Store locations in AST so that error reporting does not depend on parser.

This allows us to delay things that should be delayed (like expanding
regexps with respect to current encoding), but which may throw an error
and need location info.

diff --git a/re2c/bootstrap/src/parse/lex.cc b/re2c/bootstrap/src/parse/lex.cc
index 0c987e425fec86d9d21ded8764b02e4078b6fd5f..a368cc93d03b0680598ebd8c687c7430d3d0164d 100644 (file)
--- a/re2c/bootstrap/src/parse/lex.cc
+++ b/re2c/bootstrap/src/parse/lex.cc
@@ -1,4 +1,4 @@
-/* Generated by re2c 0.16 on Fri Mar  3 02:04:11 2017 */
+/* Generated by re2c 0.16 on Sat Mar  4 16:56:34 2017 */
 #line 1 "../src/parse/lex.re"
 #include "src/util/c99_stdint.h"
 #include <stddef.h>
@@ -983,7 +983,7 @@ start:
 yy159:
        ++YYCURSOR;
 yy160:
-#line 376 "../src/parse/lex.re"
+#line 380 "../src/parse/lex.re"
        {
                                        fatalf("unexpected character: '%c'", *tok);
                                        goto scan;
@@ -996,7 +996,7 @@ yy161:
        if (yybm[0+yych] & 16) {
                goto yy161;
        }
-#line 360 "../src/parse/lex.re"
+#line 364 "../src/parse/lex.re"
        {
                                        goto scan;
                                }
@@ -1011,7 +1011,7 @@ yy164:
                if (yych == '#') goto yy190;
        }
 yy165:
-#line 369 "../src/parse/lex.re"
+#line 373 "../src/parse/lex.re"
        {
                                        if (cur == eof) return 0;
                                        pos = cur;
@@ -1049,9 +1049,9 @@ yy174:
        goto yy168;
 yy175:
        ++YYCURSOR;
-#line 355 "../src/parse/lex.re"
+#line 359 "../src/parse/lex.re"
        {
-                                       yylval.regexp = mkDot();
+                                       yylval.regexp = RegExp::make_dot(cline, get_column(), opts, warn);
                                        return TOKEN_REGEXP;
                                }
 #line 1058 "src/parse/lex.cc"
@@ -1214,14 +1214,18 @@ yy198:
                                                const RegExp *r = NULL;
                                                const bool casing = opts->bCaseInsensitive || opts->bCaseInverted;
                                                for (char *s = tok; s < cur; ++s) {
-                                                       const uint32_t c = static_cast<uint8_t>(*s);
-                                                       r = doCat(r, casing ? ichr(c) : schr(c));
+                                                       const uint32_t
+                                                               c = static_cast<uint8_t>(*s),
+                                                               column = static_cast<uint32_t>(s - pos);
+                                                       r = RegExp::make_cat(r, casing
+                                                               ? RegExp::make_ichar(cline, column, c, opts)
+                                                               : RegExp::make_schar(cline, column, c, opts));
                                                }
-                                               yylval.regexp = r ? r : RegExp::make_nil();
+                                               yylval.regexp = r ? r : RegExp::make_nil(cline, get_column());
                                                return TOKEN_REGEXP;
                                        }
                                }
-#line 1225 "src/parse/lex.cc"
+#line 1229 "src/parse/lex.cc"
 yy200:
        ++YYCURSOR;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
@@ -1246,7 +1250,7 @@ yy202:
                                        yylval.str = new std::string (tok, tok_len ());
                                        return TOKEN_ID;
                                }
-#line 1250 "src/parse/lex.cc"
+#line 1254 "src/parse/lex.cc"
 yy204:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '>') goto yy221;
@@ -1257,7 +1261,7 @@ yy204:
                                        code_line = cline;
                                        goto code;
                                }
-#line 1261 "src/parse/lex.cc"
+#line 1265 "src/parse/lex.cc"
 yy206:
        ++YYCURSOR;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
@@ -1281,15 +1285,15 @@ yy208:
                        fatal("tags are only allowed with '-T, --tags' option");
                }
                const std::string *name = new std::string(tok + 1, tok_len() - 1);
-               yylval.regexp = RegExp::make_tag(name);
+               yylval.regexp = RegExp::make_tag(cline, get_column(), name);
                return TOKEN_REGEXP;
        }
-#line 1288 "src/parse/lex.cc"
+#line 1292 "src/parse/lex.cc"
 yy209:
        ++YYCURSOR;
 #line 263 "../src/parse/lex.re"
        { yylval.regexp = lex_cls(true);  return TOKEN_REGEXP; }
-#line 1293 "src/parse/lex.cc"
+#line 1297 "src/parse/lex.cc"
 yy211:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '2') goto yy223;
@@ -1301,7 +1305,7 @@ yy213:
        {
                                        fatal("illegal closure form, use '{n}', '{n,}', '{n,m}' where n and m are numbers");
                                }
-#line 1305 "src/parse/lex.cc"
+#line 1309 "src/parse/lex.cc"
 yy214:
        ++YYCURSOR;
        if ((YYLIMIT - YYCURSOR) < 2) YYFILL(2);
@@ -1367,7 +1371,7 @@ yy219:
                                                return TOKEN_ID;
                                        }
                                }
-#line 1371 "src/parse/lex.cc"
+#line 1375 "src/parse/lex.cc"
 yy221:
        ++YYCURSOR;
        YYCURSOR -= 2;
@@ -1375,7 +1379,7 @@ yy221:
        {
                                        return *tok;
                                }
-#line 1379 "src/parse/lex.cc"
+#line 1383 "src/parse/lex.cc"
 yy223:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == 'c') goto yy230;
@@ -1398,7 +1402,7 @@ yy225:
                                        yylval.extop.max = yylval.extop.min;
                                        return TOKEN_CLOSESIZE;
                                }
-#line 1402 "src/parse/lex.cc"
+#line 1406 "src/parse/lex.cc"
 yy227:
        ++YYCURSOR;
 #line 311 "../src/parse/lex.re"
@@ -1409,7 +1413,7 @@ yy227:
                                        yylval.str = new std::string (tok + 1, tok_len () - 2); // -2 to omit braces
                                        return TOKEN_ID;
                                }
-#line 1413 "src/parse/lex.cc"
+#line 1417 "src/parse/lex.cc"
 yy229:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == 'n') goto yy235;
@@ -1437,7 +1441,7 @@ yy233:
                                        yylval.extop.max = std::numeric_limits<uint32_t>::max();
                                        return TOKEN_CLOSESIZE;
                                }
-#line 1441 "src/parse/lex.cc"
+#line 1445 "src/parse/lex.cc"
 yy235:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == 'e') goto yy240;
@@ -1446,7 +1450,7 @@ yy236:
        ++YYCURSOR;
 #line 319 "../src/parse/lex.re"
        { lex_conf (); return TOKEN_CONF; }
-#line 1450 "src/parse/lex.cc"
+#line 1454 "src/parse/lex.cc"
 yy238:
        ++YYCURSOR;
 #line 285 "../src/parse/lex.re"
@@ -1462,7 +1466,7 @@ yy238:
                                        }
                                        return TOKEN_CLOSESIZE;
                                }
-#line 1466 "src/parse/lex.cc"
+#line 1470 "src/parse/lex.cc"
 yy240:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= '0') goto yy242;
@@ -1518,12 +1522,12 @@ yy245:
 yy247:
        ++YYCURSOR;
        YYCURSOR = YYCTXMARKER;
-#line 364 "../src/parse/lex.re"
+#line 368 "../src/parse/lex.re"
        {
                                        set_sourceline ();
                                        goto scan;
                                }
-#line 1527 "src/parse/lex.cc"
+#line 1531 "src/parse/lex.cc"
 yy249:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '\n') goto yy247;
@@ -1548,12 +1552,12 @@ yy253:
        if (yych == '\n') goto yy189;
        goto yy250;
 }
-#line 380 "../src/parse/lex.re"
+#line 384 "../src/parse/lex.re"
 
 
 flex_name:
 
-#line 1557 "src/parse/lex.cc"
+#line 1561 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        if ((YYLIMIT - YYCURSOR) < 2) YYFILL(2);
@@ -1562,32 +1566,32 @@ flex_name:
        if (yych == '\r') goto yy260;
        ++YYCURSOR;
 yy257:
-#line 391 "../src/parse/lex.re"
+#line 395 "../src/parse/lex.re"
        {
                YYCURSOR = tok;
                goto start;
        }
-#line 1571 "src/parse/lex.cc"
+#line 1575 "src/parse/lex.cc"
 yy258:
        ++YYCURSOR;
-#line 385 "../src/parse/lex.re"
+#line 389 "../src/parse/lex.re"
        {
                YYCURSOR = tok;
                lexer_state = LEX_NORMAL;
                return TOKEN_FID_END;
        }
-#line 1580 "src/parse/lex.cc"
+#line 1584 "src/parse/lex.cc"
 yy260:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '\n') goto yy258;
        goto yy257;
 }
-#line 395 "../src/parse/lex.re"
+#line 399 "../src/parse/lex.re"
 
 
 code:
 
