X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/78102347e0341b1cd4795accd0ca5ff1f2621bb2..51b1bca10c16ef14af0a0e9cb91a2879f6826041:/regnodes.h diff --git a/regnodes.h b/regnodes.h index 72c93ff..0caf86d 100644 --- a/regnodes.h +++ b/regnodes.h @@ -6,8 +6,8 @@ /* Regops and State definitions */ -#define REGNODE_MAX 108 -#define REGMATCH_STATE_MAX 148 +#define REGNODE_MAX 95 +#define REGMATCH_STATE_MAX 135 #define END 0 /* 0000 End of program. */ #define SUCCEED 1 /* 0x01 Return from a subroutine, basically. */ @@ -31,93 +31,80 @@ #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. */ +#define ANYOF_WARN_SUPER 22 /* 0x16 Match character in (or not in) this class, warn (if enabled) upon matching a char above Unicode max; */ +#define ANYOF_SYNTHETIC 23 /* 0x17 Synthetic start class */ +#define POSIXD 24 /* 0x18 Some [[:class:]] under /d; the FLAGS field gives which one */ +#define POSIXL 25 /* 0x19 Some [[:class:]] under /l; the FLAGS field gives which one */ +#define POSIXU 26 /* 0x1a Some [[:class:]] under /u; the FLAGS field gives which one */ +#define POSIXA 27 /* 0x1b Some [[:class:]] under /a; the FLAGS field gives which one */ +#define NPOSIXD 28 /* 0x1c complement of POSIXD, [[:^class:]] */ +#define NPOSIXL 29 /* 0x1d complement of POSIXL, [[:^class:]] */ +#define NPOSIXU 30 /* 0x1e complement of POSIXU, [[:^class:]] */ +#define NPOSIXA 31 /* 0x1f complement of POSIXA, [[:^class:]] */ +#define CLUMP 32 /* 0x20 Match any extended grapheme cluster sequence */ +#define BRANCH 33 /* 0x21 Match this alternative, or the next... */ +#define BACK 34 /* 0x22 Match "", "next" ptr points backward. */ +#define EXACT 35 /* 0x23 Match this string (preceded by length). */ +#define EXACTF 36 /* 0x24 Match this non-UTF-8 string (not guaranteed to be folded) using /id rules (w/len). */ +#define EXACTFL 37 /* 0x25 Match this string (not guaranteed to be folded) using /il rules (w/len). */ +#define EXACTFU 38 /* 0x26 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 EXACTFA 39 /* 0x27 Match this string (not guaranteed to be folded) using /iaa rules (w/len). */ +#define EXACTFU_SS 40 /* 0x28 Match this string (folded iff in UTF-8, length in folding may change even if not in UTF-8) using /iu rules (w/len). */ +#define EXACTFU_TRICKYFOLD 41 /* 0x29 Match this folded UTF-8 string using /iu rules */ +#define NOTHING 42 /* 0x2a Match empty string. */ +#define TAIL 43 /* 0x2b Match empty string. Can jump here from outside. */ +#define STAR 44 /* 0x2c Match this (simple) thing 0 or more times. */ +#define PLUS 45 /* 0x2d Match this (simple) thing 1 or more times. */ +#define CURLY 46 /* 0x2e Match this simple thing {n,m} times. */ +#define CURLYN 47 /* 0x2f Capture next-after-this simple thing */ +#define CURLYM 48 /* 0x30 Capture this medium-complex thing {n,m} times. */ +#define CURLYX 49 /* 0x31 Match this complex thing {n,m} times. */ +#define WHILEM 50 /* 0x32 Do curly processing and see if rest matches. */ +#define OPEN 51 /* 0x33 Mark this point in input as start of #n. */ +#define CLOSE 52 /* 0x34 Analogous to OPEN. */ +#define REF 53 /* 0x35 Match some already matched string */ +#define REFF 54 /* 0x36 Match