In perl.git, the branch blead has been updated <https://perl5.git.perl.org/perl.git/commitdiff/6beede9533e4bc8c9b4d67a42483123c6adafd78?hp=e3c1dc81bda4523bfc897fec8afe25b162e29cb6>
- Log ----------------------------------------------------------------- commit 6beede9533e4bc8c9b4d67a42483123c6adafd78 Author: Karl Williamson <[email protected]> Date: Fri May 31 08:13:05 2019 -0600 regcomp.c: White-space only Outdent since previous commit removed an enclosing block commit 6966a05b6e7353ca4162fd3f65c1f0a963a243f0 Author: Karl Williamson <[email protected]> Date: Sat Mar 30 18:45:20 2019 -0600 Split ANYOFH regnode into two types ANYOFHb will be for nodes where all the matching code points share the frst UTF-8 byte. ANYOFH will be for all others. Neither of these has a bitmap. I noticed that we can omit some execution conditionals by splitting the nodes. ----------------------------------------------------------------------- Summary of changes: pod/perldebguts.pod | 7 +- regcomp.c | 60 +++++----- regcomp.sym | 3 +- regexec.c | 69 ++++++----- regnodes.h | 337 ++++++++++++++++++++++++++-------------------------- t/re/anyof.t | 152 ++++++++++++------------ 6 files changed, 325 insertions(+), 303 deletions(-) diff --git a/pod/perldebguts.pod b/pod/perldebguts.pod index 797e2c69e0..8430e2759f 100644 --- a/pod/perldebguts.pod +++ b/pod/perldebguts.pod @@ -612,9 +612,10 @@ will be lost. charclass_ classes posixl ANYOFH sv 1 Like ANYOF, but only has "High" matches, - none in the bitmap; non-zero flags "f" - means "f" is the first UTF-8 byte shared in - common by all code points matched + none in the bitmap; + ANYOFHb sv 1 Like ANYOFH, but all matches share the same + first byte, the repurposed flags field + contains it ANYOFM byte 1 Like ANYOF, but matches an invariant byte as determined by the mask and arg NANYOFM byte 1 complement of ANYOFM diff --git a/regcomp.c b/regcomp.c index 36f5afff71..8125ce03ed 100644 --- a/regcomp.c +++ b/regcomp.c @@ -1582,7 +1582,7 @@ S_get_ANYOF_cp_list_for_ssc(pTHX_ const RExC_state_t *pRExC_state, unsigned int i; const U32 n = ARG(node); bool new_node_has_latin1 = FALSE; - const U8 flags = OP(node) == ANYOFH ? 0 : ANYOF_FLAGS(node); + const U8 flags = OP(node) == ANYOFHb ? 0 : ANYOF_FLAGS(node); PERL_ARGS_ASSERT_GET_ANYOF_CP_LIST_FOR_SSC; @@ -1635,7 +1635,7 @@ S_get_ANYOF_cp_list_for_ssc(pTHX_ const RExC_state_t *pRExC_state, } /* Add in the points from the bit map */ - if (OP(node) != ANYOFH) { + if (OP(node) != ANYOFH && OP(node) != ANYOFHb) { for (i = 0; i < NUM_ANYOF_CODE_POINTS; i++) { if (ANYOF_BITMAP_TEST(node, i)) { unsigned int start = i++; @@ -1722,7 +1722,7 @@ S_ssc_and(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc, * another SSC or a regular ANYOF class. Can create false positives. */ SV* anded_cp_list; - U8 and_with_flags = (OP(and_with) == ANYOFH) ? 0 : ANYOF_FLAGS(and_with); + U8 and_with_flags = (OP(and_with) == ANYOFHb) ? 0 : ANYOF_FLAGS(and_with); U8 anded_flags; PERL_ARGS_ASSERT_SSC_AND; @@ -1906,7 +1906,7 @@ S_ssc_or(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc, SV* ored_cp_list; U8 ored_flags; - U8 or_with_flags = (OP(or_with) == ANYOFH) ? 0 : ANYOF_FLAGS(or_with); + U8 or_with_flags = (OP(or_with) == ANYOFHb) ? 0 : ANYOF_FLAGS(or_with); PERL_ARGS_ASSERT_SSC_OR; @@ -5844,6 +5844,7 @@ Perl_re_printf( aTHX_ "LHS=%" UVuf " RHS=%" UVuf "\n", case ANYOFL: case ANYOFPOSIXL: case ANYOFH: + case ANYOFHb: case ANYOF: if (flags & SCF_DO_STCLASS_AND) ssc_and(pRExC_state, data->start_class, @@ -14787,7 +14788,7 @@ S_populate_ANYOF_from_invlist(pTHX_ regnode *node, SV** invlist_ptr) assert(PL_regkind[OP(node)] == ANYOF); /* There is no bitmap for this node type */ - if (OP(node) == ANYOFH) { + if (OP(node) == ANYOFH || OP(node) == ANYOFHb) { return; } @@ -19028,40 +19029,39 @@ S_regclass(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth, /* If the lowest and highest code point in the class have the same * UTF-8 first byte, then all do, and we can store that byte for * regexec.c to use so that it can more quickly scan the target - * string for potential matches for this class. We co-opt the the - * flags field for this. Zero means, they don't have the same - * first byte. We do accept here very large code points (for - * future use), but don't bother with this optimization for them, - * as it would cause other complications */ - if (highest_cp > IV_MAX) { - anyof_flags = 0; - } - else { + * string for potential matches for this class. We co-opt the + * flags field for this, and make the node ANYOFb. We do accept + * here very large code