-#line 1591 "src/parse/lex.cc"
+#line 1595 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        unsigned int yyaccept = 0;
@@ -1648,7 +1652,7 @@ code:
        }
 yy263:
        ++YYCURSOR;
-#line 458 "../src/parse/lex.re"
+#line 462 "../src/parse/lex.re"
        {
                                        if (cur == eof)
                                        {
@@ -1660,15 +1664,15 @@ yy263:
                                        }
                                        goto code;
                                }
-#line 1664 "src/parse/lex.cc"
+#line 1668 "src/parse/lex.cc"
 yy265:
        ++YYCURSOR;
 yy266:
-#line 472 "../src/parse/lex.re"
+#line 476 "../src/parse/lex.re"
        {
                                        goto code;
                                }
-#line 1672 "src/parse/lex.cc"
+#line 1676 "src/parse/lex.cc"
 yy267:
        yyaccept = 0;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -1687,7 +1691,7 @@ yy267:
                }
        }
 yy268:
-#line 439 "../src/parse/lex.re"
+#line 443 "../src/parse/lex.re"
        {
                                        if (depth == 0)
                                        {
@@ -1707,7 +1711,7 @@ yy268:
                                        cline++;
                                        goto code;
                                }
-#line 1711 "src/parse/lex.cc"
+#line 1715 "src/parse/lex.cc"
 yy269:
        yyaccept = 1;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -1720,7 +1724,7 @@ yy270:
        goto yy287;
 yy271:
        ++YYCURSOR;
-#line 411 "../src/parse/lex.re"
+#line 415 "../src/parse/lex.re"
        {
                                        if (depth == 0)
                                        {
@@ -1732,10 +1736,10 @@ yy271:
                                        }
                                        goto code;
                                }
-#line 1736 "src/parse/lex.cc"
+#line 1740 "src/parse/lex.cc"
 yy273:
        ++YYCURSOR;
-#line 399 "../src/parse/lex.re"
+#line 403 "../src/parse/lex.re"
        {
                                        if (depth == 0)
                                        {
@@ -1748,7 +1752,7 @@ yy273:
                                        }
                                        goto code;
                                }
-#line 1752 "src/parse/lex.cc"
+#line 1756 "src/parse/lex.cc"
 yy275:
        yyaccept = 2;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -1760,7 +1764,7 @@ yy275:
        }
 yy276:
        YYCURSOR -= 1;
-#line 426 "../src/parse/lex.re"
+#line 430 "../src/parse/lex.re"
        {
                                        if (depth == 0)
                                        {
@@ -1774,7 +1778,7 @@ yy276:
                                        cline++;
                                        goto code;
                                }
-#line 1778 "src/parse/lex.cc"
+#line 1782 "src/parse/lex.cc"
 yy277:
        ++YYCURSOR;
        goto yy276;
@@ -1809,11 +1813,11 @@ yy282:
        if (yych >= '#') goto yy285;
 yy283:
        ++YYCURSOR;
-#line 469 "../src/parse/lex.re"
+#line 473 "../src/parse/lex.re"
        {
                                        goto code;
                                }
-#line 1817 "src/parse/lex.cc"
+#line 1821 "src/parse/lex.cc"
 yy285:
        ++YYCURSOR;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
@@ -1903,12 +1907,12 @@ yy299:
 yy301:
        ++YYCURSOR;
        YYCURSOR = YYCTXMARKER;
-#line 422 "../src/parse/lex.re"
+#line 426 "../src/parse/lex.re"
        {
                                        set_sourceline ();
                                        goto code;
                                }
-#line 1912 "src/parse/lex.cc"
+#line 1916 "src/parse/lex.cc"
 yy303:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '\n') goto yy301;
@@ -1937,12 +1941,12 @@ yy307:
        if (yych == '\n') goto yy280;
        goto yy304;
 }
-#line 475 "../src/parse/lex.re"
+#line 479 "../src/parse/lex.re"
 
 
 comment:
 
-#line 1946 "src/parse/lex.cc"
+#line 1950 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        static const unsigned char yybm[] = {
@@ -1989,7 +1993,7 @@ comment:
        }
        ++YYCURSOR;
 yy311:
-#line 507 "../src/parse/lex.re"
+#line 511 "../src/parse/lex.re"
        {
                                        if (cur == eof)
                                        {
@@ -1997,7 +2001,7 @@ yy311:
                                        }
                                        goto comment;
                                }
-#line 2001 "src/parse/lex.cc"
+#line 2005 "src/parse/lex.cc"
 yy312:
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
        if (yybm[0+yych] & 32) {
@@ -2005,7 +2009,7 @@ yy312:
        }
        if (yych == '#') goto yy319;
 yy313:
-#line 498 "../src/parse/lex.re"
+#line 502 "../src/parse/lex.re"
        {
                                        if (cur == eof)
                                        {
@@ -2015,7 +2019,7 @@ yy313:
                                        cline++;
                                        goto comment;
                                }
-#line 2019 "src/parse/lex.cc"
+#line 2023 "src/parse/lex.cc"
 yy314:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '/') goto yy321;
@@ -2049,7 +2053,7 @@ yy319:
        }
 yy321:
        ++YYCURSOR;
-#line 479 "../src/parse/lex.re"
+#line 483 "../src/parse/lex.re"
        {
                                        if (--depth == 0)
                                        {
@@ -2060,16 +2064,16 @@ yy321:
                                                goto comment;
                                        }
                                }
-#line 2064 "src/parse/lex.cc"
+#line 2068 "src/parse/lex.cc"
 yy323:
        ++YYCURSOR;
-#line 489 "../src/parse/lex.re"
+#line 493 "../src/parse/lex.re"
        {
                                        ++depth;
                                        fatal("ambiguous /* found");
                                        goto comment;
                                }
-#line 2073 "src/parse/lex.cc"
+#line 2077 "src/parse/lex.cc"
 yy325:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych != 'i') goto yy318;
@@ -2126,12 +2130,12 @@ yy333:
 yy335:
        ++YYCURSOR;
        YYCURSOR = YYCTXMARKER;
-#line 494 "../src/parse/lex.re"
+#line 498 "../src/parse/lex.re"
        {
                                        set_sourceline ();
                                        goto comment;
                                }
-#line 2135 "src/parse/lex.cc"
+#line 2139 "src/parse/lex.cc"
 yy337:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych == '\n') goto yy335;
@@ -2156,28 +2160,28 @@ yy341:
        if (yych == '\n') goto yy318;
        goto yy338;
 }
-#line 514 "../src/parse/lex.re"
+#line 518 "../src/parse/lex.re"
 
 
 nextLine:
 
-#line 2165 "src/parse/lex.cc"
+#line 2169 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
        yych = (YYCTYPE)*YYCURSOR;
        if (yych == '\n') goto yy346;
        ++YYCURSOR;
-#line 525 "../src/parse/lex.re"
+#line 529 "../src/parse/lex.re"
        {  if(cur == eof) {
                   return 0;
                }
                goto nextLine;
             }
-#line 2178 "src/parse/lex.cc"
+#line 2182 "src/parse/lex.cc"
 yy346:
        ++YYCURSOR;
-#line 518 "../src/parse/lex.re"
+#line 522 "../src/parse/lex.re"
        { if(cur == eof) {
                   return 0;
                }
@@ -2185,9 +2189,9 @@ yy346:
                cline++;
                goto scan;
             }
-#line 2189 "src/parse/lex.cc"
+#line 2193 "src/parse/lex.cc"
 }
-#line 530 "../src/parse/lex.re"
+#line 534 "../src/parse/lex.re"
 
 }
 
@@ -2207,39 +2211,40 @@ static void escape (std::string & dest, const std::string & src)
 
 const RegExp *Scanner::lex_cls(bool neg)
 {
+       const uint32_t column = get_column();
        Range *r = NULL, *s;
        uint32_t u, l;
 fst:
        
-#line 2215 "src/parse/lex.cc"
+#line 2220 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
        yych = (YYCTYPE)*YYCURSOR;
        if (yych == ']') goto yy351;
-#line 554 "../src/parse/lex.re"
+#line 559 "../src/parse/lex.re"
        { l = lex_cls_chr(); goto snd; }
-#line 2223 "src/parse/lex.cc"
+#line 2228 "src/parse/lex.cc"
 yy351:
        ++YYCURSOR;
-#line 553 "../src/parse/lex.re"
+#line 558 "../src/parse/lex.re"
        { goto end; }
-#line 2228 "src/parse/lex.cc"
+#line 2233 "src/parse/lex.cc"
 }
-#line 555 "../src/parse/lex.re"
+#line 560 "../src/parse/lex.re"
 
 snd:
        
-#line 2234 "src/parse/lex.cc"
+#line 2239 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        if ((YYLIMIT - YYCURSOR) < 2) YYFILL(2);
        yych = (YYCTYPE)*(YYMARKER = YYCURSOR);
        if (yych == '-') goto yy356;
 yy355:
-#line 558 "../src/parse/lex.re"
+#line 563 "../src/parse/lex.re"
        { u = l; goto add; }
-#line 2243 "src/parse/lex.cc"
+#line 2248 "src/parse/lex.cc"
 yy356:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych != ']') goto yy358;
@@ -2248,7 +2253,7 @@ yy356:
 yy358:
        ++YYCURSOR;
        YYCURSOR -= 1;
-#line 559 "../src/parse/lex.re"
+#line 564 "../src/parse/lex.re"
        {
                        u = lex_cls_chr();
                        if (l > u) {
@@ -2257,9 +2262,9 @@ yy358:
                        }
                        goto add;
                }
-#line 2261 "src/parse/lex.cc"
+#line 2266 "src/parse/lex.cc"
 }
-#line 567 "../src/parse/lex.re"
+#line 572 "../src/parse/lex.re"
 
 add:
        if (!(s = opts->encoding.encodeRange(l, u))) {
@@ -2271,14 +2276,14 @@ end:
        if (neg) {
                r = Range::sub(opts->encoding.fullRange(), r);
        }
-       return cls(r);
+       return RegExp::make_class(cline, column, r, opts, warn);
 }
 
 uint32_t Scanner::lex_cls_chr()
 {
        tok = cur;
        
