/* Regops and State definitions */
-#define REGNODE_MAX 93
-#define REGMATCH_STATE_MAX 133
+#define REGNODE_MAX 92
+#define REGMATCH_STATE_MAX 132
#define END 0 /* 0000 End of program. */
#define SUCCEED 1 /* 0x01 Return from a subroutine, basically. */
#define MEOL 5 /* 0x05 Same, assuming multiline: /$/m */
#define EOS 6 /* 0x06 Match "" at end of string: /\z/ */
#define GPOS 7 /* 0x07 Matches where last m//g left off. */
-#define BOUND 8 /* 0x08 Match "" at any word boundary using native charset rules for non-utf8 */
-#define BOUNDL 9 /* 0x09 Match "" at any locale word boundary */
-#define BOUNDU 10 /* 0x0a Match "" at any word boundary using Unicode rules */
-#define BOUNDA 11 /* 0x0b Match "" at any word boundary using ASCII rules */
-#define NBOUND 12 /* 0x0c Match "" at any word non-boundary using native charset rules for non-utf8 */
-#define NBOUNDL 13 /* 0x0d Match "" at any locale word non-boundary */
-#define NBOUNDU 14 /* 0x0e Match "" at any word non-boundary using Unicode rules */
-#define NBOUNDA 15 /* 0x0f Match "" at any word non-boundary using ASCII rules */
+#define BOUND 8 /* 0x08 Like BOUNDA for non-utf8, otherwise match "" between any Unicode \w\W or \W\w */
+#define BOUNDL 9 /* 0x09 Like BOUND/BOUNDU, but \w and \W are defined by current locale */
+#define BOUNDU 10 /* 0x0a Match "" at any boundary of a given type using Unicode rules */
+#define BOUNDA 11 /* 0x0b Match "" at any boundary between \w\W or \W\w, where \w is [_a-zA-Z0-9] */
+#define NBOUND 12 /* 0x0c Like NBOUNDA for non-utf8, otherwise match "" between any Unicode \w\w or \W\W */
+#define NBOUNDL 13 /* 0x0d Like NBOUND/NBOUNDU, but \w and \W are defined by current locale */
+#define NBOUNDU 14 /* 0x0e Match "" at any non-boundary of a given type using using Unicode rules */
+#define NBOUNDA 15 /* 0x0f Match "" betweeen any \w\w or \W\W, where \w is [_a-zA-Z0-9] */
#define REG_ANY 16 /* 0x10 Match any one character (except newline). */
#define SANY 17 /* 0x11 Match any one character. */
-#define CANY 18 /* 0x12 Match any one byte. */
-#define ANYOF 19 /* 0x13 Match character in (or not in) this class, single char match only */
+#define ANYOF 18 /* 0x12 Match character in (or not in) this class, single char match only */
+#define ANYOFD 19 /* 0x13 Like ANYOF, but /d is in effect */
#define ANYOFL 20 /* 0x14 Like ANYOF, but /l is in effect */
#define POSIXD 21 /* 0x15 Some [[:class:]] under /d; the FLAGS field gives which one */
#define POSIXL 22 /* 0x16 Some [[:class:]] under /l; the FLAGS field gives which one */
#define CLUMP 29 /* 0x1d Match any extended grapheme cluster sequence */
#define BRANCH 30 /* 0x1e Match this alternative, or the next... */
#define EXACT 31 /* 0x1f Match this string (preceded by length). */
-#define EXACTL 32 /* 0x20 Like EXACT, but /l is in effect. */
+#define EXACTL 32 /* 0x20 Like EXACT, but /l is in effect (used so locale-related warnings can be checked for). */
#define EXACTF 33 /* 0x21 Match this non-UTF-8 string (not guaranteed to be folded) using /id rules (w/len). */
#define EXACTFL 34 /* 0x22 Match this string (not guaranteed to be folded) using /il rules (w/len). */
#define EXACTFU 35 /* 0x23 Match this string (folded iff in UTF-8, length in folding doesn't change if not in UTF-8) using /iu rules (w/len). */
