1: #include "first.h"
2:
3: /*
4: ** This file contains all sources (including headers) to the LEMON
5: ** LALR(1) parser generator. The sources have been combined into a
6: ** single file to make it easy to include LEMON in the source tree
7: ** and Makefile of another program.
8: **
9: ** The author of this program disclaims copyright.
10: */
11: #include <stdio.h>
12: #include <stdarg.h>
13: #include <string.h>
14: #include <ctype.h>
15: #include <stdlib.h>
16:
17: #ifdef HAVE_STDINT_H
18: # include <stdint.h>
19: #endif
20: #ifdef HAVE_INTTYPES_H
21: # include <inttypes.h>
22: #endif
23:
24: #define UNUSED(x) ( (void)(x) )
25:
26: extern void qsort();
27: extern double strtod();
28: extern long strtol();
29: extern void free();
30: extern int access();
31: extern int atoi();
32: extern char *getenv();
33:
34: #ifndef __WIN32__
35: # if defined(_WIN32) || defined(WIN32)
36: # define __WIN32__
37: # endif
38: #endif
39:
40: #if __GNUC__ > 2
41: #define NORETURN __attribute__ ((__noreturn__))
42: #else
43: #define NORETURN
44: #endif
45:
46: /* #define PRIVATE static */
47: #define PRIVATE static
48:
49: #ifdef TEST
50: #define MAXRHS 5 /* Set low to exercise exception code */
51: #else
52: #define MAXRHS 1000
53: #endif
54:
55: char *msort();
56: extern void *malloc();
57:
58: extern void memory_error() NORETURN;
59:
60: /******** From the file "action.h" *************************************/
61: struct action *Action_new();
62: struct action *Action_sort();
63: void Action_add();
64:
65: /********* From the file "assert.h" ************************************/
66: void myassert() NORETURN;
67: #ifndef NDEBUG
68: # define assert(X) if(!(X))myassert(__FILE__,__LINE__)
69: #else
70: # define assert(X)
71: #endif
72:
73: /********** From the file "build.h" ************************************/
74: void FindRulePrecedences();
75: void FindFirstSets();
76: void FindStates();
77: void FindLinks();
78: void FindFollowSets();
79: void FindActions();
80:
81: /********* From the file "configlist.h" *********************************/
82: void Configlist_init(/* void */);
83: struct config *Configlist_add(/* struct rule *, int */);
84: struct config *Configlist_addbasis(/* struct rule *, int */);
85: void Configlist_closure(/* void */);
86: void Configlist_sort(/* void */);
87: void Configlist_sortbasis(/* void */);
88: struct config *Configlist_return(/* void */);
89: struct config *Configlist_basis(/* void */);
90: void Configlist_eat(/* struct config * */);
91: void Configlist_reset(/* void */);
92:
93: /********* From the file "error.h" ***************************************/
94: void ErrorMsg(const char *, int,const char *, ...);
95:
96: /****** From the file "option.h" ******************************************/
97: struct s_options {
98: enum { OPT_FLAG=1, OPT_INT, OPT_DBL, OPT_STR,
99: OPT_FFLAG, OPT_FINT, OPT_FDBL, OPT_FSTR} type;
100: char *label;
101: char *arg;
102: char *message;
103: };
104: int OptInit(/* char**,struct s_options*,FILE* */);
105: int OptNArgs(/* void */);
106: char *OptArg(/* int */);
107: void OptErr(/* int */);
108: void OptPrint(/* void */);
109:
110: /******** From the file "parse.h" *****************************************/
111: void Parse(/* struct lemon *lemp */);
112:
113: /********* From the file "plink.h" ***************************************/
114: struct plink *Plink_new(/* void */);
115: void Plink_add(/* struct plink **, struct config * */);
116: void Plink_copy(/* struct plink **, struct plink * */);
117: void Plink_delete(/* struct plink * */);
118:
119: /********** From the file "report.h" *************************************/
120: void Reprint(/* struct lemon * */);
121: void ReportOutput(/* struct lemon * */);
122: void ReportTable(/* struct lemon * */);
123: void ReportHeader(/* struct lemon * */);
124: void CompressTables(/* struct lemon * */);
125:
126: /********** From the file "set.h" ****************************************/
127: void SetSize(/* int N */); /* All sets will be of size N */
128: char *SetNew(/* void */); /* A new set for element 0..N */
129: void SetFree(/* char* */); /* Deallocate a set */
130:
131: int SetAdd(/* char*,int */); /* Add element to a set */
132: int SetUnion(/* char *A,char *B */); /* A <- A U B, thru element N */
133:
134: #define SetFind(X,Y) (X[Y]) /* True if Y is in set X */
135:
136: /********** From the file "struct.h" *************************************/
137: /*
138: ** Principal data structures for the LEMON parser generator.
139: */
140:
141: typedef enum {Bo_FALSE=0, Bo_TRUE} Boolean;
142:
143: /* Symbols (terminals and nonterminals) of the grammar are stored
144: ** in the following: */
145: struct symbol {
146: char *name; /* Name of the symbol */
147: int index; /* Index number for this symbol */
148: enum {
149: TERMINAL,
150: NONTERMINAL
151: } type; /* Symbols are all either TERMINALS or NTs */
152: struct rule *rule; /* Linked list of rules of this (if an NT) */
153: struct symbol *fallback; /* fallback token in case this token doesn't parse */
154: int prec; /* Precedence if defined (-1 otherwise) */
155: enum e_assoc {
156: LEFT,
157: RIGHT,
158: NONE,
159: UNK
160: } assoc; /* Associativity if predecence is defined */
161: char *firstset; /* First-set for all rules of this symbol */
162: Boolean lambda; /* True if NT and can generate an empty string */
163: char *destructor; /* Code which executes whenever this symbol is
164: ** popped from the stack during error processing */
165: int destructorln; /* Line number of destructor code */
166: char *datatype; /* The data type of information held by this
167: ** object. Only used if type==NONTERMINAL */
168: int dtnum; /* The data type number. In the parser, the value
169: ** stack is a union. The .yy%d element of this
170: ** union is the correct data type for this object */
171: };
172:
173: /* Each production rule in the grammar is stored in the following
174: ** structure. */
175: struct rule {
176: struct symbol *lhs; /* Left-hand side of the rule */
177: char *lhsalias; /* Alias for the LHS (NULL if none) */
178: int ruleline; /* Line number for the rule */
179: int nrhs; /* Number of RHS symbols */
180: struct symbol **rhs; /* The RHS symbols */
181: char **rhsalias; /* An alias for each RHS symbol (NULL if none) */
182: int line; /* Line number at which code begins */
183: char *code; /* The code executed when this rule is reduced */
184: struct symbol *precsym; /* Precedence symbol for this rule */
185: int index; /* An index number for this rule */
186: Boolean canReduce; /* True if this rule is ever reduced */
187: struct rule *nextlhs; /* Next rule with the same LHS */
188: struct rule *next; /* Next rule in the global list */
189: };
190:
191: /* A configuration is a production rule of the grammar together with
192: ** a mark (dot) showing how much of that rule has been processed so far.
193: ** Configurations also contain a follow-set which is a list of terminal
194: ** symbols which are allowed to immediately follow the end of the rule.
195: ** Every configuration is recorded as an instance of the following: */
196: struct config {
197: struct rule *rp; /* The rule upon which the configuration is based */
198: int dot; /* The parse point */
199: char *fws; /* Follow-set for this configuration only */
200: struct plink *fplp; /* Follow-set forward propagation links */
201: struct plink *bplp; /* Follow-set backwards propagation links */
202: struct state *stp; /* Pointer to state which contains this */
203: enum {
204: COMPLETE, /* The status is used during followset and */
205: INCOMPLETE /* shift computations */
206: } status;
207: struct config *next; /* Next configuration in the state */
208: struct config *bp; /* The next basis configuration */
209: };
210:
211: /* Every shift or reduce operation is stored as one of the following */
212: struct action {
213: struct symbol *sp; /* The look-ahead symbol */
214: enum e_action {
215: SHIFT,
216: ACCEPT,
217: REDUCE,
218: ERROR,
219: CONFLICT, /* Was a reduce, but part of a conflict */
220: SH_RESOLVED, /* Was a shift. Precedence resolved conflict */
221: RD_RESOLVED, /* Was reduce. Precedence resolved conflict */
222: NOT_USED /* Deleted by compression */
223: } type;
224: union {
225: struct state *stp; /* The new state, if a shift */
226: struct rule *rp; /* The rule, if a reduce */
227: } x;
228: struct action *next; /* Next action for this state */
229: struct action *collide; /* Next action with the same hash */
230: };
231:
232: /* Each state of the generated parser's finite state machine
233: ** is encoded as an instance of the following structure. */
234: struct state {
235: struct config *bp; /* The basis configurations for this state */
236: struct config *cfp; /* All configurations in this set */
237: int index; /* Sequencial number for this state */
238: struct action *ap; /* Array of actions for this state */
239: int nTknAct, nNtAct; /* Number of actions on terminals and nonterminals */
240: int iTknOfst, iNtOfst; /* yy_action[] offset for terminals and nonterms */
241: int iDflt; /* Default action */
242: };
243: #define NO_OFFSET (-2147483647)
244:
245: /* A followset propagation link indicates that the contents of one
246: ** configuration followset should be propagated to another whenever
247: ** the first changes. */
248: struct plink {
249: struct config *cfp; /* The configuration to which linked */
250: struct plink *next; /* The next propagate link */
251: };
252:
253: /* The state vector for the entire parser generator is recorded as
254: ** follows. (LEMON uses no global variables and makes little use of
255: ** static variables. Fields in the following structure can be thought
256: ** of as begin global variables in the program.) */
257: struct lemon {
258: struct state **sorted; /* Table of states sorted by state number */
259: struct rule *rule; /* List of all rules */
260: int nstate; /* Number of states */
261: int nrule; /* Number of rules */
262: int nsymbol; /* Number of terminal and nonterminal symbols */
263: int nterminal; /* Number of terminal symbols */
264: struct symbol **symbols; /* Sorted array of pointers to symbols */
265: int errorcnt; /* Number of errors */
266: struct symbol *errsym; /* The error symbol */
267: char *name; /* Name of the generated parser */
268: char *arg; /* Declaration of the 3th argument to parser */
269: char *tokentype; /* Type of terminal symbols in the parser stack */
270: char *vartype; /* The default type of non-terminal symbols */
271: char *start; /* Name of the start symbol for the grammar */
272: char *stacksize; /* Size of the parser stack */
273: char *include; /* Code to put at the start of the C file */
274: int includeln; /* Line number for start of include code */
275: char *error; /* Code to execute when an error is seen */
276: int errorln; /* Line number for start of error code */
277: char *overflow; /* Code to execute on a stack overflow */
278: int overflowln; /* Line number for start of overflow code */
279: char *failure; /* Code to execute on parser failure */
280: int failureln; /* Line number for start of failure code */
281: char *accept; /* Code to execute when the parser excepts */
282: int acceptln; /* Line number for the start of accept code */
283: char *extracode; /* Code appended to the generated file */
284: int extracodeln; /* Line number for the start of the extra code */
285: char *tokendest; /* Code to execute to destroy token data */
286: int tokendestln; /* Line number for token destroyer code */
287: char *vardest; /* Code for the default non-terminal destructor */
288: int vardestln; /* Line number for default non-term destructor code*/
289: char *filename; /* Name of the input file */
290: char *tmplname; /* Name of the template file */
291: char *outname; /* Name of the current output file */
292: char *tokenprefix; /* A prefix added to token names in the .h file */
293: int nconflict; /* Number of parsing conflicts */
294: int tablesize; /* Size of the parse tables */
295: int basisflag; /* Print only basis configurations */
296: int has_fallback; /* True if any %fallback is seen in the grammer */
297: char *argv0; /* Name of the program */
298: };
299:
300: #define MemoryCheck(X) if((X)==0){ \
301: memory_error(); \
302: }
303:
304: /**************** From the file "table.h" *********************************/
305: /*
306: ** All code in this file has been automatically generated
307: ** from a specification in the file
308: ** "table.q"
309: ** by the associative array code building program "aagen".
310: ** Do not edit this file! Instead, edit the specification
311: ** file, then rerun aagen.
312: */
313: /*
314: ** Code for processing tables in the LEMON parser generator.
315: */
316:
317: /* Routines for handling a strings */
318:
319: char *Strsafe();
320:
321: void Strsafe_init(/* void */);
322: int Strsafe_insert(/* char * */);
323: char *Strsafe_find(/* char * */);
324:
325: /* Routines for handling symbols of the grammar */
326:
327: struct symbol *Symbol_new();
328: int Symbolcmpp(/* struct symbol **, struct symbol ** */);
329: void Symbol_init(/* void */);
330: int Symbol_insert(/* struct symbol *, char * */);
331: struct symbol *Symbol_find(/* char * */);
332: struct symbol *Symbol_Nth(/* int */);
333: int Symbol_count(/* */);
334: struct symbol **Symbol_arrayof(/* */);
335:
336: /* Routines to manage the state table */
337:
338: int Configcmp(/* struct config *, struct config * */);
339: struct state *State_new();
340: void State_init(/* void */);
341: int State_insert(/* struct state *, struct config * */);
342: struct state *State_find(/* struct config * */);
343: struct state **State_arrayof(/* */);
344:
345: /* Routines used for efficiency in Configlist_add */
346:
347: void Configtable_init(/* void */);
348: int Configtable_insert(/* struct config * */);
349: struct config *Configtable_find(/* struct config * */);
350: void Configtable_clear(/* int(*)(struct config *) */);
351: /****************** From the file "action.c" *******************************/
352: /*
353: ** Routines processing parser actions in the LEMON parser generator.
354: */
355:
356: /* Allocate a new parser action */
357: struct action *Action_new(){
358: static struct action *freelist = NULL;
359: struct action *new;
360:
361: if( freelist==NULL ){
362: int i;
363: int amt = 100;
364: freelist = (struct action *)malloc( sizeof(struct action)*amt );
365: if( freelist==0 ){
366: fprintf(stderr,"Unable to allocate memory for a new parser action.");
367: exit(1);
368: }
369: for(i=0; i<amt-1; i++) freelist[i].next = &freelist[i+1];
370: freelist[amt-1].next = 0;
371: }
372: new = freelist;
373: freelist = freelist->next;
374: return new;
375: }
376:
377: /* Compare two actions */
378: static int actioncmp(ap1,ap2)
379: struct action *ap1;
380: struct action *ap2;
381: {
382: int rc;
383: rc = ap1->sp->index - ap2->sp->index;
384: if( rc==0 ) rc = (int)ap1->type - (int)ap2->type;
385: if( rc==0 ){
386: assert( ap1->type==REDUCE || ap1->type==RD_RESOLVED || ap1->type==CONFLICT);
387: assert( ap2->type==REDUCE || ap2->type==RD_RESOLVED || ap2->type==CONFLICT);
388: rc = ap1->x.rp->index - ap2->x.rp->index;
389: }
390: return rc;
391: }
392:
393: /* Sort parser actions */
394: struct action *Action_sort(ap)
395: struct action *ap;
396: {
397: ap = (struct action *)msort(ap,&ap->next,actioncmp);
398: return ap;
399: }
400:
401: void Action_add(app,type,sp,arg)
402: struct action **app;
403: enum e_action type;
404: struct symbol *sp;
405: void *arg;
406: {
407: struct action *new;
408: new = Action_new();
409: new->next = *app;
410: *app = new;
411: new->type = type;
412: new->sp = sp;
413: if( type==SHIFT ){
414: new->x.stp = (struct state *)arg;
415: }else{
416: new->x.rp = (struct rule *)arg;
417: }
418: }
419: /********************** New code to implement the "acttab" module ***********/
420: /*
421: ** This module implements routines use to construct the yy_action[] table.
422: */
423:
424: /*
425: ** The state of the yy_action table under construction is an instance of
426: ** the following structure
427: */
428: typedef struct acttab acttab;
429: struct acttab {
430: int nAction; /* Number of used slots in aAction[] */
431: int nActionAlloc; /* Slots allocated for aAction[] */
432: struct {
433: int lookahead; /* Value of the lookahead token */
434: int action; /* Action to take on the given lookahead */
435: } *aAction, /* The yy_action[] table under construction */
436: *aLookahead; /* A single new transaction set */
437: int mnLookahead; /* Minimum aLookahead[].lookahead */
438: int mnAction; /* Action associated with mnLookahead */
439: int mxLookahead; /* Maximum aLookahead[].lookahead */
440: int nLookahead; /* Used slots in aLookahead[] */
441: int nLookaheadAlloc; /* Slots allocated in aLookahead[] */
442: };
443:
444: /* Return the number of entries in the yy_action table */
445: #define acttab_size(X) ((X)->nAction)
446:
447: /* The value for the N-th entry in yy_action */
448: #define acttab_yyaction(X,N) ((X)->aAction[N].action)
449:
450: /* The value for the N-th entry in yy_lookahead */
451: #define acttab_yylookahead(X,N) ((X)->aAction[N].lookahead)
452:
453: /* Free all memory associated with the given acttab */
454: /*
455: PRIVATE void acttab_free(acttab *p){
456: free( p->aAction );
457: free( p->aLookahead );
458: free( p );
459: }
460: */
461:
462: /* Allocate a new acttab structure */
463: PRIVATE acttab *acttab_alloc(void){
464: acttab *p = malloc( sizeof(*p) );
465: if( p==0 ){
466: fprintf(stderr,"Unable to allocate memory for a new acttab.");
467: exit(1);
468: }
469: memset(p, 0, sizeof(*p));
470: return p;
471: }
472:
473: /* Add a new action to the current transaction set
474: */
475: PRIVATE void acttab_action(acttab *p, int lookahead, int action){
476: if( p->nLookahead>=p->nLookaheadAlloc ){
477: p->nLookaheadAlloc += 25;
478: p->aLookahead = realloc( p->aLookahead,
479: sizeof(p->aLookahead[0])*p->nLookaheadAlloc );
480: if( p->aLookahead==0 ){
481: fprintf(stderr,"malloc failed\n");
482: exit(1);
483: }
484: }
485: if( p->nLookahead==0 ){
486: p->mxLookahead = lookahead;
487: p->mnLookahead = lookahead;
488: p->mnAction = action;
489: }else{
490: if( p->mxLookahead<lookahead ) p->mxLookahead = lookahead;
491: if( p->mnLookahead>lookahead ){
492: p->mnLookahead = lookahead;
493: p->mnAction = action;
494: }
495: }
496: p->aLookahead[p->nLookahead].lookahead = lookahead;
497: p->aLookahead[p->nLookahead].action = action;
498: p->nLookahead++;
499: }
500:
501: /*
502: ** Add the transaction set built up with prior calls to acttab_action()
503: ** into the current action table. Then reset the transaction set back
504: ** to an empty set in preparation for a new round of acttab_action() calls.
505: **
506: ** Return the offset into the action table of the new transaction.
507: */
508: PRIVATE int acttab_insert(acttab *p){
509: int i, j, k, n;
510: assert( p->nLookahead>0 );
511:
512: /* Make sure we have enough space to hold the expanded action table
513: ** in the worst case. The worst case occurs if the transaction set
514: ** must be appended to the current action table
515: */
516: n = p->mxLookahead + 1;
517: if( p->nAction + n >= p->nActionAlloc ){
518: int oldAlloc = p->nActionAlloc;
519: p->nActionAlloc = p->nAction + n + p->nActionAlloc + 20;
520: p->aAction = realloc( p->aAction,
521: sizeof(p->aAction[0])*p->nActionAlloc);
522: if( p->aAction==0 ){
523: fprintf(stderr,"malloc failed\n");
524: exit(1);
525: }
526: for(i=oldAlloc; i<p->nActionAlloc; i++){
527: p->aAction[i].lookahead = -1;
528: p->aAction[i].action = -1;
529: }
530: }
531:
532: /* Scan the existing action table looking for an offset where we can
533: ** insert the current transaction set. Fall out of the loop when that
534: ** offset is found. In the worst case, we fall out of the loop when
535: ** i reaches p->nAction, which means we append the new transaction set.
536: **
537: ** i is the index in p->aAction[] where p->mnLookahead is inserted.
538: */
539: for(i=0; i<p->nAction+p->mnLookahead; i++){
540: if( p->aAction[i].lookahead<0 ){
541: for(j=0; j<p->nLookahead; j++){
542: k = p->aLookahead[j].lookahead - p->mnLookahead + i;
543: if( k<0 ) break;
544: if( p->aAction[k].lookahead>=0 ) break;
545: }
546: if( j<p->nLookahead ) continue;
547: for(j=0; j<p->nAction; j++){
548: if( p->aAction[j].lookahead==j+p->mnLookahead-i ) break;
549: }
550: if( j==p->nAction ){
551: break; /* Fits in empty slots */
552: }
553: }else if( p->aAction[i].lookahead==p->mnLookahead ){
554: if( p->aAction[i].action!=p->mnAction ) continue;
555: for(j=0; j<p->nLookahead; j++){
556: k = p->aLookahead[j].lookahead - p->mnLookahead + i;
557: if( k<0 || k>=p->nAction ) break;
558: if( p->aLookahead[j].lookahead!=p->aAction[k].lookahead ) break;
559: if( p->aLookahead[j].action!=p->aAction[k].action ) break;
560: }
561: if( j<p->nLookahead ) continue;
562: n = 0;
563: for(j=0; j<p->nAction; j++){
564: if( p->aAction[j].lookahead<0 ) continue;
565: if( p->aAction[j].lookahead==j+p->mnLookahead-i ) n++;
566: }
567: if( n==p->nLookahead ){
568: break; /* Same as a prior transaction set */
569: }
570: }
571: }
572: /* Insert transaction set at index i. */
573: for(j=0; j<p->nLookahead; j++){
574: k = p->aLookahead[j].lookahead - p->mnLookahead + i;
575: p->aAction[k] = p->aLookahead[j];
576: if( k>=p->nAction ) p->nAction = k+1;
577: }
578: p->nLookahead = 0;
579:
580: /* Return the offset that is added to the lookahead in order to get the
581: ** index into yy_action of the action */
582: return i - p->mnLookahead;
583: }
584:
585: /********************** From the file "assert.c" ****************************/
586: /*
587: ** A more efficient way of handling assertions.