-#line 2282 "src/parse/lex.cc"
+#line 2287 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        unsigned int yyaccept = 0;
@@ -2287,14 +2292,14 @@ uint32_t Scanner::lex_cls_chr()
        if (yych == '\n') goto yy364;
        if (yych == '\\') goto yy366;
        ++YYCURSOR;
-#line 590 "../src/parse/lex.re"
+#line 595 "../src/parse/lex.re"
        { return static_cast<uint8_t>(tok[0]); }
-#line 2293 "src/parse/lex.cc"
+#line 2298 "src/parse/lex.cc"
 yy364:
        ++YYCURSOR;
-#line 585 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error"); }
-#line 2298 "src/parse/lex.cc"
+#line 590 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error"); }
+#line 2303 "src/parse/lex.cc"
 yy366:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= '`') {
@@ -2343,31 +2348,31 @@ yy366:
                        }
                }
        }
-#line 588 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error in escape sequence"); }
-#line 2349 "src/parse/lex.cc"
+#line 593 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error in escape sequence"); }
+#line 2354 "src/parse/lex.cc"
 yy368:
        ++YYCURSOR;
-#line 603 "../src/parse/lex.re"
+#line 608 "../src/parse/lex.re"
        {
-                       warn.useless_escape(cline, tok - pos, tok[1]);
+                       warn.useless_escape(cline, get_column(), tok[1]);
                        return static_cast<uint8_t>(tok[1]);
                }
-#line 2357 "src/parse/lex.cc"
+#line 2362 "src/parse/lex.cc"
 yy370:
        ++YYCURSOR;
-#line 601 "../src/parse/lex.re"
+#line 606 "../src/parse/lex.re"
        { return static_cast<uint8_t>('-'); }
-#line 2362 "src/parse/lex.cc"
+#line 2367 "src/parse/lex.cc"
 yy372:
        yyaccept = 0;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
        if (yych <= '/') goto yy373;
        if (yych <= '7') goto yy397;
 yy373:
-#line 587 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error in octal escape sequence"); }
-#line 2371 "src/parse/lex.cc"
+#line 592 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error in octal escape sequence"); }
+#line 2376 "src/parse/lex.cc"
 yy374:
        ++YYCURSOR;
        goto yy373;
@@ -2383,9 +2388,9 @@ yy375:
                if (yych <= 'f') goto yy399;
        }
 yy376:
-#line 586 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error in hexadecimal escape sequence"); }
-#line 2389 "src/parse/lex.cc"
+#line 591 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error in hexadecimal escape sequence"); }
+#line 2394 "src/parse/lex.cc"
 yy377:
        yyaccept = 1;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -2401,49 +2406,49 @@ yy377:
        }
 yy378:
        ++YYCURSOR;
-#line 600 "../src/parse/lex.re"
+#line 605 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\\'); }
-#line 2407 "src/parse/lex.cc"
+#line 2412 "src/parse/lex.cc"
 yy380:
        ++YYCURSOR;
-#line 602 "../src/parse/lex.re"
+#line 607 "../src/parse/lex.re"
        { return static_cast<uint8_t>(']'); }
-#line 2412 "src/parse/lex.cc"
+#line 2417 "src/parse/lex.cc"
 yy382:
        ++YYCURSOR;
-#line 593 "../src/parse/lex.re"
+#line 598 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\a'); }
-#line 2417 "src/parse/lex.cc"
+#line 2422 "src/parse/lex.cc"
 yy384:
        ++YYCURSOR;
-#line 594 "../src/parse/lex.re"
+#line 599 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\b'); }
-#line 2422 "src/parse/lex.cc"
+#line 2427 "src/parse/lex.cc"
 yy386:
        ++YYCURSOR;
-#line 595 "../src/parse/lex.re"
+#line 600 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\f'); }
-#line 2427 "src/parse/lex.cc"
+#line 2432 "src/parse/lex.cc"
 yy388:
        ++YYCURSOR;
-#line 596 "../src/parse/lex.re"
+#line 601 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\n'); }
-#line 2432 "src/parse/lex.cc"
+#line 2437 "src/parse/lex.cc"
 yy390:
        ++YYCURSOR;
-#line 597 "../src/parse/lex.re"
+#line 602 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\r'); }
-#line 2437 "src/parse/lex.cc"
+#line 2442 "src/parse/lex.cc"
 yy392:
        ++YYCURSOR;
-#line 598 "../src/parse/lex.re"
+#line 603 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\t'); }
-#line 2442 "src/parse/lex.cc"
+#line 2447 "src/parse/lex.cc"
 yy394:
        ++YYCURSOR;
-#line 599 "../src/parse/lex.re"
+#line 604 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\v'); }
-#line 2447 "src/parse/lex.cc"
+#line 2452 "src/parse/lex.cc"
 yy396:
        yyaccept = 1;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -2506,9 +2511,9 @@ yy401:
        }
 yy402:
        ++YYCURSOR;
-#line 592 "../src/parse/lex.re"
+#line 597 "../src/parse/lex.re"
        { return unesc_oct(tok, cur); }
-#line 2512 "src/parse/lex.cc"
+#line 2517 "src/parse/lex.cc"
 yy404:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= '@') {
@@ -2535,9 +2540,9 @@ yy405:
        }
 yy406:
        ++YYCURSOR;
-#line 591 "../src/parse/lex.re"
+#line 596 "../src/parse/lex.re"
        { return unesc_hex(tok, cur); }
-#line 2541 "src/parse/lex.cc"
+#line 2546 "src/parse/lex.cc"
 yy408:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= '@') {
@@ -2561,7 +2566,7 @@ yy409:
                goto yy398;
        }
 }
-#line 607 "../src/parse/lex.re"
+#line 612 "../src/parse/lex.re"
 
 }
 
@@ -2570,7 +2575,7 @@ uint32_t Scanner::lex_str_chr(char quote, bool &end)
        end = false;
        tok = cur;
        
-#line 2574 "src/parse/lex.cc"
+#line 2579 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        unsigned int yyaccept = 0;
@@ -2579,17 +2584,17 @@ uint32_t Scanner::lex_str_chr(char quote, bool &end)
        if (yych == '\n') goto yy414;
        if (yych == '\\') goto yy416;
        ++YYCURSOR;
-#line 620 "../src/parse/lex.re"
+#line 625 "../src/parse/lex.re"
        {
                        end = tok[0] == quote;
                        return static_cast<uint8_t>(tok[0]);
                }
-#line 2588 "src/parse/lex.cc"
+#line 2593 "src/parse/lex.cc"
 yy414:
        ++YYCURSOR;
-#line 615 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error"); }
-#line 2593 "src/parse/lex.cc"
+#line 620 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error"); }
+#line 2598 "src/parse/lex.cc"
 yy416:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= 'a') {
@@ -2635,28 +2640,28 @@ yy416:
                        }
                }
        }
-#line 618 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error in escape sequence"); }
-#line 2641 "src/parse/lex.cc"
+#line 623 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error in escape sequence"); }
+#line 2646 "src/parse/lex.cc"
 yy418:
        ++YYCURSOR;
-#line 634 "../src/parse/lex.re"
+#line 639 "../src/parse/lex.re"
        {
                        if (tok[1] != quote) {
-                               warn.useless_escape(cline, tok - pos, tok[1]);
+                               warn.useless_escape(cline, get_column(), tok[1]);
                        }
                        return static_cast<uint8_t>(tok[1]);
                }
-#line 2651 "src/parse/lex.cc"
+#line 2656 "src/parse/lex.cc"
 yy420:
        yyaccept = 0;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
        if (yych <= '/') goto yy421;
        if (yych <= '7') goto yy443;
 yy421:
-#line 617 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error in octal escape sequence"); }
-#line 2660 "src/parse/lex.cc"
+#line 622 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error in octal escape sequence"); }
+#line 2665 "src/parse/lex.cc"
 yy422:
        ++YYCURSOR;
        goto yy421;
@@ -2672,9 +2677,9 @@ yy423:
                if (yych <= 'f') goto yy445;
        }
 yy424:
-#line 616 "../src/parse/lex.re"
-       { fatal ((tok - pos) - tchar, "syntax error in hexadecimal escape sequence"); }
-#line 2678 "src/parse/lex.cc"
+#line 621 "../src/parse/lex.re"
+       { fatal (get_column() - tchar, "syntax error in hexadecimal escape sequence"); }
+#line 2683 "src/parse/lex.cc"
 yy425:
        yyaccept = 1;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -2690,44 +2695,44 @@ yy425:
        }
 yy426:
        ++YYCURSOR;
-#line 633 "../src/parse/lex.re"
+#line 638 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\\'); }
-#line 2696 "src/parse/lex.cc"
+#line 2701 "src/parse/lex.cc"
 yy428:
        ++YYCURSOR;
-#line 626 "../src/parse/lex.re"
+#line 631 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\a'); }
-#line 2701 "src/parse/lex.cc"
+#line 2706 "src/parse/lex.cc"
 yy430:
        ++YYCURSOR;
-#line 627 "../src/parse/lex.re"
+#line 632 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\b'); }
-#line 2706 "src/parse/lex.cc"
+#line 2711 "src/parse/lex.cc"
 yy432:
        ++YYCURSOR;
-#line 628 "../src/parse/lex.re"
+#line 633 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\f'); }
-#line 2711 "src/parse/lex.cc"
+#line 2716 "src/parse/lex.cc"
 yy434:
        ++YYCURSOR;
-#line 629 "../src/parse/lex.re"
+#line 634 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\n'); }
-#line 2716 "src/parse/lex.cc"
+#line 2721 "src/parse/lex.cc"
 yy436:
        ++YYCURSOR;
-#line 630 "../src/parse/lex.re"
+#line 635 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\r'); }
-#line 2721 "src/parse/lex.cc"
+#line 2726 "src/parse/lex.cc"
 yy438:
        ++YYCURSOR;
-#line 631 "../src/parse/lex.re"
+#line 636 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\t'); }
-#line 2726 "src/parse/lex.cc"
+#line 2731 "src/parse/lex.cc"
 yy440:
        ++YYCURSOR;
-#line 632 "../src/parse/lex.re"
+#line 637 "../src/parse/lex.re"
        { return static_cast<uint8_t>('\v'); }
-#line 2731 "src/parse/lex.cc"
+#line 2736 "src/parse/lex.cc"
 yy442:
        yyaccept = 1;
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
@@ -2790,9 +2795,9 @@ yy447:
        }
 yy448:
        ++YYCURSOR;
-#line 625 "../src/parse/lex.re"
+#line 630 "../src/parse/lex.re"
        { return unesc_oct(tok, cur); }
-#line 2796 "src/parse/lex.cc"
+#line 2801 "src/parse/lex.cc"
 yy450:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= '@') {
@@ -2819,9 +2824,9 @@ yy451:
        }
 yy452:
        ++YYCURSOR;
-#line 624 "../src/parse/lex.re"
+#line 629 "../src/parse/lex.re"
        { return unesc_hex(tok, cur); }
-#line 2825 "src/parse/lex.cc"
+#line 2830 "src/parse/lex.cc"
 yy454:
        yych = (YYCTYPE)*++YYCURSOR;
        if (yych <= '@') {
@@ -2845,7 +2850,7 @@ yy455:
                goto yy444;
        }
 }
-#line 640 "../src/parse/lex.re"
+#line 645 "../src/parse/lex.re"
 
 }
 
@@ -2855,9 +2860,11 @@ const RegExp *Scanner::lex_str(char quote, bool casing)
        for (bool end;;) {
                const uint32_t c = lex_str_chr(quote, end);
                if (end) {
-                       return r ? r : RegExp::make_nil();
+                       return r ? r : RegExp::make_nil(cline, get_column());
                }
-               r = doCat(r, casing ? ichr(c) : schr(c));
+               r = RegExp::make_cat(r, casing
+                       ? RegExp::make_ichar(cline, get_column(), c, opts)
+                       : RegExp::make_schar(cline, get_column(), c, opts));
        }
 }
 
@@ -2866,7 +2873,7 @@ void Scanner::set_sourceline ()
 sourceline:
        tok = cur;
 
