diff options
| -rw-r--r-- | regcomp.sym | 8 | ||||
| -rw-r--r-- | regnodes.h | 411 |
2 files changed, 234 insertions, 185 deletions
diff --git a/regcomp.sym b/regcomp.sym index 2fb1785aeb..286b942517 100644 --- a/regcomp.sym +++ b/regcomp.sym @@ -33,10 +33,12 @@ SEOL EOL, no ; Same, assuming singleline. BOUND BOUND, no ; Match "" at any word boundary using native charset semantics for non-utf8 BOUNDL BOUND, no ; Match "" at any locale word boundary BOUNDU BOUND, no ; Match "" at any word boundary using Unicode semantics +BOUNDA BOUND, no ; Match "" at any word boundary using ASCII semantics # All NBOUND nodes are required by a line regexec.c to be greater than all BOUND ones NBOUND NBOUND, no ; Match "" at any word non-boundary using native charset semantics for non-utf8 NBOUNDL NBOUND, no ; Match "" at any locale word non-boundary NBOUNDU NBOUND, no ; Match "" at any word non-boundary using Unicode semantics +NBOUNDA NBOUND, no ; Match "" at any word non-boundary using ASCII semantics GPOS GPOS, no ; Matches where last m//g left off. #* [Special] alternatives: @@ -49,19 +51,25 @@ ANYOFV ANYOF, sv 0 V ; Match character in (or not in) this class, c ALNUM ALNUM, no 0 S ; Match any alphanumeric character using native charset semantics for non-utf8 ALNUML ALNUM, no 0 S ; Match any alphanumeric char in locale ALNUMU ALNUM, no 0 S ; Match any alphanumeric char using Unicode semantics +ALNUMA ALNUM, no 0 S ; Match [A-Za-z_0-9] NALNUM NALNUM, no 0 S ; Match any non-alphanumeric character using native charset semantics for non-utf8 NALNUML NALNUM, no 0 S ; Match any non-alphanumeric char in locale NALNUMU NALNUM, no 0 S ; Match any non-alphanumeric char using Unicode semantics +NALNUMA NALNUM, no 0 S ; Match [^A-Za-z_0-9] SPACE SPACE, no 0 S ; Match any whitespace character using native charset semantics for non-utf8 SPACEL SPACE, no 0 S ; Match any whitespace char in locale SPACEU SPACE, no 0 S ; Match any whitespace char using Unicode semantics +SPACEA SPACE, no 0 S ; Match [ \t\n\f\r] NSPACE NSPACE, no 0 S ; Match any non-whitespace character using native charset semantics for non-utf8 NSPACEL NSPACE, no 0 S ; Match any non-whitespace char in locale NSPACEU NSPACE, no 0 S ; Match any non-whitespace char using Unicode semantics +NSPACEA NSPACE, no 0 S ; Match [^ \t\n\f\r] DIGIT DIGIT, no 0 S ; Match any numeric character using native charset semantics for non-utf8 DIGITL DIGIT, no 0 S ; Match any numeric character in locale +DIGITA DIGIT, no 0 S ; Match [0-9] NDIGIT NDIGIT, no 0 S ; Match any non-numeric character using native charset semantics for non-utf8 NDIGITL NDIGIT, no 0 S ; Match any non-numeric character in locale +NDIGITA NDIGIT, no 0 S ; Match [^0-9] CLUMP CLUMP, no 0 V ; Match any extended grapheme cluster sequence #* Alternation diff --git a/regnodes.h b/regnodes.h index 670c24b0ea..80e3ed5b32 100644 --- a/regnodes.h +++ b/regnodes.h @@ -6,8 +6,8 @@ /* Regops and State definitions */ -#define REGNODE_MAX 100 -#define REGMATCH_STATE_MAX 140 +#define REGNODE_MAX 108 +#define REGMATCH_STATE_MAX 148 #define END 0 /* 0000 End of program. */ #define SUCCEED 1 /* 0x01 Return from a subroutine, basically. */ @@ -21,95 +21,103 @@ #define BOUND 9 /* 0x09 Match "" at any word boundary using native charset semantics for non-utf8 */ #define BOUNDL 10 /* 0x0a Match "" at any locale word boundary */ #define BOUNDU 11 /* 0x0b Match "" at any word boundary using Unicode semantics */ -#define NBOUND 12 /* 0x0c Match "" at any word non-boundary using native charset semantics 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 semantics */ -#define GPOS 15 /* 0x0f Matches where last m//g left off. */ -#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 ANYOFV 20 /* 0x14 Match character in (or not in) this class, can match-multiple chars */ -#define ALNUM 21 /* 0x15 Match any alphanumeric character using native charset semantics for non-utf8 */ -#define ALNUML 22 /* 0x16 Match any alphanumeric char in locale */ -#define ALNUMU 23 /* 0x17 Match any alphanumeric char using Unicode semantics */ -#define NALNUM 24 /* 0x18 Match any non-alphanumeric character using native charset semantics for non-utf8 */ -#define NALNUML 25 /* 0x19 Match any non-alphanumeric char in locale */ -#define NALNUMU 26 /* 0x1a Match any non-alphanumeric char using Unicode semantics */ -#define SPACE 27 /* 0x1b Match any whitespace character using native charset semantics for non-utf8 */ -#define SPACEL 28 /* 0x1c Match any whitespace char in locale */ -#define SPACEU 29 /* 0x1d Match any whitespace char using Unicode semantics */ -#define NSPACE 30 /* 0x1e Match any non-whitespace character using native charset semantics for non-utf8 */ -#define NSPACEL 31 /* 0x1f Match any non-whitespace char in locale */ -#define NSPACEU 32 /* 0x20 Match any non-whitespace char using Unicode semantics */ -#define DIGIT 33 /* 0x21 Match any numeric character using native charset semantics for non-utf8 */ -#define DIGITL 34 /* 0x22 Match any numeric character in locale */ -#define NDIGIT 35 /* 0x23 Match any non-numeric character using native charset semantics for non-utf8 */ -#define NDIGITL 36 /* 0x24 Match any non-numeric character in locale */ -#define CLUMP 37 /* 0x25 Match any extended grapheme cluster sequence */ -#define BRANCH 38 /* 0x26 Match this alternative, or the next... */ -#define BACK 39 /* 0x27 Match "", "next" ptr points backward. */ -#define EXACT 40 /* 0x28 Match this string (preceded by length). */ -#define EXACTF 41 /* 0x29 Match this string, folded, native charset semantics for non-utf8 (prec. by length). */ -#define EXACTFL 42 /* 0x2a Match this string, folded in locale (w/len). */ -#define EXACTFU 43 /* 0x2b Match this string, folded, Unicode semantics for non-utf8 (prec. by length). */ -#define NOTHING 44 /* 0x2c Match empty string. */ -#define TAIL 45 /* 0x2d Match empty string. Can jump here from outside. */ -#define STAR 46 /* 0x2e Match this (simple) thing 0 or more times. */ -#define PLUS 47 /* 0x2f Match this (simple) thing 1 or more times. */ -#define CURLY 48 /* 0x30 Match this simple thing {n,m} times. */ -#define CURLYN 49 /* 0x31 Capture next-after-this simple thing */ -#define CURLYM 50 /* 0x32 