588: */
589: void myassert(file,line)
590: char *file;
591: int line;
592: {
593: fprintf(stderr,"Assertion failed on line %d of file \"%s\"\n",line,file);
594: exit(1);
595: }
596: /********************** From the file "build.c" *****************************/
597: /*
598: ** Routines to construction the finite state machine for the LEMON
599: ** parser generator.
600: */
601:
602: /* Find a precedence symbol of every rule in the grammar.
603: **
604: ** Those rules which have a precedence symbol coded in the input
605: ** grammar using the "[symbol]" construct will already have the
606: ** rp->precsym field filled. Other rules take as their precedence
607: ** symbol the first RHS symbol with a defined precedence. If there
608: ** are not RHS symbols with a defined precedence, the precedence
609: ** symbol field is left blank.
610: */
611: void FindRulePrecedences(xp)
612: struct lemon *xp;
613: {
614: struct rule *rp;
615: for(rp=xp->rule; rp; rp=rp->next){
616: if( rp->precsym==0 ){
617: int i;
618: for(i=0; i<rp->nrhs; i++){
619: if( rp->rhs[i]->prec>=0 ){
620: rp->precsym = rp->rhs[i];
621: break;
622: }
623: }
624: }
625: }
626: return;
627: }
628:
629: /* Find all nonterminals which will generate the empty string.
630: ** Then go back and compute the first sets of every nonterminal.
631: ** The first set is the set of all terminal symbols which can begin
632: ** a string generated by that nonterminal.
633: */
634: void FindFirstSets(lemp)
635: struct lemon *lemp;
636: {
637: int i;
638: struct rule *rp;
639: int progress;
640:
641: for(i=0; i<lemp->nsymbol; i++){
642: lemp->symbols[i]->lambda = Bo_FALSE;
643: }
644: for(i=lemp->nterminal; i<lemp->nsymbol; i++){
645: lemp->symbols[i]->firstset = SetNew();
646: }
647:
648: /* First compute all lambdas */
649: do{
650: progress = 0;
651: for(rp=lemp->rule; rp; rp=rp->next){
652: if( rp->lhs->lambda ) continue;
653: for(i=0; i<rp->nrhs; i++){
654: if( rp->rhs[i]->lambda==Bo_FALSE ) break;
655: }
656: if( i==rp->nrhs ){
657: rp->lhs->lambda = Bo_TRUE;
658: progress = 1;
659: }
660: }
661: }while( progress );
662:
663: /* Now compute all first sets */
664: do{
665: struct symbol *s1, *s2;
666: progress = 0;
667: for(rp=lemp->rule; rp; rp=rp->next){
668: s1 = rp->lhs;
669: for(i=0; i<rp->nrhs; i++){
670: s2 = rp->rhs[i];
671: if( s2->type==TERMINAL ){
672: progress += SetAdd(s1->firstset,s2->index);
673: break;
674: }else if( s1==s2 ){
675: if( s1->lambda==Bo_FALSE ) break;
676: }else{
677: progress += SetUnion(s1->firstset,s2->firstset);
678: if( s2->lambda==Bo_FALSE ) break;
679: }
680: }
681: }
682: }while( progress );
683: return;
684: }
685:
686: /* Compute all LR(0) states for the grammar. Links
687: ** are added to between some states so that the LR(1) follow sets
688: ** can be computed later.
689: */
690: PRIVATE struct state *getstate(/* struct lemon * */); /* forward reference */
691: void FindStates(lemp)
692: struct lemon *lemp;
693: {
694: struct symbol *sp;
695: struct rule *rp;
696:
697: Configlist_init();
698:
699: /* Find the start symbol */
700: if( lemp->start ){
701: sp = Symbol_find(lemp->start);
702: if( sp==0 ){
703: ErrorMsg(lemp->filename,0,
704: "The specified start symbol \"%s\" is not \
705: in a nonterminal of the grammar. \"%s\" will be used as the start \
706: symbol instead.",lemp->start,lemp->rule->lhs->name);
707: lemp->errorcnt++;
708: sp = lemp->rule->lhs;
709: }
710: }else{
711: sp = lemp->rule->lhs;
712: }
713:
714: /* Make sure the start symbol doesn't occur on the right-hand side of
715: ** any rule. Report an error if it does. (YACC would generate a new
716: ** start symbol in this case.) */
717: for(rp=lemp->rule; rp; rp=rp->next){
718: int i;
719: for(i=0; i<rp->nrhs; i++){
720: if( rp->rhs[i]==sp ){
721: ErrorMsg(lemp->filename,0,
722: "The start symbol \"%s\" occurs on the \
723: right-hand side of a rule. This will result in a parser which \
724: does not work properly.",sp->name);
725: lemp->errorcnt++;
726: }
727: }
728: }
729:
730: /* The basis configuration set for the first state
731: ** is all rules which have the start symbol as their
732: ** left-hand side */
733: for(rp=sp->rule; rp; rp=rp->nextlhs){
734: struct config *newcfp;
735: newcfp = Configlist_addbasis(rp,0);
736: SetAdd(newcfp->fws,0);
737: }
738:
739: /* Compute the first state. All other states will be
740: ** computed automatically during the computation of the first one.
741: ** The returned pointer to the first state is not used. */
742: (void)getstate(lemp);
743: return;
744: }
745:
746: /* Return a pointer to a state which is described by the configuration
747: ** list which has been built from calls to Configlist_add.
748: */
749: PRIVATE void buildshifts(/* struct lemon *, struct state * */); /* Forwd ref */
750: PRIVATE struct state *getstate(lemp)
751: struct lemon *lemp;
752: {
753: struct config *cfp, *bp;
754: struct state *stp;
755:
756: /* Extract the sorted basis of the new state. The basis was constructed
757: ** by prior calls to "Configlist_addbasis()". */
758: Configlist_sortbasis();
759: bp = Configlist_basis();
760:
761: /* Get a state with the same basis */
762: stp = State_find(bp);
763: if( stp ){
764: /* A state with the same basis already exists! Copy all the follow-set
765: ** propagation links from the state under construction into the
766: ** preexisting state, then return a pointer to the preexisting state */
767: struct config *x, *y;
768: for(x=bp, y=stp->bp; x && y; x=x->bp, y=y->bp){
769: Plink_copy(&y->bplp,x->bplp);
770: Plink_delete(x->fplp);
771: x->fplp = x->bplp = 0;
772: }
773: cfp = Configlist_return();
774: Configlist_eat(cfp);
775: }else{
776: /* This really is a new state. Construct all the details */
777: Configlist_closure(lemp); /* Compute the configuration closure */
778: Configlist_sort(); /* Sort the configuration closure */
779: cfp = Configlist_return(); /* Get a pointer to the config list */
780: stp = State_new(); /* A new state structure */
781: MemoryCheck(stp);
782: stp->bp = bp; /* Remember the configuration basis */
783: stp->cfp = cfp; /* Remember the configuration closure */
784: stp->index = lemp->nstate++; /* Every state gets a sequence number */
785: stp->ap = 0; /* No actions, yet. */
786: State_insert(stp,stp->bp); /* Add to the state table */
787: buildshifts(lemp,stp); /* Recursively compute successor states */
788: }
789: return stp;
790: }
791:
792: /* Construct all successor states to the given state. A "successor"
793: ** state is any state which can be reached by a shift action.
794: */
795: PRIVATE void buildshifts(lemp,stp)
796: struct lemon *lemp;
797: struct state *stp; /* The state from which successors are computed */
798: {
799: struct config *cfp; /* For looping thru the config closure of "stp" */
800: struct config *bcfp; /* For the inner loop on config closure of "stp" */
801: struct config *new; /* */
802: struct symbol *sp; /* Symbol following the dot in configuration "cfp" */
803: struct symbol *bsp; /* Symbol following the dot in configuration "bcfp" */
804: struct state *newstp; /* A pointer to a successor state */
805:
806: /* Each configuration becomes complete after it contibutes to a successor
807: ** state. Initially, all configurations are incomplete */
808: for(cfp=stp->cfp; cfp; cfp=cfp->next) cfp->status = INCOMPLETE;
809:
810: /* Loop through all configurations of the state "stp" */
811: for(cfp=stp->cfp; cfp; cfp=cfp->next){
812: if( cfp->status==COMPLETE ) continue; /* Already used by inner loop */
813: if( cfp->dot>=cfp->rp->nrhs ) continue; /* Can't shift this config */
814: Configlist_reset(); /* Reset the new config set */
815: sp = cfp->rp->rhs[cfp->dot]; /* Symbol after the dot */
816:
817: /* For every configuration in the state "stp" which has the symbol "sp"
818: ** following its dot, add the same configuration to the basis set under
819: ** construction but with the dot shifted one symbol to the right. */
820: for(bcfp=cfp; bcfp; bcfp=bcfp->next){
821: if( bcfp->status==COMPLETE ) continue; /* Already used */
822: if( bcfp->dot>=bcfp->rp->nrhs ) continue; /* Can't shift this one */
823: bsp = bcfp->rp->rhs[bcfp->dot]; /* Get symbol after dot */
824: if( bsp!=sp ) continue; /* Must be same as for "cfp" */
825: bcfp->status = COMPLETE; /* Mark this config as used */
826: new = Configlist_addbasis(bcfp->rp,bcfp->dot+1);
827: Plink_add(&new->bplp,bcfp);
828: }
829:
830: /* Get a pointer to the state described by the basis configuration set
831: ** constructed in the preceding loop */
832: newstp = getstate(lemp);
833:
834: /* The state "newstp" is reached from the state "stp" by a shift action
835: ** on the symbol "sp" */
836: Action_add(&stp->ap,SHIFT,sp,newstp);
837: }
838: }
839:
840: /*
841: ** Construct the propagation links
842: */
843: void FindLinks(lemp)
844: struct lemon *lemp;
845: {
846: int i;
847: struct config *cfp, *other;
848: struct state *stp;
849: struct plink *plp;
850:
851: /* Housekeeping detail:
852: ** Add to every propagate link a pointer back to the state to
853: ** which the link is attached. */
854: for(i=0; i<lemp->nstate; i++){
855: stp = lemp->sorted[i];
856: for(cfp=stp->cfp; cfp; cfp=cfp->next){
857: cfp->stp = stp;
858: }
859: }
860:
861: /* Convert all backlinks into forward links. Only the forward
862: ** links are used in the follow-set computation. */
863: for(i=0; i<lemp->nstate; i++){
864: stp = lemp->sorted[i];
865: for(cfp=stp->cfp; cfp; cfp=cfp->next){
866: for(plp=cfp->bplp; plp; plp=plp->next){
867: other = plp->cfp;
868: Plink_add(&other->fplp,cfp);
869: }
870: }
871: }
872: }
873:
874: /* Compute all followsets.
875: **
876: ** A followset is the set of all symbols which can come immediately
877: ** after a configuration.
878: */
879: void FindFollowSets(lemp)
880: struct lemon *lemp;
881: {
882: int i;
883: struct config *cfp;
884: struct plink *plp;
885: int progress;
886: int change;
887:
888: for(i=0; i<lemp->nstate; i++){
889: for(cfp=lemp->sorted[i]->cfp; cfp; cfp=cfp->next){
890: cfp->status = INCOMPLETE;
891: }
892: }
893:
894: do{
895: progress = 0;
896: for(i=0; i<lemp->nstate; i++){
897: for(cfp=lemp->sorted[i]->cfp; cfp; cfp=cfp->next){
898: if( cfp->status==COMPLETE ) continue;
899: for(plp=cfp->fplp; plp; plp=plp->next){
900: change = SetUnion(plp->cfp->fws,cfp->fws);
901: if( change ){
902: plp->cfp->status = INCOMPLETE;
903: progress = 1;
904: }
905: }
906: cfp->status = COMPLETE;
907: }
908: }
909: }while( progress );
910: }
911:
912: static int resolve_conflict();
913:
914: /* Compute the reduce actions, and resolve conflicts.
915: */
916: void FindActions(lemp)
917: struct lemon *lemp;
918: {
919: int i,j;
920: struct config *cfp;
921: struct symbol *sp;
922: struct rule *rp;
923:
924: /* Add all of the reduce actions
925: ** A reduce action is added for each element of the followset of
926: ** a configuration which has its dot at the extreme right.
927: */
928: for(i=0; i<lemp->nstate; i++){ /* Loop over all states */
929: struct state *stp;
930: stp = lemp->sorted[i];
931: for(cfp=stp->cfp; cfp; cfp=cfp->next){ /* Loop over all configurations */
932: if( cfp->rp->nrhs==cfp->dot ){ /* Is dot at extreme right? */
933: for(j=0; j<lemp->nterminal; j++){
934: if( SetFind(cfp->fws,j) ){
935: /* Add a reduce action to the state "stp" which will reduce by the
936: ** rule "cfp->rp" if the lookahead symbol is "lemp->symbols[j]" */
937: Action_add(&stp->ap,REDUCE,lemp->symbols[j],cfp->rp);
938: }
939: }
940: }
941: }
942: }
943:
944: /* Add the accepting token */
945: if( lemp->start ){
946: sp = Symbol_find(lemp->start);
947: if( sp==0 ) sp = lemp->rule->lhs;
948: }else{
949: sp = lemp->rule->lhs;
950: }
951: /* Add to the first state (which is always the starting state of the
952: ** finite state machine) an action to ACCEPT if the lookahead is the
953: ** start nonterminal. */
954: Action_add(&lemp->sorted[0]->ap,ACCEPT,sp,0);
955:
956: /* Resolve conflicts */
957: for(i=0; i<lemp->nstate; i++){
958: struct action *ap, *nap;
959: struct state *stp;
960: stp = lemp->sorted[i];
961: assert( stp->ap );
962: stp->ap = Action_sort(stp->ap);
963: for(ap=stp->ap; ap && ap->next; ap=ap->next){
964: for(nap=ap->next; nap && nap->sp==ap->sp; nap=nap->next){
965: /* The two actions "ap" and "nap" have the same lookahead.
966: ** Figure out which one should be used */
967: lemp->nconflict += resolve_conflict(ap,nap,lemp->errsym);
968: }
969: }
970: }
971:
972: /* Report an error for each rule that can never be reduced. */
973: for(rp=lemp->rule; rp; rp=rp->next) rp->canReduce = Bo_FALSE;
974: for(i=0; i<lemp->nstate; i++){
975: struct action *ap;
976: for(ap=lemp->sorted[i]->ap; ap; ap=ap->next){
977: if( ap->type==REDUCE ) ap->x.rp->canReduce = Bo_TRUE;
978: }
979: }
980: for(rp=lemp->rule; rp; rp=rp->next){
981: if( rp->canReduce ) continue;
982: ErrorMsg(lemp->filename,rp->ruleline,"This rule can not be reduced.\n");
983: lemp->errorcnt++;
984: }
985: }
986:
987: /* Resolve a conflict between the two given actions. If the
988: ** conflict can't be resolve, return non-zero.
989: **
990: ** NO LONGER TRUE:
991: ** To resolve a conflict, first look to see if either action
992: ** is on an error rule. In that case, take the action which
993: ** is not associated with the error rule. If neither or both
994: ** actions are associated with an error rule, then try to
995: ** use precedence to resolve the conflict.
996: **
997: ** If either action is a SHIFT, then it must be apx. This
998: ** function won't work if apx->type==REDUCE and apy->type==SHIFT.
999: */
1000: static int resolve_conflict(apx,apy,errsym)
1001: struct action *apx;
1002: struct action *apy;
1003: struct symbol *errsym; /* The error symbol (if defined. NULL otherwise) */
1004: {
1005: struct symbol *spx, *spy;
1006: int errcnt = 0;
1007: UNUSED(errsym);
1008: assert( apx->sp==apy->sp ); /* Otherwise there would be no conflict */
1009: if( apx->type==SHIFT && apy->type==REDUCE ){
1010: spx = apx->sp;
1011: spy = apy->x.rp->precsym;
1012: if( spy==0 || spx->prec<0 || spy->prec<0 ){
1013: /* Not enough precedence information. */
1014: apy->type = CONFLICT;
1015: errcnt++;
1016: }else if( spx->prec>spy->prec ){ /* Lower precedence wins */
1017: apy->type = RD_RESOLVED;
1018: }else if( spx->prec<spy->prec ){
1019: apx->type = SH_RESOLVED;
1020: }else if( spx->prec==spy->prec && spx->assoc==RIGHT ){ /* Use operator */
1021: apy->type = RD_RESOLVED; /* associativity */
1022: }else if( spx->prec==spy->prec && spx->assoc==LEFT ){ /* to break tie */
1023: apx->type = SH_RESOLVED;
1024: }else{
1025: assert( spx->prec==spy->prec && spx->assoc==NONE );
1026: apy->type = CONFLICT;
1027: errcnt++;
1028: }
1029: }else if( apx->type==REDUCE && apy->type==REDUCE ){
1030: spx = apx->x.rp->precsym;
1031: spy = apy->x.rp->precsym;
1032: if( spx==0 || spy==0 || spx->prec<0 ||
1033: spy->prec<0 || spx->prec==spy->prec ){
1034: apy->type = CONFLICT;
1035: errcnt++;
1036: }else if( spx->prec>spy->prec ){
1037: apy->type = RD_RESOLVED;
1038: }else if( spx->prec<spy->prec ){
1039: apx->type = RD_RESOLVED;
1040: }
1041: }else{
1042: assert(
1043: apx->type==SH_RESOLVED ||
1044: apx->type==RD_RESOLVED ||
1045: apx->type==CONFLICT ||
1046: apy->type==SH_RESOLVED ||
1047: apy->type==RD_RESOLVED ||
1048: apy->type==CONFLICT
1049: );
1050: /* The REDUCE/SHIFT case cannot happen because SHIFTs come before
1051: ** REDUCEs on the list. If we reach this point it must be because
1052: ** the parser conflict had already been resolved. */
1053: }
1054: return errcnt;
1055: }
1056: /********************* From the file "configlist.c" *************************/
1057: /*
1058: ** Routines to processing a configuration list and building a state
1059: ** in the LEMON parser generator.