-#line 2870 "src/parse/lex.cc"
+#line 2877 "src/parse/lex.cc"
 {
        YYCTYPE yych;
        static const unsigned char yybm[] = {
@@ -2915,14 +2922,14 @@ sourceline:
 yy458:
        ++YYCURSOR;
 yy459:
-#line 683 "../src/parse/lex.re"
+#line 690 "../src/parse/lex.re"
        {
                                        goto sourceline;
                                }
-#line 2923 "src/parse/lex.cc"
+#line 2930 "src/parse/lex.cc"
 yy460:
        ++YYCURSOR;
-#line 671 "../src/parse/lex.re"
+#line 678 "../src/parse/lex.re"
        {
                                        if (cur == eof)
                                        {
@@ -2935,7 +2942,7 @@ yy460:
                                        tok = cur;
                                        return; 
                                }
-#line 2939 "src/parse/lex.cc"
+#line 2946 "src/parse/lex.cc"
 yy462:
        yych = (YYCTYPE)*(YYMARKER = ++YYCURSOR);
        if (yych == '\n') goto yy459;
@@ -2947,7 +2954,7 @@ yy463:
        if (yybm[0+yych] & 64) {
                goto yy463;
        }
-#line 660 "../src/parse/lex.re"
+#line 667 "../src/parse/lex.re"
        {
                                        if (!s_to_u32_unsafe (tok, cur, cline))
                                        {
@@ -2955,7 +2962,7 @@ yy463:
                                        }
                                        goto sourceline; 
                                }
-#line 2959 "src/parse/lex.cc"
+#line 2966 "src/parse/lex.cc"
 yy466:
        ++YYCURSOR;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
@@ -2972,12 +2979,12 @@ yy468:
        goto yy459;
 yy469:
        ++YYCURSOR;
-#line 667 "../src/parse/lex.re"
+#line 674 "../src/parse/lex.re"
        {
                                        escape (in.file_name, std::string (tok + 1, tok_len () - 2)); // -2 to omit quotes
                                        goto sourceline; 
                                }
-#line 2981 "src/parse/lex.cc"
+#line 2988 "src/parse/lex.cc"
 yy471:
        ++YYCURSOR;
        if (YYLIMIT <= YYCURSOR) YYFILL(1);
@@ -2985,7 +2992,7 @@ yy471:
        if (yych == '\n') goto yy468;
        goto yy466;
 }
-#line 686 "../src/parse/lex.re"
+#line 693 "../src/parse/lex.re"
 
 }
 

diff --git a/re2c/bootstrap/src/parse/lex_conf.cc b/re2c/bootstrap/src/parse/lex_conf.cc
index 4d99ac58261c6b958cad589d94d9bd220ba4d270..f29dfc9ffe639ce31b510c68d4a564d3c20566a3 100644 (file)
--- a/re2c/bootstrap/src/parse/lex_conf.cc
+++ b/re2c/bootstrap/src/parse/lex_conf.cc
@@ -1,4 +1,4 @@
-/* Generated by re2c 0.16 on Fri Mar  3 02:03:51 2017 */
+/* Generated by re2c 0.16 on Sat Mar  4 15:12:07 2017 */
 #line 1 "../src/parse/lex_conf.re"
 #include "src/util/c99_stdint.h"
 #include <string>
@@ -75,7 +75,7 @@ yy2:
        ++cur;
 yy3:
 #line 40 "../src/parse/lex_conf.re"
-       { fatal ((tok - pos) - tchar, "unrecognized configuration"); }
+       { fatal(get_column() - tchar, "unrecognized configuration"); }
 #line 80 "src/parse/lex_conf.cc"
 yy4:
        yyaccept = 0;

diff --git a/re2c/bootstrap/src/parse/parser.cc b/re2c/bootstrap/src/parse/parser.cc
index bc24fc8883a5009f307d3dd9c62ad47968a38dd0..720024821d9db13ea28af56bd82b20d13a7cfe4e 100644 (file)
--- a/re2c/bootstrap/src/parse/parser.cc
+++ b/re2c/bootstrap/src/parse/parser.cc
@@ -206,7 +206,9 @@ static void prepare(specs_t &specs, const Scanner &in)
        // merge default rule with the lowest priority
        for (i = b; i != e; ++i) {
                if (!i->defs.empty()) {
-                       i->rules.push_back(RegExpRule(in.mkDefault(), i->defs[0]));
+                       const Code *c = i->defs[0];
+                       const RegExp *r = RegExp::make_default(c->fline, 0, in.opts);
+                       i->rules.push_back(RegExpRule(r, c));
                }
        }
 
@@ -603,11 +605,11 @@ static const yytype_uint8 yytranslate[] =
   /* YYRLINE[YYN] -- Source line where rule number YYN was defined.  */
 static const yytype_uint16 yyrline[] =
 {
-       0,   199,   199,   201,   202,   203,   207,   214,   219,   222,
-     226,   226,   229,   233,   237,   244,   251,   258,   263,   265,
-     271,   278,   279,   285,   291,   298,   299,   304,   312,   316,
-     323,   327,   334,   338,   345,   346,   352,   357,   358,   362,
-     363,   364,   368,   369,   379
+       0,   201,   201,   203,   204,   205,   209,   216,   221,   224,
+     228,   228,   231,   235,   239,   246,   253,   260,   266,   268,
+     274,   281,   282,   288,   294,   301,   302,   307,   315,   319,
+     326,   330,   337,   341,   348,   349,   355,   360,   361,   365,
+     366,   367,   371,   372,   382
 };
 #endif
 
@@ -1510,7 +1512,8 @@ yyreduce:
   case 17:
 
     {
-               find(context.specs, "0").rules.push_back(RegExpRule(RegExp::make_nil(), (yyvsp[0].code)));
+               const RegExp *r = RegExp::make_nil(context.input.get_cline(), 0);
+               find(context.specs, "0").rules.push_back(RegExpRule(r, (yyvsp[0].code)));
        }
 
     break;
@@ -1567,7 +1570,7 @@ yyreduce:
   case 26:
 
     {
-               (yyval.regexp) = RegExp::make_cat((yyvsp[-2].regexp), RegExp::make_cat(RegExp::make_tag(NULL), (yyvsp[0].regexp)));
+               (yyval.regexp) = RegExp::make_cat((yyvsp[-2].regexp), RegExp::make_cat(RegExp::make_tag(context.input.get_cline(), 0, NULL), (yyvsp[0].regexp)));
        }
 
     break;
@@ -1594,7 +1597,7 @@ yyreduce:
   case 29:
 
     {
-                       (yyval.regexp) = mkAlt((yyvsp[-2].regexp), (yyvsp[0].regexp));
+                       (yyval.regexp) = RegExp::make_alt((yyvsp[-2].regexp), (yyvsp[0].regexp));
                }
 
     break;
@@ -1610,7 +1613,7 @@ yyreduce:
   case 31:
 
     {
-                       (yyval.regexp) = context.input.mkDiff((yyvsp[-2].regexp), (yyvsp[0].regexp));
+                       (yyval.regexp) = RegExp::make_diff((yyvsp[-2].regexp), (yyvsp[0].regexp), context.input.opts, context.input.warn);
                }
 
     break;
@@ -1683,7 +1686,7 @@ yyreduce:
                        context.input.fatal("can't find symbol");
                }
                (yyval.regexp) = i->second;
-               if (context.input.opts->posix_captures && need_wrap((yyval.regexp))) {
+               if (context.input.opts->posix_captures && RegExp::need_wrap((yyval.regexp))) {
                        (yyval.regexp) = RegExp::make_ref((yyval.regexp), *(yyvsp[0].str));
                }
                delete (yyvsp[0].str);

diff --git a/re2c/src/conf/msg.cc b/re2c/src/conf/msg.cc
index c9cbd4e554010abb38919c48afbb66ea80a8e263..367897667dc1c1afca6e2e55e2f20182f675b42f 100644 (file)
--- a/re2c/src/conf/msg.cc
+++ b/re2c/src/conf/msg.cc
@@ -1,5 +1,6 @@
 #include <stdarg.h>
 #include <stdio.h>
+#include <stdlib.h>
 #include <string>
 
 #include "config.h"
@@ -19,6 +20,19 @@ void error (const char * fmt, ...)
        fprintf (stderr, "\n");
 }
 
+void fatal_error(uint32_t line, uint32_t column, const char *fmt, ...)
+{
+       fprintf (stderr, "re2c: error: line %u, column %u: ", line, column);
+
+       va_list args;
+       va_start (args, fmt);
+       vfprintf (stderr, fmt, args);
+       va_end (args);
+
+       fprintf (stderr, "\n");
+       exit(1);
+}
+
 void error_encoding ()
 {
        error ("only one of switches -e, -w, -x, -u and -8 must be set");

diff --git a/re2c/src/conf/msg.h b/re2c/src/conf/msg.h
index b70555239d7e6b877908ce42392dc24873365dfe..5bacc832f8dcf83618f08ed836feb67d8b61f7b4 100644 (file)
--- a/re2c/src/conf/msg.h
+++ b/re2c/src/conf/msg.h
@@ -9,6 +9,7 @@
 namespace re2c {
 
 void error (const char * fmt, ...) RE2C_GXX_ATTRIBUTE ((format (printf, 1, 2)));
+void fatal_error(uint32_t line, uint32_t column, const char *fmt, ...) RE2C_GXX_ATTRIBUTE ((format (printf, 3, 4)));
 void error_encoding ();
 void error_arg (const char * option);
 void warning_start (uint32_t line, bool error);

diff --git a/re2c/src/ir/regexp/encoding/range_suffix.cc b/re2c/src/ir/regexp/encoding/range_suffix.cc
index 70241e360cfddaeec90210dd3807db30064f4ced..6ce03e8b26d88dc0236cb3f70ab78fb700de3c36 100644 (file)
--- a/re2c/src/ir/regexp/encoding/range_suffix.cc
+++ b/re2c/src/ir/regexp/encoding/range_suffix.cc
@@ -4,28 +4,29 @@
 
 namespace re2c {
 
-static const RegExp * emit (RangeSuffix * p, const RegExp * re);
+static const RegExp *emit(uint32_t l, uint32_t c, RangeSuffix * p, const RegExp * re);
 
 free_list<RangeSuffix *> RangeSuffix::freeList;
 
