1 %{ 2 /* 3 * CDDL HEADER START 4 * 5 * The contents of this file are subject to the terms of the 6 * Common Development and Distribution License, Version 1.0 only 7 * (the "License"). You may not use this file except in compliance 8 * with the License. 9 * 10 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE 11 * or http://www.opensolaris.org/os/licensing. 12 * See the License for the specific language governing permissions 13 * and limitations under the License. 14 * 15 * When distributing Covered Code, include this CDDL HEADER in each 16 * file and include the License file at usr/src/OPENSOLARIS.LICENSE. 17 * If applicable, add the following below this CDDL HEADER, with the 18 * fields enclosed by brackets "[]" replaced with your own identifying 19 * information: Portions Copyright [yyyy] [name of copyright owner] 20 * 21 * CDDL HEADER END 22 */ 23 %} 24 /* Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T */ 25 /* All Rights Reserved */ 26 27 28 %{ 29 #ident "%Z%%M% %I% %E% SMI" /* SVr4.0 2.10 */ 30 %} 31 32 %{ 33 #include "awk.h" 34 yywrap() { return(1); } 35 #ifndef DEBUG 36 # define PUTS(x) 37 #endif 38 Node *beginloc = 0, *endloc = 0; 39 int infunc = 0; /* = 1 if in arglist or body of func */ 40 uchar *curfname = 0; 41 Node *arglist = 0; /* list of args for current function */ 42 uchar *strnode(); 43 Node *notnull(); 44 %} 45 46 %union { 47 Node *p; 48 Cell *cp; 49 int i; 50 uchar *s; 51 } 52 53 %token <i> FIRSTTOKEN /* must be first */ 54 %token <p> PROGRAM PASTAT PASTAT2 XBEGIN XEND 55 %token <i> NL ',' '{' '(' '|' ';' '/' ')' '}' '[' ']' 56 %token <i> ARRAY 57 %token <i> MATCH NOTMATCH MATCHOP 58 %token <i> FINAL DOT ALL CCL NCCL CHAR OR STAR QUEST PLUS 59 %token <i> AND BOR APPEND EQ GE GT LE LT NE IN 60 %token <i> ARG BLTIN BREAK CLOSE CONTINUE DELETE DO EXIT FOR FUNC 61 %token <i> SUB GSUB IF INDEX LSUBSTR MATCHFCN NEXT 62 %token <i> ADD MINUS MULT DIVIDE MOD 63 %token <i> ASSIGN ASGNOP ADDEQ SUBEQ MULTEQ DIVEQ MODEQ POWEQ 64 %token <i> PRINT PRINTF SPRINTF 65 %token <p> ELSE INTEST CONDEXPR 66 %token <i> POSTINCR PREINCR POSTDECR PREDECR 67 %token <cp> VAR IVAR VARNF CALL NUMBER STRING FIELD 68 %token <s> REGEXPR 69 70 %type <p> pas pattern ppattern plist pplist patlist prarg term 71 %type <p> pa_pat pa_stat pa_stats 72 %type <s> reg_expr 73 %type <p> simple_stmt opt_simple_stmt stmt stmtlist 74 %type <p> var varname funcname varlist 75 %type <p> for if while 76 %type <i> pst opt_pst lbrace rparen comma nl opt_nl and bor 77 %type <i> subop print 78 79 %right ASGNOP 80 %right '?' 