1060: */
1061:
1062: static struct config *freelist = 0; /* List of free configurations */
1063: static struct config *current = 0; /* Top of list of configurations */
1064: static struct config **currentend = 0; /* Last on list of configs */
1065: static struct config *basis = 0; /* Top of list of basis configs */
1066: static struct config **basisend = 0; /* End of list of basis configs */
1067:
1068: /* Return a pointer to a new configuration */
1069: PRIVATE struct config *newconfig(){
1070: struct config *new;
1071: if( freelist==0 ){
1072: int i;
1073: int amt = 3;
1074: freelist = (struct config *)malloc( sizeof(struct config)*amt );
1075: if( freelist==0 ){
1076: fprintf(stderr,"Unable to allocate memory for a new configuration.");
1077: exit(1);
1078: }
1079: for(i=0; i<amt-1; i++) freelist[i].next = &freelist[i+1];
1080: freelist[amt-1].next = 0;
1081: }
1082: new = freelist;
1083: freelist = freelist->next;
1084: return new;
1085: }
1086:
1087: /* The configuration "old" is no longer used */
1088: PRIVATE void deleteconfig(old)
1089: struct config *old;
1090: {
1091: old->next = freelist;
1092: freelist = old;
1093: }
1094:
1095: /* Initialized the configuration list builder */
1096: void Configlist_init(){
1097: current = 0;
1098: currentend = ¤t;
1099: basis = 0;
1100: basisend = &basis;
1101: Configtable_init();
1102: return;
1103: }
1104:
1105: /* Initialized the configuration list builder */
1106: void Configlist_reset(){
1107: current = 0;
1108: currentend = ¤t;
1109: basis = 0;
1110: basisend = &basis;
1111: Configtable_clear(0);
1112: return;
1113: }
1114:
1115: /* Add another configuration to the configuration list */
1116: struct config *Configlist_add(rp,dot)
1117: struct rule *rp; /* The rule */
1118: int dot; /* Index into the RHS of the rule where the dot goes */
1119: {
1120: struct config *cfp, model;
1121:
1122: assert( currentend!=0 );
1123: model.rp = rp;
1124: model.dot = dot;
1125: cfp = Configtable_find(&model);
1126: if( cfp==0 ){
1127: cfp = newconfig();
1128: cfp->rp = rp;
1129: cfp->dot = dot;
1130: cfp->fws = SetNew();
1131: cfp->stp = 0;
1132: cfp->fplp = cfp->bplp = 0;
1133: cfp->next = 0;
1134: cfp->bp = 0;
1135: *currentend = cfp;
1136: currentend = &cfp->next;
1137: Configtable_insert(cfp);
1138: }
1139: return cfp;
1140: }
1141:
1142: /* Add a basis configuration to the configuration list */
1143: struct config *Configlist_addbasis(rp,dot)
1144: struct rule *rp;
1145: int dot;
1146: {
1147: struct config *cfp, model;
1148:
1149: assert( basisend!=0 );
1150: assert( currentend!=0 );
1151: model.rp = rp;
1152: model.dot = dot;
1153: cfp = Configtable_find(&model);
1154: if( cfp==0 ){
1155: cfp = newconfig();
1156: cfp->rp = rp;
1157: cfp->dot = dot;
1158: cfp->fws = SetNew();
1159: cfp->stp = 0;
1160: cfp->fplp = cfp->bplp = 0;
1161: cfp->next = 0;
1162: cfp->bp = 0;
1163: *currentend = cfp;
1164: currentend = &cfp->next;
1165: *basisend = cfp;
1166: basisend = &cfp->bp;
1167: Configtable_insert(cfp);
1168: }
1169: return cfp;
1170: }
1171:
1172: /* Compute the closure of the configuration list */
1173: void Configlist_closure(lemp)
1174: struct lemon *lemp;
1175: {
1176: struct config *cfp, *newcfp;
1177: struct rule *rp, *newrp;
1178: struct symbol *sp, *xsp;
1179: int i, dot;
1180:
1181: assert( currentend!=0 );
1182: for(cfp=current; cfp; cfp=cfp->next){
1183: rp = cfp->rp;
1184: dot = cfp->dot;
1185: if( dot>=rp->nrhs ) continue;
1186: sp = rp->rhs[dot];
1187: if( sp->type==NONTERMINAL ){
1188: if( sp->rule==0 && sp!=lemp->errsym ){
1189: ErrorMsg(lemp->filename,rp->line,"Nonterminal \"%s\" has no rules.",
1190: sp->name);
1191: lemp->errorcnt++;
1192: }
1193: for(newrp=sp->rule; newrp; newrp=newrp->nextlhs){
1194: newcfp = Configlist_add(newrp,0);
1195: for(i=dot+1; i<rp->nrhs; i++){
1196: xsp = rp->rhs[i];
1197: if( xsp->type==TERMINAL ){
1198: SetAdd(newcfp->fws,xsp->index);
1199: break;
1200: }else{
1201: SetUnion(newcfp->fws,xsp->firstset);
1202: if( xsp->lambda==Bo_FALSE ) break;
1203: }
1204: }
1205: if( i==rp->nrhs ) Plink_add(&cfp->fplp,newcfp);
1206: }
1207: }
1208: }
1209: return;
1210: }
1211:
1212: /* Sort the configuration list */
1213: void Configlist_sort(){
1214: current = (struct config *)msort(current,&(current->next),Configcmp);
1215: currentend = 0;
1216: return;
1217: }
1218:
1219: /* Sort the basis configuration list */
1220: void Configlist_sortbasis(){
1221: basis = (struct config *)msort(current,&(current->bp),Configcmp);
1222: basisend = 0;
1223: return;
1224: }
1225:
1226: /* Return a pointer to the head of the configuration list and
1227: ** reset the list */
1228: struct config *Configlist_return(){
1229: struct config *old;
1230: old = current;
1231: current = 0;
1232: currentend = 0;
1233: return old;
1234: }
1235:
1236: /* Return a pointer to the head of the configuration list and
1237: ** reset the list */
1238: struct config *Configlist_basis(){
1239: struct config *old;
1240: old = basis;
1241: basis = 0;
1242: basisend = 0;
1243: return old;
1244: }
1245:
1246: /* Free all elements of the given configuration list */
1247: void Configlist_eat(cfp)
1248: struct config *cfp;
1249: {
1250: struct config *nextcfp;
1251: for(; cfp; cfp=nextcfp){
1252: nextcfp = cfp->next;
1253: assert( cfp->fplp==0 );
1254: assert( cfp->bplp==0 );
1255: if( cfp->fws ) SetFree(cfp->fws);
1256: deleteconfig(cfp);
1257: }
1258: return;
1259: }
1260: /***************** From the file "error.c" *********************************/
1261: /*
1262: ** Code for printing error message.
1263: */
1264:
1265: /* Find a good place to break "msg" so that its length is at least "min"
1266: ** but no more than "max". Make the point as close to max as possible.
1267: */
1268: static int findbreak(msg,min,max)
1269: char *msg;
1270: int min;
1271: int max;
1272: {
1273: int i,spot;
1274: char c;
1275: for(i=spot=min; i<=max; i++){
1276: c = msg[i];
1277: if( c=='\t' ) msg[i] = ' ';
1278: if( c=='\n' ){ msg[i] = ' '; spot = i; break; }
1279: if( c==0 ){ spot = i; break; }
1280: if( c=='-' && i<max-1 ) spot = i+1;
1281: if( c==' ' ) spot = i;
1282: }
1283: return spot;
1284: }
1285:
1286: /*
1287: ** The error message is split across multiple lines if necessary. The
1288: ** splits occur at a space, if there is a space available near the end
1289: ** of the line.
1290: */
1291: #define ERRMSGSIZE 10000 /* Hope this is big enough. No way to error check */
1292: #define LINEWIDTH 79 /* Max width of any output line */
1293: #define PREFIXLIMIT 30 /* Max width of the prefix on each line */
1294: void ErrorMsg(const char *filename, int lineno, const char *format, ...){
1295: char errmsg[ERRMSGSIZE];
1296: char prefix[PREFIXLIMIT+10];
1297: int errmsgsize;
1298: int prefixsize;
1299: int availablewidth;
1300: va_list ap;
1301: int end, restart, base;
1302:
1303: va_start(ap, format);
1304: /* Prepare a prefix to be prepended to every output line */
1305: if( lineno>0 ){
1306: sprintf(prefix,"%.*s:%d: ",PREFIXLIMIT-10,filename,lineno);
1307: }else{
1308: sprintf(prefix,"%.*s: ",PREFIXLIMIT-10,filename);
1309: }
1310: prefixsize = strlen(prefix);
1311: availablewidth = LINEWIDTH - prefixsize;
1312:
1313: /* Generate the error message */
1314: vsprintf(errmsg,format,ap);
1315: va_end(ap);
1316: errmsgsize = strlen(errmsg);
1317: /* Remove trailing '\n's from the error message. */
1318: while( errmsgsize>0 && errmsg[errmsgsize-1]=='\n' ){
1319: errmsg[--errmsgsize] = 0;
1320: }
1321:
1322: /* Print the error message */
1323: base = 0;
1324: while( errmsg[base]!=0 ){
1325: end = restart = findbreak(&errmsg[base],0,availablewidth);
1326: restart += base;
1327: while( errmsg[restart]==' ' ) restart++;
1328: fprintf(stdout,"%s%.*s\n",prefix,end,&errmsg[base]);
1329: base = restart;
1330: }
1331: }
1332: /**************** From the file "main.c" ************************************/
1333: /*
1334: ** Main program file for the LEMON parser generator.
1335: */
1336:
1337: /* Report an out-of-memory condition and abort. This function
1338: ** is used mostly by the "MemoryCheck" macro in struct.h
1339: */
1340: void memory_error() {
1341: fprintf(stderr,"Out of memory. Aborting...\n");
1342: exit(1);
1343: }
1344:
1345:
1346: /* The main program. Parse the command line and do it... */
1347: int main(argc,argv)
1348: int argc;
1349: char **argv;
1350: {
1351: static int version = 0;
1352: static int rpflag = 0;
1353: static int basisflag = 0;
1354: static int compress = 0;
1355: static int quiet = 0;
1356: static int statistics = 0;
1357: static int mhflag = 0;
1358: static struct s_options options[] = {
1359: {OPT_FLAG, "b", (char*)&basisflag, "Print only the basis in report."},
1360: {OPT_FLAG, "c", (char*)&compress, "Don't compress the action table."},
1361: {OPT_FLAG, "g", (char*)&rpflag, "Print grammar without actions."},
1362: {OPT_FLAG, "m", (char*)&mhflag, "Output a makeheaders compatible file"},
1363: {OPT_FLAG, "q", (char*)&quiet, "(Quiet) Don't print the report file."},
1364: {OPT_FLAG, "s", (char*)&statistics, "Print parser stats to standard output."},
1365: {OPT_FLAG, "x", (char*)&version, "Print the version number."},
1366: {OPT_FLAG,0,0,0}
1367: };
1368: int i;
1369: struct lemon lem;
1370: char *def_tmpl_name = "lempar.c";
1371:
1372: UNUSED(argc);
1373: OptInit(argv,options,stderr);
1374: if( version ){
1375: printf("Lemon version 1.0\n");
1376: exit(0);
1377: }
1378: if( OptNArgs() < 1 ){
1379: fprintf(stderr,"Exactly one filename argument is required.\n");
1380: exit(1);
1381: }
1382: lem.errorcnt = 0;
1383:
1384: /* Initialize the machine */
1385: Strsafe_init();
1386: Symbol_init();
1387: State_init();
1388: lem.argv0 = argv[0];
1389: lem.filename = OptArg(0);
1390: lem.tmplname = (OptNArgs() == 2) ? OptArg(1) : def_tmpl_name;
1391: lem.basisflag = basisflag;
1392: lem.has_fallback = 0;
1393: lem.nconflict = 0;
1394: lem.name = lem.include = lem.arg = lem.tokentype = lem.start = 0;
1395: lem.vartype = 0;
1396: lem.stacksize = 0;
1397: lem.error = lem.overflow = lem.failure = lem.accept = lem.tokendest =
1398: lem.tokenprefix = lem.outname = lem.extracode = 0;
1399: lem.vardest = 0;
1400: lem.tablesize = 0;
1401: Symbol_new("$");
1402: lem.errsym = Symbol_new("error");
1403:
1404: /* Parse the input file */
1405: Parse(&lem);
1406: if( lem.errorcnt ) exit(lem.errorcnt);
1407: if( lem.rule==0 ){
1408: fprintf(stderr,"Empty grammar.\n");
1409: exit(1);
1410: }
1411:
1412: /* Count and index the symbols of the grammar */
1413: lem.nsymbol = Symbol_count();
1414: Symbol_new("{default}");
1415: lem.symbols = Symbol_arrayof();
1416: for(i=0; i<=lem.nsymbol; i++) lem.symbols[i]->index = i;
1417: qsort(lem.symbols,lem.nsymbol+1,sizeof(struct symbol*),
1418: (int(*)())Symbolcmpp);
1419: for(i=0; i<=lem.nsymbol; i++) lem.symbols[i]->index = i;
1420: for(i=1; isupper(lem.symbols[i]->name[0]); i++);
1421: lem.nterminal = i;
1422:
1423: /* Generate a reprint of the grammar, if requested on the command line */
1424: if( rpflag ){
1425: Reprint(&lem);
1426: }else{
1427: /* Initialize the size for all follow and first sets */
1428: SetSize(lem.nterminal);
1429:
1430: /* Find the precedence for every production rule (that has one) */
1431: FindRulePrecedences(&lem);
1432:
1433: /* Compute the lambda-nonterminals and the first-sets for every
1434: ** nonterminal */
1435: FindFirstSets(&lem);
1436:
1437: /* Compute all LR(0) states. Also record follow-set propagation
1438: ** links so that the follow-set can be computed later */
1439: lem.nstate = 0;
1440: FindStates(&lem);
1441: lem.sorted = State_arrayof();
1442:
1443: /* Tie up loose ends on the propagation links */
1444: FindLinks(&lem);
1445:
1446: /* Compute the follow set of every reducible configuration */
1447: FindFollowSets(&lem);
1448:
1449: /* Compute the action tables */
1450: FindActions(&lem);
1451:
1452: /* Compress the action tables */
1453: if( compress==0 ) CompressTables(&lem);
1454:
1455: /* Generate a report of the parser generated. (the "y.output" file) */
1456: if( !quiet ) ReportOutput(&lem);
1457:
1458: /* Generate the source code for the parser */
1459: ReportTable(&lem, mhflag);
1460:
1461: /* Produce a header file for use by the scanner. (This step is
1462: ** omitted if the "-m" option is used because makeheaders will
1463: ** generate the file for us.) */
1464: if( !mhflag ) ReportHeader(&lem);
1465: }
1466: if( statistics ){
1467: printf("Parser statistics: %d terminals, %d nonterminals, %d rules\n",
1468: lem.nterminal, lem.nsymbol - lem.nterminal, lem.nrule);
1469: printf(" %d states, %d parser table entries, %d conflicts\n",
1470: lem.nstate, lem.tablesize, lem.nconflict);
1471: }
1472: if( lem.nconflict ){
1473: fprintf(stderr,"%d parsing conflicts.\n",lem.nconflict);
1474: }
1475: exit(lem.errorcnt + lem.nconflict);
1476: }
1477: /******************** From the file "msort.c" *******************************/
1478: /*
1479: ** A generic merge-sort program.
1480: **
1481: ** USAGE:
1482: ** Let "ptr" be a pointer to some structure which is at the head of
1483: ** a null-terminated list. Then to sort the list call:
1484: **
1485: ** ptr = msort(ptr,&(ptr->next),cmpfnc);
1486: **
1487: ** In the above, "cmpfnc" is a pointer to a function which compares
1488: ** two instances of the structure and returns an integer, as in
1489: ** strcmp. The second argument is a pointer to the pointer to the
1490: ** second element of the linked list. This address is used to compute
1491: ** the offset to the "next" field within the structure. The offset to
1492: ** the "next" field must be constant for all structures in the list.
1493: **
1494: ** The function returns a new pointer which is the head of the list
1495: ** after sorting.
1496: **
1497: ** ALGORITHM:
1498: ** Merge-sort.
1499: */
1500:
1501: /*
1502: ** Return a pointer to the next structure in the linked list.
1503: */
1504: #define NEXT(A) (*(char**)(((unsigned long)A)+offset))
1505:
1506: /*
1507: ** Inputs:
1508: ** a: A sorted, null-terminated linked list. (May be null).
1509: ** b: A sorted, null-terminated linked list. (May be null).
1510: ** cmp: A pointer to the comparison function.
1511: ** offset: Offset in the structure to the "next" field.
1512: **
1513: ** Return Value:
1514: ** A pointer to the head of a sorted list containing the elements
1515: ** of both a and b.
1516: **
1517: ** Side effects:
1518: ** The "next" pointers for elements in the lists a and b are
1519: ** changed.
1520: */
1521: static char *merge(a,b,cmp,offset)
1522: char *a;
1523: char *b;
1524: int (*cmp)();
1525: int offset;
1526: {
1527: char *ptr, *head;
1528:
1529: if( a==0 ){
1530: head = b;
1531: }else if( b==0 ){
1532: head = a;
1533: }else{
1534: if( (*cmp)(a,b)<0 ){
1535: ptr = a;
1536: a = NEXT(a);
1537: }else{
1538: ptr = b;
1539: b = NEXT(b);
1540: }
1541: head = ptr;
1542: while( a && b ){
1543: if( (*cmp)(a,b)<0 ){
1544: NEXT(ptr) = a;
1545: ptr = a;
1546: a = NEXT(a);
1547: }else{
1548: NEXT(ptr) = b;
1549: ptr = b;
1550: b = NEXT(b);
1551: }
1552: }
1553: if( a ) NEXT(ptr) = a;
1554: else NEXT(ptr) = b;
1555: }
1556: return head;
1557: }
1558:
1559: /*
1560: ** Inputs:
1561: ** list: Pointer to a singly-linked list of structures.
1562: ** next: Pointer to pointer to the second element of the list.
1563: ** cmp: A comparison function.
1564: **
1565: ** Return Value:
1566: ** A pointer to the head of a sorted list containing the elements
1567: ** orginally in list.
1568: **
1569: ** Side effects:
1570: ** The "next" pointers for elements in list are changed.
1571: */
1572: #define LISTSIZE 30
1573: char *msort(list,next,cmp)
1574: char *list;
1575: char **next;
1576: int (*cmp)();
1577: {
1578: unsigned long offset;
1579: char *ep;
1580: char *set[LISTSIZE];
1581: int i;
1582: offset = (unsigned long)next - (unsigned long)list;
1583: for(i=0; i<LISTSIZE; i++) set[i] = 0;
1584: while( list ){
1585: ep = list;
1586: list = NEXT(list);
1587: NEXT(ep) = 0;
1588: for(i=0; i<LISTSIZE-1 && set[i]!=0; i++){
1589: ep = merge(ep,set[i],cmp,offset);
1590: set[i] = 0;
1591: }
1592: set[i] = ep;
1593: }
1594: ep = 0;
1595: for(i=0; i<LISTSIZE; i++) if( set[i] ) ep = merge(ep,set[i],cmp,offset);
1596: return ep;
1597: }
1598: /************************ From the file "option.c" **************************/
1599: static char **argv;
1600: static struct s_options *op;
1601: static FILE *errstream;
1602:
1603: #define ISOPT(X) ((X)[0]=='-'||(X)[0]=='+'||strchr((X),'=')!=0)
1604:
1605: /*
1606: ** Print the command line with a carrot pointing to the k-th character
1607: ** of the n-th field.
1608: */
1609: static void errline(n,k,err)
1610: int n;
1611: int k;
1612: FILE *err;
1613: {
1614: int spcnt = 0, i;
1615: if( argv[0] ) {
1616: fprintf(err,"%s",argv[0]);
1617: spcnt += strlen(argv[0]) + 1;
1618: }
1619: for(i=1; i<n && argv[i]; i++){
1620: fprintf(err," %s",argv[i]);
1621: spcnt += strlen(argv[i]) + 1;
1622: }
1623: spcnt += k;
1624: for(; argv[i]; i++) fprintf(err," %s",argv[i]);
1625: if( spcnt<20 ){
1626: fprintf(err,"\n%*s^-- here\n",spcnt,"");
1627: }else{
1628: fprintf(err,"\n%*shere --^\n",spcnt-7,"");
1629: }
1630: }
1631:
1632: /*
1633: ** Return the index of the N-th non-switch argument. Return -1
1634: ** if N is out of range.
1635: */
1636: static int argindex(n)
1637: int n;
1638: {
1639: int i;
1640: int dashdash = 0;
1641: if( argv!=0 && *argv!=0 ){
1642: for(i=1; argv[i]; i++){
1643: if( dashdash || !ISOPT(argv[i]) ){
1644: if( n==0 ) return i;
1645: n--;
1646: }
1647: if( strcmp(argv[i],"--")==0 ) dashdash = 1;
1648: }
1649: }
1650: return -1;
1651: }
1652:
1653: static char emsg[] = "Command line syntax error: ";
1654:
1655: /*
1656: ** Process a flag command line argument.
1657: */
1658: static int handleflags(i,err)
1659: int i;
1660: FILE *err;
1661: {
1662: int v;
1663: int errcnt = 0;
1664: int j;
1665: for(j=0; op[j].label; j++){
1666: if( strcmp(&argv[i][1],op[j].label)==0 ) break;
1667: }
1668: v = argv[i][0]=='-' ? 1 : 0;
1669: if( op[j].label==0 ){
1670: if( err ){
1671: fprintf(err,"%sundefined option.\n",emsg);
1672: errline(i,1,err);
1673: }
1674: errcnt++;
1675: }else if( op[j].type==OPT_FLAG ){
1676: *((int*)op[j].arg) = v;
1677: }else if( op[j].type==OPT_FFLAG ){
1678: (*(void(*)())(intptr_t)(op[j].arg))(v);
1679: }else{
1680: if( err ){
1681: fprintf(err,"%smissing argument on switch.\n",emsg);
1682: errline(i,1,err);
1683: }
1684: errcnt++;
1685: }
1686: return errcnt;
1687: }
1688:
1689: /*
1690: ** Process a command line switch which has an argument.
