/* -*- buffer-read-only: t -*-
!!!!!!! DO NOT EDIT THIS FILE !!!!!!!
- This file is built by regcomp.pl from regcomp.sym.
+ This file is built by regen/regcomp.pl from regcomp.sym.
Any changes made here will be lost!
*/
/* Regops and State definitions */
-#define REGNODE_MAX 89
-#define REGMATCH_STATE_MAX 129
+#define REGNODE_MAX 90
+#define REGMATCH_STATE_MAX 130
#define END 0 /* 0000 End of program. */
#define SUCCEED 1 /* 0x01 Return from a subroutine, basically. */
#define REG_ANY 14 /* 0x0e Match any one character (except newline). */
#define SANY 15 /* 0x0f Match any one character. */
#define CANY 16 /* 0x10 Match any one byte. */
-#define ANYOF 17 /* 0x11 Match character in (or not in) this class. */
+#define ANYOF 17 /* 0x11 Match character in (or not in) this class, folding is native charset for non-utf8. */
#define ALNUM 18 /* 0x12 Match any alphanumeric character */
#define ALNUML 19 /* 0x13 Match any alphanumeric char in locale */
#define NALNUM 20 /* 0x14 Match any non-alphanumeric character */
#define DIGITL 27 /* 0x1b Match any numeric character in locale */
#define NDIGIT 28 /* 0x1c Match any non-numeric character */
#define NDIGITL 29 /* 0x1d Match any non-numeric character in locale */
-#define CLUMP 30 /* 0x1e Match any combining character sequence */
+#define CLUMP 30 /* 0x1e Match any extended grapheme cluster sequence */
#define BRANCH 31 /* 0x1f Match this alternative, or the next... */
#define BACK 32 /* 0x20 Match "", "next" ptr points backward. */
#define EXACT 33 /* 0x21 Match this string (preceded by length). */
-#define EXACTF 34 /* 0x22 Match this string, folded (prec. by length). */
+#define EXACTF 34 /* 0x22 Match this string, folded, native charset semantics for non-utf8 (prec. by length). */
#define EXACTFL 35 /* 0x23 Match this string, folded in locale (w/len). */
#define NOTHING 36 /* 0x24 Match empty string. */
#define TAIL 37 /* 0x25 Match empty string. Can jump here from outside. */
#define OPEN 45 /* 0x2d Mark this point in input as start of */
#define CLOSE 46 /* 0x2e Analogous to OPEN. */
#define REF 47 /* 0x2f Match some already matched string */
-#define REFF 48 /* 0x30 Match already matched string, folded */
+#define REFF 48 /* 0x30 Match already matched string, folded using native charset semantics for non-utf8 */
#define REFFL 49 /* 0x31 Match already matched string, folded in loc. */
#define IFMATCH 50 /* 0x32 Succeeds if the following matches. */
#define UNLESSM 51 /* 0x33 Fails if the following matches. */
#define GOSUB 65 /* 0x41 recurse to paren arg1 at (signed) ofs arg2 */
#define GOSTART 66 /* 0x42 recurse to start of pattern */
#define NREF 67 /* 0x43 Match some already matched string */
-#define NREFF 68 /* 0x44 Match already matched string, folded */
+#define NREFF 68 /* 0x44 Match already matched string, folded using native charset semantics for non-utf8 */
#define NREFFL 69 /* 0x45 Match already matched string, folded in loc. */
#define NGROUPP 70 /* 0x46 Whether the group matched. */
#define INSUBP 71 /* 0x47 Whether we are in a specific recurse. */
#define ENDLIKE 73 /* 0x49 Used only for the type field of verbs */
#define OPFAIL 74 /* 0x4a Same as (?!) */
#define ACCEPT 75 /* 0x4b Accepts the current matched string. */
-#define VERB 76 /* 0x4c no-sv 1 Used only for the type field of verbs */
+#define VERB 76 /* 0x4c Used only for the type field of verbs */
#define PRUNE 77 /* 0x4d Pattern fails at this startpoint if no-backtracking through this */
#define MARKPOINT 78 /* 0x4e Push the current location for rollback by cut. */
#define SKIP 79 /* 0x4f On failure skip forward (to the mark) before retrying */
#define NVERTWS 85 /* 0x55 not vertical whitespace (Perl 6) */
#define HORIZWS 86 /* 0x56 horizontal whitespace (Perl 6) */
#define NHORIZWS 87 /* 0x57 not horizontal whitespace (Perl 6) */
-#define OPTIMIZED 88 /* 0x58 Placeholder for dump. */
-#define PSEUDO 89 /* 0x59 Pseudo opcode for internal use. */
+#define FOLDCHAR 88 /* 0x58 codepoint with tricky case folding properties. */