points (for future use), but don't do + * this optimization for them, as it would cause other + * complications */ + op = ANYOFH; + if (highest_cp <= IV_MAX) { U8 low_utf8[UTF8_MAXBYTES+1]; U8 high_utf8[UTF8_MAXBYTES+1]; (void) uvchr_to_utf8(low_utf8, start[0]); (void) uvchr_to_utf8(high_utf8, invlist_highest(cp_list)); - anyof_flags = (low_utf8[0] == high_utf8[0]) - ? low_utf8[0] - : 0; + if (low_utf8[0] == high_utf8[0]) { + anyof_flags = low_utf8[0]; + op = ANYOFHb; + } } - op = ANYOFH; + goto done_finding_op; } } /* End of seeing if can optimize it into a different node */ is_anyof: /* It's going to be an ANYOF node. */ - if (op != ANYOFH) { - op = (has_runtime_dependency & HAS_D_RUNTIME_DEPENDENCY) - ? ANYOFD - : ((posixl) - ? ANYOFPOSIXL - : ((LOC) - ? ANYOFL - : ANYOF)); - } + op = (has_runtime_dependency & HAS_D_RUNTIME_DEPENDENCY) + ? ANYOFD + : ((posixl) + ? ANYOFPOSIXL + : ((LOC) + ? ANYOFL + : ANYOF)); + + done_finding_op: ret = regnode_guts(pRExC_state, op, regarglen[op], "anyof"); FILL_NODE(ret, op); /* We set the argument later */ @@ -20348,7 +20348,7 @@ Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o, const regmatch_ /* 2: embedded, otherwise 1 */ Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); else if (k == ANYOF) { - const U8 flags = (OP(o) == ANYOFH) ? 0 : ANYOF_FLAGS(o); + const U8 flags = (OP(o) == ANYOFHb) ? 0 : ANYOF_FLAGS(o); bool do_sep = FALSE; /* Do we need to separate various components of the output? */ /* Set if there is still an unresolved user-defined property */ @@ -20404,7 +20404,7 @@ Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o, const regmatch_ /* Ready to start outputting. First, the initial left bracket */ Perl_sv_catpvf(aTHX_ sv, "[%s", PL_colors[0]); - if (OP(o) != ANYOFH) { + if (OP(o) != ANYOFH && OP(o) != ANYOFHb) { /* Then all the things that could fit in the bitmap */ do_sep = put_charclass_bitmap_innards(sv, ANYOF_BITMAP(o), @@ -20502,7 +20502,7 @@ Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o, const regmatch_ /* And finally the matching, closing ']' */ Perl_sv_catpvf(aTHX_ sv, "%s]", PL_colors[1]); - if (OP(o) == ANYOFH && FLAGS(o) != 0) { + if (OP(o) == ANYOFHb) { Perl_sv_catpvf(aTHX_ sv, " (First UTF-8 byte=\\x%02x)", FLAGS(o)); } diff --git a/regcomp.sym b/regcomp.sym index 51f3176bcf..b702c437a1 100644 --- a/regcomp.sym +++ b/regcomp.sym @@ -62,7 +62,8 @@ ANYOF ANYOF, sv charclass S ; Match character in (or not in) this ANYOFD ANYOF, sv charclass S ; Like ANYOF, but /d is in effect ANYOFL ANYOF, sv charclass S ; Like ANYOF, but /l is in effect ANYOFPOSIXL ANYOF, sv charclass_posixl S ; Like ANYOFL, but matches [[:posix:]] classes -ANYOFH ANYOF, sv 1 S ; Like ANYOF, but only has "High" matches, none in the bitmap; non-zero flags "f" means "f" is the first UTF-8 byte shared in common by all code points matched +ANYOFH ANYOF, sv 1 S ; Like ANYOF, but only has "High" matches, none in the bitmap; +ANYOFHb ANYOF, sv 1 S ; Like ANYOFH, but all matches share the same first byte, the repurposed flags field contains it ANYOFM ANYOFM byte 1 S ; Like ANYOF, but matches an invariant byte as determined by the mask and arg NANYOFM ANYOFM byte 1 S ; complement of ANYOFM diff --git a/regexec.c b/regexec.c index 02c2d623c1..eaaef94881 100644 --- a/regexec.c +++ b/regexec.c @@ -2178,17 +2178,19 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, case ANYOFH: if (utf8_target) { /* Can't possibly match a non-UTF-8 target */ + REXEC_FBC_CLASS_SCAN(TRUE, + reginclass(prog, c, (U8*)s, (U8*) strend, utf8_target)); + } + break; + + case ANYOFHb: + if (utf8_target) { /* Can't possibly match a non-UTF-8 target */ + + /* We know what the first byte of any matched string should be */ U8 first_byte = FLAGS(c); - if (first_byte) { /* We know what the first byte of any matched - string should be */ - REXEC_FBC_FIND_NEXT_UTF8_BYTE_SCAN(first_byte, - reginclass(prog, c, (U8*)s, (U8*) strend, utf8_target)); - } - else { - REXEC_FBC_CLASS_SCAN(TRUE, + REXEC_FBC_FIND_NEXT_UTF8_BYTE_SCAN(first_byte, reginclass(prog, c, (U8*)s, (U8*) strend, utf8_target)); - } } break; @@ -6803,8 +6805,6 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) case ANYOFH: if ( ! utf8_target || NEXTCHR_IS_EOS - || ( ANYOF_FLAGS(scan) != 0 - && ANYOF_FLAGS(scan) != (U8) *locinput) || ! reginclass(rex, scan, (U8*)locinput, (U8*) loceol, utf8_target)) { @@ -6813,6 +6813,18 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) goto increment_locinput; break; + case ANYOFHb: + if ( ! utf8_target + || NEXTCHR_IS_EOS + || ANYOF_FLAGS(scan) != (U8) *locinput + || ! reginclass(rex, scan, (U8*)locinput, (U8*) loceol, + utf8_target)) + { + sayNO; + } + goto increment_locinput; + break; + /* The argument (FLAGS) to all the POSIX node types is the class number * */ @@ -9578,22 +9590,25 @@ S_regrepeat(pTHX_ regexp *prog, char **startposp, const regnode *p, case ANYOFH: if (utf8_target) { /* ANYOFH only can match UTF-8 targets */ - if (ANYOF_FLAGS(p)) { /* If we know the first byte of what - matches, we can avoid calling reginclass - */ - while ( hardcount < max - && scan < this_eol - && (U8) *scan == ANYOF_FLAGS(p) - && reginclass(prog, p, (U8*)scan, (U8*) this_eol, - TRUE)) - { - scan += UTF8SKIP(scan); - hardcount++; - } + while ( hardcount < max + && scan < this_eol + && reginclass(prog, p, (U8*)scan, (U8*) this_eol, TRUE)) + { + scan += UTF8SKIP(scan); + hardcount++; } - else while ( hardcount < max - && scan < this_eol - && reginclass(prog, p, (U8*)scan, (U8*) this_eol, TRUE)) + } + break; + + case ANYOFHb: + if (utf8_target) { /* ANYOFHb only can match UTF-8 targets */ + + /* we know the first byte must be the FLAGS field */ + while ( hardcount < max + && scan < this_eol + && (U8) *scan == ANYOF_FLAGS(p) + && reginclass(prog, p, (U8*)scan, (U8*) this_eol, + TRUE)) { scan += UTF8SKIP(scan); hardcount++; @@ -9844,7 +9859,7 @@ STATIC bool S_reginclass(pTHX_ regexp * const prog, const regnode * const n, const U8* const p, const U8* const p_end, const bool utf8_target) { dVAR; - const char flags = (OP(n) == ANYOFH) ? 0 : ANYOF_FLAGS(n); + const char flags = (OP(n) == ANYOFHb) ? 0 : ANYOF_FLAGS(n); bool match = FALSE; UV c = *p; @@ -9871,7 +9886,7 @@ S_reginclass(pTHX_ regexp * const prog, const regnode * const n, const U8* const } /* If this character is potentially in the bitmap, check it */ - if (c < NUM_ANYOF_CODE_POINTS && OP(n) != ANYOFH) { + if (c < NUM_ANYOF_CODE_POINTS && OP(n) != ANYOFH && OP(n) != ANYOFHb) { if (ANYOF_BITMAP_TEST(n, c)) match = TRUE; else if ((flags diff --git a/regnodes.h b/regnodes.h index dd23029a2c..add19473a1 100644 --- a/regnodes.h +++ b/regnodes.h @@ -6,8 +6,8 @@ /* Regops and State definitions */ -#define REGNODE_MAX 101 -#define REGMATCH_STATE_MAX 141 +#define REGNODE_MAX 102 +#define REGMATCH_STATE_MAX 142 #define END 0 /* 0000 End of program. */ #define SUCCEED 1 /* 0x01 Return from a subroutine, basically. */ @@ -33,86 +33,87 @@ #define ANYOFD 19 /* 0x13 Like ANYOF, but /d is in effect */ #define ANYOFL 20 /* 0x14 Like ANYOF, but /l is in effect */ #define ANYOFPOSIXL 21 /* 0x15 Like ANYOFL, but matches [[:posix:]] classes */ -#define ANYOFH 22 /* 0x16 Like ANYOF, but only has "High" matches, none in the bitmap; non-zero flags "f" means "f" is the first UTF-8 byte shared in common by all code points matched */ -#define ANYOFM 23 /* 0x17 Like ANYOF, but matches an invariant byte as determined by the mask and arg */ -#define NANYOFM 24 /* 0x18 complement of ANYOFM */ -#define POSIXD 25 /* 0x19 Some [[:class:]] under /d; the FLAGS field gives which one */ -#define POSIXL 26 /* 0x1a Some [[:class:]] under /l; the FLAGS field gives which one */ -#define POSIXU 27 /* 0x1b Some [[:class:]] under /u; the FLAGS field gives which one */ -#define POSIXA 28 /* 0x1c Some [[:class:]] under /a; the FLAGS field gives which one */ -#define NPOSIXD 29 /* 0x1d complement of POSIXD, [[:^class:]] */ -#define NPOSIXL 30 /* 0x1e complement of POSIXL, [[:^class:]] */ -#define NPOSIXU 31 /* 0x1f complement of POSIXU, [[:^class:]] */ -#define NPOSIXA 32 /* 0x20 complement of POSIXA, [[:^class:]] */ -#define CLUMP 33 /* 0x21 Match any extended grapheme cluster sequence */ -#define BRANCH 34 /* 0x22 Match this alternative, or the next... */ -#define EXACT 35 /* 0x23 Match this string (preceded by length). */ -#define EXACTL 36 /* 0x24 Like EXACT, but /l is in effect (used so locale-related warnings can be checked for). */ -#define EXACTF 37 /* 0x25 Match this string using /id rules (w/len); (string not UTF-8, not guaranteed to be folded). */ -#define EXACTFL 38 /* 0x26 Match this string using /il rules (w/len); (string not guaranteed to be folded). */ -#define EXACTFU 39 /* 0x27 Match this string using /iu rules (w/len); (string folded iff in UTF-8; non-UTF8 folded length <= unfolded). */ -#define EXACTFAA 40 /* 0x28 Match this string using /iaa rules (w/len) (string folded iff in UTF-8; non-UTF8 folded length <= unfolded). */ -#define EXACTFUP 41 /* 0x29 Match this string using /iu rules (w/len); (string not UTF-8, not guaranteed to be folded; and its Problematic). */ -#define EXACTFLU8 42 /* 0x2a Like EXACTFU, but use /il, UTF-8, folded, and everything in