already matched string, folded using native charset semantics for non-utf8 */ +#define REFFL 55 /* 0x37 Match already matched string, folded in loc. */ +#define REFFU 56 /* 0x38 Match already matched string, folded using unicode semantics for non-utf8 */ +#define REFFA 57 /* 0x39 Match already matched string, folded using unicode semantics for non-utf8, no mixing ASCII, non-ASCII */ +#define NREF 58 /* 0x3a Match some already matched string */ +#define NREFF 59 /* 0x3b Match already matched string, folded using native charset semantics for non-utf8 */ +#define NREFFL 60 /* 0x3c Match already matched string, folded in loc. */ +#define NREFFU 61 /* 0x3d Match already matched string, folded using unicode semantics for non-utf8 */ +#define NREFFA 62 /* 0x3e Match already matched string, folded using unicode semantics for non-utf8, no mixing ASCII, non-ASCII */ +#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[ALU]?)? 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 OPTIMIZED 94 /* 0x5e Placeholder for dump. */ +#define PSEUDO 95 /* 0x5f 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 */ @@ -188,29 +175,16 @@ EXTCONST U8 PL_regkind[] = { REG_ANY, /* SANY */ REG_ANY, /* CANY */ ANYOF, /* ANYOF */ - ANYOF, /* ANYOFV */ - 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 */ + ANYOF, /* ANYOF_WARN_SUPER */ + ANYOF, /* ANYOF_SYNTHETIC */ + POSIXD, /* POSIXD */ + POSIXD, /* POSIXL */ + POSIXD, /* POSIXU */ + POSIXD, /* POSIXA */ + NPOSIXD, /* NPOSIXD */ + NPOSIXD, /* NPOSIXL */ + NPOSIXD, /* NPOSIXU */ + NPOSIXD, /* NPOSIXA */ CLUMP, /* CLUMP */ BRANCH, /* BRANCH */ BACK, /* BACK */ @@ -218,6 +192,9 @@ EXTCONST U8 PL_regkind[] = { EXACT, /* EXACTF */ EXACT, /* EXACTFL */ EXACT, /* EXACTFU */ + EXACT, /* EXACTFA */ + EXACT, /* EXACTFU_SS */ + EXACT, /* EXACTFU_TRICKYFOLD */ NOTHING, /* NOTHING */ NOTHING, /* TAIL */ STAR, /* STAR */ @@ -233,10 +210,12 @@ EXTCONST U8 PL_regkind[] = { REF, /* REFF */ REF, /* REFFL */ REF, /* REFFU */ + REF, /* REFFA */ REF, /* NREF */ REF, /* NREFF */ REF, /* NREFFL */ REF, /* NREFFU */ + REF, /* NREFFA */ BRANCHJ, /* IFMATCH */ BRANCHJ, /* UNLESSM */ BRANCHJ, /* SUSPEND */ @@ -268,11 +247,6 @@ EXTCONST U8 PL_regkind[] = { VERB, /* CUTGROUP */ KEEPS, /* KEEPS */ LNBREAK, /* LNBREAK */ - VERTWS, /* VERTWS */ - NVERTWS, /* NVERTWS */ - HORIZWS, /* HORIZWS */ - NHORIZWS, /* NHORIZWS */ - FOLDCHAR, /* FOLDCHAR */ NOTHING, /* OPTIMIZED */ PSEUDO, /* PSEUDO */ /* ------------ States ------------- */ @@ -345,29 +319,16 @@ static const U8 regarglen[] = { 0, /* SANY */ 0, /* CANY */ 0, /* ANYOF */ - 0, /* ANYOFV */ - 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, /* ANYOF_WARN_SUPER */ + 0, /* ANYOF_SYNTHETIC */ + 0, /* POSIXD */ + 0, /* POSIXL */ + 0, /* POSIXU */ + 0, /* POSIXA */ + 0, /* NPOSIXD */ + 0, /* NPOSIXL */ + 0, /* NPOSIXU */ + 0, /* NPOSIXA */ 0, /* CLUMP */ 0, /* BRANCH */ 0, /* BACK */ @@ -375,6 +336,9 @@ static const U8 regarglen[] = { 0, /* EXACTF */ 0, /* EXACTFL */ 0, /* EXACTFU */ + 0, /* EXACTFA */ + 0, /* EXACTFU_SS */ + 0, /* EXACTFU_TRICKYFOLD */ 0, /* NOTHING */ 0, /* TAIL */ 0, /* STAR */ @@ -390,10 +354,12 @@ static const U8 regarglen[] = { EXTRA_SIZE(struct