dlls/vbscript/lex.c

   1 /*
   2  * Copyright 2011 Jacek Caban for CodeWeavers
   3  *
   4  * This library is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU Lesser General Public
   6  * License as published by the Free Software Foundation; either
   7  * version 2.1 of the License, or (at your option) any later version.
   8  *
   9  * This library is distributed in the hope that it will be useful,
  10  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12  * Lesser General Public License for more details.
  13  *
  14  * You should have received a copy of the GNU Lesser General Public
  15  * License along with this library; if not, write to the Free Software
  16  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
  17  */
  18
  19 #include <assert.h>
  20
  21 #include "vbscript.h"
  22 #include "parse.h"
  23 #include "parser.tab.h"
  24
  25 #include "wine/debug.h"
  26
  27 WINE_DEFAULT_DEBUG_CHANNEL(vbscript);
  28
  29 static const WCHAR andW[] = {'a','n','d',0};
  30 static const WCHAR byrefW[] = {'b','y','r','e','f',0};
  31 static const WCHAR byvalW[] = {'b','y','v','a','l',0};
  32 static const WCHAR callW[] = {'c','a','l','l',0};
  33 static const WCHAR classW[] = {'c','l','a','s','s',0};
  34 static const WCHAR constW[] = {'c','o','n','s','t',0};
  35 static const WCHAR defaultW[] = {'d','e','f','a','u','l','t',0};
  36 static const WCHAR dimW[] = {'d','i','m',0};
  37 static const WCHAR doW[] = {'d','o',0};
  38 static const WCHAR elseW[] = {'e','l','s','e',0};
  39 static const WCHAR elseifW[] = {'e','l','s','e','i','f',0};
  40 static const WCHAR emptyW[] = {'e','m','p','t','y',0};
  41 static const WCHAR endW[] = {'e','n','d',0};
  42 static const WCHAR eqvW[] = {'e','q','v',0};
  43 static const WCHAR errorW[] = {'e','r','r','o','r',0};
  44 static const WCHAR exitW[] = {'e','x','i','t',0};
  45 static const WCHAR explicitW[] = {'e','x','p','l','i','c','i','t',0};
  46 static const WCHAR falseW[] = {'f','a','l','s','e',0};
  47 static const WCHAR forW[] = {'f','o','r',0};
  48 static const WCHAR functionW[] = {'f','u','n','c','t','i','o','n',0};
  49 static const WCHAR getW[] = {'g','e','t',0};
  50 static const WCHAR gotoW[] = {'g','o','t','o',0};
  51 static const WCHAR ifW[] = {'i','f',0};
  52 static const WCHAR impW[] = {'i','m','p',0};
  53 static const WCHAR isW[] = {'i','s',0};
  54 static const WCHAR letW[] = {'l','e','t',0};
  55 static const WCHAR loopW[] = {'l','o','o','p',0};
  56 static const WCHAR meW[] = {'m','e',0};
  57 static const WCHAR modW[] = {'m','o','d',0};
  58 static const WCHAR newW[] = {'n','e','w',0};
  59 static const WCHAR nextW[] = {'n','e','x','t',0};
  60 static const WCHAR notW[] = {'n','o','t',0};
  61 static const WCHAR nothingW[] = {'n','o','t','h','i','n','g',0};
  62 static const WCHAR nullW[] = {'n','u','l','l',0};
  63 static const WCHAR onW[] = {'o','n',0};
  64 static const WCHAR optionW[] = {'o','p','t','i','o','n',0};
  65 static const WCHAR orW[] = {'o','r',0};
  66 static const WCHAR privateW[] = {'p','r','i','v','a','t','e',0};
  67 static const WCHAR propertyW[] = {'p','r','o','p','e','r','t','y',0};
  68 static const WCHAR publicW[] = {'p','u','b','l','i','c',0};
  69 static const WCHAR remW[] = {'r','e','m',0};