#define AHOCORASICK 74 /* 0x4a Aho Corasick stclass. flags==type */
#define AHOCORASICKC 75 /* 0x4b Same as AHOCORASICK, but with embedded charclass data */
#define GOSUB 76 /* 0x4c recurse to paren arg1 at (signed) ofs arg2 */
-#define GOSTART 77 /* 0x4d recurse to start of pattern */
-#define NGROUPP 78 /* 0x4e Whether the group matched. */
-#define INSUBP 79 /* 0x4f Whether we are in a specific recurse. */
-#define DEFINEP 80 /* 0x50 Never execute directly. */
-#define ENDLIKE 81 /* 0x51 Used only for the type field of verbs */
-#define OPFAIL 82 /* 0x52 Same as (?!) */
-#define ACCEPT 83 /* 0x53 Accepts the current matched string. */
-#define VERB 84 /* 0x54 Used only for the type field of verbs */
-#define PRUNE 85 /* 0x55 Pattern fails at this startpoint if no-backtracking through this */
-#define MARKPOINT 86 /* 0x56 Push the current location for rollback by cut. */
-#define SKIP 87 /* 0x57 On failure skip forward (to the mark) before retrying */
-#define COMMIT 88 /* 0x58 Pattern fails outright if backtracking through this */
-#define CUTGROUP 89 /* 0x59 On failure go to the next alternation in the group */
-#define KEEPS 90 /* 0x5a $& begins here. */
-#define LNBREAK 91 /* 0x5b generic newline pattern */
-#define OPTIMIZED 92 /* 0x5c Placeholder for dump. */
-#define PSEUDO 93 /* 0x5d Pseudo opcode for internal use. */
+#define NGROUPP 77 /* 0x4d Whether the group matched. */
+#define INSUBP 78 /* 0x4e Whether we are in a specific recurse. */
+#define DEFINEP 79 /* 0x4f Never execute directly. */
+#define ENDLIKE 80 /* 0x50 Used only for the type field of verbs */
+#define OPFAIL 81 /* 0x51 Same as (?!), but with verb arg */
+#define ACCEPT 82 /* 0x52 Accepts the current matched string, with verbar */
+#define VERB 83 /* 0x53 Used only for the type field of verbs */
+#define PRUNE 84 /* 0x54 Pattern fails at this startpoint if no-backtracking through this */
+#define MARKPOINT 85 /* 0x55 Push the current location for rollback by cut. */
+#define SKIP 86 /* 0x56 On failure skip forward (to the mark) before retrying */
+#define COMMIT 87 /* 0x57 Pattern fails outright if backtracking through this */
+#define CUTGROUP 88 /* 0x58 On failure go to the next alternation in the group */
+#define KEEPS 89 /* 0x59 $& begins here. */
+#define LNBREAK 90 /* 0x5a generic newline pattern */
+#define OPTIMIZED 91 /* 0x5b Placeholder for dump. */
+#define PSEUDO 92 /* 0x5c Pseudo opcode for internal use. */
/* ------------ States ------------- */
#define TRIE_next (REGNODE_MAX + 1) /* state for TRIE */
#define TRIE_next_fail (REGNODE_MAX + 2) /* state for TRIE */
NBOUND, /* NBOUNDA */
REG_ANY, /* REG_ANY */
REG_ANY, /* SANY */
- REG_ANY, /* CANY */
ANYOF, /* ANYOF */
+ ANYOF, /* ANYOFD */
ANYOF, /* ANYOFL */
POSIXD, /* POSIXD */
POSIXD, /* POSIXL */
TRIE, /* AHOCORASICK */
TRIE, /* AHOCORASICKC */
GOSUB, /* GOSUB */
- GOSTART, /* GOSTART */
NGROUPP, /* NGROUPP */
INSUBP, /* INSUBP */
DEFINEP, /* DEFINEP */
};
#endif
+#ifdef REG_COMP_C
+
/* regarglen[] - How large is the argument part of the node (in regnodes) */
-#ifdef REG_COMP_C
static const U8 regarglen[] = {
0, /* END */
0, /* SUCCEED */
0, /* NBOUNDA */
0, /* REG_ANY */