1691: */
1692: static int handleswitch(i,err)
1693: int i;
1694: FILE *err;
1695: {
1696: int lv = 0;
1697: double dv = 0.0;
1698: char *sv = 0, *end;
1699: char *cp;
1700: int j;
1701: int errcnt = 0;
1702: cp = strchr(argv[i],'=');
1703: *cp = 0;
1704: for(j=0; op[j].label; j++){
1705: if( strcmp(argv[i],op[j].label)==0 ) break;
1706: }
1707: *cp = '=';
1708: if( op[j].label==0 ){
1709: if( err ){
1710: fprintf(err,"%sundefined option.\n",emsg);
1711: errline(i,0,err);
1712: }
1713: errcnt++;
1714: }else{
1715: cp++;
1716: switch( op[j].type ){
1717: case OPT_FLAG:
1718: case OPT_FFLAG:
1719: if( err ){
1720: fprintf(err,"%soption requires an argument.\n",emsg);
1721: errline(i,0,err);
1722: }
1723: errcnt++;
1724: break;
1725: case OPT_DBL:
1726: case OPT_FDBL:
1727: dv = strtod(cp,&end);
1728: if( *end ){
1729: if( err ){
1730: fprintf(err,"%sillegal character in floating-point argument.\n",emsg);
1731: errline(i,((unsigned long)end)-(unsigned long)argv[i],err);
1732: }
1733: errcnt++;
1734: }
1735: break;
1736: case OPT_INT:
1737: case OPT_FINT:
1738: lv = strtol(cp,&end,0);
1739: if( *end ){
1740: if( err ){
1741: fprintf(err,"%sillegal character in integer argument.\n",emsg);
1742: errline(i,((unsigned long)end)-(unsigned long)argv[i],err);
1743: }
1744: errcnt++;
1745: }
1746: break;
1747: case OPT_STR:
1748: case OPT_FSTR:
1749: sv = cp;
1750: break;
1751: }
1752: switch( op[j].type ){
1753: case OPT_FLAG:
1754: case OPT_FFLAG:
1755: break;
1756: case OPT_DBL:
1757: *(double*)(op[j].arg) = dv;
1758: break;
1759: case OPT_FDBL:
1760: (*(void(*)())(intptr_t)(op[j].arg))(dv);
1761: break;
1762: case OPT_INT:
1763: *(int*)(op[j].arg) = lv;
1764: break;
1765: case OPT_FINT:
1766: (*(void(*)())(intptr_t)(op[j].arg))((int)lv);
1767: break;
1768: case OPT_STR:
1769: *(char**)(op[j].arg) = sv;
1770: break;
1771: case OPT_FSTR:
1772: (*(void(*)())(intptr_t)(op[j].arg))(sv);
1773: break;
1774: }
1775: }
1776: return errcnt;
1777: }
1778:
1779: int OptInit(a,o,err)
1780: char **a;
1781: struct s_options *o;
1782: FILE *err;
1783: {
1784: int errcnt = 0;
1785: argv = a;
1786: op = o;
1787: errstream = err;
1788: if( argv && *argv && op ){
1789: int i;
1790: for(i=1; argv[i]; i++){
1791: if( argv[i][0]=='+' || argv[i][0]=='-' ){
1792: errcnt += handleflags(i,err);
1793: }else if( strchr(argv[i],'=') ){
1794: errcnt += handleswitch(i,err);
1795: }
1796: }
1797: }
1798: if( errcnt>0 ){
1799: fprintf(err,"Valid command line options for \"%s\" are:\n",*a);
1800: OptPrint();
1801: exit(1);
1802: }
1803: return 0;
1804: }
1805:
1806: int OptNArgs(){
1807: int cnt = 0;
1808: int dashdash = 0;
1809: int i;
1810: if( argv!=0 && argv[0]!=0 ){
1811: for(i=1; argv[i]; i++){
1812: if( dashdash || !ISOPT(argv[i]) ) cnt++;
1813: if( strcmp(argv[i],"--")==0 ) dashdash = 1;
1814: }
1815: }
1816: return cnt;
1817: }
1818:
1819: char *OptArg(n)
1820: int n;
1821: {
1822: int i;
1823: i = argindex(n);
1824: return i>=0 ? argv[i] : 0;
1825: }
1826:
1827: void OptErr(n)
1828: int n;
1829: {
1830: int i;
1831: i = argindex(n);
1832: if( i>=0 ) errline(i,0,errstream);
1833: }
1834:
1835: void OptPrint(){
1836: int i;
1837: int max, len;
1838: max = 0;
1839: for(i=0; op[i].label; i++){
1840: len = strlen(op[i].label) + 1;
1841: switch( op[i].type ){
1842: case OPT_FLAG:
1843: case OPT_FFLAG:
1844: break;
1845: case OPT_INT:
1846: case OPT_FINT:
1847: len += 9; /* length of "<integer>" */
1848: break;
1849: case OPT_DBL:
1850: case OPT_FDBL:
1851: len += 6; /* length of "<real>" */
1852: break;
1853: case OPT_STR:
1854: case OPT_FSTR:
1855: len += 8; /* length of "<string>" */
1856: break;
1857: }
1858: if( len>max ) max = len;
1859: }
1860: for(i=0; op[i].label; i++){
1861: switch( op[i].type ){
1862: case OPT_FLAG:
1863: case OPT_FFLAG:
1864: fprintf(errstream," -%-*s %s\n",max,op[i].label,op[i].message);
1865: break;
1866: case OPT_INT:
1867: case OPT_FINT:
1868: fprintf(errstream," %s=<integer>%*s %s\n",op[i].label,
1869: (int)(max-strlen(op[i].label)-9),"",op[i].message);
1870: break;
1871: case OPT_DBL:
1872: case OPT_FDBL:
1873: fprintf(errstream," %s=<real>%*s %s\n",op[i].label,
1874: (int)(max-strlen(op[i].label)-6),"",op[i].message);
1875: break;
1876: case OPT_STR:
1877: case OPT_FSTR:
1878: fprintf(errstream," %s=<string>%*s %s\n",op[i].label,
1879: (int)(max-strlen(op[i].label)-8),"",op[i].message);
1880: break;
1881: }
1882: }
1883: }
1884: /*********************** From the file "parse.c" ****************************/
1885: /*
1886: ** Input file parser for the LEMON parser generator.
1887: */
1888:
1889: /* The state of the parser */
1890: struct pstate {
1891: char *filename; /* Name of the input file */
1892: int tokenlineno; /* Linenumber at which current token starts */
1893: int errorcnt; /* Number of errors so far */
1894: char *tokenstart; /* Text of current token */
1895: struct lemon *gp; /* Global state vector */
1896: enum e_state {
1897: INITIALIZE,
1898: WAITING_FOR_DECL_OR_RULE,
1899: WAITING_FOR_DECL_KEYWORD,
1900: WAITING_FOR_DECL_ARG,
1901: WAITING_FOR_PRECEDENCE_SYMBOL,
1902: WAITING_FOR_ARROW,
1903: IN_RHS,
1904: LHS_ALIAS_1,
1905: LHS_ALIAS_2,
1906: LHS_ALIAS_3,
1907: RHS_ALIAS_1,
1908: RHS_ALIAS_2,
1909: PRECEDENCE_MARK_1,
1910: PRECEDENCE_MARK_2,
1911: RESYNC_AFTER_RULE_ERROR,
1912: RESYNC_AFTER_DECL_ERROR,
1913: WAITING_FOR_DESTRUCTOR_SYMBOL,
1914: WAITING_FOR_DATATYPE_SYMBOL,
1915: WAITING_FOR_FALLBACK_ID
1916: } state; /* The state of the parser */
1917: struct symbol *fallback; /* The fallback token */
1918: struct symbol *lhs; /* Left-hand side of current rule */
1919: char *lhsalias; /* Alias for the LHS */
1920: int nrhs; /* Number of right-hand side symbols seen */
1921: struct symbol *rhs[MAXRHS]; /* RHS symbols */
1922: char *alias[MAXRHS]; /* Aliases for each RHS symbol (or NULL) */
1923: struct rule *prevrule; /* Previous rule parsed */
1924: char *declkeyword; /* Keyword of a declaration */
1925: char **declargslot; /* Where the declaration argument should be put */
1926: int *decllnslot; /* Where the declaration linenumber is put */
1927: enum e_assoc declassoc; /* Assign this association to decl arguments */
1928: int preccounter; /* Assign this precedence to decl arguments */
1929: struct rule *firstrule; /* Pointer to first rule in the grammar */
1930: struct rule *lastrule; /* Pointer to the most recently parsed rule */
1931: };
1932:
1933: /* Parse a single token */
1934: static void parseonetoken(psp)
1935: struct pstate *psp;
1936: {
1937: char *x;
1938: x = Strsafe(psp->tokenstart); /* Save the token permanently */
1939: #if 0
1940: printf("%s:%d: Token=[%s] state=%d\n",psp->filename,psp->tokenlineno,
1941: x,psp->state);
1942: #endif
1943: switch( psp->state ){
1944: case INITIALIZE:
1945: psp->prevrule = 0;
1946: psp->preccounter = 0;
1947: psp->firstrule = psp->lastrule = 0;
1948: psp->gp->nrule = 0;
1949: /* Fall thru to next case */
1950: case WAITING_FOR_DECL_OR_RULE:
1951: if( x[0]=='%' ){
1952: psp->state = WAITING_FOR_DECL_KEYWORD;
1953: }else if( islower(x[0]) ){
1954: psp->lhs = Symbol_new(x);
1955: psp->nrhs = 0;
1956: psp->lhsalias = 0;
1957: psp->state = WAITING_FOR_ARROW;
1958: }else if( x[0]=='{' ){
1959: if( psp->prevrule==0 ){
1960: ErrorMsg(psp->filename,psp->tokenlineno,
1961: "There is not prior rule opon which to attach the code \
1962: fragment which begins on this line.");
1963: psp->errorcnt++;
1964: }else if( psp->prevrule->code!=0 ){
1965: ErrorMsg(psp->filename,psp->tokenlineno,
1966: "Code fragment beginning on this line is not the first \
1967: to follow the previous rule.");
1968: psp->errorcnt++;
1969: }else{
1970: psp->prevrule->line = psp->tokenlineno;
1971: psp->prevrule->code = &x[1];
1972: }
1973: }else if( x[0]=='[' ){
1974: psp->state = PRECEDENCE_MARK_1;
1975: }else{
1976: ErrorMsg(psp->filename,psp->tokenlineno,
1977: "Token \"%s\" should be either \"%%\" or a nonterminal name.",
1978: x);
1979: psp->errorcnt++;
1980: }
1981: break;
1982: case PRECEDENCE_MARK_1:
1983: if( !isupper(x[0]) ){
1984: ErrorMsg(psp->filename,psp->tokenlineno,
1985: "The precedence symbol must be a terminal.");
1986: psp->errorcnt++;
1987: }else if( psp->prevrule==0 ){
1988: ErrorMsg(psp->filename,psp->tokenlineno,
1989: "There is no prior rule to assign precedence \"[%s]\".",x);
1990: psp->errorcnt++;
1991: }else if( psp->prevrule->precsym!=0 ){
1992: ErrorMsg(psp->filename,psp->tokenlineno,
1993: "Precedence mark on this line is not the first \
1994: to follow the previous rule.");
1995: psp->errorcnt++;
1996: }else{
1997: psp->prevrule->precsym = Symbol_new(x);
1998: }
1999: psp->state = PRECEDENCE_MARK_2;
2000: break;
2001: case PRECEDENCE_MARK_2:
2002: if( x[0]!=']' ){
2003: ErrorMsg(psp->filename,psp->tokenlineno,
2004: "Missing \"]\" on precedence mark.");
2005: psp->errorcnt++;
2006: }
2007: psp->state = WAITING_FOR_DECL_OR_RULE;
2008: break;
2009: case WAITING_FOR_ARROW:
2010: if( x[0]==':' && x[1]==':' && x[2]=='=' ){
2011: psp->state = IN_RHS;
2012: }else if( x[0]=='(' ){
2013: psp->state = LHS_ALIAS_1;
2014: }else{
2015: ErrorMsg(psp->filename,psp->tokenlineno,
2016: "Expected to see a \":\" following the LHS symbol \"%s\".",
2017: psp->lhs->name);
2018: psp->errorcnt++;
2019: psp->state = RESYNC_AFTER_RULE_ERROR;
2020: }
2021: break;
2022: case LHS_ALIAS_1:
2023: if( isalpha(x[0]) ){
2024: psp->lhsalias = x;
2025: psp->state = LHS_ALIAS_2;
2026: }else{
2027: ErrorMsg(psp->filename,psp->tokenlineno,
2028: "\"%s\" is not a valid alias for the LHS \"%s\"\n",
2029: x,psp->lhs->name);
2030: psp->errorcnt++;
2031: psp->state = RESYNC_AFTER_RULE_ERROR;
2032: }
2033: break;
2034: case LHS_ALIAS_2:
2035: if( x[0]==')' ){
2036: psp->state = LHS_ALIAS_3;
2037: }else{
2038: ErrorMsg(psp->filename,psp->tokenlineno,
2039: "Missing \")\" following LHS alias name \"%s\".",psp->lhsalias);
2040: psp->errorcnt++;
2041: psp->state = RESYNC_AFTER_RULE_ERROR;
2042: }
2043: break;
2044: case LHS_ALIAS_3:
2045: if( x[0]==':' && x[1]==':' && x[2]=='=' ){
2046: psp->state = IN_RHS;
2047: }else{
2048: ErrorMsg(psp->filename,psp->tokenlineno,
2049: "Missing \"->\" following: \"%s(%s)\".",
2050: psp->lhs->name,psp->lhsalias);
2051: psp->errorcnt++;
2052: psp->state = RESYNC_AFTER_RULE_ERROR;
2053: }
2054: break;
2055: case IN_RHS:
2056: if( x[0]=='.' ){
2057: struct rule *rp;
2058: rp = (struct rule *)malloc( sizeof(struct rule) +
2059: sizeof(struct symbol*)*psp->nrhs + sizeof(char*)*psp->nrhs );
2060: if( rp==0 ){
2061: ErrorMsg(psp->filename,psp->tokenlineno,
2062: "Can't allocate enough memory for this rule.");
2063: psp->errorcnt++;
2064: psp->prevrule = 0;
2065: }else{
2066: int i;
2067: rp->ruleline = psp->tokenlineno;
2068: rp->rhs = (struct symbol**)&rp[1];
2069: rp->rhsalias = (char**)&(rp->rhs[psp->nrhs]);
2070: for(i=0; i<psp->nrhs; i++){
2071: rp->rhs[i] = psp->rhs[i];
2072: rp->rhsalias[i] = psp->alias[i];
2073: }
2074: rp->lhs = psp->lhs;
2075: rp->lhsalias = psp->lhsalias;
2076: rp->nrhs = psp->nrhs;
2077: rp->code = 0;
2078: rp->precsym = 0;
2079: rp->index = psp->gp->nrule++;
2080: rp->nextlhs = rp->lhs->rule;
2081: rp->lhs->rule = rp;
2082: rp->next = 0;
2083: if( psp->firstrule==0 ){
2084: psp->firstrule = psp->lastrule = rp;
2085: }else{
2086: psp->lastrule->next = rp;
2087: psp->lastrule = rp;
2088: }
2089: psp->prevrule = rp;
2090: }
2091: psp->state = WAITING_FOR_DECL_OR_RULE;
2092: }else if( isalpha(x[0]) ){
2093: if( psp->nrhs>=MAXRHS ){
2094: ErrorMsg(psp->filename,psp->tokenlineno,
2095: "Too many symbol on RHS or rule beginning at \"%s\".",
2096: x);
2097: psp->errorcnt++;
2098: psp->state = RESYNC_AFTER_RULE_ERROR;
2099: }else{
2100: psp->rhs[psp->nrhs] = Symbol_new(x);
2101: psp->alias[psp->nrhs] = 0;
2102: psp->nrhs++;
2103: }
2104: }else if( x[0]=='(' && psp->nrhs>0 ){
2105: psp->state = RHS_ALIAS_1;
2106: }else{
2107: ErrorMsg(psp->filename,psp->tokenlineno,
2108: "Illegal character on RHS of rule: \"%s\".",x);
2109: psp->errorcnt++;
2110: psp->state = RESYNC_AFTER_RULE_ERROR;
2111: }
2112: break;
2113: case RHS_ALIAS_1:
2114: if( isalpha(x[0]) ){
2115: psp->alias[psp->nrhs-1] = x;
2116: psp->state = RHS_ALIAS_2;
2117: }else{
2118: ErrorMsg(psp->filename,psp->tokenlineno,
2119: "\"%s\" is not a valid alias for the RHS symbol \"%s\"\n",
2120: x,psp->rhs[psp->nrhs-1]->name);
2121: psp->errorcnt++;
2122: psp->state = RESYNC_AFTER_RULE_ERROR;
2123: }
2124: break;
2125: case RHS_ALIAS_2:
2126: if( x[0]==')' ){
2127: psp->state = IN_RHS;
2128: }else{
2129: ErrorMsg(psp->filename,psp->tokenlineno,
2130: "Missing \")\" following LHS alias name \"%s\".",psp->lhsalias);
2131: psp->errorcnt++;
2132: psp->state = RESYNC_AFTER_RULE_ERROR;
2133: }
2134: break;
2135: case WAITING_FOR_DECL_KEYWORD:
2136: if( isalpha(x[0]) ){
2137: psp->declkeyword = x;
2138: psp->declargslot = 0;
2139: psp->decllnslot = 0;
2140: psp->state = WAITING_FOR_DECL_ARG;
2141: if( strcmp(x,"name")==0 ){
2142: psp->declargslot = &(psp->gp->name);
2143: }else if( strcmp(x,"include")==0 ){
2144: psp->declargslot = &(psp->gp->include);
2145: psp->decllnslot = &psp->gp->includeln;
2146: }else if( strcmp(x,"code")==0 ){
2147: psp->declargslot = &(psp->gp->extracode);
2148: psp->decllnslot = &psp->gp->extracodeln;
2149: }else if( strcmp(x,"token_destructor")==0 ){
2150: psp->declargslot = &psp->gp->tokendest;
2151: psp->decllnslot = &psp->gp->tokendestln;
2152: }else if( strcmp(x,"default_destructor")==0 ){
2153: psp->declargslot = &psp->gp->vardest;
2154: psp->decllnslot = &psp->gp->vardestln;
2155: }else if( strcmp(x,"token_prefix")==0 ){
2156: psp->declargslot = &psp->gp->tokenprefix;
2157: }else if( strcmp(x,"syntax_error")==0 ){
2158: psp->declargslot = &(psp->gp->error);
2159: psp->decllnslot = &psp->gp->errorln;
2160: }else if( strcmp(x,"parse_accept")==0 ){
2161: psp->declargslot = &(psp->gp->accept);
2162: psp->decllnslot = &psp->gp->acceptln;
2163: }else if( strcmp(x,"parse_failure")==0 ){
2164: psp->declargslot = &(psp->gp->failure);
2165: psp->decllnslot = &psp->gp->failureln;
2166: }else if( strcmp(x,"stack_overflow")==0 ){
2167: psp->declargslot = &(psp->gp->overflow);
2168: psp->decllnslot = &psp->gp->overflowln;
2169: }else if( strcmp(x,"extra_argument")==0 ){
2170: psp->declargslot = &(psp->gp->arg);
2171: }else if( strcmp(x,"token_type")==0 ){
2172: psp->declargslot = &(psp->gp->tokentype);
2173: }else if( strcmp(x,"default_type")==0 ){
2174: psp->declargslot = &(psp->gp->vartype);
2175: }else if( strcmp(x,"stack_size")==0 ){
2176: psp->declargslot = &(psp->gp->stacksize);
2177: }else if( strcmp(x,"start_symbol")==0 ){
2178: psp->declargslot = &(psp->gp->start);
2179: }else if( strcmp(x,"left")==0 ){
2180: psp->preccounter++;
2181: psp->declassoc = LEFT;
2182: psp->state = WAITING_FOR_PRECEDENCE_SYMBOL;
2183: }else if( strcmp(x,"right")==0 ){
2184: psp->preccounter++;
2185: psp->declassoc = RIGHT;
2186: psp->state = WAITING_FOR_PRECEDENCE_SYMBOL;
2187: }else if( strcmp(x,"nonassoc")==0 ){
2188: psp->preccounter++;
2189: psp->declassoc = NONE;
2190: psp->state = WAITING_FOR_PRECEDENCE_SYMBOL;
2191: }else if( strcmp(x,"destructor")==0 ){
2192: psp->state = WAITING_FOR_DESTRUCTOR_SYMBOL;
2193: }else if( strcmp(x,"type")==0 ){
2194: psp->state = WAITING_FOR_DATATYPE_SYMBOL;
2195: }else if( strcmp(x,"fallback")==0 ){
2196: psp->fallback = 0;
2197: psp->state = WAITING_FOR_FALLBACK_ID;
2198: }else{
2199: ErrorMsg(psp->filename,psp->tokenlineno,
2200: "Unknown declaration keyword: \"%%%s\".",x);
2201: psp->errorcnt++;
2202: psp->state = RESYNC_AFTER_DECL_ERROR;
2203: }
2204: }else{
2205: ErrorMsg(psp->filename,psp->tokenlineno,
2206: "Illegal declaration keyword: \"%s\".",x);
2207: psp->errorcnt++;
2208: psp->state = RESYNC_AFTER_DECL_ERROR;
2209: }
2210: break;
2211: case WAITING_FOR_DESTRUCTOR_SYMBOL:
2212: if( !isalpha(x[0]) ){
2213: ErrorMsg(psp->filename,psp->tokenlineno,
2214: "Symbol name missing after %destructor keyword");
2215: psp->errorcnt++;
2216: psp->state = RESYNC_AFTER_DECL_ERROR;
2217: }else{
2218: struct symbol *sp = Symbol_new(x);
2219: psp->declargslot = &sp->destructor;
2220: psp->decllnslot = &sp->destructorln;
2221: psp->state = WAITING_FOR_DECL_ARG;
2222: }
2223: break;
2224: case WAITING_FOR_DATATYPE_SYMBOL:
2225: if( !isalpha(x[0]) ){
2226: ErrorMsg(psp->filename,psp->tokenlineno,
2227: "Symbol name missing after %destructor keyword");
2228: psp->errorcnt++;
2229: psp->state = RESYNC_AFTER_DECL_ERROR;
2230: }else{
2231: struct symbol *sp = Symbol_new(x);
2232: psp->declargslot = &sp->datatype;
2233: psp->decllnslot = 0;
2234: psp->state = WAITING_FOR_DECL_ARG;
2235: }
2236: break;
2237: case WAITING_FOR_PRECEDENCE_SYMBOL:
2238: if( x[0]=='.' ){
2239: psp->state = WAITING_FOR_DECL_OR_RULE;
2240: }else if( isupper(x[0]) ){
2241: struct symbol *sp;
2242: sp = Symbol_new(x);
2243: if( sp->prec>=0 ){
2244: ErrorMsg(psp->filename,psp->tokenlineno,
2245: "Symbol \"%s\" has already be given a precedence.",x);
2246: psp->errorcnt++;
2247: }else{
2248: sp->prec = psp->preccounter;
2249: sp->assoc = psp->declassoc;
2250: }
2251: }else{
2252: ErrorMsg(psp->filename,psp->tokenlineno,
2253: "Can't assign a precedence to \"%s\".",x);
2254: psp->errorcnt++;
2255: }
2256: break;
2257: case WAITING_FOR_DECL_ARG:
2258: if( (x[0]=='{' || x[0]=='\"' || isalnum(x[0])) ){
2259: if( *(psp->declargslot)!=0 ){
2260: ErrorMsg(psp->filename,psp->tokenlineno,
2261: "The argument \"%s\" to declaration \"%%%s\" is not the first.",
2262: x[0]=='\"' ? &x[1] : x,psp->declkeyword);
2263: psp->errorcnt++;
2264: psp->state = RESYNC_AFTER_DECL_ERROR;
2265: }else{
2266: *(psp->declargslot) = (x[0]=='\"' || x[0]=='{') ? &x[1] : x;
2267: if( psp->decllnslot ) *psp->decllnslot = psp->tokenlineno;
2268: psp->state = WAITING_FOR_DECL_OR_RULE;
2269: }
2270: }else{
2271: ErrorMsg(psp->filename,psp->tokenlineno,
2272: "Illegal argument to %%%s: %s",psp->declkeyword,x);
2273: psp->errorcnt++;
2274: psp->state = RESYNC_AFTER_DECL_ERROR;
2275: }
2276: break;
2277: case WAITING_FOR_FALLBACK_ID:
2278: if( x[0]=='.' ){
2279: psp->state = WAITING_FOR_DECL_OR_RULE;
2280: }else if( !isupper(x[0]) ){
2281: ErrorMsg(psp->filename, psp->tokenlineno,
2282: "%%fallback argument \"%s\" should be a token", x);
2283: psp->errorcnt++;
2284: }else{
2285: struct symbol *sp = Symbol_new(x);
2286: if( psp->fallback==0 ){
2287: psp->fallback = sp;
2288: }else if( sp->fallback ){
2289: ErrorMsg(psp->filename, psp->tokenlineno,
2290: "More than one fallback assigned to token %s", x);
2291: psp->errorcnt++;
2292: }else{
2293: sp->fallback = psp->fallback;
2294: psp->gp->has_fallback = 1;
2295: }
2296: }
2297: break;
2298: case RESYNC_AFTER_RULE_ERROR:
2299: /* if( x[0]=='.' ) psp->state = WAITING_FOR_DECL_OR_RULE;
2300: ** break; */
2301: case RESYNC_AFTER_DECL_ERROR:
2302: if( x[0]=='.' ) psp->state = WAITING_FOR_DECL_OR_RULE;
2303: if( x[0]=='%' ) psp->state = WAITING_FOR_DECL_KEYWORD;
2304: break;
2305: }
2306: }
2307:
2308: /* In spite of its name, this function is really a scanner. It read
2309: ** in the entire input file (all at once) then tokenizes it. Each
2310: ** token is passed to the function "parseonetoken" which builds all
2311: ** the appropriate data structures in the global state vector "gp".