  70 static const WCHAR resumeW[] = {'r','e','s','u','m','e',0};
  71 static const WCHAR setW[] = {'s','e','t',0};
  72 static const WCHAR stepW[] = {'s','t','e','p',0};
  73 static const WCHAR stopW[] = {'s','t','o','p',0};
  74 static const WCHAR subW[] = {'s','u','b',0};
  75 static const WCHAR thenW[] = {'t','h','e','n',0};
  76 static const WCHAR toW[] = {'t','o',0};
  77 static const WCHAR trueW[] = {'t','r','u','e',0};
  78 static const WCHAR untilW[] = {'u','n','t','i','l',0};
  79 static const WCHAR wendW[] = {'w','e','n','d',0};
  80 static const WCHAR whileW[] = {'w','h','i','l','e',0};
  81 static const WCHAR xorW[] = {'x','o','r',0};
  82
  83 static const struct {
  84     const WCHAR *word;
  85     int token;
  86 } keywords[] = {
  87     {andW,       tAND},
  88     {byrefW,     tBYREF},
  89     {byvalW,     tBYVAL},
  90     {callW,      tCALL},
  91     {classW,     tCLASS},
  92     {constW,     tCONST},
  93     {defaultW,   tDEFAULT},
  94     {dimW,       tDIM},
  95     {doW,        tDO},
  96     {elseW,      tELSE},
  97     {elseifW,    tELSEIF},
  98     {emptyW,     tEMPTY},
  99     {endW,       tEND},
 100     {eqvW,       tEQV},
 101     {errorW,     tERROR},
 102     {exitW,      tEXIT},
 103     {explicitW,  tEXPLICIT},
 104     {falseW,     tFALSE},
 105     {forW,       tFOR},
 106     {functionW,  tFUNCTION},
 107     {getW,       tGET},
 108     {gotoW,      tGOTO},
 109     {ifW,        tIF},
 110     {impW,       tIMP},
 111     {isW,        tIS},
 112     {letW,       tLET},
 113     {loopW,      tLOOP},
 114     {meW,        tME},
 115     {modW,       tMOD},
 116     {newW,       tNEW},
 117     {nextW,      tNEXT},
 118     {notW,       tNOT},
 119     {nothingW,   tNOTHING},
 120     {nullW,      tNULL},
 121     {onW,        tON},
 122     {optionW,    tOPTION},
 123     {orW,        tOR},
 124     {privateW,   tPRIVATE},
 125     {propertyW,  tPROPERTY},
 126     {publicW,    tPUBLIC},
 127     {remW,       tREM},
 128     {resumeW,    tRESUME},
 129     {setW,       tSET},
 130     {stepW,      tSTEP},
 131     {stopW,      tSTOP},
 132     {subW,       tSUB},
 133     {thenW,      tTHEN},
 134     {toW,        tTO},
 135     {trueW,      tTRUE},
 136     {untilW,     tUNTIL},
 137     {wendW,      tWEND},
 138     {whileW,     tWHILE},
 139     {xorW,       tXOR}
 140 };
 141
 142 static inline BOOL is_identifier_char(WCHAR c)
 143 {
 144     return isalnumW(c) || c == '_';
 145 }
 146
 147 static int check_keyword(parser_ctx_t *ctx, const WCHAR *word)
 148 {
 149     const WCHAR *p1 = ctx->ptr;
 150     const WCHAR *p2 = word;
 151     WCHAR c;
 152
 153     while(p1 < ctx->end && *p2) {
 154         c = tolowerW(*p1);
 155         if(c != *p2)
 156             return c - *p2;
 157         p1++;
 158         p2++;
 159     }
 160
 161     if(*p2 || (p1 < ctx->end && is_identifier_char(*p1)))
 162         return 1;
 163
 164     ctx->ptr = p1;
 165     return 0;
 166 }
 167
 168 static int check_keywords(parser_ctx_t *ctx)
 169 {
 170     int min = 0, max = sizeof(keywords)/sizeof(keywords[0])-1, r, i;
 171
 172     while(min <= max) {
 173         i = (min+max)/2;
 174
 175         r = check_keyword(ctx, keywords[i].word);
 176         if(!r)