-const RegExp * to_regexp (RangeSuffix * p)
+const RegExp *to_regexp(uint32_t l, uint32_t c, RangeSuffix * p)
 {
-       return p ? emit(p, NULL)
-               : RegExp::make_sym(NULL);
+       return p ? emit(l, c, p, NULL)
+               : RegExp::make_sym(l, c, NULL);
 }
 
 /*
  * Build regexp from suffix tree.
  */
-const RegExp * emit(RangeSuffix * p, const RegExp * re)
+const RegExp *emit(uint32_t l, uint32_t c, RangeSuffix * p, const RegExp * re)
 {
        if (p == NULL) {
                return re;
        } else {
                const RegExp *regexp = NULL;
                for (; p != NULL; p = p->next) {
-                       const RegExp *re1 = doCat(RegExp::make_sym(Range::ran(p->l, p->h + 1)), re);
-                       regexp = doAlt(regexp, emit(p->child, re1));
+                       const RegExp *re1 = RegExp::make_cat(
+                               RegExp::make_sym(l, c, Range::ran(p->l, p->h + 1)), re);
+                       regexp = RegExp::make_alt(regexp, emit(l, c, p->child, re1));
                }
                return regexp;
        }

diff --git a/re2c/src/ir/regexp/encoding/range_suffix.h b/re2c/src/ir/regexp/encoding/range_suffix.h
index 66672d20e6935423ec468a26217fb480b66db802..3ad1fac1629ef3adda5f8e9ee7a34855e704a3e7 100644 (file)
--- a/re2c/src/ir/regexp/encoding/range_suffix.h
+++ b/re2c/src/ir/regexp/encoding/range_suffix.h
@@ -32,7 +32,7 @@ struct RangeSuffix
        FORBID_COPY (RangeSuffix);
 };
 
-const RegExp * to_regexp (RangeSuffix * p);
+const RegExp *to_regexp(uint32_t l, uint32_t c, RangeSuffix * p);
 
 } // namespace re2c
 

diff --git a/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.cc b/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.cc
index 43b46122ca17a957bebe1c5ac7ac9b20d647e1e6..fe80c5ab7805443f715f405970a45f4846b4b08a 100644 (file)
--- a/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.cc
+++ b/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.cc
@@ -8,15 +8,15 @@
 
 namespace re2c {
 
-const RegExp * UTF16Symbol(utf16::rune r)
+const RegExp *UTF16Symbol(uint32_t l, uint32_t c, utf16::rune r)
 {
        if (r <= utf16::MAX_1WORD_RUNE) {
-               return RegExp::make_sym(Range::sym(r));
+               return RegExp::make_sym(l, c, Range::sym(r));
        } else {
                const uint32_t ld = utf16::lead_surr(r);
                const uint32_t tr = utf16::trail_surr(r);
-               return RegExp::make_cat(RegExp::make_sym(Range::sym(ld)),
-                       RegExp::make_sym(Range::sym(tr)));
+               return RegExp::make_cat(RegExp::make_sym(l, c, Range::sym(ld)),
+                       RegExp::make_sym(l, c, Range::sym(tr)));
        }
 }
 
@@ -26,12 +26,12 @@ const RegExp * UTF16Symbol(utf16::rune r)
  * them. We store partially built range in suffix tree, which
  * allows to eliminate common suffixes while building.
  */
-const RegExp * UTF16Range(const Range * r)
+const RegExp *UTF16Range(uint32_t l, uint32_t c, const Range * r)
 {
        RangeSuffix * root = NULL;
        for (; r != NULL; r = r->next ())
                UTF16splitByRuneLength(root, r->lower (), r->upper () - 1);
-       return to_regexp (root);
+       return to_regexp(l, c, root);
 }
 
 } // namespace re2c

diff --git a/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.h b/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.h
index 9d1345c11212c589cd698562fc1887536dcecdec..87dbf8d1074c1ec3ea18fc350e24ce7a56a2bd42 100644 (file)
--- a/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.h
+++ b/re2c/src/ir/regexp/encoding/utf16/utf16_regexp.h
@@ -8,8 +8,8 @@ namespace re2c {
 class Range;
 struct RegExp;
 
-const RegExp * UTF16Symbol(utf16::rune r);
-const RegExp * UTF16Range(const Range * r);
+const RegExp * UTF16Symbol(uint32_t l, uint32_t c, utf16::rune r);
+const RegExp * UTF16Range(uint32_t l, uint32_t c, const Range * r);
 
 } // namespace re2c
 

diff --git a/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.cc b/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.cc
index 7d4ab93e1c41b878b58b7e0d25e9f6f26d0da261..7ac062b2207fc4b6b8034f0d129c383a929d1230 100644 (file)
--- a/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.cc
+++ b/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.cc
@@ -8,13 +8,13 @@
 
 namespace re2c {
 
-const RegExp * UTF8Symbol(utf8::rune r)
+const RegExp * UTF8Symbol(uint32_t l, uint32_t c, utf8::rune r)
 {
        uint32_t chars[utf8::MAX_RUNE_LENGTH];
        const uint32_t chars_count = utf8::rune_to_bytes(chars, r);
-       const RegExp *re = RegExp::make_sym(Range::sym(chars[0]));
+       const RegExp *re = RegExp::make_sym(l, c, Range::sym(chars[0]));
        for (uint32_t i = 1; i < chars_count; ++i) {
-               re = RegExp::make_cat(re, RegExp::make_sym(Range::sym(chars[i])));
+               re = RegExp::make_cat(re, RegExp::make_sym(l, c, Range::sym(chars[i])));
        }
        return re;
 }
@@ -25,12 +25,12 @@ const RegExp * UTF8Symbol(utf8::rune r)
  * them. We store partially built range in suffix tree, which
  * allows to eliminate common suffixes while building.
  */
-const RegExp * UTF8Range(const Range * r)
+const RegExp * UTF8Range(uint32_t l, uint32_t c, const Range * r)
 {
        RangeSuffix * root = NULL;
        for (; r != NULL; r = r->next ())
                UTF8splitByRuneLength(root, r->lower (), r->upper () - 1);
-       return to_regexp (root);
+       return to_regexp (l, c, root);
 }
 
 } // namespace re2c

diff --git a/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.h b/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.h
index 0d327014e39a7fa3a608ab652f41be8f928e5a90..ae638d470b1db405555f2f6ce880bea97f08d73b 100644 (file)
--- a/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.h
+++ b/re2c/src/ir/regexp/encoding/utf8/utf8_regexp.h
@@ -8,8 +8,8 @@ namespace re2c {
 class Range;
 struct RegExp;
 
-const RegExp * UTF8Symbol(utf8::rune r);
-const RegExp * UTF8Range(const Range * r);
+const RegExp * UTF8Symbol(uint32_t l, uint32_t c, utf8::rune r);
+const RegExp * UTF8Range(uint32_t l, uint32_t c, const Range * r);
 
 } // namespace re2c
 

diff --git a/re2c/src/ir/regexp/regexp.cc b/re2c/src/ir/regexp/regexp.cc
index 7f306c22d6add7ec331895f2737ab36079ce6115..95b4b231271008fa28d0994b13cee5f944bb109e 100644 (file)
--- a/re2c/src/ir/regexp/regexp.cc
+++ b/re2c/src/ir/regexp/regexp.cc
@@ -1,6 +1,8 @@
 #include <limits>
 #include <stddef.h>
 
+#include "src/conf/msg.h"
+#include "src/conf/opt.h"
 #include "src/ir/regexp/empty_class_policy.h"
 #include "src/ir/regexp/encoding/case.h"
 #include "src/ir/regexp/encoding/enc.h"
@@ -17,102 +19,72 @@ free_list<RegExp*> RegExp::flist;
 
 const uint32_t RegExp::MANY = std::numeric_limits<uint32_t>::max();
 
-const RegExp *doAlt(const RegExp *re1, const RegExp *re2)
-{
-       if (!re1) {
-               return re2;
-       }
-       if (!re2) {
-               return re1;
-       }
-       return RegExp::make_alt(re1, re2);
-}
-
-const RegExp *mkAlt(const RegExp *re1, const RegExp *re2)
-{
-       if (!re1) return re2;
-       if (!re2) return re1;
-       if (re1->type == RegExp::SYM && re2->type == RegExp::SYM) {
-               return RegExp::make_sym(Range::add(re1->sym, re2->sym));
-       }
-       return RegExp::make_alt(re1, re2);
-}
-
-const RegExp *doCat(const RegExp *re1, const RegExp *re2)
-{
-       if (!re1) {
-               return re2;
-       }
-       if (!re2) {
-               return re1;
-       }
-       return RegExp::make_cat(re1, re2);
-}
-
-const RegExp *Scanner::schr(uint32_t c) const
+const RegExp *RegExp::make_schar(uint32_t line, uint32_t column, uint32_t c, Opt &opts)
 {
        if (!opts->encoding.encode(c)) {
-               fatalf("Bad code point: '0x%X'", c);
+               fatal_error(line, column, "bad code point: '0x%X'", c);
        }
        switch (opts->encoding.type ()) {
-               case Enc::UTF16: return UTF16Symbol(c);
-               case Enc::UTF8:  return UTF8Symbol(c);
-               default:         return RegExp::make_sym(Range::sym(c));
+               case Enc::UTF16: return UTF16Symbol(line, column, c);
+               case Enc::UTF8:  return UTF8Symbol(line, column, c);
+               default:         return RegExp::make_sym(line, column, Range::sym(c));
        }
 }
 
-const RegExp *Scanner::ichr(uint32_t c) const
+const RegExp *RegExp::make_ichar(uint32_t line, uint32_t column, uint32_t c, Opt &opts)
 {
        if (is_alpha(c)) {
-               const RegExp *l = schr(to_lower_unsafe(c));
-               const RegExp *u = schr(to_upper_unsafe(c));
-               return mkAlt(l, u);
+               const RegExp *l = RegExp::make_schar(line, column, to_lower_unsafe(c), opts);
+               const RegExp *u = RegExp::make_schar(line, column, to_upper_unsafe(c), opts);
+               return RegExp::make_alt(l, u);
        } else {
-               return schr(c);
+               return RegExp::make_schar(line, column, c, opts);
        }
 }
 