2312: */
2313: struct pstate ps;
2314: void Parse(gp)
2315: struct lemon *gp;
2316: {
2317: FILE *fp;
2318: char *filebuf;
2319: size_t filesize;
2320: int lineno;
2321: int c;
2322: char *cp, *nextcp;
2323: int startline = 0;
2324:
2325: ps.gp = gp;
2326: ps.filename = gp->filename;
2327: ps.errorcnt = 0;
2328: ps.state = INITIALIZE;
2329:
2330: /* Begin by reading the input file */
2331: fp = fopen(ps.filename,"rb");
2332: if( fp==0 ){
2333: ErrorMsg(ps.filename,0,"Can't open this file for reading.");
2334: gp->errorcnt++;
2335: return;
2336: }
2337: fseek(fp,0,2);
2338: filesize = ftell(fp);
2339: rewind(fp);
2340: filebuf = (char *)malloc( filesize+1 );
2341: if( filebuf==0 ){
2342: ErrorMsg(ps.filename,0,"Can't allocate %d of memory to hold this file.",
2343: filesize+1);
2344: fclose(fp);
2345: gp->errorcnt++;
2346: return;
2347: }
2348: if( fread(filebuf,1,filesize,fp)!=filesize ){
2349: ErrorMsg(ps.filename,0,"Can't read in all %d bytes of this file.",
2350: filesize);
2351: free(filebuf);
2352: fclose(fp);
2353: gp->errorcnt++;
2354: return;
2355: }
2356: fclose(fp);
2357: filebuf[filesize] = 0;
2358:
2359: /* Now scan the text of the input file */
2360: lineno = 1;
2361: for(cp=filebuf; (c= *cp)!=0; ){
2362: if( c=='\n' ) lineno++; /* Keep track of the line number */
2363: if( isspace(c) ){ cp++; continue; } /* Skip all white space */
2364: if( c=='/' && cp[1]=='/' ){ /* Skip C++ style comments */
2365: cp+=2;
2366: while( (c= *cp)!=0 && c!='\n' ) cp++;
2367: continue;
2368: }
2369: if( c=='/' && cp[1]=='*' ){ /* Skip C style comments */
2370: cp+=2;
2371: while( (c= *cp)!=0 && (c!='/' || cp[-1]!='*') ){
2372: if( c=='\n' ) lineno++;
2373: cp++;
2374: }
2375: if( c ) cp++;
2376: continue;
2377: }
2378: ps.tokenstart = cp; /* Mark the beginning of the token */
2379: ps.tokenlineno = lineno; /* Linenumber on which token begins */
2380: if( c=='\"' ){ /* String literals */
2381: cp++;
2382: while( (c= *cp)!=0 && c!='\"' ){
2383: if( c=='\n' ) lineno++;
2384: cp++;
2385: }
2386: if( c==0 ){
2387: ErrorMsg(ps.filename,startline,
2388: "String starting on this line is not terminated before the end of the file.");
2389: ps.errorcnt++;
2390: nextcp = cp;
2391: }else{
2392: nextcp = cp+1;
2393: }
2394: }else if( c=='{' ){ /* A block of C code */
2395: int level;
2396: cp++;
2397: for(level=1; (c= *cp)!=0 && (level>1 || c!='}'); cp++){
2398: if( c=='\n' ) lineno++;
2399: else if( c=='{' ) level++;
2400: else if( c=='}' ) level--;
2401: else if( c=='/' && cp[1]=='*' ){ /* Skip comments */
2402: int prevc;
2403: cp = &cp[2];
2404: prevc = 0;
2405: while( (c= *cp)!=0 && (c!='/' || prevc!='*') ){
2406: if( c=='\n' ) lineno++;
2407: prevc = c;
2408: cp++;
2409: }
2410: }else if( c=='/' && cp[1]=='/' ){ /* Skip C++ style comments too */
2411: cp = &cp[2];
2412: while( (c= *cp)!=0 && c!='\n' ) cp++;
2413: if( c ) lineno++;
2414: }else if( c=='\'' || c=='\"' ){ /* String a character literals */
2415: int startchar, prevc;
2416: startchar = c;
2417: prevc = 0;
2418: for(cp++; (c= *cp)!=0 && (c!=startchar || prevc=='\\'); cp++){
2419: if( c=='\n' ) lineno++;
2420: if( prevc=='\\' ) prevc = 0;
2421: else prevc = c;
2422: }
2423: }
2424: }
2425: if( c==0 ){
2426: ErrorMsg(ps.filename,ps.tokenlineno,
2427: "C code starting on this line is not terminated before the end of the file.");
2428: ps.errorcnt++;
2429: nextcp = cp;
2430: }else{
2431: nextcp = cp+1;
2432: }
2433: }else if( isalnum(c) ){ /* Identifiers */
2434: while( (c= *cp)!=0 && (isalnum(c) || c=='_') ) cp++;
2435: nextcp = cp;
2436: }else if( c==':' && cp[1]==':' && cp[2]=='=' ){ /* The operator "::=" */
2437: cp += 3;
2438: nextcp = cp;
2439: }else{ /* All other (one character) operators */
2440: cp++;
2441: nextcp = cp;
2442: }
2443: c = *cp;
2444: *cp = 0; /* Null terminate the token */
2445: parseonetoken(&ps); /* Parse the token */
2446: *cp = c; /* Restore the buffer */
2447: cp = nextcp;
2448: }
2449: free(filebuf); /* Release the buffer after parsing */
2450: gp->rule = ps.firstrule;
2451: gp->errorcnt = ps.errorcnt;
2452: }
2453: /*************************** From the file "plink.c" *********************/
2454: /*
2455: ** Routines processing configuration follow-set propagation links
2456: ** in the LEMON parser generator.
2457: */
2458: static struct plink *plink_freelist = 0;
2459:
2460: /* Allocate a new plink */
2461: struct plink *Plink_new(){
2462: struct plink *new;
2463:
2464: if( plink_freelist==0 ){
2465: int i;
2466: int amt = 100;
2467: plink_freelist = (struct plink *)malloc( sizeof(struct plink)*amt );
2468: if( plink_freelist==0 ){
2469: fprintf(stderr,
2470: "Unable to allocate memory for a new follow-set propagation link.\n");
2471: exit(1);
2472: }
2473: for(i=0; i<amt-1; i++) plink_freelist[i].next = &plink_freelist[i+1];
2474: plink_freelist[amt-1].next = 0;
2475: }
2476: new = plink_freelist;
2477: plink_freelist = plink_freelist->next;
2478: return new;
2479: }
2480:
2481: /* Add a plink to a plink list */
2482: void Plink_add(plpp,cfp)
2483: struct plink **plpp;
2484: struct config *cfp;
2485: {
2486: struct plink *new;
2487: new = Plink_new();
2488: new->next = *plpp;
2489: *plpp = new;
2490: new->cfp = cfp;
2491: }
2492:
2493: /* Transfer every plink on the list "from" to the list "to" */
2494: void Plink_copy(to,from)
2495: struct plink **to;
2496: struct plink *from;
2497: {
2498: struct plink *nextpl;
2499: while( from ){
2500: nextpl = from->next;
2501: from->next = *to;
2502: *to = from;
2503: from = nextpl;
2504: }
2505: }
2506:
2507: /* Delete every plink on the list */
2508: void Plink_delete(plp)
2509: struct plink *plp;
2510: {
2511: struct plink *nextpl;
2512:
2513: while( plp ){
2514: nextpl = plp->next;
2515: plp->next = plink_freelist;
2516: plink_freelist = plp;
2517: plp = nextpl;
2518: }
2519: }
2520: /*********************** From the file "report.c" **************************/
2521: /*
2522: ** Procedures for generating reports and tables in the LEMON parser generator.
2523: */
2524:
2525: /* Generate a filename with the given suffix. Space to hold the
2526: ** name comes from malloc() and must be freed by the calling
2527: ** function.
2528: */
2529: PRIVATE char *file_makename(lemp,suffix)
2530: struct lemon *lemp;
2531: char *suffix;
2532: {
2533: char *name;
2534: char *cp;
2535:
2536: name = malloc( strlen(lemp->filename) + strlen(suffix) + 5 );
2537: if( name==0 ){
2538: fprintf(stderr,"Can't allocate space for a filename.\n");
2539: exit(1);
2540: }
2541: /* skip directory, JK */
2542: if (NULL == (cp = strrchr(lemp->filename, '/'))) {
2543: cp = lemp->filename;
2544: } else {
2545: cp++;
2546: }
2547: strcpy(name,cp);
2548: cp = strrchr(name,'.');
2549: if( cp ) *cp = 0;
2550: strcat(name,suffix);
2551: return name;
2552: }
2553:
2554: /* Open a file with a name based on the name of the input file,
2555: ** but with a different (specified) suffix, and return a pointer
2556: ** to the stream */
2557: PRIVATE FILE *file_open(lemp,suffix,mode)
2558: struct lemon *lemp;
2559: char *suffix;
2560: char *mode;
2561: {
2562: FILE *fp;
2563:
2564: if( lemp->outname ) free(lemp->outname);
2565: lemp->outname = file_makename(lemp, suffix);
2566: fp = fopen(lemp->outname,mode);
2567: if( fp==0 && *mode=='w' ){
2568: fprintf(stderr,"Can't open file \"%s\".\n",lemp->outname);
2569: lemp->errorcnt++;
2570: return 0;
2571: }
2572: return fp;
2573: }
2574:
2575: /* Duplicate the input file without comments and without actions
2576: ** on rules */
2577: void Reprint(lemp)
2578: struct lemon *lemp;
2579: {
2580: struct rule *rp;
2581: struct symbol *sp;
2582: int i, j, maxlen, len, ncolumns, skip;
2583: printf("// Reprint of input file \"%s\".\n// Symbols:\n",lemp->filename);
2584: maxlen = 10;
2585: for(i=0; i<lemp->nsymbol; i++){
2586: sp = lemp->symbols[i];
2587: len = strlen(sp->name);
2588: if( len>maxlen ) maxlen = len;
2589: }
2590: ncolumns = 76/(maxlen+5);
2591: if( ncolumns<1 ) ncolumns = 1;
2592: skip = (lemp->nsymbol + ncolumns - 1)/ncolumns;
2593: for(i=0; i<skip; i++){
2594: printf("//");
2595: for(j=i; j<lemp->nsymbol; j+=skip){
2596: sp = lemp->symbols[j];
2597: assert( sp->index==j );
2598: printf(" %3d %-*.*s",j,maxlen,maxlen,sp->name);
2599: }
2600: printf("\n");
2601: }
2602: for(rp=lemp->rule; rp; rp=rp->next){
2603: printf("%s",rp->lhs->name);
2604: /* if( rp->lhsalias ) printf("(%s)",rp->lhsalias); */
2605: printf(" ::=");
2606: for(i=0; i<rp->nrhs; i++){
2607: printf(" %s",rp->rhs[i]->name);
2608: /* if( rp->rhsalias[i] ) printf("(%s)",rp->rhsalias[i]); */
2609: }
2610: printf(".");
2611: if( rp->precsym ) printf(" [%s]",rp->precsym->name);
2612: /* if( rp->code ) printf("\n %s",rp->code); */
2613: printf("\n");
2614: }
2615: }
2616:
2617: PRIVATE void ConfigPrint(fp,cfp)
2618: FILE *fp;
2619: struct config *cfp;
2620: {
2621: struct rule *rp;
2622: int i;
2623: rp = cfp->rp;
2624: fprintf(fp,"%s ::=",rp->lhs->name);
2625: for(i=0; i<=rp->nrhs; i++){
2626: if( i==cfp->dot ) fprintf(fp," *");
2627: if( i==rp->nrhs ) break;
2628: fprintf(fp," %s",rp->rhs[i]->name);
2629: }
2630: }
2631:
2632: /* #define TEST */
2633: #ifdef TEST
2634: /* Print a set */
2635: PRIVATE void SetPrint(out,set,lemp)
2636: FILE *out;
2637: char *set;
2638: struct lemon *lemp;
2639: {
2640: int i;
2641: char *spacer;
2642: spacer = "";
2643: fprintf(out,"%12s[","");
2644: for(i=0; i<lemp->nterminal; i++){
2645: if( SetFind(set,i) ){
2646: fprintf(out,"%s%s",spacer,lemp->symbols[i]->name);
2647: spacer = " ";
2648: }
2649: }
2650: fprintf(out,"]\n");
2651: }
2652:
2653: /* Print a plink chain */
2654: void PlinkPrint(out,plp,tag)
2655: FILE *out;
2656: struct plink *plp;
2657: char *tag;
2658: {
2659: while( plp ){
2660: fprintf(out,"%12s%s (state %2d) ","",tag,plp->cfp->stp->index);
2661: ConfigPrint(out,plp->cfp);
2662: fprintf(out,"\n");
2663: plp = plp->next;
2664: }
2665: }
2666: #endif
2667:
2668: /* Print an action to the given file descriptor. Return FALSE if
2669: ** nothing was actually printed.
2670: */
2671: PRIVATE int PrintAction(struct action *ap, FILE *fp, int indent){
2672: int result = 1;
2673: switch( ap->type ){
2674: case SHIFT:
2675: fprintf(fp,"%*s shift %d",indent,ap->sp->name,ap->x.stp->index);
2676: break;
2677: case REDUCE:
2678: fprintf(fp,"%*s reduce %d",indent,ap->sp->name,ap->x.rp->index);
2679: break;
2680: case ACCEPT:
2681: fprintf(fp,"%*s accept",indent,ap->sp->name);
2682: break;
2683: case ERROR:
2684: fprintf(fp,"%*s error",indent,ap->sp->name);
2685: break;
2686: case CONFLICT:
2687: fprintf(fp,"%*s reduce %-3d ** Parsing conflict **",
2688: indent,ap->sp->name,ap->x.rp->index);
2689: break;
2690: case SH_RESOLVED:
2691: case RD_RESOLVED:
2692: case NOT_USED:
2693: result = 0;
2694: break;
2695: }
2696: return result;
2697: }
2698:
2699: /* Generate the "y.output" log file */
2700: void ReportOutput(lemp)
2701: struct lemon *lemp;
2702: {
2703: int i;
2704: struct state *stp;
2705: struct config *cfp;
2706: struct action *ap;
2707: FILE *fp;
2708:
2709: fp = file_open(lemp,".out","w");
2710: if( fp==0 ) return;
2711: fprintf(fp," \b");
2712: for(i=0; i<lemp->nstate; i++){
2713: stp = lemp->sorted[i];
2714: fprintf(fp,"State %d:\n",stp->index);
2715: if( lemp->basisflag ) cfp=stp->bp;
2716: else cfp=stp->cfp;
2717: while( cfp ){
2718: char buf[20];
2719: if( cfp->dot==cfp->rp->nrhs ){
2720: sprintf(buf,"(%d)",cfp->rp->index);
2721: fprintf(fp," %5s ",buf);
2722: }else{
2723: fprintf(fp," ");
2724: }
2725: ConfigPrint(fp,cfp);
2726: fprintf(fp,"\n");
2727: #ifdef TEST
2728: SetPrint(fp,cfp->fws,lemp);
2729: PlinkPrint(fp,cfp->fplp,"To ");
2730: PlinkPrint(fp,cfp->bplp,"From");
2731: #endif
2732: if( lemp->basisflag ) cfp=cfp->bp;
2733: else cfp=cfp->next;
2734: }
2735: fprintf(fp,"\n");
2736: for(ap=stp->ap; ap; ap=ap->next){
2737: if( PrintAction(ap,fp,30) ) fprintf(fp,"\n");
2738: }
2739: fprintf(fp,"\n");
2740: }
2741: fclose(fp);
2742: return;
2743: }
2744:
2745: extern int access();
2746: /* Search for the file "name" which is in the same directory as
2747: ** the exacutable */
2748: PRIVATE char *pathsearch(argv0,name,modemask)
2749: char *argv0;
2750: char *name;
2751: int modemask;
2752: {
2753: char *pathlist;
2754: char *path,*cp;
2755: char c;
2756:
2757: #ifdef __WIN32__
2758: cp = strrchr(argv0,'\\');
2759: #else
2760: cp = strrchr(argv0,'/');
2761: #endif
2762: if( cp ){
2763: c = *cp;
2764: *cp = 0;
2765: path = (char *)malloc( strlen(argv0) + strlen(name) + 2 );
2766: if( path ) sprintf(path,"%s/%s",argv0,name);
2767: *cp = c;
2768: }else{
2769: pathlist = getenv("PATH");
2770: if( pathlist==0 ) pathlist = ".:/bin:/usr/bin";
2771: path = (char *)malloc( strlen(pathlist)+strlen(name)+2 );
2772: if( path!=0 ){
2773: while( *pathlist ){
2774: cp = strchr(pathlist,':');
2775: if( cp==0 ) cp = &pathlist[strlen(pathlist)];
2776: c = *cp;
2777: *cp = 0;
2778: sprintf(path,"%s/%s",pathlist,name);
2779: *cp = c;
2780: if( c==0 ) pathlist = "";
2781: else pathlist = &cp[1];
2782: if( access(path,modemask)==0 ) break;
2783: }
2784: }
2785: }
2786: return path;
2787: }
2788:
2789: /* Given an action, compute the integer value for that action
2790: ** which is to be put in the action table of the generated machine.
2791: ** Return negative if no action should be generated.