 177             return keywords[i].token;
 178
 179         if(r > 0)
 180             min = i+1;
 181         else
 182             max = i-1;
 183     }
 184
 185     return 0;
 186 }
 187
 188 static int parse_identifier(parser_ctx_t *ctx, const WCHAR **ret)
 189 {
 190     const WCHAR *ptr = ctx->ptr++;
 191     WCHAR *str;
 192     int len;
 193
 194     while(ctx->ptr < ctx->end && is_identifier_char(*ctx->ptr))
 195         ctx->ptr++;
 196     len = ctx->ptr-ptr;
 197
 198     str = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
 199     if(!str)
 200         return 0;
 201
 202     memcpy(str, ptr, (len+1)*sizeof(WCHAR));
 203     str[len] = 0;
 204     *ret = str;
 205     return tIdentifier;
 206 }
 207
 208 static int parse_string_literal(parser_ctx_t *ctx, const WCHAR **ret)
 209 {
 210     const WCHAR *ptr = ++ctx->ptr;
 211     WCHAR *rptr;
 212     int len = 0;
 213
 214     while(ctx->ptr < ctx->end) {
 215         if(*ctx->ptr == '\n') {
 216             FIXME("newline inside string literal\n");
 217             return 0;
 218         }
 219
 220        if(*ctx->ptr == '"') {
 221             if(ctx->ptr[1] != '"')
 222                 break;
 223             len--;
 224             ctx->ptr++;
 225         }
 226         ctx->ptr++;
 227     }
 228
 229     if(ctx->ptr == ctx->end) {
 230         FIXME("unterminated string literal\n");
 231         return 0;
 232     }
 233
 234     len += ctx->ptr-ptr;
 235
 236     *ret = rptr = parser_alloc(ctx, (len+1)*sizeof(WCHAR));
 237     if(!rptr)
 238         return 0;
 239
 240     while(ptr < ctx->ptr) {
 241         if(*ptr == '"')
 242             ptr++;
 243         *rptr++ = *ptr++;
 244     }
 245
 246     *rptr = 0;
 247     ctx->ptr++;
 248     return tString;
 249 }
 250
 251 static int parse_numeric_literal(parser_ctx_t *ctx, void **ret)
 252 {
 253     double n = 0;
 254
 255     if(*ctx->ptr == '0' && !('0' <= ctx->ptr[1] && ctx->ptr[1] <= '9') && ctx->ptr[1] != '.')
 256         return *ctx->ptr++;
 257
 258     do {
 259         n = n*10 + *ctx->ptr++ - '0';
 260     }while('0' <= *ctx->ptr && *ctx->ptr <= '9');
 261
 262     if(*ctx->ptr != '.') {
 263         if((LONG)n == n) {
 264             LONG l = n;
 265             *(LONG*)ret = l;
 266             return (short)l == l ? tShort : tLong;
 267         }
 268     }else {
 269         double e = 1.0;
 270         while('0' <= *++ctx->ptr && *ctx->ptr <= '9')
 271             n += (e /= 10.0)*(*ctx->ptr-'0');
 272     }
 273
 274     *(double*)ret = n;
 275     return tDouble;
 276 }
 277
 278 static int hex_to_int(WCHAR c)
 279 {
 280     if('0' <= c && c <= '9')
 281         return c-'0';
 282     if('a' <= c && c <= 'f')
 283         return c+10-'a';
 284     if('A' <= c && c <= 'F')
 285         return c+10-'A';
 286     return -1;
 287 }
 288
 289 static int parse_hex_literal(parser_ctx_t *ctx, LONG *ret)
 290 {
 291     const WCHAR *begin = ctx->ptr;
 292     LONG l = 0, d;
 293
 294     while((d = hex_to_int(*++ctx->ptr)) != -1)
 295         l = l*16 + d;
 296
 297     if(begin + 9 /* max digits+1 */ < ctx->ptr || (*ctx->ptr != '&' && is_identifier_char(*ctx->ptr))) {
 298         FIXME("invalid literal\n");
 299         return 0;
 300     }
 301
 302     if(*ctx->ptr == '&')