+#define OPTIMIZED 89 /* 0x59 Placeholder for dump. */
+#define PSEUDO 90 /* 0x5a 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 */
NVERTWS, /* NVERTWS */
HORIZWS, /* HORIZWS */
NHORIZWS, /* NHORIZWS */
+ FOLDCHAR, /* FOLDCHAR */
NOTHING, /* OPTIMIZED */
PSEUDO, /* PSEUDO */
/* ------------ States ------------- */
0, /* ENDLIKE */
0, /* OPFAIL */
EXTRA_SIZE(struct regnode_1), /* ACCEPT */
- 0, /* VERB */
+ EXTRA_SIZE(struct regnode_1), /* VERB */
EXTRA_SIZE(struct regnode_1), /* PRUNE */
EXTRA_SIZE(struct regnode_1), /* MARKPOINT */
EXTRA_SIZE(struct regnode_1), /* SKIP */
0, /* NVERTWS */
0, /* HORIZWS */
0, /* NHORIZWS */
+ EXTRA_SIZE(struct regnode_1), /* FOLDCHAR */
0, /* OPTIMIZED */
0, /* PSEUDO */
};
0, /* NVERTWS */
0, /* HORIZWS */
0, /* NHORIZWS */
+ 0, /* FOLDCHAR */
0, /* OPTIMIZED */
0, /* PSEUDO */
};
"NVERTWS", /* 0x55 */
"HORIZWS", /* 0x56 */
"NHORIZWS", /* 0x57 */
- "OPTIMIZED", /* 0x58 */
- "PSEUDO", /* 0x59 */
+ "FOLDCHAR", /* 0x58 */
+ "OPTIMIZED", /* 0x59 */
+ "PSEUDO", /* 0x5a */
/* ------------ States ------------- */
"TRIE_next", /* REGNODE_MAX +0x01 */
"TRIE_next_fail", /* REGNODE_MAX +0x02 */
};
#endif /* DOINIT */
+/* PL_reg_extflags_name[] - Opcode/state names in string form, for debugging */
+
+#ifndef DOINIT
+EXTCONST char * PL_reg_extflags_name[];
+#else
+EXTCONST char * const PL_reg_extflags_name[] = {
+ /* Bits in extflags defined: 11111111111111111111111001111111 */
+ "MULTILINE", /* 0x00000001 */
+ "SINGLELINE", /* 0x00000002 */
+ "FOLD", /* 0x00000004 */
+ "EXTENDED", /* 0x00000008 */
+ "KEEPCOPY", /* 0x00000010 */
+ "LOCALE", /* 0x00000020 */
+ "UNICODE", /* 0x00000040 */
+ "UNUSED_BIT_7", /* 0x00000080 */
+ "UNUSED_BIT_8", /* 0x00000100 */
+ "ANCH_BOL", /* 0x00000200 */
+ "ANCH_MBOL", /* 0x00000400 */
+ "ANCH_SBOL", /* 0x00000800 */
+ "ANCH_GPOS", /* 0x00001000 */
+ "GPOS_SEEN", /* 0x00002000 */
+ "GPOS_FLOAT", /* 0x00004000 */
+ "LOOKBEHIND_SEEN", /* 0x00008000 */
+ "EVAL_SEEN", /* 0x00010000 */
+ "CANY_SEEN", /* 0x00020000 */
+ "NOSCAN", /* 0x00040000 */
+ "CHECK_ALL", /* 0x00080000 */
+ "MATCH_UTF8", /* 0x00100000 */
+ "USE_INTUIT_NOML", /* 0x00200000 */
+ "USE_INTUIT_ML", /* 0x00400000 */
+ "INTUIT_TAIL", /* 0x00800000 */
+ "SPLIT", /* 0x01000000 */
+ "COPY_DONE", /* 0x02000000 */
+ "TAINTED_SEEN", /* 0x04000000 */
+ "TAINTED", /* 0x08000000 */
+ "START_ONLY", /* 0x10000000 */
+ "SKIPWHITE", /* 0x20000000 */
+ "WHITE", /* 0x40000000 */
+ "NULL", /* 0x80000000 */
+};
+#endif /* DOINIT */
+
+/* The following have no fixed length. U8 so we can do strchr() on it. */
+#define REGNODE_VARIES(node) (PL_varies_bitmask[(node) >> 3] & (1 << ((node) & 7)))
+
+#ifndef DOINIT
+EXTCONST U8 PL_varies[] __attribute__deprecated__;
+#else
+EXTCONST U8 PL_varies[] __attribute__deprecated__ = {
+ CLUMP, BRANCH, BACK, STAR, PLUS, CURLY, CURLYN, CURLYM, CURLYX, WHILEM,
+ REF, REFF, REFFL, SUSPEND, IFTHEN, BRANCHJ, NREF, NREFF, NREFFL,
+ 0
+};
+#endif /* DOINIT */
+
+#ifndef DOINIT
+EXTCONST U8 PL_varies_bitmask[];
+#else
+EXTCONST U8 PL_varies_bitmask[] = {
+ 0x00, 0x00, 0x00, 0xC0, 0xC1, 0x9F, 0x33, 0x01, 0x38, 0x00, 0x00, 0x00
+};
+#endif /* DOINIT */
+
+/* The following always have a length of 1. U8 we can do strchr() on it. */
+/* (Note that length 1 means "one character" under UTF8, not "one octet".) */
+#define REGNODE_SIMPLE(node) (PL_simple_bitmask[(node) >> 3] & (1 << ((node) & 7)))
+
+#ifndef DOINIT
+EXTCONST U8 PL_simple[] __attribute__deprecated__;
+#else
+EXTCONST U8 PL_simple[] __attribute__deprecated__ = {
+ REG_ANY, SANY, CANY, ANYOF, ALNUM, ALNUML, NALNUM, NALNUML, SPACE,
+ SPACEL, NSPACE, NSPACEL, DIGIT, NDIGIT, VERTWS, NVERTWS, HORIZWS,
+ NHORIZWS,
+ 0
+};
+#endif /* DOINIT */
+
+#ifndef DOINIT
+EXTCONST U8 PL_simple_bitmask[];
+#else
+EXTCONST U8 PL_simple_bitmask[] = {
+ 0x00, 0xC0, 0xFF, 0x17, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xF0, 0x00
+};
+#endif /* DOINIT */
+
/* ex: set ro: */
https://perl5.git.perl.org / perl5.git / blobdiff