Capture this medium-complex thing {n,m} times. */ -#define CURLYX 51 /* 0x33 Match this complex thing {n,m} times. */ -#define WHILEM 52 /* 0x34 Do curly processing and see if rest matches. */ -#define OPEN 53 /* 0x35 Mark this point in input as start of */ -#define CLOSE 54 /* 0x36 Analogous to OPEN. */ -#define REF 55 /* 0x37 Match some already matched string */ -#define REFF 56 /* 0x38 Match already matched string, folded using native charset semantics for non-utf8 */ -#define REFFL 57 /* 0x39 Match already matched string, folded in loc. */ -#define REFFU 58 /* 0x3a Match already matched string, folded using unicode semantics for non-utf8 */ -#define NREF 59 /* 0x3b Match some already matched string */ -#define NREFF 60 /* 0x3c Match already matched string, folded using native charset semantics for non-utf8 */ -#define NREFFL 61 /* 0x3d Match already matched string, folded in loc. */ -#define NREFFU 62 /* 0x3e Match already matched string, folded using unicode semantics for non-utf8 */ -#define IFMATCH 63 /* 0x3f Succeeds if the following matches. */ -#define UNLESSM 64 /* 0x40 Fails if the following matches. */ -#define SUSPEND 65 /* 0x41 "Independent" sub-RE. */ -#define IFTHEN 66 /* 0x42 Switch, should be preceded by switcher . */ -#define GROUPP 67 /* 0x43 Whether the group matched. */ -#define LONGJMP 68 /* 0x44 Jump far away. */ -#define BRANCHJ 69 /* 0x45 BRANCH with long offset. */ -#define EVAL 70 /* 0x46 Execute some Perl code. */ -#define MINMOD 71 /* 0x47 Next operator is not greedy. */ -#define LOGICAL 72 /* 0x48 Next opcode should set the flag only. */ -#define RENUM 73 /* 0x49 Group with independently numbered parens. */ -#define TRIE 74 /* 0x4a Match many EXACT(F[LU]?)? at once. flags==type */ -#define TRIEC 75 /* 0x4b Same as TRIE, but with embedded charclass data */ -#define AHOCORASICK 76 /* 0x4c Aho Corasick stclass. flags==type */ -#define AHOCORASICKC 77 /* 0x4d Same as AHOCORASICK, but with embedded charclass data */ -#define GOSUB 78 /* 0x4e recurse to paren arg1 at (signed) ofs arg2 */ -#define GOSTART 79 /* 0x4f recurse to start of pattern */ -#define NGROUPP 80 /* 0x50 Whether the group matched. */ -#define INSUBP 81 /* 0x51 Whether we are in a specific recurse. */ -#define DEFINEP 82 /* 0x52 Never execute directly. */ -#define ENDLIKE 83 /* 0x53 Used only for the type field of verbs */ -#define OPFAIL 84 /* 0x54 Same as (?!) */ -#define ACCEPT 85 /* 0x55 Accepts the current matched string. */ -#define VERB 86 /* 0x56 Used only for the type field of verbs */ -#define PRUNE 87 /* 0x57 Pattern fails at this startpoint if no-backtracking through this */ -#define MARKPOINT 88 /* 0x58 Push the current location for rollback by cut. */ -#define SKIP 89 /* 0x59 On failure skip forward (to the mark) before retrying */ -#define COMMIT 90 /* 0x5a Pattern fails outright if backtracking through this */ -#define CUTGROUP 91 /* 0x5b On failure go to the next alternation in the group */ -#define KEEPS 92 /* 0x5c $& begins here. */ -#define LNBREAK 93 /* 0x5d generic newline pattern */ -#define VERTWS 94 /* 0x5e vertical whitespace (Perl 6) */ -#define NVERTWS 95 /* 0x5f not vertical whitespace (Perl 6) */ -#define HORIZWS 96 /* 0x60 horizontal whitespace (Perl 6) */ -#define NHORIZWS 97 /* 0x61 not horizontal whitespace (Perl 6) */ -#define FOLDCHAR 98 /* 0x62 codepoint with tricky case folding properties. */ -#define OPTIMIZED 99 /* 0x63 Placeholder for dump. */ -#define PSEUDO 100 /* 0x64 Pseudo opcode for internal use. */ +#define BOUNDA 12 /* 0x0c Match "" at any word boundary using ASCII semantics */ +#define NBOUND 13 /* 0x0d Match "" at any word non-boundary using native charset semantics for non-utf8 */ +#define NBOUNDL 14 /* 0x0e Match "" at any locale word non-boundary */ +#define NBOUNDU 15 /* 0x0f Match "" at any word non-boundary using Unicode semantics */ +#define NBOUNDA 16 /* 0x10 Match "" at any word non-boundary using ASCII semantics */ +#define GPOS 17 /* 0x11 Matches where last m//g left off. */ +#define REG_ANY 18 /* 0x12 Match any one character (except newline). */ +#define SANY 19 /* 0x13 Match any one character. */ +#define CANY 20 /* 0x14 Match any one byte. */ +#define ANYOF 21 /* 0x15 Match character in (or not in) this class, single char match only */ +#define ANYOFV 22 /* 0x16 Match character in (or not in) this class, can match-multiple chars */ +#define ALNUM 23 /* 0x17 Match any alphanumeric character using native charset semantics for non-utf8 */ +#define ALNUML 24 /* 0x18 Match any alphanumeric char in locale */ +#define ALNUMU 25 /* 0x19 Match any alphanumeric char using Unicode semantics */ +#define ALNUMA 26 /* 0x1a Match [A-Za-z_0-9] */ +#define NALNUM 27 /* 0x1b Match any non-alphanumeric character using native charset semantics for non-utf8 */ +#define NALNUML 28 /* 0x1c Match any non-alphanumeric char in locale */ +#define NALNUMU 29 /* 0x1d Match any non-alphanumeric char using Unicode semantics */ +#define NALNUMA 30 /* 0x1e Match [^A-Za-z_0-9] */ +#define SPACE 31 /* 0x1f Match any whitespace character using native charset semantics for non-utf8 */ +#define SPACEL 32 /* 0x20 Match any whitespace char in locale */ +#define SPACEU 33 /* 0x21 Match any whitespace char using Unicode semantics */ +#define SPACEA 34 /* 0x22 Match [ \t\n\f\r] */ +#define NSPACE 35 /* 0x23 Match any non-whitespace character using native charset semantics for non-utf8 */ +#define NSPACEL 36 /* 0x24 Match any non-whitespace char in locale */ +#define NSPACEU 37 /* 0x25 Match any non-whitespace char using Unicode semantics */ +#define NSPACEA 38 /* 0x26 Match [^ \t\n\f\r] */ +#define DIGIT 39 /* 0x27 Match any numeric character using native charset semantics for non-utf8 */ +#define DIGITL 40 /* 0x28 Match any numeric character in locale */ +#define DIGITA 41 /* 0x29 Match [0-9] */ +#define NDIGIT 42 /* 0x2a Match any non-numeric character using native charset semantics for non-utf8 */ +#define NDIGITL 43 /* 0x2b Match any non-numeric character in locale */ +#define NDIGITA 44 /* 0x2c Match [^0-9] */ +#define CLUMP 45 /* 0x2d Match any extended grapheme cluster sequence */ +#define BRANCH 46 /* 0x2e Match this alternative, or the next... */ +#define