81 %right ':' 82 %left BOR 83 %left AND 84 %left GETLINE 85 %nonassoc APPEND EQ GE GT LE LT NE MATCHOP IN '|' 86 %left ARG BLTIN BREAK CALL CLOSE CONTINUE DELETE DO EXIT FOR FIELD FUNC 87 %left GSUB IF INDEX LSUBSTR MATCHFCN NEXT NUMBER 88 %left PRINT PRINTF RETURN SPLIT SPRINTF STRING SUB SUBSTR 89 %left REGEXPR VAR VARNF IVAR WHILE '(' 90 %left CAT 91 %left '+' '-' 92 %left '*' '/' '%' 93 %left NOT UMINUS 94 %right POWER 95 %right DECR INCR 96 %left INDIRECT 97 %token LASTTOKEN /* must be last */ 98 99 %% 100 101 program: 102 pas { if (errorflag==0) 103 winner = (Node *)stat3(PROGRAM, beginloc, $1, endloc); } 104 | error { yyclearin; bracecheck(); ERROR "bailing out" SYNTAX; } 105 ; 106 107 and: 108 AND | and NL 109 ; 110 111 bor: 112 BOR | bor NL 113 ; 114 115 comma: 116 ',' | comma NL 117 ; 118 119 do: 120 DO | do NL 121 ; 122 123 else: 124 ELSE | else NL 125 ; 126 127 for: 128 FOR '(' opt_simple_stmt ';' pattern ';' opt_simple_stmt rparen stmt 129 { $$ = stat4(FOR, $3, notnull($5), $7, $9); } 130 | FOR '(' opt_simple_stmt ';' ';' opt_simple_stmt rparen stmt 131 { $$ = stat4(FOR, $3, NIL, $6, $8); } 132 | FOR '(' varname IN varname rparen stmt 133 { $$ = stat3(IN, $3, makearr($5), $7); } 134 ; 135 136 funcname: 137 VAR { setfname($1); } 138 | CALL { setfname($1); } 139 ; 140 141 if: 142 IF '(' pattern rparen { $$ = notnull($3); } 143 ; 144 145 lbrace: 146 '{' | lbrace NL 147 ; 148 149 nl: 150 NL | nl NL 151 ; 152 153 opt_nl: 154 /* empty */ { $$ = 0; } 155 | nl 156 ; 157 158 opt_pst: 159 /* empty */ { $$ = 0; } 160 | pst 161 ; 162 163 164 opt_simple_stmt: 165 /* empty */ { $$ = 0; } 166 | simple_stmt 167 ; 168 169 pas: 170 opt_pst { $$ = 0; } 171 | opt_pst pa_stats opt_pst { $$ = $2; } 172 ; 173 174 pa_pat: 175 pattern { $$ = notnull($1); } 176 ; 177 178 pa_stat: 179 pa_pat { $$ = stat2(PASTAT, $1, stat2(PRINT, rectonode(), NIL)); } 180 | pa_pat lbrace stmtlist '}' { $$ = stat2(PASTAT, $1, $3); } 181 | pa_pat ',' pa_pat { $$ = pa2stat($1, $3, stat2(PRINT, rectonode(), NIL)); } 182 | pa_pat ',' pa_pat lbrace stmtlist '}' { $$ = pa2stat($1, $3, $5); } 183 | lbrace stmtlist '}' { $$ = stat2(PASTAT, NIL, $2); } 184 | XBEGIN lbrace stmtlist '}' 185 { beginloc = linkum(beginloc, $3); $$ = 0; } 186 | XEND lbrace stmtlist '}' 187 { endloc = linkum(endloc, $3); $$ = 0; } 188 | FUNC funcname '(' varlist rparen {infunc++;} lbrace stmtlist '}' 189 { infunc--; curfname=0; defn((Cell *)$2, $4, $8); $$ = 0; } 190 ; 191 192 pa_stats: 193 pa_stat 194 | pa_stats opt_pst pa_stat { $$ = linkum($1, $3); } 195 ; 196 197 patlist: 198 pattern 199 | patlist comma pattern { $$ = linkum($1, $3); } 200 ; 201 202 ppattern: 203 var ASGNOP ppattern { $$ = op2($2, $1, $3); } 204 | ppattern '?' ppattern ':' ppattern %prec '?' 