2792: */
2793: PRIVATE int compute_action(lemp,ap)
2794: struct lemon *lemp;
2795: struct action *ap;
2796: {
2797: int act;
2798: switch( ap->type ){
2799: case SHIFT: act = ap->x.stp->index; break;
2800: case REDUCE: act = ap->x.rp->index + lemp->nstate; break;
2801: case ERROR: act = lemp->nstate + lemp->nrule; break;
2802: case ACCEPT: act = lemp->nstate + lemp->nrule + 1; break;
2803: default: act = -1; break;
2804: }
2805: return act;
2806: }
2807:
2808: #define LINESIZE 1000
2809: /* The next cluster of routines are for reading the template file
2810: ** and writing the results to the generated parser */
2811: /* The first function transfers data from "in" to "out" until
2812: ** a line is seen which begins with "%%". The line number is
2813: ** tracked.
2814: **
2815: ** if name!=0, then any word that begin with "Parse" is changed to
2816: ** begin with *name instead.
2817: */
2818: PRIVATE void tplt_xfer(name,in,out,lineno)
2819: char *name;
2820: FILE *in;
2821: FILE *out;
2822: int *lineno;
2823: {
2824: int i, iStart;
2825: char line[LINESIZE];
2826: while( fgets(line,LINESIZE,in) && (line[0]!='%' || line[1]!='%') ){
2827: (*lineno)++;
2828: iStart = 0;
2829: if( name ){
2830: for(i=0; line[i]; i++){
2831: if( line[i]=='P' && strncmp(&line[i],"Parse",5)==0
2832: && (i==0 || !isalpha(line[i-1]))
2833: ){
2834: if( i>iStart ) fprintf(out,"%.*s",i-iStart,&line[iStart]);
2835: fprintf(out,"%s",name);
2836: i += 4;
2837: iStart = i+1;
2838: }
2839: }
2840: }
2841: fprintf(out,"%s",&line[iStart]);
2842: }
2843: }
2844:
2845: /* The next function finds the template file and opens it, returning
2846: ** a pointer to the opened file. */
2847: PRIVATE FILE *tplt_open(lemp)
2848: struct lemon *lemp;
2849: {
2850:
2851: char buf[1000];
2852: FILE *in;
2853: char *tpltname;
2854: char *cp;
2855:
2856: cp = strrchr(lemp->filename,'.');
2857: if( cp ){
2858: sprintf(buf,"%.*s.lt",(int)(cp-lemp->filename),lemp->filename);
2859: }else{
2860: sprintf(buf,"%s.lt",lemp->filename);
2861: }
2862: if( access(buf,004)==0 ){
2863: tpltname = buf;
2864: }else if( access(lemp->tmplname,004)==0 ){
2865: tpltname = lemp->tmplname;
2866: }else{
2867: tpltname = pathsearch(lemp->argv0,lemp->tmplname,0);
2868: }
2869: if( tpltname==0 ){
2870: fprintf(stderr,"Can't find the parser driver template file \"%s\".\n",
2871: lemp->tmplname);
2872: lemp->errorcnt++;
2873: return 0;
2874: }
2875: in = fopen(tpltname,"r");
2876: if( in==0 ){
2877: fprintf(stderr,"Can't open the template file \"%s\".\n",lemp->tmplname);
2878: lemp->errorcnt++;
2879: return 0;
2880: }
2881: return in;
2882: }
2883:
2884: /* Print a string to the file and keep the linenumber up to date */
2885: PRIVATE void tplt_print(out,lemp,str,strln,lineno)
2886: FILE *out;
2887: struct lemon *lemp;
2888: char *str;
2889: int strln;
2890: int *lineno;
2891: {
2892: if( str==0 ) return;
2893: fprintf(out,"#line %d \"%s\"\n",strln,lemp->filename); (*lineno)++;
2894: while( *str ){
2895: if( *str=='\n' ) (*lineno)++;
2896: putc(*str,out);
2897: str++;
2898: }
2899: fprintf(out,"\n#line %d \"%s\"\n",*lineno+2,lemp->outname); (*lineno)+=2;
2900: return;
2901: }
2902:
2903: /*
2904: ** The following routine emits code for the destructor for the
2905: ** symbol sp
2906: */
2907: PRIVATE void emit_destructor_code(out,sp,lemp,lineno)
2908: FILE *out;
2909: struct symbol *sp;
2910: struct lemon *lemp;
2911: int *lineno;
2912: {
2913: char *cp = 0;
2914:
2915: int linecnt = 0;
2916: if( sp->type==TERMINAL ){
2917: cp = lemp->tokendest;
2918: if( cp==0 ) return;
2919: fprintf(out,"#line %d \"%s\"\n{",lemp->tokendestln,lemp->filename);
2920: }else if( sp->destructor ){
2921: cp = sp->destructor;
2922: fprintf(out,"#line %d \"%s\"\n{",sp->destructorln,lemp->filename);
2923: }else{
2924: cp = lemp->vardest;
2925: if( cp==0 ) return;
2926: fprintf(out,"#line %d \"%s\"\n{",lemp->vardestln,lemp->filename);
2927: }
2928: for(; *cp; cp++){
2929: if( *cp=='$' && cp[1]=='$' ){
2930: fprintf(out,"(yypminor->yy%d)",sp->dtnum);
2931: cp++;
2932: continue;
2933: }
2934: if( *cp=='\n' ) linecnt++;
2935: fputc(*cp,out);
2936: }
2937: (*lineno) += 3 + linecnt;
2938: fprintf(out,"}\n#line %d \"%s\"\n",*lineno,lemp->outname);
2939: return;
2940: }
2941:
2942: /*
2943: ** Return TRUE (non-zero) if the given symbol has a destructor.
2944: */
2945: PRIVATE int has_destructor(sp, lemp)
2946: struct symbol *sp;
2947: struct lemon *lemp;
2948: {
2949: int ret;
2950: if( sp->type==TERMINAL ){
2951: ret = lemp->tokendest!=0;
2952: }else{
2953: ret = lemp->vardest!=0 || sp->destructor!=0;
2954: }
2955: return ret;
2956: }
2957:
2958: /*
2959: ** Generate code which executes when the rule "rp" is reduced. Write
2960: ** the code to "out". Make sure lineno stays up-to-date.
2961: */
2962: PRIVATE void emit_code(out,rp,lemp,lineno)
2963: FILE *out;
2964: struct rule *rp;
2965: struct lemon *lemp;
2966: int *lineno;
2967: {
2968: char *cp, *xp;
2969: int linecnt = 0;
2970: int i;
2971: char lhsused = 0; /* True if the LHS element has been used */
2972: char used[MAXRHS]; /* True for each RHS element which is used */
2973:
2974: for(i=0; i<rp->nrhs; i++) used[i] = 0;
2975: lhsused = 0;
2976:
2977: /* Generate code to do the reduce action */
2978: if( rp->code ){
2979: fprintf(out,"#line %d \"%s\"\n{",rp->line,lemp->filename);
2980: for(cp=rp->code; *cp; cp++){
2981: if( isalpha(*cp) && (cp==rp->code || (!isalnum(cp[-1]) && cp[-1]!='_')) ){
2982: char saved;
2983: for(xp= &cp[1]; isalnum(*xp) || *xp=='_'; xp++);
2984: saved = *xp;
2985: *xp = 0;
2986: if( rp->lhsalias && strcmp(cp,rp->lhsalias)==0 ){
2987: fprintf(out,"yygotominor.yy%d",rp->lhs->dtnum);
2988: cp = xp;
2989: lhsused = 1;
2990: }else{
2991: for(i=0; i<rp->nrhs; i++){
2992: if( rp->rhsalias[i] && strcmp(cp,rp->rhsalias[i])==0 ){
2993: fprintf(out,"yymsp[%d].minor.yy%d",i-rp->nrhs+1,rp->rhs[i]->dtnum);
2994: cp = xp;
2995: used[i] = 1;
2996: break;
2997: }
2998: }
2999: }
3000: *xp = saved;
3001: }
3002: if( *cp=='\n' ) linecnt++;
3003: fputc(*cp,out);
3004: } /* End loop */
3005: (*lineno) += 3 + linecnt;
3006: fprintf(out,"}\n#line %d \"%s\"\n",*lineno,lemp->outname);
3007: } /* End if( rp->code ) */
3008:
3009: /* Check to make sure the LHS has been used */
3010: if( rp->lhsalias && !lhsused ){
3011: ErrorMsg(lemp->filename,rp->ruleline,
3012: "Label \"%s\" for \"%s(%s)\" is never used.",
3013: rp->lhsalias,rp->lhs->name,rp->lhsalias);
3014: lemp->errorcnt++;
3015: }
3016:
3017: /* Generate destructor code for RHS symbols which are not used in the
3018: ** reduce code */
3019: for(i=0; i<rp->nrhs; i++){
3020: if( rp->rhsalias[i] && !used[i] ){
3021: ErrorMsg(lemp->filename,rp->ruleline,
3022: "Label %s for \"%s(%s)\" is never used.",
3023: rp->rhsalias[i],rp->rhs[i]->name,rp->rhsalias[i]);
3024: lemp->errorcnt++;
3025: }else if( rp->rhsalias[i]==0 ){
3026: if( has_destructor(rp->rhs[i],lemp) ){
3027: fprintf(out," yy_destructor(%d,&yymsp[%d].minor);\n",
3028: rp->rhs[i]->index,i-rp->nrhs+1); (*lineno)++;
3029: }else{
3030: fprintf(out," /* No destructor defined for %s */\n",
3031: rp->rhs[i]->name);
3032: (*lineno)++;
3033: }
3034: }
3035: }
3036: return;
3037: }
3038:
3039: /*
3040: ** Print the definition of the union used for the parser's data stack.
3041: ** This union contains fields for every possible data type for tokens
3042: ** and nonterminals. In the process of computing and printing this
3043: ** union, also set the ".dtnum" field of every terminal and nonterminal
3044: ** symbol.
3045: */
3046: PRIVATE void print_stack_union(out,lemp,plineno,mhflag)
3047: FILE *out; /* The output stream */
3048: struct lemon *lemp; /* The main info structure for this parser */
3049: int *plineno; /* Pointer to the line number */
3050: int mhflag; /* True if generating makeheaders output */
3051: {
3052: int lineno; /* The line number of the output */
3053: char **types; /* A hash table of datatypes */
3054: int arraysize; /* Size of the "types" array */
3055: int maxdtlength; /* Maximum length of any ".datatype" field. */
3056: char *stddt; /* Standardized name for a datatype */
3057: int i,j; /* Loop counters */
3058: int hash; /* For hashing the name of a type */
3059: char *name; /* Name of the parser */
3060:
3061: /* Allocate and initialize types[] and allocate stddt[] */
3062: arraysize = lemp->nsymbol * 2;
3063: types = (char**)malloc( arraysize * sizeof(char*) );
3064: for(i=0; i<arraysize; i++) types[i] = 0;
3065: maxdtlength = 0;
3066: if( lemp->vartype ){
3067: maxdtlength = strlen(lemp->vartype);
3068: }
3069: for(i=0; i<lemp->nsymbol; i++){
3070: int len;
3071: struct symbol *sp = lemp->symbols[i];
3072: if( sp->datatype==0 ) continue;
3073: len = strlen(sp->datatype);
3074: if( len>maxdtlength ) maxdtlength = len;
3075: }
3076: stddt = (char*)malloc( maxdtlength*2 + 1 );
3077: if( types==0 || stddt==0 ){
3078: fprintf(stderr,"Out of memory.\n");
3079: exit(1);
3080: }
3081:
3082: /* Build a hash table of datatypes. The ".dtnum" field of each symbol
3083: ** is filled in with the hash index plus 1. A ".dtnum" value of 0 is
3084: ** used for terminal symbols. If there is no %default_type defined then
3085: ** 0 is also used as the .dtnum value for nonterminals which do not specify
3086: ** a datatype using the %type directive.
3087: */
3088: for(i=0; i<lemp->nsymbol; i++){
3089: struct symbol *sp = lemp->symbols[i];
3090: char *cp;
3091: if( sp==lemp->errsym ){
3092: sp->dtnum = arraysize+1;
3093: continue;
3094: }
3095: if( sp->type!=NONTERMINAL || (sp->datatype==0 && lemp->vartype==0) ){
3096: sp->dtnum = 0;
3097: continue;
3098: }
3099: cp = sp->datatype;
3100: if( cp==0 ) cp = lemp->vartype;
3101: j = 0;
3102: while( isspace(*cp) ) cp++;
3103: while( *cp ) stddt[j++] = *cp++;
3104: while( j>0 && isspace(stddt[j-1]) ) j--;
3105: stddt[j] = 0;
3106: hash = 0;
3107: for(j=0; stddt[j]; j++){
3108: hash = (unsigned int)hash*53u + (unsigned int) stddt[j];
3109: }
3110: hash = (hash & 0x7fffffff)%arraysize;
3111: while( types[hash] ){
3112: if( strcmp(types[hash],stddt)==0 ){
3113: sp->dtnum = hash + 1;
3114: break;
3115: }
3116: hash++;
3117: if( hash>=arraysize ) hash = 0;
3118: }
3119: if( types[hash]==0 ){
3120: sp->dtnum = hash + 1;
3121: types[hash] = (char*)malloc( strlen(stddt)+1 );
3122: if( types[hash]==0 ){
3123: fprintf(stderr,"Out of memory.\n");
3124: exit(1);
3125: }
3126: strcpy(types[hash],stddt);
3127: }
3128: }
3129:
3130: /* Print out the definition of YYTOKENTYPE and YYMINORTYPE */
3131: name = lemp->name ? lemp->name : "Parse";
3132: lineno = *plineno;
3133: if( mhflag ){ fprintf(out,"#if INTERFACE\n"); lineno++; }
3134: fprintf(out,"#define %sTOKENTYPE %s\n",name,
3135: lemp->tokentype?lemp->tokentype:"void*"); lineno++;
3136: if( mhflag ){ fprintf(out,"#endif\n"); lineno++; }
3137: fprintf(out,"typedef union {\n"); lineno++;
3138: fprintf(out," %sTOKENTYPE yy0;\n",name); lineno++;
3139: for(i=0; i<arraysize; i++){
3140: if( types[i]==0 ) continue;
3141: fprintf(out," %s yy%d;\n",types[i],i+1); lineno++;
3142: free(types[i]);
3143: }
3144: fprintf(out," int yy%d;\n",lemp->errsym->dtnum); lineno++;
3145: free(stddt);
3146: free(types);
3147: fprintf(out,"} YYMINORTYPE;\n"); lineno++;
3148: *plineno = lineno;
3149: }
3150:
3151: /*
3152: ** Return the name of a C datatype able to represent values between
3153: ** lwr and upr, inclusive.
3154: */
3155: static const char *minimum_size_type(int lwr, int upr){
3156: if( lwr>=0 ){
3157: if( upr<=255 ){
3158: return "unsigned char";
3159: }else if( upr<65535 ){
3160: return "unsigned short int";
3161: }else{
3162: return "unsigned int";
3163: }
3164: }else if( lwr>=-127 && upr<=127 ){
3165: return "signed char";
3166: }else if( lwr>=-32767 && upr<32767 ){
3167: return "short";
3168: }else{
3169: return "int";
3170: }
3171: }
3172:
3173: /*
3174: ** Each state contains a set of token transaction and a set of
3175: ** nonterminal transactions. Each of these sets makes an instance
3176: ** of the following structure. An array of these structures is used
3177: ** to order the creation of entries in the yy_action[] table.
3178: */
3179: struct axset {
3180: struct state *stp; /* A pointer to a state */
3181: int isTkn; /* True to use tokens. False for non-terminals */
3182: int nAction; /* Number of actions */
3183: };
3184:
3185: /*
3186: ** Compare to axset structures for sorting purposes
3187: */
3188: static int axset_compare(const void *a, const void *b){
3189: struct axset *p1 = (struct axset*)a;
3190: struct axset *p2 = (struct axset*)b;
3191: return p2->nAction - p1->nAction;
3192: }
3193:
3194: /* Generate C source code for the parser */
3195: void ReportTable(lemp, mhflag)
3196: struct lemon *lemp;
3197: int mhflag; /* Output in makeheaders format if true */
3198: {
3199: FILE *out, *in;
3200: char line[LINESIZE];
3201: int lineno;
3202: struct state *stp;
3203: struct action *ap;
3204: struct rule *rp;
3205: struct acttab *pActtab;
3206: int i, j, n;
3207: int mnTknOfst, mxTknOfst;
3208: int mnNtOfst, mxNtOfst;
3209: struct axset *ax;
3210: char *name;
3211:
3212: in = tplt_open(lemp);
3213: if( in==0 ) return;
3214: out = file_open(lemp,".c","w");
3215: if( out==0 ){
3216: fclose(in);
3217: return;
3218: }
3219: lineno = 1;
3220: tplt_xfer(lemp->name,in,out,&lineno);
3221:
3222: /* Generate the include code, if any */
3223: tplt_print(out,lemp,lemp->include,lemp->includeln,&lineno);
3224: if( mhflag ){
3225: name = file_makename(lemp, ".h");
3226: fprintf(out,"#include \"%s\"\n", name); lineno++;
3227: free(name);
3228: }
3229: tplt_xfer(lemp->name,in,out,&lineno);
3230:
3231: /* Generate #defines for all tokens */
3232: if( mhflag ){
3233: char *prefix;
3234: fprintf(out,"#if INTERFACE\n"); lineno++;
3235: if( lemp->tokenprefix ) prefix = lemp->tokenprefix;
3236: else prefix = "";
3237: for(i=1; i<lemp->nterminal; i++){
3238: fprintf(out,"#define %s%-30s %2d\n",prefix,lemp->symbols[i]->name,i);
3239: lineno++;
3240: }
3241: fprintf(out,"#endif\n"); lineno++;
3242: }
3243: tplt_xfer(lemp->name,in,out,&lineno);
3244:
3245: /* Generate the defines */
3246: fprintf(out,"/* \001 */\n");
3247: fprintf(out,"#define YYCODETYPE %s\n",
3248: minimum_size_type(0, lemp->nsymbol+5)); lineno++;
3249: fprintf(out,"#define YYNOCODE %d\n",lemp->nsymbol+1); lineno++;
3250: fprintf(out,"#define YYACTIONTYPE %s\n",
3251: minimum_size_type(0, lemp->nstate+lemp->nrule+5)); lineno++;
3252: print_stack_union(out,lemp,&lineno,mhflag);
3253: if( lemp->stacksize ){
3254: if( atoi(lemp->stacksize)<=0 ){
3255: ErrorMsg(lemp->filename,0,
3256: "Illegal stack size: [%s]. The stack size should be an integer constant.",
3257: lemp->stacksize);
3258: lemp->errorcnt++;
3259: lemp->stacksize = "100";
3260: }
3261: fprintf(out,"#define YYSTACKDEPTH %s\n",lemp->stacksize); lineno++;
3262: }else{
3263: fprintf(out,"#define YYSTACKDEPTH 100\n"); lineno++;
3264: }
3265: if( mhflag ){
3266: fprintf(out,"#if INTERFACE\n"); lineno++;
3267: }
3268: name = lemp->name ? lemp->name : "Parse";
3269: if( lemp->arg && lemp->arg[0] ){
3270: i = strlen(lemp->arg);
3271: while( i>=1 && isspace(lemp->arg[i-1]) ) i--;
3272: while( i>=1 && (isalnum(lemp->arg[i-1]) || lemp->arg[i-1]=='_') ) i--;
3273: fprintf(out,"#define %sARG_SDECL %s;\n",name,lemp->arg); lineno++;
3274: fprintf(out,"#define %sARG_PDECL ,%s\n",name,lemp->arg); lineno++;
3275: fprintf(out,"#define %sARG_FETCH %s = yypParser->%s\n",
3276: name,lemp->arg,&lemp->arg[i]); lineno++;
3277: fprintf(out,"#define %sARG_STORE yypParser->%s = %s\n",
3278: name,&lemp->arg[i],&lemp->arg[i]); lineno++;
3279: }else{
3280: fprintf(out,"#define %sARG_SDECL\n",name); lineno++;
3281: fprintf(out,"#define %sARG_PDECL\n",name); lineno++;
3282: fprintf(out,"#define %sARG_FETCH\n",name); lineno++;
3283: fprintf(out,"#define %sARG_STORE\n",name); lineno++;
3284: }
3285: if( mhflag ){
3286: fprintf(out,"#endif\n"); lineno++;
3287: }
3288: fprintf(out,"#define YYNSTATE %d\n",lemp->nstate); lineno++;
3289: fprintf(out,"#define YYNRULE %d\n",lemp->nrule); lineno++;
3290: fprintf(out,"#define YYERRORSYMBOL %d\n",lemp->errsym->index); lineno++;
3291: fprintf(out,"#define YYERRSYMDT yy%d\n",lemp->errsym->dtnum); lineno++;
3292: if( lemp->has_fallback ){
3293: fprintf(out,"#define YYFALLBACK 1\n"); lineno++;
3294: }
3295: tplt_xfer(lemp->name,in,out,&lineno);
3296:
3297: /* Generate the action table and its associates:
3298: **
3299: ** yy_action[] A single table containing all actions.
3300: ** yy_lookahead[] A table containing the lookahead for each entry in
3301: ** yy_action. Used to detect hash collisions.