BACK 47 /* 0x2f Match "", "next" ptr points backward. */ +#define EXACT 48 /* 0x30 Match this string (preceded by length). */ +#define EXACTF 49 /* 0x31 Match this string, folded, native charset semantics for non-utf8 (prec. by length). */ +#define EXACTFL 50 /* 0x32 Match this string, folded in locale (w/len). */ +#define EXACTFU 51 /* 0x33 Match this string, folded, Unicode semantics for non-utf8 (prec. by length). */ +#define NOTHING 52 /* 0x34 Match empty string. */ +#define TAIL 53 /* 0x35 Match empty string. Can jump here from outside. */ +#define STAR 54 /* 0x36 Match this (simple) thing 0 or more times. */ +#define PLUS 55 /* 0x37 Match this (simple) thing 1 or more times. */ +#define CURLY 56 /* 0x38 Match this simple thing {n,m} times. */ +#define CURLYN 57 /* 0x39 Capture next-after-this simple thing */ +#define CURLYM 58 /* 0x3a Capture this medium-complex thing {n,m} times. */ +#define CURLYX 59 /* 0x3b Match this complex thing {n,m} times. */ +#define WHILEM 60 /* 0x3c Do curly processing and see if rest matches. */ +#define OPEN 61 /* 0x3d Mark this point in input as start of */ +#define CLOSE 62 /* 0x3e Analogous to OPEN. */ +#define REF 63 /* 0x3f Match some already matched string */ +#define REFF 64 /* 0x40 Match already matched string, folded using native charset semantics for non-utf8 */ +#define REFFL 65 /* 0x41 Match already matched string, folded in loc. */ +#define REFFU 66 /* 0x42 Match already matched string, folded using unicode semantics for non-utf8 */ +#define NREF 67 /* 0x43 Match some already matched string */ +#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 NREFFU 70 /* 0x46 Match already matched string, folded using unicode semantics for non-utf8 */ +#define IFMATCH 71 /* 0x47 Succeeds if the following matches. */ +#define UNLESSM 72 /* 0x48 Fails if the following matches. */ +#define SUSPEND 73 /* 0x49 "Independent" sub-RE. */ +#define IFTHEN 74 /* 0x4a Switch, should be preceded by switcher . */ +#define GROUPP 75 /* 0x4b Whether the group matched. */ +#define LONGJMP 76 /* 0x4c Jump far away. */ +#define BRANCHJ 77 /* 0x4d BRANCH with long offset. */ +#define EVAL 78 /* 0x4e Execute some Perl code. */ +#define MINMOD 79 /* 0x4f Next operator is not greedy. */ +#define LOGICAL 80 /* 0x50 Next opcode should set the flag only. */ +#define RENUM 81 /* 0x51 Group with independently numbered parens. */ +#define TRIE 82 /* 0x52 Match many EXACT(F[LU]?)? at once. flags==type */ +#define TRIEC 83 /* 0x53 Same as TRIE, but with embedded charclass data */ +#define AHOCORASICK 84 /* 0x54 Aho Corasick stclass. flags==type */ +#define AHOCORASICKC 85 /* 0x55 Same as AHOCORASICK, but with embedded charclass data */ +#define GOSUB 86 /* 0x56 recurse to paren arg1 at (signed) ofs arg2 */ +#define GOSTART 87 /* 0x57 recurse to start of pattern */ +#define NGROUPP 88 /* 0x58 Whether the group matched. */ +#define INSUBP 89 /* 0x59 Whether we are in a specific recurse. */ +#define DEFINEP 90 /* 0x5a Never execute directly. */ +#define ENDLIKE 91 /* 0x5b Used only for the type field of verbs */ +#define OPFAIL 92 /* 