it is above 255. */ -#define EXACTFAA_NO_TRIE 43 /* 0x2b Match this string using /iaa rules (w/len) (string not UTF-8, not guaranteed to be folded, not currently trie-able). */ -#define EXACT_ONLY8 44 /* 0x2c Like EXACT, but only UTF-8 encoded targets can match */ -#define EXACTFU_ONLY8 45 /* 0x2d Like EXACTFU, but only UTF-8 encoded targets can match */ -#define EXACTFU_S_EDGE 46 /* 0x2e /di rules, but nothing in it precludes /ui, except begins and/or ends with [Ss]; (string not UTF-8; compile-time only). */ -#define NOTHING 47 /* 0x2f Match empty string. */ -#define TAIL 48 /* 0x30 Match empty string. Can jump here from outside. */ -#define STAR 49 /* 0x31 Match this (simple) thing 0 or more times. */ -#define PLUS 50 /* 0x32 Match this (simple) thing 1 or more times. */ -#define CURLY 51 /* 0x33 Match this simple thing {n,m} times. */ -#define CURLYN 52 /* 0x34 Capture next-after-this simple thing */ -#define CURLYM 53 /* 0x35 Capture this medium-complex thing {n,m} times. */ -#define CURLYX 54 /* 0x36 Match this complex thing {n,m} times. */ -#define WHILEM 55 /* 0x37 Do curly processing and see if rest matches. */ -#define OPEN 56 /* 0x38 Mark this point in input as start of #n. */ -#define CLOSE 57 /* 0x39 Close corresponding OPEN of #n. */ -#define SROPEN 58 /* 0x3a Same as OPEN, but for script run */ -#define SRCLOSE 59 /* 0x3b Close preceding SROPEN */ -#define REF 60 /* 0x3c Match some already matched string */ -#define REFF 61 /* 0x3d Match already matched string, using /di rules. */ -#define REFFL 62 /* 0x3e Match already matched string, using /li rules. */ -#define REFFU 63 /* 0x3f Match already matched string, usng /ui. */ -#define REFFA 64 /* 0x40 Match already matched string, using /aai rules. */ -#define REFN 65 /* 0x41 Match some already matched string */ -#define REFFN 66 /* 0x42 Match already matched string, using /di rules. */ -#define REFFLN 67 /* 0x43 Match already matched string, using /li rules. */ -#define REFFUN 68 /* 0x44 Match already matched string, using /ui rules. */ -#define REFFAN 69 /* 0x45 Match already matched string, using /aai rules. */ -#define LONGJMP 70 /* 0x46 Jump far away. */ -#define BRANCHJ 71 /* 0x47 BRANCH with long offset. */ -#define IFMATCH 72 /* 0x48 Succeeds if the following matches; non-zero flags "f", next_off "o" means lookbehind assertion starting "f..(f-o)" characters before current */ -#define UNLESSM 73 /* 0x49 Fails if the following matches; non-zero flags "f", next_off "o" means lookbehind assertion starting "f..(f-o)" characters before current */ -#define SUSPEND 74 /* 0x4a "Independent" sub-RE. */ -#define IFTHEN 75 /* 0x4b Switch, should be preceded by switcher. */ -#define GROUPP 76 /* 0x4c Whether the group matched. */ -#define EVAL 77 /* 0x4d Execute some Perl code. */ -#define MINMOD 78 /* 0x4e Next operator is not greedy. */ -#define LOGICAL 79 /* 0x4f Next opcode should set the flag only. */ -#define RENUM 80 /* 0x50 Group with independently numbered parens. */ -#define TRIE 81 /* 0x51 Match many EXACT(F[ALU]?)? at once. flags==type */ -#define TRIEC 82 /* 0x52 Same as TRIE, but with embedded charclass data */ -#define AHOCORASICK 83 /* 0x53 Aho Corasick stclass. flags==type */ -#define AHOCORASICKC 84 /* 0x54 Same as AHOCORASICK, but with embedded charclass data */ -#define GOSUB 85 /* 0x55 recurse to paren arg1 at (signed) ofs arg2 */ -#define GROUPPN 86 /* 0x56 Whether the group matched. */ -#define INSUBP 87 /* 0x57 Whether we are in a specific recurse. */ -#define DEFINEP 88 /* 0x58 Never execute directly. */ -#define ENDLIKE 89 /* 0x59 Used only for the type field of verbs */ -#define OPFAIL 90 /* 0x5a Same as (?!), but with verb arg */ -#define ACCEPT 91 /* 0x5b Accepts the current matched string, with verbar */ -#define VERB 92 /* 0x5c Used only for the type field of verbs */ -#define PRUNE 93 /* 0x5d Pattern fails at this startpoint if no-backtracking through this */ -#define MARKPOINT 94 /* 0x5e Push the current location for rollback by cut. */ -#define SKIP 95 /* 0x5f On failure skip forward (to the mark) before retrying */ -#define COMMIT 96 /* 0x60 Pattern fails outright if backtracking through this */ -#define CUTGROUP 97 /* 0x61 On failure go to the next alternation in the group */ -#define KEEPS 98 /* 0x62 $& begins here. */ -#define LNBREAK 99 /* 0x63 generic newline pattern */ -#define OPTIMIZED 100 /* 0x64 Placeholder for dump. */ -#define PSEUDO 101 /* 0x65 Pseudo opcode for internal use. */ +#define ANYOFH 22 /* 0x16 Like ANYOF, but only has "High" matches, none in the bitmap; */ +#define ANYOFHb 23 /* 0x17 Like ANYOFH, but all matches share the same first byte, the repurposed flags field contains it */ +#define ANYOFM 24 /* 0x18 Like ANYOF, but matches an invariant byte as determined by the mask and arg */ +#define NANYOFM 25 /* 0x19 complement of ANYOFM */ +#define POSIXD 26 /* 0x1a Some [[:class:]] under /d; the FLAGS field gives which one */ +#define POSIXL 27 /* 0x1b Some [[:class:]] under /l; the FLAGS field gives which one */ +#define POSIXU 28 /* 0x1c Some [[:class:]] under /u; the FLAGS field gives which one */ +#define POSIXA 29 /* 0x1d Some [[:class:]] under /a; the FLAGS field gives which one */ +#define NPOSIXD 30 /* 0x1e complement of POSIXD, [[:^class:]] */ +#define NPOSIXL 31 /* 0x1f complement of POSIXL, [[:^class:]] */ +#define NPOSIXU 32 /* 0x20 complement of POSIXU, [[:^class:]] */ +#define NPOSIXA 33 /* 0x21 complement of POSIXA, [[:^class:]] */ +#define CLUMP 34 /* 0x22 Match any extended grapheme cluster sequence */ +#define BRANCH 35 /* 0x23 Match this alternative, or the next... */ +#define EXACT 36 /* 0x24 Match this string (preceded by length). */ +#define EXACTL 37 /* 0x25 Like EXACT, but /l is in effect (used so locale-related warnings can be checked for). */ +#define EXACTF 38 /* 0x26 Match this string using /id rules (w/len); (string not UTF-8, not guaranteed to be folded). */ +#define EXACTFL 39 /* 0x27 Match this string using /il rules (w/len); (string not guaranteed to be folded). */ +#define EXACTFU 40 /* 0x28 Match this string using /iu rules (w/len); (string folded iff in UTF-8; non-UTF8 folded length <= unfolded). */ +#define EXACTFAA 41 /* 0x29 Match this string using /iaa rules (w/len) (string folded iff in UTF-8; non-UTF8 folded length <= unfolded). */ +#define EXACTFUP 42 /* 0x2a Match this string using /iu rules (w/len); (string not UTF-8, not guaranteed to be folded; and its Problematic). */ +#define EXACTFLU8 43 /* 0x2b Like EXACTFU, but use /il, UTF-8, folded, and everything in it is above 255. */ +#define EXACTFAA_NO_TRIE 44 /* 0x2c Match this string using /iaa rules (w/len) (string not UTF-8, not guaranteed to be folded, not currently trie-able). */ +#define EXACT_ONLY8 45 /* 0x2d Like EXACT, but only UTF-8 encoded targets can match */ +#define EXACTFU_ONLY8 46 /* 0x2e Like EXACTFU, but only UTF-8 encoded targets can match */ +#define EXACTFU_S_EDGE 47 /* 0x2f /di rules, but nothing in it precludes /ui, except begins and/or ends with [Ss]; (string not UTF-8; compile-time only). */ +#define NOTHING 48 /* 0x30 Match empty string. */ +#define TAIL 49 /* 0x31 Match empty string. Can jump here from outside. */ +#define STAR 50 /* 0x32 Match this (simple) thing 0 or more times. */ +#define PLUS 51 /* 0x33 Match this (simple) thing 1 or more times. */ +#define CURLY 52 /* 0x34 Match this simple thing {n,m} times. */ +#define CURLYN 53 /* 0x35 Capture next-after-this simple thing */ +#define CURLYM 54 /* 0x36 Capture this medium-complex thing {n,m} times. */ +#define CURLYX 55 /* 0x37 Match this complex thing {n,m} times. */ +#define WHILEM 56 /* 0x38 Do curly processing and see if rest matches. */ +#define OPEN 57 /* 0x39 Mark this point in input as start of #n. */ +#define CLOSE 58 /* 0x3a Close corresponding OPEN of #n. */ +#define SROPEN 59 /* 0x3b Same as OPEN, but for script run */ +#define SRCLOSE 60 /* 0x3c Close preceding SROPEN */ +#define REF 61 /* 0x3d Match some already matched string */ +#define REFF 62 /* 0x3e Match already matched string, using /di rules. */ +#define REFFL 63 /* 0x3f Match already matched string, using /li rules. */ +#define REFFU 64 /* 0x40 Match already matched string, usng /ui. */ +#define REFFA 65 /* 0x41 Match already matched string, using /aai rules. */ +#define REFN 66 /* 0x42 Match some already matched string */ +#define REFFN 67 /* 0x43 Match already matched string, using /di rules. */ +#define REFFLN 68 /* 0x44 Match already matched string, using /li rules. */ +#define REFFUN 69 /* 0x45 Match already matched string, using /ui rules. */ +#define REFFAN 70 /* 0x46 Match already matched string, using /aai rules. */ +#define LONGJMP 71 /* 0x47 Jump far away. */ +#define BRANCHJ 72 /* 0x48 BRANCH with long offset. */ +#define IFMATCH 73 /* 0x49 Succeeds if the following matches; non-zero flags "f", next_off "o" means lookbehind assertion starting "f..(f-o)" characters before current */ +#define UNLESSM 74 /* 0x4a Fails if the following matches; non-zero flags "f", next_off "o" means lookbehind assertion starting "f..