3302: ** yy_shift_ofst[] For each state, the offset into yy_action for
3303: ** shifting terminals.
3304: ** yy_reduce_ofst[] For each state, the offset into yy_action for
3305: ** shifting non-terminals after a reduce.
3306: ** yy_default[] Default action for each state.
3307: */
3308:
3309: /* Compute the actions on all states and count them up */
3310: ax = malloc( sizeof(ax[0])*lemp->nstate*2 );
3311: if( ax==0 ){
3312: fprintf(stderr,"malloc failed\n");
3313: exit(1);
3314: }
3315: for(i=0; i<lemp->nstate; i++){
3316: stp = lemp->sorted[i];
3317: stp->nTknAct = stp->nNtAct = 0;
3318: stp->iDflt = lemp->nstate + lemp->nrule;
3319: stp->iTknOfst = NO_OFFSET;
3320: stp->iNtOfst = NO_OFFSET;
3321: for(ap=stp->ap; ap; ap=ap->next){
3322: if( compute_action(lemp,ap)>=0 ){
3323: if( ap->sp->index<lemp->nterminal ){
3324: stp->nTknAct++;
3325: }else if( ap->sp->index<lemp->nsymbol ){
3326: stp->nNtAct++;
3327: }else{
3328: stp->iDflt = compute_action(lemp, ap);
3329: }
3330: }
3331: }
3332: ax[i*2].stp = stp;
3333: ax[i*2].isTkn = 1;
3334: ax[i*2].nAction = stp->nTknAct;
3335: ax[i*2+1].stp = stp;
3336: ax[i*2+1].isTkn = 0;
3337: ax[i*2+1].nAction = stp->nNtAct;
3338: }
3339: mxTknOfst = mnTknOfst = 0;
3340: mxNtOfst = mnNtOfst = 0;
3341:
3342: /* Compute the action table. In order to try to keep the size of the
3343: ** action table to a minimum, the heuristic of placing the largest action
3344: ** sets first is used.
3345: */
3346: qsort(ax, lemp->nstate*2, sizeof(ax[0]), axset_compare);
3347: pActtab = acttab_alloc();
3348: for(i=0; i<lemp->nstate*2 && ax[i].nAction>0; i++){
3349: stp = ax[i].stp;
3350: if( ax[i].isTkn ){
3351: for(ap=stp->ap; ap; ap=ap->next){
3352: int action;
3353: if( ap->sp->index>=lemp->nterminal ) continue;
3354: action = compute_action(lemp, ap);
3355: if( action<0 ) continue;
3356: acttab_action(pActtab, ap->sp->index, action);
3357: }
3358: stp->iTknOfst = acttab_insert(pActtab);
3359: if( stp->iTknOfst<mnTknOfst ) mnTknOfst = stp->iTknOfst;
3360: if( stp->iTknOfst>mxTknOfst ) mxTknOfst = stp->iTknOfst;
3361: }else{
3362: for(ap=stp->ap; ap; ap=ap->next){
3363: int action;
3364: if( ap->sp->index<lemp->nterminal ) continue;
3365: if( ap->sp->index==lemp->nsymbol ) continue;
3366: action = compute_action(lemp, ap);
3367: if( action<0 ) continue;
3368: acttab_action(pActtab, ap->sp->index, action);
3369: }
3370: stp->iNtOfst = acttab_insert(pActtab);
3371: if( stp->iNtOfst<mnNtOfst ) mnNtOfst = stp->iNtOfst;
3372: if( stp->iNtOfst>mxNtOfst ) mxNtOfst = stp->iNtOfst;
3373: }
3374: }
3375: free(ax);
3376:
3377: /* Output the yy_action table */
3378: fprintf(out,"static YYACTIONTYPE yy_action[] = {\n"); lineno++;
3379: n = acttab_size(pActtab);
3380: for(i=j=0; i<n; i++){
3381: int action = acttab_yyaction(pActtab, i);
3382: if( action<0 ) action = lemp->nsymbol + lemp->nrule + 2;
3383: if( j==0 ) fprintf(out," /* %5d */ ", i);
3384: fprintf(out, " %4d,", action);
3385: if( j==9 || i==n-1 ){
3386: fprintf(out, "\n"); lineno++;
3387: j = 0;
3388: }else{
3389: j++;
3390: }
3391: }
3392: fprintf(out, "};\n"); lineno++;
3393:
3394: /* Output the yy_lookahead table */
3395: fprintf(out,"static YYCODETYPE yy_lookahead[] = {\n"); lineno++;
3396: for(i=j=0; i<n; i++){
3397: int la = acttab_yylookahead(pActtab, i);
3398: if( la<0 ) la = lemp->nsymbol;
3399: if( j==0 ) fprintf(out," /* %5d */ ", i);
3400: fprintf(out, " %4d,", la);
3401: if( j==9 || i==n-1 ){
3402: fprintf(out, "\n"); lineno++;
3403: j = 0;
3404: }else{
3405: j++;
3406: }
3407: }
3408: fprintf(out, "};\n"); lineno++;
3409:
3410: /* Output the yy_shift_ofst[] table */
3411: fprintf(out, "#define YY_SHIFT_USE_DFLT (%d)\n", mnTknOfst-1); lineno++;
3412: fprintf(out, "static %s yy_shift_ofst[] = {\n",
3413: minimum_size_type(mnTknOfst-1, mxTknOfst)); lineno++;
3414: n = lemp->nstate;
3415: for(i=j=0; i<n; i++){
3416: int ofst;
3417: stp = lemp->sorted[i];
3418: ofst = stp->iTknOfst;
3419: if( ofst==NO_OFFSET ) ofst = mnTknOfst - 1;
3420: if( j==0 ) fprintf(out," /* %5d */ ", i);
3421: fprintf(out, " %4d,", ofst);
3422: if( j==9 || i==n-1 ){
3423: fprintf(out, "\n"); lineno++;
3424: j = 0;
3425: }else{
3426: j++;
3427: }
3428: }
3429: fprintf(out, "};\n"); lineno++;
3430:
3431: /* Output the yy_reduce_ofst[] table */
3432: fprintf(out, "#define YY_REDUCE_USE_DFLT (%d)\n", mnNtOfst-1); lineno++;
3433: fprintf(out, "static %s yy_reduce_ofst[] = {\n",
3434: minimum_size_type(mnNtOfst-1, mxNtOfst)); lineno++;
3435: n = lemp->nstate;
3436: for(i=j=0; i<n; i++){
3437: int ofst;
3438: stp = lemp->sorted[i];
3439: ofst = stp->iNtOfst;
3440: if( ofst==NO_OFFSET ) ofst = mnNtOfst - 1;
3441: if( j==0 ) fprintf(out," /* %5d */ ", i);
3442: fprintf(out, " %4d,", ofst);
3443: if( j==9 || i==n-1 ){
3444: fprintf(out, "\n"); lineno++;
3445: j = 0;
3446: }else{
3447: j++;
3448: }
3449: }
3450: fprintf(out, "};\n"); lineno++;
3451:
3452: /* Output the default action table */
3453: fprintf(out, "static YYACTIONTYPE yy_default[] = {\n"); lineno++;
3454: n = lemp->nstate;
3455: for(i=j=0; i<n; i++){
3456: stp = lemp->sorted[i];
3457: if( j==0 ) fprintf(out," /* %5d */ ", i);
3458: fprintf(out, " %4d,", stp->iDflt);
3459: if( j==9 || i==n-1 ){
3460: fprintf(out, "\n"); lineno++;
3461: j = 0;
3462: }else{
3463: j++;
3464: }
3465: }
3466: fprintf(out, "};\n"); lineno++;
3467: tplt_xfer(lemp->name,in,out,&lineno);
3468:
3469: /* Generate the table of fallback tokens.
3470: */
3471: if( lemp->has_fallback ){
3472: for(i=0; i<lemp->nterminal; i++){
3473: struct symbol *p = lemp->symbols[i];
3474: if( p->fallback==0 ){
3475: fprintf(out, " 0, /* %10s => nothing */\n", p->name);
3476: }else{
3477: fprintf(out, " %3d, /* %10s => %s */\n", p->fallback->index,
3478: p->name, p->fallback->name);
3479: }
3480: lineno++;
3481: }
3482: }
3483: tplt_xfer(lemp->name, in, out, &lineno);
3484:
3485: /* Generate a table containing the symbolic name of every symbol
3486: */
3487: for(i=0; i<lemp->nsymbol; i++){
3488: sprintf(line,"\"%s\",",lemp->symbols[i]->name);
3489: fprintf(out," %-15s",line);
3490: if( (i&3)==3 ){ fprintf(out,"\n"); lineno++; }
3491: }
3492: if( (i&3)!=0 ){ fprintf(out,"\n"); lineno++; }
3493: tplt_xfer(lemp->name,in,out,&lineno);
3494:
3495: /* Generate a table containing a text string that describes every
3496: ** rule in the rule set of the grammer. This information is used
3497: ** when tracing REDUCE actions.
3498: */
3499: for(i=0, rp=lemp->rule; rp; rp=rp->next, i++){
3500: assert( rp->index==i );
3501: fprintf(out," /* %3d */ \"%s ::=", i, rp->lhs->name);
3502: for(j=0; j<rp->nrhs; j++) fprintf(out," %s",rp->rhs[j]->name);
3503: fprintf(out,"\",\n"); lineno++;
3504: }
3505: tplt_xfer(lemp->name,in,out,&lineno);
3506:
3507: /* Generate code which executes every time a symbol is popped from
3508: ** the stack while processing errors or while destroying the parser.
3509: ** (In other words, generate the %destructor actions)
3510: */
3511: if( lemp->tokendest ){
3512: for(i=0; i<lemp->nsymbol; i++){
3513: struct symbol *sp = lemp->symbols[i];
3514: if( sp==0 || sp->type!=TERMINAL ) continue;
3515: fprintf(out," case %d:\n",sp->index); lineno++;
3516: }
3517: for(i=0; i<lemp->nsymbol && lemp->symbols[i]->type!=TERMINAL; i++);
3518: if( i<lemp->nsymbol ){
3519: emit_destructor_code(out,lemp->symbols[i],lemp,&lineno);
3520: fprintf(out," break;\n"); lineno++;
3521: }
3522: }
3523: for(i=0; i<lemp->nsymbol; i++){
3524: struct symbol *sp = lemp->symbols[i];
3525: if( sp==0 || sp->type==TERMINAL || sp->destructor==0 ) continue;
3526: fprintf(out," case %d:\n",sp->index); lineno++;
3527: emit_destructor_code(out,lemp->symbols[i],lemp,&lineno);
3528: fprintf(out," break;\n"); lineno++;
3529: }
3530: if( lemp->vardest ){
3531: struct symbol *dflt_sp = 0;
3532: for(i=0; i<lemp->nsymbol; i++){
3533: struct symbol *sp = lemp->symbols[i];
3534: if( sp==0 || sp->type==TERMINAL ||
3535: sp->index<=0 || sp->destructor!=0 ) continue;
3536: fprintf(out," case %d:\n",sp->index); lineno++;
3537: dflt_sp = sp;
3538: }
3539: if( dflt_sp!=0 ){
3540: emit_destructor_code(out,dflt_sp,lemp,&lineno);
3541: fprintf(out," break;\n"); lineno++;
3542: }
3543: }
3544: tplt_xfer(lemp->name,in,out,&lineno);
3545:
3546: /* Generate code which executes whenever the parser stack overflows */
3547: tplt_print(out,lemp,lemp->overflow,lemp->overflowln,&lineno);
3548: tplt_xfer(lemp->name,in,out,&lineno);
3549:
3550: /* Generate the table of rule information
3551: **
3552: ** Note: This code depends on the fact that rules are number
3553: ** sequentually beginning with 0.
3554: */
3555: for(rp=lemp->rule; rp; rp=rp->next){
3556: fprintf(out," { %d, %d },\n",rp->lhs->index,rp->nrhs); lineno++;
3557: }
3558: tplt_xfer(lemp->name,in,out,&lineno);
3559:
3560: /* Generate code which execution during each REDUCE action */
3561: for(rp=lemp->rule; rp; rp=rp->next){
3562: fprintf(out," case %d:\n",rp->index); lineno++;
3563: emit_code(out,rp,lemp,&lineno);
3564: fprintf(out," break;\n"); lineno++;
3565: }
3566: tplt_xfer(lemp->name,in,out,&lineno);
3567:
3568: /* Generate code which executes if a parse fails */
3569: tplt_print(out,lemp,lemp->failure,lemp->failureln,&lineno);
3570: tplt_xfer(lemp->name,in,out,&lineno);
3571:
3572: /* Generate code which executes when a syntax error occurs */
3573: tplt_print(out,lemp,lemp->error,lemp->errorln,&lineno);
3574: tplt_xfer(lemp->name,in,out,&lineno);
3575:
3576: /* Generate code which executes when the parser accepts its input */
3577: tplt_print(out,lemp,lemp->accept,lemp->acceptln,&lineno);
3578: tplt_xfer(lemp->name,in,out,&lineno);
3579:
3580: /* Append any addition code the user desires */
3581: tplt_print(out,lemp,lemp->extracode,lemp->extracodeln,&lineno);
3582:
3583: fclose(in);
3584: fclose(out);
3585: return;
3586: }
3587:
3588: /* Generate a header file for the parser */
3589: void ReportHeader(lemp)
3590: struct lemon *lemp;
3591: {
3592: FILE *out, *in;
3593: char *prefix;
3594: char line[LINESIZE];
3595: char pattern[LINESIZE];
3596: int i;
3597:
3598: if( lemp->tokenprefix ) prefix = lemp->tokenprefix;
3599: else prefix = "";
3600: in = file_open(lemp,".h","r");
3601: if( in ){
3602: for(i=1; i<lemp->nterminal && fgets(line,LINESIZE,in); i++){
3603: sprintf(pattern,"#define %s%-30s %2d\n",prefix,lemp->symbols[i]->name,i);
3604: if( strcmp(line,pattern) ) break;
3605: }
3606: fclose(in);
3607: if( i==lemp->nterminal ){
3608: /* No change in the file. Don't rewrite it. */
3609: return;
3610: }
3611: }
3612: out = file_open(lemp,".h","w");
3613: if( out ){
3614: for(i=1; i<lemp->nterminal; i++){
3615: fprintf(out,"#define %s%-30s %2d\n",prefix,lemp->symbols[i]->name,i);
3616: }
3617: fclose(out);
3618: }
3619: return;
3620: }
3621:
3622: /* Reduce the size of the action tables, if possible, by making use
3623: ** of defaults.
3624: **
3625: ** In this version, we take the most frequent REDUCE action and make
3626: ** it the default. Only default a reduce if there are more than one.
3627: */
3628: void CompressTables(lemp)
3629: struct lemon *lemp;
3630: {
3631: struct state *stp;
3632: struct action *ap, *ap2;
3633: struct rule *rp, *rp2, *rbest;
3634: int nbest, n;
3635: int i;
3636:
3637: for(i=0; i<lemp->nstate; i++){
3638: stp = lemp->sorted[i];
3639: nbest = 0;
3640: rbest = 0;
3641:
3642: for(ap=stp->ap; ap; ap=ap->next){
3643: if( ap->type!=REDUCE ) continue;
3644: rp = ap->x.rp;
3645: if( rp==rbest ) continue;
3646: n = 1;
3647: for(ap2=ap->next; ap2; ap2=ap2->next){
3648: if( ap2->type!=REDUCE ) continue;
3649: rp2 = ap2->x.rp;
3650: if( rp2==rbest ) continue;
3651: if( rp2==rp ) n++;
3652: }
3653: if( n>nbest ){
3654: nbest = n;
3655: rbest = rp;
3656: }
3657: }
3658:
3659: /* Do not make a default if the number of rules to default
3660: ** is not at least 2 */
3661: if( nbest<2 ) continue;
3662:
3663:
3664: /* Combine matching REDUCE actions into a single default */
3665: for(ap=stp->ap; ap; ap=ap->next){
3666: if( ap->type==REDUCE && ap->x.rp==rbest ) break;
3667: }
3668: assert( ap );
3669: ap->sp = Symbol_new("{default}");
3670: for(ap=ap->next; ap; ap=ap->next){
3671: if( ap->type==REDUCE && ap->x.rp==rbest ) ap->type = NOT_USED;
3672: }
3673: stp->ap = Action_sort(stp->ap);
3674: }
3675: }
3676:
3677: /***************** From the file "set.c" ************************************/
3678: /*
3679: ** Set manipulation routines for the LEMON parser generator.
3680: */
3681:
3682: static int global_size = 0;
3683:
3684: /* Set the set size */
3685: void SetSize(n)
3686: int n;
3687: {
3688: global_size = n+1;
3689: }
3690:
3691: /* Allocate a new set */
3692: char *SetNew(){
3693: char *s;
3694: int i;
3695: s = (char*)malloc( global_size );
3696: if( s==0 ){
3697: memory_error();
3698: }
3699: for(i=0; i<global_size; i++) s[i] = 0;
3700: return s;
3701: }
3702:
3703: /* Deallocate a set */
3704: void SetFree(s)
3705: char *s;
3706: {
3707: free(s);
3708: }
3709:
3710: /* Add a new element to the set. Return TRUE if the element was added
3711: ** and FALSE if it was already there. */
3712: int SetAdd(s,e)
3713: char *s;
3714: int e;
3715: {
3716: int rv;
3717: rv = s[e];
3718: s[e] = 1;
3719: return !rv;
3720: }
3721:
3722: /* Add every element of s2 to s1. Return TRUE if s1 changes. */
3723: int SetUnion(s1,s2)
3724: char *s1;
3725: char *s2;
3726: {
3727: int i, progress;
3728: progress = 0;
3729: for(i=0; i<global_size; i++){
3730: if( s2[i]==0 ) continue;
3731: if( s1[i]==0 ){
3732: progress = 1;
3733: s1[i] = 1;
3734: }
3735: }
3736: return progress;
3737: }
3738: /********************** From the file "table.c" ****************************/
3739: /*
3740: ** All code in this file has been automatically generated
3741: ** from a specification in the file
3742: ** "table.q"
3743: ** by the associative array code building program "aagen".
3744: ** Do not edit this file! Instead, edit the specification
3745: ** file, then rerun aagen.
3746: */
3747: /*
3748: ** Code for processing tables in the LEMON parser generator.
3749: */
3750:
3751: PRIVATE int strhash(x)
3752: char *x;
3753: {
3754: unsigned int h = 0;
3755: while( *x) h = h*13u + (unsigned int) *(x++);
3756: return h;
3757: }
3758:
3759: /* Works like strdup, sort of. Save a string in malloced memory, but
3760: ** keep strings in a table so that the same string is not in more
3761: ** than one place.
3762: */
3763: char *Strsafe(y)
3764: char *y;
3765: {
3766: char *z;
3767:
3768: z = Strsafe_find(y);
3769: if( z==0 && (z=malloc( strlen(y)+1 ))!=0 ){
3770: strcpy(z,y);
3771: Strsafe_insert(z);
3772: }
3773: MemoryCheck(z);
3774: return z;
3775: }
3776:
3777: /* There is one instance of the following structure for each
3778: ** associative array of type "x1".
3779: */
3780: struct s_x1 {
3781: int size; /* The number of available slots. */
3782: /* Must be a power of 2 greater than or */
3783: /* equal to 1 */
3784: int count; /* Number of currently slots filled */
3785: struct s_x1node *tbl; /* The data stored here */
3786: struct s_x1node **ht; /* Hash table for lookups */
3787: };
3788:
3789: /* There is one instance of this structure for every data element
3790: ** in an associative array of type "x1".
3791: */
3792: typedef struct s_x1node {
3793: char *data; /* The data */
3794: struct s_x1node *next; /* Next entry with the same hash */
3795: struct s_x1node **from; /* Previous link */
3796: } x1node;
3797:
3798: /* There is only one instance of the array, which is the following */
3799: static struct s_x1 *x1a;
3800:
3801: /* Allocate a new associative array */
3802: void Strsafe_init(){
3803: if( x1a ) return;
3804: x1a = (struct s_x1*)malloc( sizeof(struct s_x1) );
3805: if( x1a ){
3806: x1a->size = 1024;
3807: x1a->count = 0;
3808: x1a->tbl = (x1node*)malloc(
3809: (sizeof(x1node) + sizeof(x1node*))*1024 );
3810: if( x1a->tbl==0 ){
3811: free(x1a);
3812: x1a = 0;
3813: }else{
3814: int i;
3815: x1a->ht = (x1node**)&(x1a->tbl[1024]);
3816: for(i=0; i<1024; i++) x1a->ht[i] = 0;
3817: }
3818: }
3819: }
3820: /* Insert a new record into the array. Return TRUE if successful.