0, /* SANY */
- 0, /* CANY */
EXTRA_SIZE(struct regnode_1), /* ANYOF */
+ EXTRA_SIZE(struct regnode_1), /* ANYOFD */
EXTRA_SIZE(struct regnode_1), /* ANYOFL */
0, /* POSIXD */
0, /* POSIXL */
EXTRA_SIZE(struct regnode_1), /* AHOCORASICK */
EXTRA_SIZE(struct regnode_charclass), /* AHOCORASICKC */
EXTRA_SIZE(struct regnode_2L), /* GOSUB */
- 0, /* GOSTART */
EXTRA_SIZE(struct regnode_1), /* NGROUPP */
EXTRA_SIZE(struct regnode_1), /* INSUBP */
EXTRA_SIZE(struct regnode_1), /* DEFINEP */
0, /* ENDLIKE */
- 0, /* OPFAIL */
- EXTRA_SIZE(struct regnode_1), /* ACCEPT */
+ EXTRA_SIZE(struct regnode_1), /* OPFAIL */
+ EXTRA_SIZE(struct regnode_2L), /* ACCEPT */
EXTRA_SIZE(struct regnode_1), /* VERB */
EXTRA_SIZE(struct regnode_1), /* PRUNE */
EXTRA_SIZE(struct regnode_1), /* MARKPOINT */
0, /* NBOUNDA */
0, /* REG_ANY */
0, /* SANY */
- 0, /* CANY */
0, /* ANYOF */
+ 0, /* ANYOFD */
0, /* ANYOFL */
0, /* POSIXD */
0, /* POSIXL */
0, /* AHOCORASICK */
0, /* AHOCORASICKC */
0, /* GOSUB */
- 0, /* GOSTART */
0, /* NGROUPP */
0, /* INSUBP */
0, /* DEFINEP */
#endif /* REG_COMP_C */
+
/* reg_name[] - Opcode/state names in string form, for debugging */
#ifndef DOINIT
"NBOUNDA", /* 0x0f */
"REG_ANY", /* 0x10 */
"SANY", /* 0x11 */
- "CANY", /* 0x12 */
- "ANYOF", /* 0x13 */
+ "ANYOF", /* 0x12 */
+ "ANYOFD", /* 0x13 */
"ANYOFL", /* 0x14 */
"POSIXD", /* 0x15 */
"POSIXL", /* 0x16 */
"AHOCORASICK", /* 0x4a */
"AHOCORASICKC", /* 0x4b */
"GOSUB", /* 0x4c */
- "GOSTART", /* 0x4d */
- "NGROUPP", /* 0x4e */
- "INSUBP", /* 0x4f */
- "DEFINEP", /* 0x50 */
- "ENDLIKE", /* 0x51 */
- "OPFAIL", /* 0x52 */
- "ACCEPT", /* 0x53 */
- "VERB", /* 0x54 */
- "PRUNE", /* 0x55 */
- "MARKPOINT", /* 0x56 */
- "SKIP", /* 0x57 */
- "COMMIT", /* 0x58 */
- "CUTGROUP", /* 0x59 */
- "KEEPS", /* 0x5a */
- "LNBREAK", /* 0x5b */
- "OPTIMIZED", /* 0x5c */
- "PSEUDO", /* 0x5d */
+ "NGROUPP", /* 0x4d */
+ "INSUBP", /* 0x4e */
+ "DEFINEP", /* 0x4f */
+ "ENDLIKE", /* 0x50 */
+ "OPFAIL", /* 0x51 */
+ "ACCEPT", /* 0x52 */
+ "VERB", /* 0x53 */
+ "PRUNE", /* 0x54 */
+ "MARKPOINT", /* 0x55 */
+ "SKIP", /* 0x56 */
+ "COMMIT", /* 0x57 */
+ "CUTGROUP", /* 0x58 */
+ "KEEPS", /* 0x59 */
+ "LNBREAK", /* 0x5a */
+ "OPTIMIZED", /* 0x5b */
+ "PSEUDO", /* 0x5c */
/* ------------ States ------------- */
"TRIE_next", /* REGNODE_MAX +0x01 */
"TRIE_next_fail", /* REGNODE_MAX +0x02 */
"CUTGROUP_SEEN", /* 0x00000010 - PREGf_CUTGROUP_SEEN */
"USE_RE_EVAL", /* 0x00000020 - PREGf_USE_RE_EVAL - compiled with "use re 'eval'" */
"NOSCAN", /* 0x00000040 - PREGf_NOSCAN */
- "CANY_SEEN", /* 0x00000080 - PREGf_CANY_SEEN */
"GPOS_SEEN", /* 0x00000100 - PREGf_GPOS_SEEN */
"GPOS_FLOAT", /* 0x00000200 - PREGf_GPOS_FLOAT */
"ANCH_MBOL", /* 0x00000400 - PREGf_ANCH_MBOL */
"ANCH_SBOL", /* 0x00000800 - PREGf_ANCH_SBOL */
"ANCH_GPOS", /* 0x00001000 - PREGf_ANCH_GPOS */
+ "RECURSE_SEEN", /* 0x00002000 - PREGf_RECURSE_SEEN */
};
#endif /* DOINIT */
EXTCONST U8 PL_simple[] __attribute__deprecated__;
#else
EXTCONST U8 PL_simple[] __attribute__deprecated__ = {
- REG_ANY, SANY, CANY, ANYOF, ANYOFL, POSIXD, POSIXL, POSIXU, POSIXA,
+ REG_ANY, SANY, ANYOF, ANYOFD, ANYOFL, POSIXD, POSIXL, POSIXU, POSIXA,
NPOSIXD, NPOSIXL, NPOSIXU, NPOSIXA,
0
};
https://perl5.git.perl.org / perl5.git / blobdiff