0x5c Same as (?!) */ +#define ACCEPT 93 /* 0x5d Accepts the current matched string. */ +#define VERB 94 /* 0x5e Used only for the type field of verbs */ +#define PRUNE 95 /* 0x5f Pattern fails at this startpoint if no-backtracking through this */ +#define MARKPOINT 96 /* 0x60 Push the current location for rollback by cut. */ +#define SKIP 97 /* 0x61 On failure skip forward (to the mark) before retrying */ +#define COMMIT 98 /* 0x62 Pattern fails outright if backtracking through this */ +#define CUTGROUP 99 /* 0x63 On failure go to the next alternation in the group */ +#define KEEPS 100 /* 0x64 $& begins here. */ +#define LNBREAK 101 /* 0x65 generic newline pattern */ +#define VERTWS 102 /* 0x66 vertical whitespace (Perl 6) */ +#define NVERTWS 103 /* 0x67 not vertical whitespace (Perl 6) */ +#define HORIZWS 104 /* 0x68 horizontal whitespace (Perl 6) */ +#define NHORIZWS 105 /* 0x69 not horizontal whitespace (Perl 6) */ +#define FOLDCHAR 106 /* 0x6a codepoint with tricky case folding properties. */ +#define OPTIMIZED 107 /* 0x6b Placeholder for dump. */ +#define PSEUDO 108 /* 0x6c 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 */ @@ -170,9 +178,11 @@ EXTCONST U8 PL_regkind[] = { BOUND, /* BOUND */ BOUND, /* BOUNDL */ BOUND, /* BOUNDU */ + BOUND, /* BOUNDA */ NBOUND, /* NBOUND */ NBOUND, /* NBOUNDL */ NBOUND, /* NBOUNDU */ + NBOUND, /* NBOUNDA */ GPOS, /* GPOS */ REG_ANY, /* REG_ANY */ REG_ANY, /* SANY */ @@ -182,19 +192,25 @@ EXTCONST U8 PL_regkind[] = { ALNUM, /* ALNUM */ ALNUM, /* ALNUML */ ALNUM, /* ALNUMU */ + ALNUM, /* ALNUMA */ NALNUM, /* NALNUM */ NALNUM, /* NALNUML */ NALNUM, /* NALNUMU */ + NALNUM, /* NALNUMA */ SPACE, /* SPACE */ SPACE, /* SPACEL */ SPACE, /* SPACEU */ + SPACE, /* SPACEA */ NSPACE, /* NSPACE */ NSPACE, /* NSPACEL */ NSPACE, /* NSPACEU */ + NSPACE, /* NSPACEA */ DIGIT, /* DIGIT */ DIGIT, /* DIGITL */ + DIGIT, /* DIGITA */ NDIGIT, /* NDIGIT */ NDIGIT, /* NDIGITL */ + NDIGIT, /* NDIGITA */ CLUMP, /* CLUMP */ BRANCH, /* BRANCH */ BACK, /* BACK */ @@ -319,9 +335,11 @@ static const U8 regarglen[] = { 0, /* BOUND */ 0, /* BOUNDL */ 0, /* BOUNDU */ + 0, /* BOUNDA */ 0, /* NBOUND */ 0, /* NBOUNDL */ 0, /* NBOUNDU */ + 0, /* NBOUNDA */ 0, /* GPOS */ 0, /* REG_ANY */ 0, /* SANY */ @@ -331,19 +349,25 @@ static const U8 regarglen[] = { 0, /* ALNUM */ 0, /* ALNUML */ 0, /* ALNUMU */ + 0, /* ALNUMA */ 0, /* NALNUM */ 0, /* NALNUML */ 0, /* NALNUMU */ + 0, /* NALNUMA */ 0, /* SPACE */ 0, /* SPACEL */ 0, /* SPACEU */ + 0, /* SPACEA */ 0, /* NSPACE */ 0, /* NSPACEL */ 0, /* NSPACEU */ + 0, /* NSPACEA */ 0, /* DIGIT */ 0, /* DIGITL */ + 0, /* DIGITA */ 0, /* NDIGIT */ 0, /* NDIGITL */ + 0, /* NDIGITA */ 0, /* CLUMP */ 0, /* BRANCH */ 0, /* BACK */ @@ -425,9 +449,11 @@ static const char reg_off_by_arg[] = { 0, /* BOUND */ 0, /* BOUNDL */ 0, /* BOUNDU */ + 0, /* BOUNDA */ 0, /* NBOUND */ 0, /* NBOUNDL */ 0, /* NBOUNDU */ + 0, /* NBOUNDA */ 0, /* GPOS */ 0, /* REG_ANY */ 0, /* SANY */ @@ -437,19 +463,25 @@ static const char reg_off_by_arg[] = { 0, /* ALNUM */ 0, /* ALNUML */ 0, /* ALNUMU */ + 0, /* ALNUMA */ 0, /* NALNUM */ 0, /* NALNUML */ 0, /* NALNUMU */ + 0, /* NALNUMA */ 0, /* SPACE */ 0, /* SPACEL */ 0, /* SPACEU */ + 0, /* SPACEA */ 0, /* NSPACE */ 0, /* NSPACEL */ 0, /* NSPACEU */ + 0, /* NSPACEA */ 0, /* DIGIT */ 0, /* DIGITL */ + 0, /* DIGITA */ 0, /* NDIGIT */ 0, /* NDIGITL */ + 0, /* NDIGITA */ 0, /* CLUMP */ 0, /* BRANCH */ 0, /* BACK */ @@ -536,95 +568,103 @@ EXTCONST char * const PL_reg_name[] = { "BOUND", /* 0x09 */ "BOUNDL", /* 0x0a */ "BOUNDU", /* 0x0b */ - "NBOUND", /* 0x0c */ - "NBOUNDL", /* 0x0d */ - "NBOUNDU", /* 0x0e */ - "GPOS", /* 0x0f */ - "REG_ANY", /* 0x10 */ - "SANY", /* 0x11 */ - "CANY", /* 0x12 */ - "ANYOF", /* 0x13 */ - "ANYOFV", /* 0x14 */ - "ALNUM", /* 0x15 */ - "ALNUML", /* 0x16 */ - "ALNUMU", /* 0x17 */ - "NALNUM", /* 0x18 */ - "NALNUML", /* 0x19 */ - "NALNUMU", /* 0x1a */ - "SPACE", /* 0x1b */ - "SPACEL", /* 0x1c */ - "SPACEU", /* 0x1d */ - "NSPACE", /* 0x1e */ - "NSPACEL", /* 0x1f */ - "NSPACEU", /* 0x20 */ - "DIGIT", /* 0x21 */ - "DIGITL", /* 0x22 */ - "NDIGIT", /* 0x23 */ - "NDIGITL", /* 0x24 */ - "CLUMP", /* 0x25 */ - "BRANCH", /* 0x26 */ - "BACK", /* 0x27 */ - "EXACT", /* 0x28 */ - "EXACTF", /* 0x29 */ - "EXACTFL", /* 0x2a */ - "EXACTFU", /* 0x2b */ - "NOTHING", /* 0x2c */ - "TAIL", /* 0x2d */ - "STAR", /* 0x2e */ - "PLUS", /* 0x2f */ - "CURLY", /* 0x30 */ - "CURLYN", /* 0x31 */ - "CURLYM", /* 0x32 */ - "CURLYX", /* 0x33 */ - "WHILEM", /* 0x34 */ - "OPEN", /* 0x35 */ - "CLOSE", /* 0x36 */ - "REF", /* 0x37 */ - "REFF", /* 0x38 */ - "REFFL", /* 0x39 */ - "REFFU", /* 0x3a */ - "NREF", /* 0x3b */ - "NREFF", /* 0x3c */ - "NREFFL", /* 0x3d */ - "NREFFU", /* 0x3e */ - "IFMATCH", /* 0x3f */ - "UNLESSM", /* 0x40 */ - "SUSPEND", /* 0x41 */ - "IFTHEN", /* 0x42 */ - "GROUPP", /* 0x43 */ - "LONGJMP", /* 0x44 */ - "BRANCHJ", /* 0x45 */ - "EVAL", /* 0x46 */ - "MINMOD", /* 0x47 */ - "LOGICAL", /* 0x48 */ - "RENUM", /* 0x49 */ - "TRIE", /* 0x4a */ - "TRIEC", /* 0x4b */ - "AHOCORASICK", /* 0x4c */ - "AHOCORASICKC", /* 0x4d */ - "GOSUB", /* 0x4e */ - "GOSTART", /* 0x4f */ - "NGROUPP", /* 0x50 */ - "INSUBP", /* 0x51 */ - "DEFINEP", /* 0x52 */ - "ENDLIKE", /* 0x53 */ - "OPFAIL", /* 0x54 */ - "ACCEPT", /* 0x55 */ - "VERB", /* 0x56 */ - "PRUNE", /* 0x57 */ - "MARKPOINT", /* 0x58 */ - "SKIP", /* 0x59 */ - "COMMIT", /* 0x5a */ - "CUTGROUP", /* 0x5b */ - "KEEPS", /* 0x5c */ - "LNBREAK", /* 0x5d */ - "VERTWS", /* 0x5e */ - "NVERTWS", /* 0x5f */ - "HORIZWS", /* 0x60 */ - "NHORIZWS", /* 0x61 */ - "FOLDCHAR", /* 0x62 */ - "OPTIMIZED", /* 0x63 */ - "PSEUDO", /* 0x64 */ + "BOUNDA", /* 0x0c */ + "NBOUND", /* 0x0d */ + "NBOUNDL", /* 0x0e */ + "NBOUNDU", /* 0x0f */ + "NBOUNDA", /* 0x10 */ + "GPOS", /* 0x11 */ + "REG_ANY", /* 0x12 */ + "SANY", /* 0x13 */ + "CANY", /* 0x14 */ + "ANYOF", /* 0x15 */ + "ANYOFV", /* 0x16 */ + "ALNUM", /* 0x17 */ + "ALNUML", /* 0x18 */ + "ALNUMU", /* 0x19 */ + "ALNUMA", /* 0x1a */ + "NALNUM", /* 0x1b */ + "NALNUML", /* 0x1c */ + "NALNUMU", /* 0x1d */ + "NALNUMA", /* 0x1e */ + "SPACE", /* 0x1f */ + "SPACEL", /* 0x20 */ + "SPACEU", /* 0x21 */ + "SPACEA", /* 0x22 */ + "NSPACE", /* 0x23 */ + "NSPACEL", /* 0x24 */ + "NSPACEU", /* 0x25 */ + "NSPACEA", /* 0x26 */ + "DIGIT", /* 0x27 */ + "DIGITL", /* 0x28 */ + "DIGITA", /* 0x29 */ + "NDIGIT", /* 0x2a */ + "NDIGITL", /* 0x2b */ + "NDIGITA", /* 0x2c */ + "CLUMP", /* 0x2d */ + "BRANCH", /* 0x2e */ + "BACK", /* 0x2f */ + "EXACT", /* 0x30 */ + "EXACTF", /* 0x31 */ + "EXACTFL", /* 0x32 */ + "EXACTFU", /* 0x33 */ + "NOTHING", /* 0x34 */ + "TAIL", /* 0x35 */ + "STAR", /* 0x36 */ + "PLUS", /* 0x37 */ + "CURLY", /* 0x38 */ + "CURLYN", /* 0x39 */ + "CURLYM", /* 0x3a */ + "CURLYX", /* 0x3b */ + "WHILEM", /* 0x3c */ + "OPEN", /* 0x3d */ + "CLOSE", /* 0x3e */ + "REF", /* 0x3f */ + "REFF", /* 0x40 */ + "REFFL", /* 0x41 */ + "REFFU", /* 0x42 */ + "NREF", /* 0x43 */ + "NREFF", /* 0x44 */ + "NREFFL", /* 0x45 */ + "NREFFU", /* 0x46 */ + "IFMATCH", /* 0x47 */ + "UNLESSM", /* 0x48 */ + "SUSPEND", /* 0x49 */ + "IFTHEN", /* 0x4a */ + "GROUPP", /* 0x4b */ + "LONGJMP", /* 0x4c */ + "BRANCHJ", /* 0x4d */ + "EVAL", /* 0x4e */ + "MINMOD", /* 0x4f */ + "LOGICAL", /* 0x50 */ + "RENUM", /* 0x51 */ + "TRIE", /* 0x52 */ + "TRIEC", /* 0x53 */ + "AHOCORASICK", /* 0x54 */ + "AHOCORASICKC", /* 0x55 */ + "GOSUB", /* 0x56 */ + "GOSTART", /* 0x57 */ + "NGROUPP", /* 0x58 */ + "INSUBP", /* 0x59 */ + "DEFINEP", /* 0x5a */ + "ENDLIKE", /* 0x5b */ + "OPFAIL", /* 0x5c */ + "ACCEPT", /* 0x5d */ + "VERB", /* 0x5e */ + "PRUNE", /* 0x5f */ + "MARKPOINT", /* 0x60 */ + "SKIP", /* 0x61 */ + "COMMIT", /* 0x62 */ + "CUTGROUP", /* 0x63 */ + "KEEPS", /* 0x64 */ + "LNBREAK", /* 0x65 */ + "VERTWS", /* 0x66 */ + "NVERTWS", /* 0x67 */ + "HORIZWS", /* 0x68 */ + "NHORIZWS", /* 0x69 */ + "FOLDCHAR", /* 0x6a */ + "OPTIMIZED", /* 0x6b */ + "PSEUDO", /* 0x6c */ /* ------------ States ------------- */ "TRIE_next", /* REGNODE_MAX +0x01 */ "TRIE_next_fail", /* REGNODE_MAX +0x02 */ @@ -729,7 +769,7 @@ EXTCONST U8 PL_varies[] __attribute__deprecated__ = { EXTCONST U8 PL_varies_bitmask[]; #else EXTCONST U8 PL_varies_bitmask[] = { - 0x00, 0x00, 0x10, 0x00, 0xE0, 0xC0, 0x9F, 0x7F, 0x26, 0x00, 0x00, 0x00, 0x00 + 0x00, 0x00, 0x40, 0x00, 0x00, 0xE0, 0xC0, 0x9F, 0x7F, 0x26, 0x00, 0x00, 0x00, 0x00 }; #endif /* DOINIT */ @@ -741,9 +781,10 @@ EXTCONST U8 PL_varies_bitmask[] = { EXTCONST U8 PL_simple[] __attribute__deprecated__; #else EXTCONST U8 PL_simple[] __attribute__deprecated__ = { - REG_ANY, SANY, CANY, ANYOF, ALNUM, ALNUML, ALNUMU, NALNUM, NALNUML, - NALNUMU, SPACE, SPACEL, SPACEU, NSPACE, NSPACEL, NSPACEU, DIGIT, - DIGITL, NDIGIT, NDIGITL, VERTWS, NVERTWS, HORIZWS, NHORIZWS, + REG_ANY, SANY, CANY, ANYOF, ALNUM, ALNUML, ALNUMU, ALNUMA, NALNUM, + NALNUML, NALNUMU, NALNUMA, SPACE, SPACEL, SPACEU, SPACEA, NSPACE, + NSPACEL, NSPACEU, NSPACEA, DIGIT, DIGITL, DIGITA, NDIGIT, NDIGITL, + NDIGITA, VERTWS, NVERTWS, HORIZWS, NHORIZWS, 0 }; #endif /* DOINIT */ @@ -752,7 +793,7 @@ EXTCONST U8 PL_simple[] __attribute__deprecated__ = { EXTCONST U8 PL_simple_bitmask[]; #else EXTCONST U8 PL_simple_bitmask[] = { - 0x00, 0x00, 0xEF, 0xFF, 0x1F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xC0, 0x03 + 0x00, 0x00, 0xBC, 0xFF, 0xFF, 0x1F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xC0, 0x03 }; #endif /* DOINIT */ |