(f-o)" characters before current */ +#define SUSPEND 75 /* 0x4b "Independent" sub-RE. */ +#define IFTHEN 76 /* 0x4c Switch, should be preceded by switcher. */ +#define GROUPP 77 /* 0x4d Whether the group matched. */ +#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[ALU]?)? 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 GROUPPN 87 /* 0x57 Whether the group matched. */ +#define INSUBP 88 /* 0x58 Whether we are in a specific recurse. */ +#define DEFINEP 89 /* 0x59 Never execute directly. */ +#define ENDLIKE 90 /* 0x5a Used only for the type field of verbs */ +#define OPFAIL 91 /* 0x5b Same as (?!), but with verb arg */ +#define ACCEPT 92 /* 0x5c Accepts the current matched string, with verbar */ +#define VERB 93 /* 0x5d Used only for the type field of verbs */ +#define PRUNE 94 /* 0x5e Pattern fails at this startpoint if no-backtracking through this */ +#define MARKPOINT 95 /* 0x5f Push the current location for rollback by cut. */ +#define SKIP 96 /* 0x60 On failure skip forward (to the mark) before retrying */ +#define COMMIT 97 /* 0x61 Pattern fails outright if backtracking through this */ +#define CUTGROUP 98 /* 0x62 On failure go to the next alternation in the group */ +#define KEEPS 99 /* 0x63 $& begins here. */ +#define LNBREAK 100 /* 0x64 generic newline pattern */ +#define OPTIMIZED 101 /* 0x65 Placeholder for dump. */ +#define PSEUDO 102 /* 0x66 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 */ @@ -184,6 +185,7 @@ EXTCONST U8 PL_regkind[] = { ANYOF, /* ANYOFL */ ANYOF, /* ANYOFPOSIXL */ ANYOF, /* ANYOFH */ + ANYOF, /* ANYOFHb */ ANYOFM, /* ANYOFM */ ANYOFM, /* NANYOFM */ POSIXD, /* POSIXD */ @@ -335,6 +337,7 @@ static const U8 regarglen[] = { EXTRA_SIZE(struct regnode_charclass), /* ANYOFL */ EXTRA_SIZE(struct regnode_charclass_posixl), /* ANYOFPOSIXL */ EXTRA_SIZE(struct regnode_1), /* ANYOFH */ + EXTRA_SIZE(struct regnode_1), /* ANYOFHb */ EXTRA_SIZE(struct regnode_1), /* ANYOFM */ EXTRA_SIZE(struct regnode_1), /* NANYOFM */ 0, /* POSIXD */ @@ -442,6 +445,7 @@ static const char reg_off_by_arg[] = { 0, /* ANYOFL */ 0, /* ANYOFPOSIXL */ 0, /* ANYOFH */ + 0, /* ANYOFHb */ 0, /* ANYOFM */ 0, /* NANYOFM */ 0, /* POSIXD */ @@ -555,85 +559,86 @@ EXTCONST char * const PL_reg_name[] = { "ANYOFL", /* 0x14 */ "ANYOFPOSIXL", /* 0x15 */ "ANYOFH", /* 0x16 */ - "ANYOFM", /* 0x17 */ - "NANYOFM", /* 0x18 */ - "POSIXD", /* 0x19 */ - "POSIXL", /* 0x1a */ - "POSIXU", /* 0x1b */ - "POSIXA", /* 0x1c */ - "NPOSIXD", /* 0x1d */ - "NPOSIXL", /* 0x1e */ - "NPOSIXU", /* 0x1f */ - "NPOSIXA", /* 0x20 */ - "CLUMP", /* 0x21 */ - "BRANCH", /* 0x22 */ - "EXACT", /* 0x23 */ - "EXACTL", /* 0x24 */ - "EXACTF", /* 0x25 */ - "EXACTFL", /* 0x26 */ - "EXACTFU", /* 0x27 */ - "EXACTFAA", /* 0x28 */ - "EXACTFUP", /* 0x29 */ - "EXACTFLU8", /* 0x2a */ - "EXACTFAA_NO_TRIE", /* 0x2b */ - "EXACT_ONLY8", /* 0x2c */ - "EXACTFU_ONLY8", /* 0x2d */ - "EXACTFU_S_EDGE", /* 0x2e */ - "NOTHING", /* 0x2f */ - "TAIL", /* 0x30 */ - "STAR", /* 0x31 */ - "PLUS", /* 0x32 */ - "CURLY", /* 0x33 */ - "CURLYN", /* 0x34 */ - "CURLYM", /* 0x35 */ - "CURLYX", /* 0x36 */ - "WHILEM", /* 0x37 */ - "OPEN", /* 0x38 */ - "CLOSE", /* 0x39 */ - "SROPEN", /* 0x3a */ - "SRCLOSE", /* 0x3b */ - "REF", /* 0x3c */ - "REFF", /* 0x3d */ - "REFFL", /* 0x3e */ - "REFFU", /* 0x3f */ - "REFFA", /* 0x40 */ - "REFN", /* 0x41 */ - "REFFN", /* 0x42 */ - "REFFLN", /* 0x43 */ - "REFFUN", /* 0x44 */ - "REFFAN", /* 0x45 */ - "LONGJMP", /* 0x46 */ - "BRANCHJ", /* 0x47 */ - "IFMATCH", /* 0x48 */ - "UNLESSM", /* 0x49 */ - "SUSPEND", /* 0x4a */ - "IFTHEN", /* 0x4b */ - "GROUPP", /* 0x4c */ - "EVAL", /* 0x4d */ - "MINMOD", /* 0x4e */ - "LOGICAL", /* 0x4f */ - "RENUM", /* 0x50 */ - "TRIE", /* 0x51 */ - "TRIEC", /* 0x52 */ - "AHOCORASICK", /* 0x53 */ - "AHOCORASICKC", /* 0x54 */ - "GOSUB", /* 0x55 */ - "GROUPPN", /* 0x56 */ - "INSUBP", /* 0x57 */ - "DEFINEP", /* 0x58 */ - "ENDLIKE", /* 0x59 */ - "OPFAIL", /* 0x5a */ - "ACCEPT", /* 0x5b */ - "VERB", /* 0x5c */ - "PRUNE", /* 0x5d */ - "MARKPOINT", /* 0x5e */ - "SKIP", /* 0x5f */ - "COMMIT", /* 0x60 */ - "CUTGROUP", /* 0x61 */ - "KEEPS", /* 0x62 */ - "LNBREAK", /* 0x63 */ - "OPTIMIZED", /* 0x64 */ - "PSEUDO", /* 0x65 */ + "ANYOFHb", /* 0x17 */ + "ANYOFM", /* 0x18 */ + "NANYOFM", /* 0x19 */ + "POSIXD", /* 0x1a */ + "POSIXL", /* 0x1b */ + "POSIXU", /* 0x1c */ + "POSIXA", /* 0x1d */ + "NPOSIXD", /* 0x1e */ + "NPOSIXL", /* 0x1f */ + "NPOSIXU", /* 0x20 */ + "NPOSIXA", /* 0x21 */ + "CLUMP", /* 0x22 */ + "BRANCH", /* 0x23 */ + "EXACT", /* 0x24 */ + "EXACTL", /* 0x25 */ + "EXACTF", /* 0x26 */ + "EXACTFL", /* 0x27 */ + "EXACTFU", /* 0x28 */ + "EXACTFAA", /* 0x29 */ + "EXACTFUP", /* 0x2a */ + "EXACTFLU8", /* 0x2b */ + "EXACTFAA_NO_TRIE", /* 0x2c */ + "EXACT_ONLY8", /* 0x2d */ + "EXACTFU_ONLY8", /* 0x2e */ + "EXACTFU_S_EDGE", /* 0x2f */ + "NOTHING", /* 0x30 */ + "TAIL", /* 0x31 */ + "STAR", /* 0x32 */ + "PLUS", /* 0x33 */ + "CURLY", /* 0x34 */ + "CURLYN", /* 0x35 */ + "CURLYM", /* 0x36 */ + "CURLYX", /* 0x37 */ + "WHILEM", /* 0x38 */ + "OPEN", /* 0x39 */ + "CLOSE", /* 0x3a */ + "SROPEN", /* 0x3b */ + "SRCLOSE", /* 0x3c */ + "REF", /* 0x3d */ + "REFF", /* 0x3e */ + "REFFL", /* 0x3f */ + "REFFU", /* 0x40 */ + "REFFA", /* 0x41 */ + "REFN", /* 0x42 */ + "REFFN", /* 0x43 */ + "REFFLN", /* 0x44 */ + "REFFUN", /* 0x45 */ + "REFFAN", /* 0x46 */ + "LONGJMP", /* 0x47 */ + "BRANCHJ", /* 0x48 */ + "IFMATCH", /* 0x49 */ + "UNLESSM", /* 0x4a */ + "SUSPEND", /* 0x4b */ + "IFTHEN", /* 0x4c */ + "GROUPP", /* 0x4d */ + "EVAL", /* 0x4e */ + "MINMOD", /* 0x4f */ + "LOGICAL", /* 0x50 */ + "RENUM", /* 0x51 */ + "TRIE", /* 0x52 */ + "TRIEC", /* 0x53 */ + "AHOCORASICK", /* 0x54 */ + "AHOCORASICKC", /* 0x55 */ + "GOSUB", /* 0x56 */ + "GROUPPN", /* 0x57 */ + "INSUBP", /* 0x58 */ + "DEFINEP", /* 0x59 */ + "ENDLIKE", /* 0x5a */ + "OPFAIL", /* 0x5b */ + "ACCEPT", /* 0x5c */ + "VERB", /* 0x5d */ + "PRUNE", /* 0x5e */ + "MARKPOINT", /* 0x5f */ + "SKIP", /* 0x60 */ + "COMMIT", /* 0x61 */ + "CUTGROUP", /* 0x62 */ + "KEEPS", /* 0x63 */ + "LNBREAK", /* 0x64 */ + "OPTIMIZED", /* 0x65 */ + "PSEUDO", /* 0x66 */ /* ------------ States ------------- */ "TRIE_next", /* REGNODE_MAX +0x01 */ "TRIE_next_fail", /* REGNODE_MAX +0x02 */ @@ -768,7 +773,7 @@ EXTCONST U8 PL_varies[] __attribute__deprecated__ = { EXTCONST U8 PL_varies_bitmask[]; #else EXTCONST U8 PL_varies_bitmask[] = { - 0x00, 0x00, 0x00, 0x00, 0x06, 0x00, 0xFE, 0xF0, 0xBF, 0x0C, 0x00, 0x00, 0x00 + 0x00, 0x00, 0x00, 0x00, 0x0C, 0x00, 0xFC, 0xE1, 0x7F, 0x19, 0x00, 0x00, 0x00 }; #endif /* DOINIT */ @@ -780,9 +785,9 @@ EXTCONST U8 PL_varies_bitmask[] = { EXTCONST U8 PL_simple[] __attribute__deprecated__; #else EXTCONST U8 PL_simple[] __attribute__deprecated__ = { - REG_ANY, SANY, ANYOF, ANYOFD, ANYOFL, ANYOFPOSIXL, ANYOFH, ANYOFM, - NANYOFM, POSIXD, POSIXL, POSIXU, POSIXA, NPOSIXD, NPOSIXL, NPOSIXU, - NPOSIXA, + REG_ANY, SANY, ANYOF, ANYOFD, ANYOFL, ANYOFPOSIXL, ANYOFH, ANYOFHb, + ANYOFM, NANYOFM, POSIXD, POSIXL, POSIXU, POSIXA, NPOSIXD, NPOSIXL, + NPOSIXU, NPOSIXA, 0 }; #endif /* DOINIT */ @@ -791,7 +796,7 @@ EXTCONST U8 PL_simple[] __attribute__deprecated__ = { EXTCONST U8 PL_simple_bitmask[]; #else EXTCONST U8 PL_simple_bitmask[] = { - 0x00, 0x00, 0xFF, 0xFF, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 + 0x00, 0x00, 0xFF, 0xFF, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }; #endif /* DOINIT */ diff --git a/t/re/anyof.t b/t/re/anyof.t index d33cbb2abe..aafcdb090c 100644 --- a/t/re/anyof.t +++ b/t/re/anyof.t @@ -478,109 +478,109 @@ my @tests = ( '[\x{00}-{INFTY_minus_1}]' => 'ANYOF[\x00-\xFF][0100-INFTY_minus_1]', '[\x{00}-{INFTY}]' => 'SANY', - '(?i)[\x{100}]' => 'ANYOFH[0100-0101]', + '(?i)[\x{100}]' => 'ANYOFHb[0100-0101]', '[\x{101}-{INFTY}]' => 'ANYOFH[0101-INFTY]', '[\x{101}-{INFTY_minus_1}]' => 'ANYOFH[0101-INFTY_minus_1]', - '[\x{102}\x{104}]' => 'ANYOFH[0102 0104]', + '[\x{102}\x{104}]' => 'ANYOFHb[0102 0104]', '[\x{102}-\x{104}{INFTY}]' => 'ANYOFH[0102-0104 INFTY-INFTY]', '[\x{102}-\x{104}{INFTY_minus_1}]' => 'ANYOFH[0102-0104 INFTY_minus_1]', - '[\x{102}-\x{104}\x{101}]' => 'ANYOFH[0101-0104]', + '[\x{102}-\x{104}\x{101}]' => 'ANYOFHb[0101-0104]', '[\x{102}-\x{104}\x{101}-{INFTY}]' => 'ANYOFH[0101-INFTY]', '[\x{102}-\x{104}\x{101}-{INFTY_minus_1}]' => 'ANYOFH[0101-INFTY_minus_1]', - '[\x{102}-\x{104}\x{102}]' => 'ANYOFH[0102-0104]', + '[\x{102}-\x{104}\x{102}]' => 'ANYOFHb[0102-0104]', '[\x{102}-\x{104}\x{102}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{102}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{103}]' => 'ANYOFH[0102-0104]', + '[\x{102}-\x{104}\x{103}]' => 'ANYOFHb[0102-0104]', '[\x{102}-\x{104}\x{103}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{103}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{104}]' => 'ANYOFH[0102-0104]', + '[\x{102}-\x{104}\x{104}]' => 'ANYOFHb[0102-0104]', '[\x{102}-\x{104}\x{104}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{104}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{105}]' => 'ANYOFH[0102-0105]', + '[\x{102}-\x{104}\x{105}]' => 'ANYOFHb[0102-0105]', '[\x{102}-\x{104}\x{105}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{105}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{106}]' => 'ANYOFH[0102-0104 0106]', + '[\x{102}-\x{104}\x{106}]' => 'ANYOFHb[0102-0104 