regnode_1), /* REFF */ EXTRA_SIZE(struct regnode_1), /* REFFL */ EXTRA_SIZE(struct regnode_1), /* REFFU */ + EXTRA_SIZE(struct regnode_1), /* REFFA */ EXTRA_SIZE(struct regnode_1), /* NREF */ EXTRA_SIZE(struct regnode_1), /* NREFF */ EXTRA_SIZE(struct regnode_1), /* NREFFL */ EXTRA_SIZE(struct regnode_1), /* NREFFU */ + EXTRA_SIZE(struct regnode_1), /* NREFFA */ EXTRA_SIZE(struct regnode_1), /* IFMATCH */ EXTRA_SIZE(struct regnode_1), /* UNLESSM */ EXTRA_SIZE(struct regnode_1), /* SUSPEND */ @@ -425,11 +391,6 @@ static const U8 regarglen[] = { EXTRA_SIZE(struct regnode_1), /* CUTGROUP */ 0, /* KEEPS */ 0, /* LNBREAK */ - 0, /* VERTWS */ - 0, /* NVERTWS */ - 0, /* HORIZWS */ - 0, /* NHORIZWS */ - EXTRA_SIZE(struct regnode_1), /* FOLDCHAR */ 0, /* OPTIMIZED */ 0, /* PSEUDO */ }; @@ -459,29 +420,16 @@ static const char reg_off_by_arg[] = { 0, /* SANY */ 0, /* CANY */ 0, /* ANYOF */ - 0, /* ANYOFV */ - 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, /* ANYOF_WARN_SUPER */ + 0, /* ANYOF_SYNTHETIC */ + 0, /* POSIXD */ + 0, /* POSIXL */ + 0, /* POSIXU */ + 0, /* POSIXA */ + 0, /* NPOSIXD */ + 0, /* NPOSIXL */ + 0, /* NPOSIXU */ + 0, /* NPOSIXA */ 0, /* CLUMP */ 0, /* BRANCH */ 0, /* BACK */ @@ -489,6 +437,9 @@ static const char reg_off_by_arg[] = { 0, /* EXACTF */ 0, /* EXACTFL */ 0, /* EXACTFU */ + 0, /* EXACTFA */ + 0, /* EXACTFU_SS */ + 0, /* EXACTFU_TRICKYFOLD */ 0, /* NOTHING */ 0, /* TAIL */ 0, /* STAR */ @@ -504,10 +455,12 @@ static const char reg_off_by_arg[] = { 0, /* REFF */ 0, /* REFFL */ 0, /* REFFU */ + 0, /* REFFA */ 0, /* NREF */ 0, /* NREFF */ 0, /* NREFFL */ 0, /* NREFFU */ + 0, /* NREFFA */ 2, /* IFMATCH */ 2, /* UNLESSM */ 1, /* SUSPEND */ @@ -539,11 +492,6 @@ static const char reg_off_by_arg[] = { 0, /* CUTGROUP */ 0, /* KEEPS */ 0, /* LNBREAK */ - 0, /* VERTWS */ - 0, /* NVERTWS */ - 0, /* HORIZWS */ - 0, /* NHORIZWS */ - 0, /* FOLDCHAR */ 0, /* OPTIMIZED */ 0, /* PSEUDO */ }; @@ -578,93 +526,80 @@ EXTCONST char * const PL_reg_name[] = { "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 */ + "ANYOF_WARN_SUPER", /* 0x16 */ + "ANYOF_SYNTHETIC", /* 0x17 */ + "POSIXD", /* 0x18 */ + "POSIXL", /* 0x19 */ + "POSIXU", /* 0x1a */ + "POSIXA", /* 0x1b */ + "NPOSIXD", /* 0x1c */ + "NPOSIXL", /* 0x1d */ + "NPOSIXU", /* 0x1e */ + "NPOSIXA", /* 0x1f */ + "CLUMP", /* 0x20 */ + "BRANCH", /* 0x21 */ + "BACK", /* 0x22 */ + "EXACT", /* 0x23 */ + "EXACTF", /* 0x24 */ + "EXACTFL", /* 0x25 */ + "EXACTFU", /* 0x26 */ + "EXACTFA", /* 0x27 */ + "EXACTFU_SS", /* 0x28 */ + "EXACTFU_TRICKYFOLD", /* 0x29 */ + "NOTHING", /* 0x2a */ + "TAIL", /* 0x2b */ + "STAR", /* 0x2c */ + "PLUS", /* 0x2d */ + "CURLY", /* 0x2e */ + "CURLYN", /* 0x2f */ + "CURLYM", /* 0x30 */ + "CURLYX", /* 0x31 */ + "WHILEM", /* 0x32 */ + "OPEN", /* 0x33 */ + "CLOSE", /* 0x34 */ + "REF", /* 0x35 */ + "REFF", /* 0x36 */ + "REFFL", /* 0x37 */ + "REFFU", /* 0x38 */ + "REFFA", /* 0x39 */ + "NREF", /* 0x3a */ + "NREFF", /* 0x3b */ + "NREFFL", /* 0x3c */ + "NREFFU", /* 0x3d */ + "NREFFA", /* 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 */ + "OPTIMIZED", /* 0x5e */ + "PSEUDO", /* 0x5f */ /* ------------ States ------------- */ "TRIE_next", /* REGNODE_MAX +0x01 */ "TRIE_next_fail", /* REGNODE_MAX +0x02 */ @@ -715,23 +650,23 @@ EXTCONST char * const PL_reg_name[] = { EXTCONST char * PL_reg_extflags_name[]; #else EXTCONST char * const PL_reg_extflags_name[] = { - /* Bits in extflags defined: 11111111111111111111111001111111 */ + /* Bits in extflags defined: 11111110111111111111111111111111 */ "MULTILINE", /* 0x00000001 */ "SINGLELINE", /* 0x00000002 */ "FOLD", /* 0x00000004 */ "EXTENDED", /* 0x00000008 */ "KEEPCOPY", /* 0x00000010 */ - "CHARSET", /* 0x00000060 */ - "CHARSET", /* 0x00000060 */ - "UNUSED_BIT_7", /* 0x00000080 */ - "UNUSED_BIT_8", /* 0x00000100 */ + "CHARSET0", /* 0x00000020 : "CHARSET" - 0x000000e0 */ + "CHARSET1", /* 0x00000040 : "CHARSET" - 0x000000e0 */ + "CHARSET2", /* 0x00000080 : "CHARSET" - 0x000000e0 */ + "SPLIT", /* 0x00000100 */ "ANCH_BOL", /* 0x00000200 */ "ANCH_MBOL", /* 0x00000400 */ "ANCH_SBOL", /* 0x00000800 */ "ANCH_GPOS", /* 0x00001000 */ "GPOS_SEEN", /* 0x00002000 */ "GPOS_FLOAT", /* 0x00004000 */ - "LOOKBEHIND_SEEN", /* 0x00008000 */ + "NO_INPLACE_SUBST", /* 0x00008000 */ "EVAL_SEEN", /* 0x00010000 */ "CANY_SEEN", /* 0x00020000 */ "NOSCAN", /* 0x00040000 */ @@ -740,7 +675,7 @@ EXTCONST char * const PL_reg_extflags_name[] = { "USE_INTUIT_NOML", /* 0x00200000 */ "USE_INTUIT_ML", /* 0x00400000 */ "INTUIT_TAIL", /* 0x00800000 */ - "SPLIT", /* 0x01000000 */ + "UNUSED_BIT_24", /* 0x01000000 */ "COPY_DONE", /* 0x02000000 */ "TAINTED_SEEN", /* 0x04000000 */ "TAINTED", /* 0x08000000 */ @@ -758,9 +693,9 @@ EXTCONST char * const PL_reg_extflags_name[] = { EXTCONST U8 PL_varies[] __attribute__deprecated__; #else EXTCONST U8 PL_varies[] __attribute__deprecated__ = { - ANYOFV, CLUMP, BRANCH, BACK, STAR, PLUS, CURLY, CURLYN, CURLYM, CURLYX, - WHILEM, REF, REFF, REFFL, REFFU, NREF, NREFF, NREFFL, NREFFU, SUSPEND, - IFTHEN, BRANCHJ, + CLUMP, BRANCH, BACK, STAR, PLUS, CURLY, CURLYN, CURLYM, CURLYX, WHILEM, + REF, REFF, REFFL, REFFU, REFFA, NREF, NREFF, NREFFL, NREFFU, NREFFA, + SUSPEND, IFTHEN, BRANCHJ, 0 }; #endif /* DOINIT */ @@ -769,7 +704,7 @@ EXTCONST U8 PL_varies[] __attribute__deprecated__ = { EXTCONST U8 PL_varies_bitmask[]; #else EXTCONST U8 PL_varies_bitmask[] = { - 0x00, 0x00, 0x40, 0x00, 0x00, 0xE0, 0xC0, 0x9F, 0x7F, 0x26, 0x00, 0x00, 0x00, 0x00 + 0x00, 0x00, 0x00, 0x00, 0x07, 0xF0, 0xE7, 0x7F, 0x26, 0x00, 0x00, 0x00 }; #endif /* DOINIT */ @@ -781,10 +716,8 @@ 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, ALNUMA, NALNUM, - NALNUML, NALNUMU, NALNUMA, SPACE, SPACEL, SPACEU, SPACEA, NSPACE, - NSPACEL, NSPACEU, NSPACEA, DIGIT, DIGITL, DIGITA, NDIGIT, NDIGITL, - NDIGITA, VERTWS, NVERTWS, HORIZWS, NHORIZWS, + REG_ANY, SANY, CANY, ANYOF, ANYOF_WARN_SUPER, ANYOF_SYNTHETIC, POSIXD, + POSIXL, POSIXU, POSIXA, NPOSIXD, NPOSIXL, NPOSIXU, NPOSIXA, 0 }; #endif /* DOINIT */ @@ -793,7 +726,7 @@ EXTCONST U8 PL_simple[] __attribute__deprecated__ = { EXTCONST U8 PL_simple_bitmask[]; #else EXTCONST U8 PL_simple_bitmask[] = { - 0x00, 0x00, 0xBC, 0xFF, 0xFF, 0x1F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xC0, 0x03 + 0x00, 0x00, 0xFC, 0xFF, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }; #endif /* DOINIT */