 303         ctx->ptr++;
 304
 305     *ret = l;
 306     return (short)l == l ? tShort : tLong;
 307 }
 308
 309 static void skip_spaces(parser_ctx_t *ctx)
 310 {
 311     while(*ctx->ptr == ' ' || *ctx->ptr == '\t' || *ctx->ptr == '\r')
 312         ctx->ptr++;
 313 }
 314
 315 static int parse_next_token(void *lval, parser_ctx_t *ctx)
 316 {
 317     WCHAR c;
 318
 319     skip_spaces(ctx);
 320     if(ctx->ptr == ctx->end)
 321         return ctx->last_token == tNL ? tEOF : tNL;
 322
 323     c = *ctx->ptr;
 324
 325     if('0' <= c && c <= '9')
 326         return parse_numeric_literal(ctx, lval);
 327
 328     if(isalphaW(c)) {
 329         int ret = check_keywords(ctx);
 330         if(!ret)
 331             return parse_identifier(ctx, lval);
 332         if(ret != tREM)
 333             return ret;
 334         c = '\'';
 335     }
 336
 337     switch(c) {
 338     case '\n':
 339         ctx->ptr++;
 340         return tNL;
 341     case '\'':
 342         ctx->ptr = strchrW(ctx->ptr, '\n');
 343         if(ctx->ptr)
 344             ctx->ptr++;
 345         else
 346             ctx->ptr = ctx->end;
 347         return tNL;
 348     case ':':
 349     case ')':
 350     case ',':
 351     case '=':
 352     case '+':
 353     case '-':
 354     case '*':
 355     case '/':
 356     case '^':
 357     case '\\':
 358     case '.':
 359     case '_':
 360         return *ctx->ptr++;
 361     case '(':
 362         /* NOTE:
 363          * We resolve empty brackets in lexer instead of parser to avoid complex conflicts
 364          * in call statement special case |f()| without 'call' keyword
 365          */
 366         ctx->ptr++;
 367         skip_spaces(ctx);
 368         if(*ctx->ptr == ')') {
 369             ctx->ptr++;
 370             return tEMPTYBRACKETS;
 371         }
 372         return '(';
 373     case '"':
 374         return parse_string_literal(ctx, lval);
 375     case '&':
 376         if(*++ctx->ptr == 'h' || *ctx->ptr == 'H')
 377             return parse_hex_literal(ctx, lval);
 378         return '&';
 379     case '<':
 380         switch(*++ctx->ptr) {
 381         case '>':
 382             ctx->ptr++;
 383             return tNEQ;
 384         case '=':
 385             ctx->ptr++;
 386             return tLTEQ;
 387         }
 388         return '<';
 389     case '>':
 390         if(*++ctx->ptr == '=') {
 391             ctx->ptr++;
 392             return tGTEQ;
 393         }
 394         return '>';
 395     default:
 396         FIXME("Unhandled char %c in %s\n", *ctx->ptr, debugstr_w(ctx->ptr));
 397     }
 398
 399     return 0;
 400 }
 401
 402 int parser_lex(void *lval, parser_ctx_t *ctx)
 403 {
 404     int ret;
 405
 406     while(1) {
 407         ret = parse_next_token(lval, ctx);
 408         if(ret == '_') {
 409             skip_spaces(ctx);
 410             if(*ctx->ptr != '\n') {
 411                 FIXME("'_' not followed by newline\n");
 412                 return 0;
 413             }
 414             ctx->ptr++;
 415             continue;
 416         }
 417         if(ret != tNL || ctx->last_token != tNL)
 418             break;
 419
 420         ctx->last_nl = ctx->ptr-ctx->code;
 421     }
 422
 423     return (ctx->last_token = ret);
 424 }