-const RegExp *Scanner::cls(const Range *r) const
+const RegExp *RegExp::make_class(uint32_t line, uint32_t column, const Range *r, Opt &opts, Warn &warn)
 {
        if (!r) {
                switch (opts->empty_class_policy) {
                        case EMPTY_CLASS_MATCH_EMPTY:
-                               warn.empty_class(cline);
-                               return RegExp::make_nil();
+                               warn.empty_class(line);
+                               return RegExp::make_nil(line, column);
                        case EMPTY_CLASS_MATCH_NONE:
-                               warn.empty_class(cline);
+                               warn.empty_class(line);
                                break;
                        case EMPTY_CLASS_ERROR:
-                               fatal("empty character class");
+                               fatal_error(line, column, "empty character class");
+                               break;
                }
        }
 
        switch (opts->encoding.type()) {
-               case Enc::UTF16: return UTF16Range(r);
-               case Enc::UTF8:  return UTF8Range(r);
-               default:         return RegExp::make_sym(r);
+               case Enc::UTF16: return UTF16Range(line, column, r);
+               case Enc::UTF8:  return UTF8Range(line, column, r);
+               default:         return RegExp::make_sym(line, column, r);
        }
 }
 
-const RegExp *Scanner::mkDiff(const RegExp *re1, const RegExp *re2) const
+const RegExp *RegExp::make_diff(const RegExp *re1, const RegExp *re2, Opt &opts, Warn &warn)
 {
        if (re1 && re2
                && re1->type == RegExp::SYM
                && re2->type == RegExp::SYM) {
-               return cls(Range::sub(re1->sym, re2->sym));
+               return RegExp::make_class(re1->line, re1->column,
+                       Range::sub(re1->sym, re2->sym), opts, warn);
        }
-       fatal("can only difference char sets");
+       fatal_error(re1->line, re1->column, "can only difference char sets");
        return NULL;
 }
 
-const RegExp *Scanner::mkDot() const
+const RegExp *RegExp::make_dot(uint32_t line, uint32_t column, Opt &opts, Warn &warn)
 {
        uint32_t c = '\n';
        if (!opts->encoding.encode(c)) {
-               fatalf("Bad code point: '0x%X'", c);
+               fatal_error(line, column, "bad code point: '0x%X'", c);
        }
-       return cls(Range::sub(opts->encoding.fullRange(),
-               Range::sym(c)));
+       return RegExp::make_class(line, column,
+               Range::sub(opts->encoding.fullRange(), Range::sym(c)), opts, warn);
 }
 
 /*
@@ -125,13 +97,13 @@ const RegExp *Scanner::mkDot() const
  * Also note that default range doesn't respect encoding policy
  * (the way invalid code points are treated).
  */
-const RegExp *Scanner::mkDefault() const
+const RegExp *RegExp::make_default(uint32_t line, uint32_t column, Opt &opts)
 {
-       return RegExp::make_sym(Range::ran(0,
+       return RegExp::make_sym(line, column, Range::ran(0,
                opts->encoding.nCodeUnits()));
 }
 
-bool need_wrap(const RegExp *re)
+bool RegExp::need_wrap(const RegExp *re)
 {
        switch (re->type) {
                case RegExp::ITER:

diff --git a/re2c/src/ir/regexp/regexp.h b/re2c/src/ir/regexp/regexp.h
index b1271814d1996a91fc0c999410c74fc914170b3a..bd08bc0315c142693f3075ebcc812a6966f388a2 100644 (file)
--- a/re2c/src/ir/regexp/regexp.h
+++ b/re2c/src/ir/regexp/regexp.h
@@ -14,6 +14,8 @@
 namespace re2c
 {
 
+struct Opt;
+
 struct RegExp
 {
        static free_list<RegExp*> flist;
@@ -47,54 +49,64 @@ struct RegExp
                        const std::string *name;
                } ref;
        };
+       uint32_t line;
+       uint32_t column;
 
-       static const RegExp *make_nil()
+       static const RegExp *make_nil(uint32_t l, uint32_t c)
        {
-               return new RegExp(NIL);
+               return new RegExp(l, c, NIL);
        }
-       static const RegExp *make_sym(const Range *r)
+       static const RegExp *make_sym(uint32_t l, uint32_t c, const Range *r)
        {
-               RegExp *re = new RegExp(SYM);
+               RegExp *re = new RegExp(l, c, SYM);
                re->sym = r;
                return re;
        }
        static const RegExp *make_alt(const RegExp *r1, const RegExp *r2)
        {
-               RegExp *re = new RegExp(ALT);
+               if (!r1) return r2;
+               if (!r2) return r1;
+               if (r1->type == RegExp::SYM && r2->type == RegExp::SYM) {
+                       return RegExp::make_sym(r1->line, r1->column,
+                               Range::add(r1->sym, r2->sym));
+               }
+               RegExp *re = new RegExp(r1->line, r1->column, ALT);
                re->alt.re1 = r1;
                re->alt.re2 = r2;
                return re;
        }
        static const RegExp *make_cat(const RegExp *r1, const RegExp *r2)
        {
-               RegExp *re = new RegExp(CAT);
+               if (!r1) return r2;
+               if (!r2) return r1;
+               RegExp *re = new RegExp(r1->line, r1->column, CAT);
                re->cat.re1 = r1;
                re->cat.re2 = r2;
                return re;
        }
        static const RegExp *make_iter(const RegExp *r, uint32_t n, uint32_t m)
        {
-               RegExp *re = new RegExp(ITER);
+               RegExp *re = new RegExp(r->line, r->column, ITER);
                re->iter.re = r;
                re->iter.min = n;
                re->iter.max = m;
                return re;
        }
-       static const RegExp *make_tag(const std::string *t)
+       static const RegExp *make_tag(uint32_t l, uint32_t c, const std::string *t)
        {
-               RegExp *re = new RegExp(TAG);
+               RegExp *re = new RegExp(l, c, TAG);
                re->tag = t;
                return re;
        }
        static const RegExp *make_cap(const RegExp *r)
        {
-               RegExp *re = new RegExp(CAP);
+               RegExp *re = new RegExp(r->line, r->column, CAP);
                re->cap = r;
                return re;
        }
        static const RegExp *make_ref(const RegExp *r, const std::string &n)
        {
-               RegExp *re = new RegExp(REF);
+               RegExp *re = new RegExp(r->line, r->column, REF);
                re->ref.re = r;
                re->ref.name = new std::string(n);
                return re;
@@ -108,9 +120,17 @@ struct RegExp
                        delete ref.name;
                }
        }
+       static const RegExp *make_schar(uint32_t line, uint32_t column, uint32_t c, Opt &opts);
+       static const RegExp *make_ichar(uint32_t line, uint32_t column, uint32_t c, Opt &opts);
+       static const RegExp *make_class(uint32_t line, uint32_t column, const Range *r, Opt &opts, Warn &warn);
+       static const RegExp *make_diff(const RegExp * e1, const RegExp * e2, Opt &opts, Warn &warn);
+       static const RegExp *make_dot(uint32_t line, uint32_t column, Opt &opts, Warn &warn);
+       static const RegExp *make_default(uint32_t line, uint32_t column, Opt &opts);
+       static bool need_wrap(const RegExp *re);
 
 private:
-       inline RegExp(type_t t) : type(t)
+       inline RegExp(uint32_t l, uint32_t c, type_t t)
+               : type(t), line(l), column(c)
        {
                flist.insert(this);
        }
@@ -127,11 +147,6 @@ struct RegExpRule
        {}
 };
 
-const RegExp *mkAlt(const RegExp *re1, const RegExp *re2);
-const RegExp *doAlt(const RegExp *re1, const RegExp *re2);
-const RegExp *doCat(const RegExp *re1, const RegExp *re2);
-bool need_wrap(const RegExp *re);
-
 } // end namespace re2c
 
 #endif // _RE2C_IR_REGEXP_REGEXP_

diff --git a/re2c/src/parse/lex.re b/re2c/src/parse/lex.re
index 8f1533d1771dce30ba4e14062c56a2334639716d..0c33a3b4f3884a51e6636d923476ed406b5f08ed 100644 (file)
--- a/re2c/src/parse/lex.re
+++ b/re2c/src/parse/lex.re
@@ -267,7 +267,7 @@ start:
                        fatal("tags are only allowed with '-T, --tags' option");
                }
                const std::string *name = new std::string(tok + 1, tok_len() - 1);
-               yylval.regexp = RegExp::make_tag(name);
+               yylval.regexp = RegExp::make_tag(cline, get_column(), name);
                return TOKEN_REGEXP;
        }
 
@@ -344,16 +344,20 @@ start:
                                                const RegExp *r = NULL;
                                                const bool casing = opts->bCaseInsensitive || opts->bCaseInverted;
                                                for (char *s = tok; s < cur; ++s) {
-                                                       const uint32_t c = static_cast<uint8_t>(*s);
-                                                       r = doCat(r, casing ? ichr(c) : schr(c));
+                                                       const uint32_t
+                                                               c = static_cast<uint8_t>(*s),
+                                                               column = static_cast<uint32_t>(s - pos);
+                                                       r = RegExp::make_cat(r, casing
+                                                               ? RegExp::make_ichar(cline, column, c, opts)
+                                                               : RegExp::make_schar(cline, column, c, opts));
                                                }
-                                               yylval.regexp = r ? r : RegExp::make_nil();
+                                               yylval.regexp = r ? r : RegExp::make_nil(cline, get_column());
                                                return TOKEN_REGEXP;
                                        }
                                }
 
        "."                     {
-                                       yylval.regexp = mkDot();
+                                       yylval.regexp = RegExp::make_dot(cline, get_column(), opts, warn);
                                        return TOKEN_REGEXP;
                                }
 
@@ -546,6 +550,7 @@ static void escape (std::string & dest, const std::string & src)
 
 const RegExp *Scanner::lex_cls(bool neg)
 {
+       const uint32_t column = get_column();
        Range *r = NULL, *s;
        uint32_t u, l;
 fst:
@@ -575,17 +580,17 @@ end:
        if (neg) {
                r = Range::sub(opts->encoding.fullRange(), r);
        }
-       return cls(r);
+       return RegExp::make_class(cline, column, r, opts, warn);
 }
 
 uint32_t Scanner::lex_cls_chr()
 {
        tok = cur;
        /*!re2c
-               *          { fatal ((tok - pos) - tchar, "syntax error"); }
-               esc [xXuU] { fatal ((tok - pos) - tchar, "syntax error in hexadecimal escape sequence"); }
-               esc [0-7]  { fatal ((tok - pos) - tchar, "syntax error in octal escape sequence"); }
-               esc        { fatal ((tok - pos) - tchar, "syntax error in escape sequence"); }
+               *          { fatal (get_column() - tchar, "syntax error"); }
+               esc [xXuU] { fatal (get_column() - tchar, "syntax error in hexadecimal escape sequence"); }
+               esc [0-7]  { fatal (get_column() - tchar, "syntax error in octal escape sequence"); }
+               esc        { fatal (get_column() - tchar, "syntax error in escape sequence"); }
 