3821: ** Prior data with the same key is NOT overwritten */
3822: int Strsafe_insert(data)
3823: char *data;
3824: {
3825: x1node *np;
3826: int h;
3827: int ph;
3828:
3829: if( x1a==0 ) return 0;
3830: ph = strhash(data);
3831: h = ph & (x1a->size-1);
3832: np = x1a->ht[h];
3833: while( np ){
3834: if( strcmp(np->data,data)==0 ){
3835: /* An existing entry with the same key is found. */
3836: /* Fail because overwrite is not allows. */
3837: return 0;
3838: }
3839: np = np->next;
3840: }
3841: if( x1a->count>=x1a->size ){
3842: /* Need to make the hash table bigger */
3843: int i,size;
3844: struct s_x1 array;
3845: array.size = size = x1a->size*2;
3846: array.count = x1a->count;
3847: array.tbl = (x1node*)malloc(
3848: (sizeof(x1node) + sizeof(x1node*))*size );
3849: if( array.tbl==0 ) return 0; /* Fail due to malloc failure */
3850: array.ht = (x1node**)&(array.tbl[size]);
3851: for(i=0; i<size; i++) array.ht[i] = 0;
3852: for(i=0; i<x1a->count; i++){
3853: x1node *oldnp, *newnp;
3854: oldnp = &(x1a->tbl[i]);
3855: h = strhash(oldnp->data) & (size-1);
3856: newnp = &(array.tbl[i]);
3857: if( array.ht[h] ) array.ht[h]->from = &(newnp->next);
3858: newnp->next = array.ht[h];
3859: newnp->data = oldnp->data;
3860: newnp->from = &(array.ht[h]);
3861: array.ht[h] = newnp;
3862: }
3863: free(x1a->tbl);
3864: *x1a = array;
3865: }
3866: /* Insert the new data */
3867: h = ph & (x1a->size-1);
3868: np = &(x1a->tbl[x1a->count++]);
3869: np->data = data;
3870: if( x1a->ht[h] ) x1a->ht[h]->from = &(np->next);
3871: np->next = x1a->ht[h];
3872: x1a->ht[h] = np;
3873: np->from = &(x1a->ht[h]);
3874: return 1;
3875: }
3876:
3877: /* Return a pointer to data assigned to the given key. Return NULL
3878: ** if no such key. */
3879: char *Strsafe_find(key)
3880: char *key;
3881: {
3882: int h;
3883: x1node *np;
3884:
3885: if( x1a==0 ) return 0;
3886: h = strhash(key) & (x1a->size-1);
3887: np = x1a->ht[h];
3888: while( np ){
3889: if( strcmp(np->data,key)==0 ) break;
3890: np = np->next;
3891: }
3892: return np ? np->data : 0;
3893: }
3894:
3895: /* Return a pointer to the (terminal or nonterminal) symbol "x".
3896: ** Create a new symbol if this is the first time "x" has been seen.
3897: */
3898: struct symbol *Symbol_new(x)
3899: char *x;
3900: {
3901: struct symbol *sp;
3902:
3903: sp = Symbol_find(x);
3904: if( sp==0 ){
3905: sp = (struct symbol *)malloc( sizeof(struct symbol) );
3906: MemoryCheck(sp);
3907: sp->name = Strsafe(x);
3908: sp->type = isupper(*x) ? TERMINAL : NONTERMINAL;
3909: sp->rule = 0;
3910: sp->fallback = 0;
3911: sp->prec = -1;
3912: sp->assoc = UNK;
3913: sp->firstset = 0;
3914: sp->lambda = Bo_FALSE;
3915: sp->destructor = 0;
3916: sp->datatype = 0;
3917: Symbol_insert(sp,sp->name);
3918: }
3919: return sp;
3920: }
3921:
3922: /* Compare two symbols for working purposes
3923: **
3924: ** Symbols that begin with upper case letters (terminals or tokens)
3925: ** must sort before symbols that begin with lower case letters
3926: ** (non-terminals). Other than that, the order does not matter.
3927: **
3928: ** We find experimentally that leaving the symbols in their original
3929: ** order (the order they appeared in the grammar file) gives the
3930: ** smallest parser tables in SQLite.
3931: */
3932: int Symbolcmpp(struct symbol **a, struct symbol **b){
3933: int i1 = (**a).index + 10000000*((**a).name[0]>'Z');
3934: int i2 = (**b).index + 10000000*((**b).name[0]>'Z');
3935: return i1-i2;
3936: }
3937:
3938: /* There is one instance of the following structure for each
3939: ** associative array of type "x2".
3940: */
3941: struct s_x2 {
3942: int size; /* The number of available slots. */
3943: /* Must be a power of 2 greater than or */
3944: /* equal to 1 */
3945: int count; /* Number of currently slots filled */
3946: struct s_x2node *tbl; /* The data stored here */
3947: struct s_x2node **ht; /* Hash table for lookups */
3948: };
3949:
3950: /* There is one instance of this structure for every data element
3951: ** in an associative array of type "x2".
3952: */
3953: typedef struct s_x2node {
3954: struct symbol *data; /* The data */
3955: char *key; /* The key */
3956: struct s_x2node *next; /* Next entry with the same hash */
3957: struct s_x2node **from; /* Previous link */
3958: } x2node;
3959:
3960: /* There is only one instance of the array, which is the following */
3961: static struct s_x2 *x2a;
3962:
3963: /* Allocate a new associative array */
3964: void Symbol_init(){
3965: if( x2a ) return;
3966: x2a = (struct s_x2*)malloc( sizeof(struct s_x2) );
3967: if( x2a ){
3968: x2a->size = 128;
3969: x2a->count = 0;
3970: x2a->tbl = (x2node*)malloc(
3971: (sizeof(x2node) + sizeof(x2node*))*128 );
3972: if( x2a->tbl==0 ){
3973: free(x2a);
3974: x2a = 0;
3975: }else{
3976: int i;
3977: x2a->ht = (x2node**)&(x2a->tbl[128]);
3978: for(i=0; i<128; i++) x2a->ht[i] = 0;
3979: }
3980: }
3981: }
3982: /* Insert a new record into the array. Return TRUE if successful.
3983: ** Prior data with the same key is NOT overwritten */
3984: int Symbol_insert(data,key)
3985: struct symbol *data;
3986: char *key;
3987: {
3988: x2node *np;
3989: int h;
3990: int ph;
3991:
3992: if( x2a==0 ) return 0;
3993: ph = strhash(key);
3994: h = ph & (x2a->size-1);
3995: np = x2a->ht[h];
3996: while( np ){
3997: if( strcmp(np->key,key)==0 ){
3998: /* An existing entry with the same key is found. */
3999: /* Fail because overwrite is not allows. */
4000: return 0;
4001: }
4002: np = np->next;
4003: }
4004: if( x2a->count>=x2a->size ){
4005: /* Need to make the hash table bigger */
4006: int i,size;
4007: struct s_x2 array;
4008: array.size = size = x2a->size*2;
4009: array.count = x2a->count;
4010: array.tbl = (x2node*)malloc(
4011: (sizeof(x2node) + sizeof(x2node*))*size );
4012: if( array.tbl==0 ) return 0; /* Fail due to malloc failure */
4013: array.ht = (x2node**)&(array.tbl[size]);
4014: for(i=0; i<size; i++) array.ht[i] = 0;
4015: for(i=0; i<x2a->count; i++){
4016: x2node *oldnp, *newnp;
4017: oldnp = &(x2a->tbl[i]);
4018: h = strhash(oldnp->key) & (size-1);
4019: newnp = &(array.tbl[i]);
4020: if( array.ht[h] ) array.ht[h]->from = &(newnp->next);
4021: newnp->next = array.ht[h];
4022: newnp->key = oldnp->key;
4023: newnp->data = oldnp->data;
4024: newnp->from = &(array.ht[h]);
4025: array.ht[h] = newnp;
4026: }
4027: free(x2a->tbl);
4028: *x2a = array;
4029: }
4030: /* Insert the new data */
4031: h = ph & (x2a->size-1);
4032: np = &(x2a->tbl[x2a->count++]);
4033: np->key = key;
4034: np->data = data;
4035: if( x2a->ht[h] ) x2a->ht[h]->from = &(np->next);
4036: np->next = x2a->ht[h];
4037: x2a->ht[h] = np;
4038: np->from = &(x2a->ht[h]);
4039: return 1;
4040: }
4041:
4042: /* Return a pointer to data assigned to the given key. Return NULL
4043: ** if no such key. */
4044: struct symbol *Symbol_find(key)
4045: char *key;
4046: {
4047: int h;
4048: x2node *np;
4049:
4050: if( x2a==0 ) return 0;
4051: h = strhash(key) & (x2a->size-1);
4052: np = x2a->ht[h];
4053: while( np ){
4054: if( strcmp(np->key,key)==0 ) break;
4055: np = np->next;
4056: }
4057: return np ? np->data : 0;
4058: }
4059:
4060: /* Return the n-th data. Return NULL if n is out of range. */
4061: struct symbol *Symbol_Nth(n)
4062: int n;
4063: {
4064: struct symbol *data;
4065: if( x2a && n>0 && n<=x2a->count ){
4066: data = x2a->tbl[n-1].data;
4067: }else{
4068: data = 0;
4069: }
4070: return data;
4071: }
4072:
4073: /* Return the size of the array */
4074: int Symbol_count()
4075: {
4076: return x2a ? x2a->count : 0;
4077: }
4078:
4079: /* Return an array of pointers to all data in the table.
4080: ** The array is obtained from malloc. Return NULL if memory allocation
4081: ** problems, or if the array is empty. */
4082: struct symbol **Symbol_arrayof()
4083: {
4084: struct symbol **array;
4085: int i,size;
4086: if( x2a==0 ) return 0;
4087: size = x2a->count;
4088: array = (struct symbol **)malloc( sizeof(struct symbol *)*size );
4089: if( array ){
4090: for(i=0; i<size; i++) array[i] = x2a->tbl[i].data;
4091: }
4092: return array;
4093: }
4094:
4095: /* Compare two configurations */
4096: int Configcmp(a,b)
4097: struct config *a;
4098: struct config *b;
4099: {
4100: int x;
4101: x = a->rp->index - b->rp->index;
4102: if( x==0 ) x = a->dot - b->dot;
4103: return x;
4104: }
4105:
4106: /* Compare two states */
4107: PRIVATE int statecmp(a,b)
4108: struct config *a;
4109: struct config *b;
4110: {
4111: int rc;
4112: for(rc=0; rc==0 && a && b; a=a->bp, b=b->bp){
4113: rc = a->rp->index - b->rp->index;
4114: if( rc==0 ) rc = a->dot - b->dot;
4115: }
4116: if( rc==0 ){
4117: if( a ) rc = 1;
4118: if( b ) rc = -1;
4119: }
4120: return rc;
4121: }
4122:
4123: /* Hash a state */
4124: PRIVATE int statehash(a)
4125: struct config *a;
4126: {
4127: unsigned int h=0;
4128: while( a ){
4129: h = h*571u + (unsigned int)a->rp->index*37u + (unsigned int)a->dot;
4130: a = a->bp;
4131: }
4132: return h;
4133: }
4134:
4135: /* Allocate a new state structure */
4136: struct state *State_new()
4137: {
4138: struct state *new;
4139: new = (struct state *)malloc( sizeof(struct state) );
4140: MemoryCheck(new);
4141: return new;
4142: }
4143:
4144: /* There is one instance of the following structure for each
4145: ** associative array of type "x3".
4146: */
4147: struct s_x3 {
4148: int size; /* The number of available slots. */
4149: /* Must be a power of 2 greater than or */
4150: /* equal to 1 */
4151: int count; /* Number of currently slots filled */
4152: struct s_x3node *tbl; /* The data stored here */
4153: struct s_x3node **ht; /* Hash table for lookups */
4154: };
4155:
4156: /* There is one instance of this structure for every data element
4157: ** in an associative array of type "x3".
4158: */
4159: typedef struct s_x3node {
4160: struct state *data; /* The data */
4161: struct config *key; /* The key */
4162: struct s_x3node *next; /* Next entry with the same hash */
4163: struct s_x3node **from; /* Previous link */
4164: } x3node;
4165:
4166: /* There is only one instance of the array, which is the following */
4167: static struct s_x3 *x3a;
4168:
4169: /* Allocate a new associative array */
4170: void State_init(){
4171: if( x3a ) return;
4172: x3a = (struct s_x3*)malloc( sizeof(struct s_x3) );
4173: if( x3a ){
4174: x3a->size = 128;
4175: x3a->count = 0;
4176: x3a->tbl = (x3node*)malloc(
4177: (sizeof(x3node) + sizeof(x3node*))*128 );
4178: if( x3a->tbl==0 ){
4179: free(x3a);
4180: x3a = 0;
4181: }else{
4182: int i;
4183: x3a->ht = (x3node**)&(x3a->tbl[128]);
4184: for(i=0; i<128; i++) x3a->ht[i] = 0;
4185: }
4186: }
4187: }
4188: /* Insert a new record into the array. Return TRUE if successful.
4189: ** Prior data with the same key is NOT overwritten */
4190: int State_insert(data,key)
4191: struct state *data;
4192: struct config *key;
4193: {
4194: x3node *np;
4195: int h;
4196: int ph;
4197:
4198: if( x3a==0 ) return 0;
4199: ph = statehash(key);
4200: h = ph & (x3a->size-1);
4201: np = x3a->ht[h];
4202: while( np ){
4203: if( statecmp(np->key,key)==0 ){
4204: /* An existing entry with the same key is found. */
4205: /* Fail because overwrite is not allows. */
4206: return 0;
4207: }
4208: np = np->next;
4209: }
4210: if( x3a->count>=x3a->size ){
4211: /* Need to make the hash table bigger */
4212: int i,size;
4213: struct s_x3 array;
4214: array.size = size = x3a->size*2;
4215: array.count = x3a->count;
4216: array.tbl = (x3node*)malloc(
4217: (sizeof(x3node) + sizeof(x3node*))*size );
4218: if( array.tbl==0 ) return 0; /* Fail due to malloc failure */
4219: array.ht = (x3node**)&(array.tbl[size]);
4220: for(i=0; i<size; i++) array.ht[i] = 0;
4221: for(i=0; i<x3a->count; i++){
4222: x3node *oldnp, *newnp;
4223: oldnp = &(x3a->tbl[i]);
4224: h = statehash(oldnp->key) & (size-1);
4225: newnp = &(array.tbl[i]);
4226: if( array.ht[h] ) array.ht[h]->from = &(newnp->next);
4227: newnp->next = array.ht[h];
4228: newnp->key = oldnp->key;
4229: newnp->data = oldnp->data;
4230: newnp->from = &(array.ht[h]);
4231: array.ht[h] = newnp;
4232: }
4233: free(x3a->tbl);
4234: *x3a = array;
4235: }
4236: /* Insert the new data */
4237: h = ph & (x3a->size-1);
4238: np = &(x3a->tbl[x3a->count++]);
4239: np->key = key;
4240: np->data = data;
4241: if( x3a->ht[h] ) x3a->ht[h]->from = &(np->next);
4242: np->next = x3a->ht[h];
4243: x3a->ht[h] = np;
4244: np->from = &(x3a->ht[h]);
4245: return 1;
4246: }
4247:
4248: /* Return a pointer to data assigned to the given key. Return NULL
4249: ** if no such key. */
4250: struct state *State_find(key)
4251: struct config *key;
4252: {
4253: int h;
4254: x3node *np;
4255:
4256: if( x3a==0 ) return 0;
4257: h = statehash(key) & (x3a->size-1);
4258: np = x3a->ht[h];
4259: while( np ){
4260: if( statecmp(np->key,key)==0 ) break;
4261: np = np->next;
4262: }
4263: return np ? np->data : 0;
4264: }
4265:
4266: /* Return an array of pointers to all data in the table.
4267: ** The array is obtained from malloc. Return NULL if memory allocation
4268: ** problems, or if the array is empty. */
4269: struct state **State_arrayof()
4270: {
4271: struct state **array;
4272: int i,size;
4273: if( x3a==0 ) return 0;
4274: size = x3a->count;
4275: array = (struct state **)malloc( sizeof(struct state *)*size );
4276: if( array ){
4277: for(i=0; i<size; i++) array[i] = x3a->tbl[i].data;
4278: }
4279: return array;
4280: }
4281:
4282: /* Hash a configuration */
4283: PRIVATE int confighash(a)
4284: struct config *a;
4285: {
4286: int h=0;
4287: h = h*571 + a->rp->index*37 + a->dot;
4288: return h;
4289: }
4290:
4291: /* There is one instance of the following structure for each
4292: ** associative array of type "x4".
4293: */
4294: struct s_x4 {
4295: int size; /* The number of available slots. */
4296: /* Must be a power of 2 greater than or */
4297: /* equal to 1 */
4298: int count; /* Number of currently slots filled */
4299: struct s_x4node *tbl; /* The data stored here */
4300: struct s_x4node **ht; /* Hash table for lookups */
4301: };
4302:
4303: /* There is one instance of this structure for every data element
4304: ** in an associative array of type "x4".
4305: */
4306: typedef struct s_x4node {
4307: struct config *data; /* The data */
4308: struct s_x4node *next; /* Next entry with the same hash */
4309: struct s_x4node **from; /* Previous link */
4310: } x4node;
4311:
4312: /* There is only one instance of the array, which is the following */
4313: static struct s_x4 *x4a;
4314:
4315: /* Allocate a new associative array */
4316: void Configtable_init(){
4317: if( x4a ) return;
4318: x4a = (struct s_x4*)malloc( sizeof(struct s_x4) );
4319: if( x4a ){
4320: x4a->size = 64;
4321: x4a->count = 0;
4322: x4a->tbl = (x4node*)malloc(
4323: (sizeof(x4node) + sizeof(x4node*))*64 );
4324: if( x4a->tbl==0 ){
4325: free(x4a);
4326: x4a = 0;
4327: }else{
4328: int i;
4329: x4a->ht = (x4node**)&(x4a->tbl[64]);
4330: for(i=0; i<64; i++) x4a->ht[i] = 0;
4331: }
4332: }
4333: }
4334: /* Insert a new record into the array. Return TRUE if successful.
4335: ** Prior data with the same key is NOT overwritten */
4336: int Configtable_insert(data)
4337: struct config *data;
4338: {
4339: x4node *np;
4340: int h;
4341: int ph;
4342:
4343: if( x4a==0 ) return 0;
4344: ph = confighash(data);
4345: h = ph & (x4a->size-1);
4346: np = x4a->ht[h];
4347: while( np ){
4348: if( Configcmp(np->data,data)==0 ){
4349: /* An existing entry with the same key is found. */
4350: /* Fail because overwrite is not allows. */
4351: return 0;
4352: }
4353: np = np->next;
4354: }
4355: if( x4a->count>=x4a->size ){
4356: /* Need to make the hash table bigger */
4357: int i,size;
4358: struct s_x4 array;
4359: array.size = size = x4a->size*2;
4360: array.count = x4a->count;
4361: array.tbl = (x4node*)malloc(
4362: (sizeof(x4node) + sizeof(x4node*))*size );
4363: if( array.tbl==0 ) return 0; /* Fail due to malloc failure */
4364: array.ht = (x4node**)&(array.tbl[size]);
4365: for(i=0; i<size; i++) array.ht[i] = 0;
4366: for(i=0; i<x4a->count; i++){
4367: x4node *oldnp, *newnp;
4368: oldnp = &(x4a->tbl[i]);
4369: h = confighash(oldnp->data) & (size-1);
4370: newnp = &(array.tbl[i]);
4371: if( array.ht[h] ) array.ht[h]->from = &(newnp->next);
4372: newnp->next = array.ht[h];
4373: newnp->data = oldnp->data;
4374: newnp->from = &(array.ht[h]);
4375: array.ht[h] = newnp;
4376: }
4377: free(x4a->tbl);
4378: *x4a = array;
4379: }
4380: /* Insert the new data */
4381: h = ph & (x4a->size-1);
4382: np = &(x4a->tbl[x4a->count++]);
4383: np->data = data;
4384: if( x4a->ht[h] ) x4a->ht[h]->from = &(np->next);
4385: np->next = x4a->ht[h];
4386: x4a->ht[h] = np;
4387: np->from = &(x4a->ht[h]);
4388: return 1;
4389: }
4390:
4391: /* Return a pointer to data assigned to the given key. Return NULL
4392: ** if no such key. */
4393: struct config *Configtable_find(key)
4394: struct config *key;
4395: {
4396: int h;
4397: x4node *np;
4398:
4399: if( x4a==0 ) return 0;
4400: h = confighash(key) & (x4a->size-1);
4401: np = x4a->ht[h];
4402: while( np ){
4403: if( Configcmp(np->data,key)==0 ) break;
4404: np = np->next;
4405: }
4406: return np ? np->data : 0;
4407: }
4408:
4409: /* Remove all data from the table. Pass each data to the function "f"
4410: ** as it is removed. ("f" may be null to avoid this step.) */
4411: void Configtable_clear(f)
4412: int(*f)(/* struct config * */);
4413: {
4414: int i;
4415: if( x4a==0 || x4a->count==0 ) return;
4416: if( f ) for(i=0; i<x4a->count; i++) (*f)(x4a->tbl[i].data);
4417: for(i=0; i<x4a->size; i++) x4a->ht[i] = 0;
4418: x4a->count = 0;
4419: return;
4420: }
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/cvsweb/back.gif){kind=icon}
Return to lemon.c CVS log ![[TXT]](/icons/cvsweb/text.gif){kind=icon}
![[DIR]](/icons/cvsweb/dir.gif){kind=icon}
Up to [ELWIX - Embedded LightWeight unIX -] / embedaddon / lighttpd / src