205 { $$ = op3(CONDEXPR, notnull($1), $3, $5); } 206 | ppattern bor ppattern %prec BOR 207 { $$ = op2(BOR, notnull($1), notnull($3)); } 208 | ppattern and ppattern %prec AND 209 { $$ = op2(AND, notnull($1), notnull($3)); } 210 | NOT ppattern 211 { $$ = op1(NOT, notnull($2)); } 212 | ppattern MATCHOP reg_expr { $$ = op3($2, NIL, $1, (Node*)makedfa($3, 0)); } 213 | ppattern MATCHOP ppattern 214 { if (constnode($3)) 215 $$ = op3($2, NIL, $1, (Node*)makedfa(strnode($3), 0)); 216 else 217 $$ = op3($2, (Node *)1, $1, $3); } 218 | ppattern IN varname { $$ = op2(INTEST, $1, makearr($3)); } 219 | '(' plist ')' IN varname { $$ = op2(INTEST, $2, makearr($5)); } 220 | ppattern term %prec CAT { $$ = op2(CAT, $1, $2); } 221 | reg_expr 222 { $$ = op3(MATCH, NIL, rectonode(), (Node*)makedfa($1, 0)); } 223 | term 224 ; 225 226 pattern: 227 var ASGNOP pattern { $$ = op2($2, $1, $3); } 228 | pattern '?' pattern ':' pattern %prec '?' 229 { $$ = op3(CONDEXPR, notnull($1), $3, $5); } 230 | pattern bor pattern %prec BOR 231 { $$ = op2(BOR, notnull($1), notnull($3)); } 232 | pattern and pattern %prec AND 233 { $$ = op2(AND, notnull($1), notnull($3)); } 234 | NOT pattern 235 { $$ = op1(NOT, op2(NE,$2,valtonode(lookup("$zero&null",symtab),CCON))); } 236 | pattern EQ pattern { $$ = op2($2, $1, $3); } 237 | pattern GE pattern { $$ = op2($2, $1, $3); } 238 | pattern GT pattern { $$ = op2($2, $1, $3); } 239 | pattern LE pattern { $$ = op2($2, $1, $3); } 240 | pattern LT pattern { $$ = op2($2, $1, $3); } 241 | pattern NE pattern { $$ = op2($2, $1, $3); } 242 | pattern MATCHOP reg_expr { $$ = op3($2, NIL, $1, (Node*)makedfa($3, 0)); } 243 | pattern MATCHOP pattern 244 { if (constnode($3)) 245 $$ = op3($2, NIL, $1, (Node*)makedfa(strnode($3), 0)); 246 else 247 $$ = op3($2, (Node *)1, $1, $3); } 248 | pattern IN varname { $$ = op2(INTEST, $1, makearr($3)); } 249 | '(' plist ')' IN varname { $$ = op2(INTEST, $2, makearr($5)); } 250 | pattern '|' GETLINE var { $$ = op3(GETLINE, $4, (Node*)$2, $1); } 251 | pattern '|' GETLINE { $$ = op3(GETLINE, (Node*)0, (Node*)$2, $1); } 252 | pattern term %prec CAT { $$ = op2(CAT, $1, $2); } 253 | reg_expr 254 { $$ = op3(MATCH, NIL, rectonode(), (Node*)makedfa($1, 0)); } 255 | term 256 ; 257 258 plist: 259 pattern comma pattern { $$ = linkum($1, $3); } 260 | plist comma pattern { $$ = linkum($1, $3); } 261 ; 262 263 pplist: 264 ppattern 265 | pplist comma ppattern { $$ = linkum($1, $3); } 266 267 prarg: 268 /* empty */ { $$ = rectonode(); } 269 | pplist 270 | '(' plist ')' { $$ = $2; } 271 ; 272 273 print: 274 PRINT | PRINTF 275 ; 276 277 pst: 278 NL | ';' | pst NL | pst ';' 279 ; 280 281 rbrace: 282 '}' | rbrace NL 283 ; 284 285 reg_expr: 286 '/' {startreg();} REGEXPR '/' { $$ = $3; } 287 ; 288 289 rparen: 290 ')' | rparen NL 291 ; 292 293 simple_stmt: 294 print prarg '|' term { $$ = stat3($1, $2, (Node *) $3, $4); } 295 | print prarg APPEND term { $$ = stat3($1, $2, (Node *) $3, $4); } 296 | print prarg GT term { $$ = stat3($1, $2, (Node *) $3, $4); } 297 | print prarg { $$ = stat3($1, $2, NIL, NIL); } 298 | DELETE varname '[' patlist ']' { $$ = stat2(DELETE, makearr($2), $4); } 299 | DELETE varname { yyclearin; ERROR "you can only delete array[element]" SYNTAX; $$ = stat1(DELETE, $2); } 300 | pattern { $$ = exptostat($1); } 301 | error { yyclearin; ERROR "illegal statement" SYNTAX; } 302 ; 303 304 st: 305 nl | ';' opt_nl 306 ; 307 308 stmt: 309 BREAK st { $$ = stat1(BREAK, NIL); } 310 | CLOSE pattern st { $$ = stat1(CLOSE, $2); } 311 | CONTINUE st { $$ = stat1(CONTINUE, NIL); } 312 | do stmt WHILE '(' pattern ')' st 313 { $$ = stat2(DO, $2, notnull($5)); } 314 | EXIT pattern st { $$ = stat1(EXIT, $2); } 315 | EXIT st { $$ = stat1(EXIT, NIL); } 316 | for 317 | if stmt else stmt { $$ = stat3(IF, $1, $2, $4); } 318 | if stmt { $$ = stat3(IF, $1, $2, NIL); } 319 | lbrace stmtlist rbrace { $$ = $2; } 320 | NEXT st { if (infunc) 321 ERROR "next is illegal inside a function" SYNTAX; 322 $$ = stat1(NEXT, NIL); } 323 | RETURN pattern st { $$ = stat1(RETURN, $2); } 324 | RETURN st { $$ = stat1(RETURN, NIL); } 325 | simple_stmt st 326 | while stmt { $$ = stat2(WHILE, $1, $2); } 327 | ';' opt_nl { $$ = 0; } 328 ; 329 330 stmtlist: 331 stmt 332 | stmtlist stmt { $$ = linkum($1, $2); } 333 ; 334 335 subop: 336 SUB | GSUB 337 ; 338 339 term: 340 term '+' term { $$ = op2(ADD, $1, $3); } 341 | term '-' term { $$ = op2(MINUS, $1, $3); } 342 | term '*' term { $$ = op2(MULT, $1, $3); } 343 | term '/' term { $$ = op2(DIVIDE, $1, $3); } 344 | term '%' term { $$ = op2(MOD, $1, $3); } 345 | term POWER term { $$ = op2(POWER, $1, $3); } 346 | '-' term %prec UMINUS { $$ = op1(UMINUS, $2); } 347 | '+' term %prec UMINUS { $$ = $2; } 348 | BLTIN '(' ')' { $$ = op2(BLTIN, (Node *) $1, rectonode()); } 349 | BLTIN '(' patlist ')' { $$ = op2(BLTIN, (Node *) $1, $3); } 350 | BLTIN { $$ = op2(BLTIN, (Node *) $1, rectonode()); } 351 | CALL '(' ')' { $$ = op2(CALL, valtonode($1,CVAR), NIL); } 352 | CALL '(' patlist ')' { $$ = op2(CALL, valtonode($1,CVAR), $3); } 353 | DECR var { $$ = op1(PREDECR, $2); } 354 | INCR var { $$ = op1(PREINCR, $2); } 355 | var DECR { $$ = op1(POSTDECR, $1); } 356 | var INCR { $$ = op1(POSTINCR, $1); } 357 | GETLINE var LT term { $$ = op3(GETLINE, $2, (Node *)$3, $4); } 358 | GETLINE LT term { $$ = op3(GETLINE, NIL, (Node *)$2, $3); } 359 | GETLINE var { $$ = op3(GETLINE, $2, NIL, NIL); } 360 | GETLINE { $$ = op3(GETLINE, NIL, NIL, NIL); } 361 | INDEX '(' pattern comma pattern ')' 362 { $$ = op2(INDEX, $3, $5); } 363 | INDEX '(' pattern comma reg_expr ')' 364 { ERROR "index() doesn't permit regular expressions" SYNTAX; 365 $$ = op2(INDEX, $3, (Node*)$5); } 366 | '(' pattern ')' { $$ = $2; } 367 | MATCHFCN '(' pattern comma reg_expr ')' 368 { $$ = op3(MATCHFCN, NIL, $3, (Node*)makedfa($5, 1)); } 369 | MATCHFCN '(' pattern comma pattern ')' 370 { if (constnode($5)) 371 $$ = op3(MATCHFCN, NIL, $3, (Node*)makedfa(strnode($5), 1)); 372 else 373 $$ = op3(MATCHFCN, (Node *)1, $3, $5); } 374 | NUMBER { $$ = valtonode($1, CCON); } 375 | SPLIT '(' pattern comma varname comma pattern ')' /* string */ 376 { $$ = op4(SPLIT, $3, makearr($5), $7, (Node*)STRING); } 377 | SPLIT '(' pattern comma varname comma reg_expr ')' /* const /regexp/ */ 378 { $$ = op4(SPLIT, $3, makearr($5), (Node*)makedfa($7, 1), (Node *)REGEXPR); } 379 | SPLIT '(' pattern comma varname ')' 380 { $$ = op4(SPLIT, $3, makearr($5), NIL, (Node*)STRING); } /* default */ 381 | SPRINTF '(' patlist ')' { $$ = op1($1, $3); } 382 | STRING { $$ = valtonode($1, CCON); } 383 | subop '(' reg_expr comma pattern ')' 384 { $$ = op4($1, NIL, (Node*)makedfa($3, 1), $5, rectonode()); } 385 | subop '(' pattern comma pattern ')' 386 { if (constnode($3)) 387 $$ = op4($1, NIL, (Node*)makedfa(strnode($3), 1), $5, rectonode()); 388 else 389 $$ = op4($1, (Node *)1, $3, $5, rectonode()); } 390 | subop '(' reg_expr comma pattern comma var ')' 391 { $$ = op4($1, NIL, (Node*)makedfa($3, 1), $5, $7); } 392 | subop '(' pattern comma pattern comma var ')' 393 { if (constnode($3)) 394 $$ = op4($1, NIL, (Node*)makedfa(strnode($3), 1), $5, $7); 395 else 396 $$ = op4($1, (Node *)1, $3, $5, $7); } 397 | SUBSTR '(' pattern comma pattern comma pattern ')' 398 { $$ = op3(SUBSTR, $3, $5, $7); } 399 | SUBSTR '(' pattern comma pattern ')' 400 { $$ = op3(SUBSTR, $3, $5, NIL); } 401 | var 402 ; 403 404 var: 405 varname 406 | varname '[' patlist ']' { $$ = op2(ARRAY, makearr($1), $3); } 407 | FIELD { $$ = valtonode($1, CFLD); } 408 | IVAR { $$ = op1(INDIRECT, valtonode($1, CVAR)); } 409 | INDIRECT term { $$ = op1(INDIRECT, $2); } 410 ; 411 412 varlist: 413 /* nothing */ { arglist = $$ = 0; } 414 | VAR { arglist = $$ = valtonode($1,CVAR); } 415 | varlist comma VAR { arglist = $$ = linkum($1,valtonode($3,CVAR)); } 416 ; 417 418 varname: 419 VAR { $$ = valtonode($1, CVAR); } 420 | ARG { $$ = op1(ARG, (Node *) $1); } 421 | VARNF { $$ = op1(VARNF, (Node *) $1); } 422 ; 423 424 425 while: 426 WHILE '(' pattern rparen { $$ = notnull($3); } 427 ; 428 429 %% 430 431 setfname(p) 432 Cell *p; 433 { 434 if (isarr(p)) 435 ERROR "%s is an array, not a function", p->nval SYNTAX; 436 else if (isfunc(p)) 437 ERROR "you can't define function %s more than once", p->nval SYNTAX; 438 curfname = p->nval; 439 } 440 441 constnode(p) 442 Node *p; 443 { 444 return p->ntype == NVALUE && ((Cell *) (p->narg[0]))->csub == CCON; 445 } 446 447 uchar *strnode(p) 448 Node *p; 449 { 450 return ((Cell *)(p->narg[0]))->sval; 451 } 452 453 Node *notnull(n) 454 Node *n; 455 { 456 switch (n->nobj) { 457 case LE: case LT: case EQ: case NE: case GT: case GE: 458 case BOR: case AND: case NOT: 459 return n; 460 default: 461 return op2(NE, n, nullnode); 462 } 463 } 464