                . \ esc    { return static_cast<uint8_t>(tok[0]); }
                esc_hex    { return unesc_hex(tok, cur); }
@@ -601,7 +606,7 @@ uint32_t Scanner::lex_cls_chr()
                esc "-"    { return static_cast<uint8_t>('-'); }
                esc "]"    { return static_cast<uint8_t>(']'); }
                esc .      {
-                       warn.useless_escape(cline, tok - pos, tok[1]);
+                       warn.useless_escape(cline, get_column(), tok[1]);
                        return static_cast<uint8_t>(tok[1]);
                }
        */
@@ -612,10 +617,10 @@ uint32_t Scanner::lex_str_chr(char quote, bool &end)
        end = false;
        tok = cur;
        /*!re2c
-               *          { fatal ((tok - pos) - tchar, "syntax error"); }
-               esc [xXuU] { fatal ((tok - pos) - tchar, "syntax error in hexadecimal escape sequence"); }
-               esc [0-7]  { fatal ((tok - pos) - tchar, "syntax error in octal escape sequence"); }
-               esc        { fatal ((tok - pos) - tchar, "syntax error in escape sequence"); }
+               *          { fatal (get_column() - tchar, "syntax error"); }
+               esc [xXuU] { fatal (get_column() - tchar, "syntax error in hexadecimal escape sequence"); }
+               esc [0-7]  { fatal (get_column() - tchar, "syntax error in octal escape sequence"); }
+               esc        { fatal (get_column() - tchar, "syntax error in escape sequence"); }
 
                . \ esc    {
                        end = tok[0] == quote;
@@ -633,7 +638,7 @@ uint32_t Scanner::lex_str_chr(char quote, bool &end)
                esc "\\"   { return static_cast<uint8_t>('\\'); }
                esc .      {
                        if (tok[1] != quote) {
-                               warn.useless_escape(cline, tok - pos, tok[1]);
+                               warn.useless_escape(cline, get_column(), tok[1]);
                        }
                        return static_cast<uint8_t>(tok[1]);
                }
@@ -646,9 +651,11 @@ const RegExp *Scanner::lex_str(char quote, bool casing)
        for (bool end;;) {
                const uint32_t c = lex_str_chr(quote, end);
                if (end) {
-                       return r ? r : RegExp::make_nil();
+                       return r ? r : RegExp::make_nil(cline, get_column());
                }
-               r = doCat(r, casing ? ichr(c) : schr(c));
+               r = RegExp::make_cat(r, casing
+                       ? RegExp::make_ichar(cline, get_column(), c, opts)
+                       : RegExp::make_schar(cline, get_column(), c, opts));
        }
 }
 

diff --git a/re2c/src/parse/lex_conf.re b/re2c/src/parse/lex_conf.re
index 5a5cf27914eb200ef9929ccdfd83c99af02103d5..2e25a5c596070789d92928db49a44930d1839935 100644 (file)
--- a/re2c/src/parse/lex_conf.re
+++ b/re2c/src/parse/lex_conf.re
@@ -37,7 +37,7 @@ void Scanner::lex_conf ()
 {
        tok = cur;
 /*!re2c
-       * { fatal ((tok - pos) - tchar, "unrecognized configuration"); }
+       * { fatal(get_column() - tchar, "unrecognized configuration"); }
 
        "flags:" ("D" | "emit-dot") { opts.set_target(opt_t::DOT);      lex_conf_semicolon(); return; }
        "flags:" ("S" | "skeleton") { opts.set_target(opt_t::SKELETON); lex_conf_semicolon(); return; }

diff --git a/re2c/src/parse/parser.ypp b/re2c/src/parse/parser.ypp
index 82f785c72e7e1ace9fefd3d228543ebe534addc1..4900343f96f37e4e35675cd57cb107c9fbc1317a 100644 (file)
--- a/re2c/src/parse/parser.ypp
+++ b/re2c/src/parse/parser.ypp
@@ -141,7 +141,9 @@ static void prepare(specs_t &specs, const Scanner &in)
        // merge default rule with the lowest priority
        for (i = b; i != e; ++i) {
                if (!i->defs.empty()) {
-                       i->rules.push_back(RegExpRule(in.mkDefault(), i->defs[0]));
+                       const Code *c = i->defs[0];
+                       const RegExp *r = RegExp::make_default(c->fline, 0, in.opts);
+                       i->rules.push_back(RegExpRule(r, c));
                }
        }
 
@@ -256,7 +258,8 @@ rule
        }
 
        | '<' '>' ccode {
-               find(context.specs, "0").rules.push_back(RegExpRule(RegExp::make_nil(), $3));
+               const RegExp *r = RegExp::make_nil(context.input.get_cline(), 0);
+               find(context.specs, "0").rules.push_back(RegExpRule(r, $3));
        };
 
 ccode
@@ -297,7 +300,7 @@ conds
 trailexpr
        : capexpr
        | capexpr '/' expr {
-               $$ = RegExp::make_cat($1, RegExp::make_cat(RegExp::make_tag(NULL), $3));
+               $$ = RegExp::make_cat($1, RegExp::make_cat(RegExp::make_tag(context.input.get_cline(), 0, NULL), $3));
        };
 
 capexpr
@@ -315,7 +318,7 @@ expr:
                }
        |       expr '|' diff
                {
-                       $$ = mkAlt($1, $3);
+                       $$ = RegExp::make_alt($1, $3);
                }
 ;
 
@@ -326,7 +329,7 @@ diff:
                }
        |       diff '\\' term
                {
-                       $$ = context.input.mkDiff($1, $3);
+                       $$ = RegExp::make_diff($1, $3, context.input.opts, context.input.warn);
                }
 ;
 
@@ -372,7 +375,7 @@ primary
                        context.input.fatal("can't find symbol");
                }
                $$ = i->second;
-               if (context.input.opts->posix_captures && need_wrap($$)) {
+               if (context.input.opts->posix_captures && RegExp::need_wrap($$)) {
                        $$ = RegExp::make_ref($$, *$1);
                }
                delete $1;

diff --git a/re2c/src/parse/scanner.h b/re2c/src/parse/scanner.h
index 0fa8024907e389c6b9df8b3e4fbcfc8f6a8eb735..c4610b49b455d1a43e489feeb0e01e2f46ea906f 100644 (file)
--- a/re2c/src/parse/scanner.h
+++ b/re2c/src/parse/scanner.h
@@ -68,10 +68,6 @@ private:
        uint32_t lex_str_chr(char quote, bool &end);
        const RegExp *lex_cls(bool neg);
        const RegExp *lex_str(char quote, bool casing);
-       const RegExp *schr(uint32_t c) const;
-       const RegExp *ichr(uint32_t c) const;
-       const RegExp *cls(const Range *r) const;
-
        void lex_conf ();
        void lex_conf_encoding_policy();
        void lex_conf_input();
@@ -105,17 +101,13 @@ public:
        void restore_state(const ScannerState&);
 
        uint32_t get_cline() const;
+       uint32_t get_column() const;
        const std::string & get_fname () const;
        void fatal_at(uint32_t line, ptrdiff_t ofs, const char *msg) const RE2C_GXX_ATTRIBUTE ((noreturn));
        void fatalf_at(uint32_t line, const char*, ...) const RE2C_GXX_ATTRIBUTE ((format (printf, 3, 4))) RE2C_GXX_ATTRIBUTE ((noreturn));
        void fatalf(const char*, ...) const RE2C_GXX_ATTRIBUTE ((format (printf, 2, 3))) RE2C_GXX_ATTRIBUTE ((noreturn));
        void fatal(const char*) const RE2C_GXX_ATTRIBUTE ((noreturn));
        void fatal(ptrdiff_t, const char*) const RE2C_GXX_ATTRIBUTE ((noreturn));
-
-       const RegExp * mkDiff (const RegExp * e1, const RegExp * e2) const;
-       const RegExp * mkDot () const;
-       const RegExp * mkDefault () const;
-
        FORBID_COPY (Scanner);
 };
 
@@ -135,6 +127,11 @@ inline uint32_t Scanner::get_cline() const
        return cline;
 }
 
+inline uint32_t Scanner::get_column() const
+{
+       return static_cast<uint32_t>(tok - pos);
+}
+
 inline void Scanner::save_state(ScannerState& state) const
 {
        state = *this;

diff --git a/re2c/test/bug61.i--empty-class(error).c b/re2c/test/bug61.i--empty-class(error).c
index 9bce6164d02b74538c28347c533e1efe78caf216..3b7847dee39db06bddf646dfa922a905903d30eb 100644 (file)
--- a/re2c/test/bug61.i--empty-class(error).c
+++ b/re2c/test/bug61.i--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 2, column 5: empty character class
+re2c: error: line 2, column 4: empty character class

diff --git a/re2c/test/bug61.i--empty-class(match-empty).c b/re2c/test/bug61.i--empty-class(match-empty).c
index 5b2fa4a26e7b2f6b8b918e09761740c0f5ec8c69..9cb24cef93c47e12c0c10c9a77368ad58101f62a 100644 (file)
--- a/re2c/test/bug61.i--empty-class(match-empty).c
+++ b/re2c/test/bug61.i--empty-class(match-empty).c
@@ -6,4 +6,4 @@ re2c: warning: line 10: empty character class [-Wempty-character-class]
 re2c: warning: line 10: rule matches empty string [-Wmatch-empty-string]
 re2c: warning: line 14: empty character class [-Wempty-character-class]
 re2c: warning: line 14: empty character class [-Wempty-character-class]
-re2c: error: line 14, column 11: can only difference char sets
+re2c: error: line 14, column 4: can only difference char sets

diff --git a/re2c/test/bug61.i.c b/re2c/test/bug61.i.c
index 5b2fa4a26e7b2f6b8b918e09761740c0f5ec8c69..9cb24cef93c47e12c0c10c9a77368ad58101f62a 100644 (file)
--- a/re2c/test/bug61.i.c
+++ b/re2c/test/bug61.i.c
@@ -6,4 +6,4 @@ re2c: warning: line 10: empty character class [-Wempty-character-class]
 re2c: warning: line 10: rule matches empty string [-Wmatch-empty-string]
 re2c: warning: line 14: empty character class [-Wempty-character-class]
 re2c: warning: line 14: empty character class [-Wempty-character-class]
-re2c: error: line 14, column 11: can only difference char sets
+re2c: error: line 14, column 4: can only difference char sets

