilename - name of include to #include 
			 (default=headerfilename)
 * headerfile - output header file
 * outputfilename - output  file name
 * action_file_name - name of the temporary 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 
 * lexer_name - name of lexer class and prefix for associated symbols
 * name_defined - flag : 1 if %name given or if default name is used
 * skelname - name of skeleton file
 * skelheaderfilename - name of skeleton header file
 */

extern int datapos, dataline, linenum;
extern FILE *skelfile, *yyin, *temp_action_file, *backtrack_file;
extern char *skelname;
extern char *infilename;
extern char *headerfilename;
extern char *outputfilename;
extern char *includefilename;
extern FILE *headerfile;
extern char *skelheaderfilename;
extern FILE *skelheaderfile;
extern char *action_file_name;
extern char **input_files;
extern int num_input_files;
extern char *program_name;
extern char lexer_name[256];
extern int name_defined;
/* 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
 * 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 a ho-hum
 *             normal rule or one which has variable head & trailing
 *             context
 * rule_linenum - line number associated with rule
 */

extern int current_mns, num_rules, current_max_rules, lastnfa;
extern int *firstst, *lastst, *finalst, *transchar, *trans1, *trans2;
extern int *accptnum, *assoc_rule, *state_type, *rule_type, *rule_linenum;

/* 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
 * xlation - maps character codes to their translations, or nil if no %t table
 * num_xlations - number of different xlation values
 */

/* 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];

extern int *xlation;
extern int num_xlations;


/* 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
 */

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