diff --git a/re2c/test/bug61_difference_full.i--empty-class(error).c b/re2c/test/bug61_difference_full.i--empty-class(error).c
index 70094ae3842f3d6d930cb74bb87df1a44b194012..3b7847dee39db06bddf646dfa922a905903d30eb 100644 (file)
--- a/re2c/test/bug61_difference_full.i--empty-class(error).c
+++ b/re2c/test/bug61_difference_full.i--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 2, column 29: empty character class
+re2c: error: line 2, column 4: empty character class

diff --git a/re2c/test/bug61_difference_negative.i--empty-class(error).c b/re2c/test/bug61_difference_negative.i--empty-class(error).c
index 9bce6164d02b74538c28347c533e1efe78caf216..3b7847dee39db06bddf646dfa922a905903d30eb 100644 (file)
--- a/re2c/test/bug61_difference_negative.i--empty-class(error).c
+++ b/re2c/test/bug61_difference_negative.i--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 2, column 5: empty character class
+re2c: error: line 2, column 4: empty character class

diff --git a/re2c/test/bug61_difference_negative.i--empty-class(match-empty).c b/re2c/test/bug61_difference_negative.i--empty-class(match-empty).c
index 049b9a7721c013eb00b94ba6541f893017635119..a0aee7284d8909fc59064e6d366ae48c294999df 100644 (file)
--- a/re2c/test/bug61_difference_negative.i--empty-class(match-empty).c
+++ b/re2c/test/bug61_difference_negative.i--empty-class(match-empty).c
@@ -1,3 +1,3 @@
 re2c: warning: line 2: empty character class [-Wempty-character-class]
 re2c: warning: line 2: empty character class [-Wempty-character-class]
-re2c: error: line 2, column 31: can only difference char sets
+re2c: error: line 2, column 4: can only difference char sets

diff --git a/re2c/test/bug61_difference_negative.i.c b/re2c/test/bug61_difference_negative.i.c
index 049b9a7721c013eb00b94ba6541f893017635119..a0aee7284d8909fc59064e6d366ae48c294999df 100644 (file)
--- a/re2c/test/bug61_difference_negative.i.c
+++ b/re2c/test/bug61_difference_negative.i.c
@@ -1,3 +1,3 @@
 re2c: warning: line 2: empty character class [-Wempty-character-class]
 re2c: warning: line 2: empty character class [-Wempty-character-class]
-re2c: error: line 2, column 31: can only difference char sets
+re2c: error: line 2, column 4: can only difference char sets

diff --git a/re2c/test/bug61_difference_positive.i--empty-class(error).c b/re2c/test/bug61_difference_positive.i--empty-class(error).c
index 9bce6164d02b74538c28347c533e1efe78caf216..3b7847dee39db06bddf646dfa922a905903d30eb 100644 (file)
--- a/re2c/test/bug61_difference_positive.i--empty-class(error).c
+++ b/re2c/test/bug61_difference_positive.i--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 2, column 5: empty character class
+re2c: error: line 2, column 4: empty character class

diff --git a/re2c/test/bug61_difference_positive.i--empty-class(match-empty).c b/re2c/test/bug61_difference_positive.i--empty-class(match-empty).c
index 463f74d44aae2b580f38b46ec25920ac882f5de9..a0aee7284d8909fc59064e6d366ae48c294999df 100644 (file)
--- a/re2c/test/bug61_difference_positive.i--empty-class(match-empty).c
+++ b/re2c/test/bug61_difference_positive.i--empty-class(match-empty).c
@@ -1,3 +1,3 @@
 re2c: warning: line 2: empty character class [-Wempty-character-class]
 re2c: warning: line 2: empty character class [-Wempty-character-class]
-re2c: error: line 2, column 11: can only difference char sets
+re2c: error: line 2, column 4: can only difference char sets

diff --git a/re2c/test/bug61_difference_positive.i.c b/re2c/test/bug61_difference_positive.i.c
index 463f74d44aae2b580f38b46ec25920ac882f5de9..a0aee7284d8909fc59064e6d366ae48c294999df 100644 (file)
--- a/re2c/test/bug61_difference_positive.i.c
+++ b/re2c/test/bug61_difference_positive.i.c
@@ -1,3 +1,3 @@
 re2c: warning: line 2: empty character class [-Wempty-character-class]
 re2c: warning: line 2: empty character class [-Wempty-character-class]
-re2c: error: line 2, column 11: can only difference char sets
+re2c: error: line 2, column 4: can only difference char sets

diff --git a/re2c/test/bug61_negative.i--empty-class(error).c b/re2c/test/bug61_negative.i--empty-class(error).c
index 9bce6164d02b74538c28347c533e1efe78caf216..3b7847dee39db06bddf646dfa922a905903d30eb 100644 (file)
--- a/re2c/test/bug61_negative.i--empty-class(error).c
+++ b/re2c/test/bug61_negative.i--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 2, column 5: empty character class
+re2c: error: line 2, column 4: empty character class

diff --git a/re2c/test/bug61_positive.i--empty-class(error).c b/re2c/test/bug61_positive.i--empty-class(error).c
index 9bce6164d02b74538c28347c533e1efe78caf216..3b7847dee39db06bddf646dfa922a905903d30eb 100644 (file)
--- a/re2c/test/bug61_positive.i--empty-class(error).c
+++ b/re2c/test/bug61_positive.i--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 2, column 5: empty character class
+re2c: error: line 2, column 4: empty character class

diff --git a/re2c/test/code_points_error.8.c b/re2c/test/code_points_error.8.c
index 271b2074281b0f4e4f1a0fbaa71452eb3a9d419a..10f0abdb9b6840f6f7e08dd8df11c193f1b084ca 100644 (file)
--- a/re2c/test/code_points_error.8.c
+++ b/re2c/test/code_points_error.8.c
@@ -1 +1 @@
-re2c: error: line 2, column 1: Bad code point: '0x110000'
+re2c: error: line 2, column 21: bad code point: '0x110000'

diff --git a/re2c/test/code_points_error.c b/re2c/test/code_points_error.c
index 4b4f6b4bf3c7c04c142f721c986013214ca593dc..d9f68c5866bd1bf7ee5448ac39e73c3aef5e3c3d 100644 (file)
--- a/re2c/test/code_points_error.c
+++ b/re2c/test/code_points_error.c
@@ -1 +1 @@
-re2c: error: line 2, column 1: Bad code point: '0xFFFF'
+re2c: error: line 2, column 5: bad code point: '0xFFFF'

diff --git a/re2c/test/code_points_error.e.c b/re2c/test/code_points_error.e.c
index 4b4f6b4bf3c7c04c142f721c986013214ca593dc..d9f68c5866bd1bf7ee5448ac39e73c3aef5e3c3d 100644 (file)
--- a/re2c/test/code_points_error.e.c
+++ b/re2c/test/code_points_error.e.c
@@ -1 +1 @@
-re2c: error: line 2, column 1: Bad code point: '0xFFFF'
+re2c: error: line 2, column 5: bad code point: '0xFFFF'

diff --git a/re2c/test/code_points_error.u.c b/re2c/test/code_points_error.u.c
index 271b2074281b0f4e4f1a0fbaa71452eb3a9d419a..10f0abdb9b6840f6f7e08dd8df11c193f1b084ca 100644 (file)
--- a/re2c/test/code_points_error.u.c
+++ b/re2c/test/code_points_error.u.c
@@ -1 +1 @@
-re2c: error: line 2, column 1: Bad code point: '0x110000'
+re2c: error: line 2, column 21: bad code point: '0x110000'

diff --git a/re2c/test/code_points_error.w.c b/re2c/test/code_points_error.w.c
index 66e9e9067dbb6d9e930e7f8a3a4608a18baa89ae..54128ffe28825a524b66f088afe2a5906589a2aa 100644 (file)
--- a/re2c/test/code_points_error.w.c
+++ b/re2c/test/code_points_error.w.c
@@ -1 +1 @@
-re2c: error: line 2, column 1: Bad code point: '0x10FFFF'
+re2c: error: line 2, column 11: bad code point: '0x10FFFF'

diff --git a/re2c/test/code_points_error.x.c b/re2c/test/code_points_error.x.c
index 271b2074281b0f4e4f1a0fbaa71452eb3a9d419a..10f0abdb9b6840f6f7e08dd8df11c193f1b084ca 100644 (file)
--- a/re2c/test/code_points_error.x.c
+++ b/re2c/test/code_points_error.x.c
@@ -1 +1 @@
-re2c: error: line 2, column 1: Bad code point: '0x110000'
+re2c: error: line 2, column 21: bad code point: '0x110000'

diff --git a/re2c/test/input4.--empty-class(error).c b/re2c/test/input4.--empty-class(error).c
index ed80caf73e5fa05367e2b15b03a46b7047d0eaf9..87fd50010b66f847a29cfb2dee279a0cb3ee22b3 100644 (file)
--- a/re2c/test/input4.--empty-class(error).c
+++ b/re2c/test/input4.--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 3, column 1: empty character class
+re2c: error: line 3, column 0: empty character class

diff --git a/re2c/test/input6.--empty-class(error).c b/re2c/test/input6.--empty-class(error).c
index ed80caf73e5fa05367e2b15b03a46b7047d0eaf9..87fd50010b66f847a29cfb2dee279a0cb3ee22b3 100644 (file)
--- a/re2c/test/input6.--empty-class(error).c
+++ b/re2c/test/input6.--empty-class(error).c
@@ -1 +1 @@
-re2c: error: line 3, column 1: empty character class
+re2c: error: line 3, column 0: empty character class

diff --git a/re2c/test/rexx.--empty-class(error).c b/re2c/test/rexx.--empty-class(error).c
index b83b89e9a8ccccf7e75697a92000fc72d1a4e441..cf69714b1a7eabfce4e079b7bc55883fe5e31cb1 100644 (file)
--- a/re2c/test/rexx.--empty-class(error).c
+++ b/re2c/test/rexx.--empty-class(error).c
@@ -1,2 +1,2 @@
 re2c: warning: line 253: rule matches empty string [-Wmatch-empty-string]
-re2c: error: line 288, column 1: empty character class
+re2c: error: line 288, column 0: empty character class