0106]', '[\x{102}-\x{104}\x{106}-{INFTY}]' => 'ANYOFH[0102-0104 0106-INFTY]', '[\x{102}-\x{104}\x{106}-{INFTY_minus_1}]' => 'ANYOFH[0102-0104 0106-INFTY_minus_1]', '[\x{102}-\x{104}\x{108}-\x{10A}{INFTY}]' => 'ANYOFH[0102-0104 0108-010A INFTY-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}{INFTY_minus_1}]' => 'ANYOFH[0102-0104 0108-010A INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}]' => 'ANYOFH[0101-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}]' => 'ANYOFHb[0101-0104 0108-010A]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-{INFTY}]' => 'ANYOFH[0101-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-{INFTY_minus_1}]' => 'ANYOFH[0101-INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{102}]' => 'ANYOFH[0101-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{103}]' => 'ANYOFH[0101-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{104}]' => 'ANYOFH[0101-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{105}]' => 'ANYOFH[0101-0105 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{106}]' => 'ANYOFH[0101-0106 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{107}]' => 'ANYOFH[0101-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{108}]' => 'ANYOFH[0101-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{109}]' => 'ANYOFH[0101-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{10A}]' => 'ANYOFH[0101-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{10B}]' => 'ANYOFH[0101-010B]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}]' => 'ANYOFH[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{102}]' => 'ANYOFHb[0101-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{103}]' => 'ANYOFHb[0101-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{104}]' => 'ANYOFHb[0101-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{105}]' => 'ANYOFHb[0101-0105 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{106}]' => 'ANYOFHb[0101-0106 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{107}]' => 'ANYOFHb[0101-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{108}]' => 'ANYOFHb[0101-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{109}]' => 'ANYOFHb[0101-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{10A}]' => 'ANYOFHb[0101-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{101}-\x{10B}]' => 'ANYOFHb[0101-010B]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}]' => 'ANYOFHb[0102-0104 0108-010A]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{102}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{103}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{104}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{105}]' => 'ANYOFH[0102-0105 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{106}]' => 'ANYOFH[0102-0106 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{107}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{108}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{109}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{10A}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{10B}]' => 'ANYOFH[0102-010B]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{10C}]' => 'ANYOFH[0102-010C]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}]' => 'ANYOFH[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{102}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{103}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{104}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{105}]' => 'ANYOFHb[0102-0105 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{106}]' => 'ANYOFHb[0102-0106 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{107}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{108}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{109}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{10A}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{10B}]' => 'ANYOFHb[0102-010B]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{102}-\x{10C}]' => 'ANYOFHb[0102-010C]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}]' => 'ANYOFHb[0102-0104 0108-010A]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{104}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{105}]' => 'ANYOFH[0102-0105 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{106}]' => 'ANYOFH[0102-0106 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{107}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{108}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{109}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{10A}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{10B}]' => 'ANYOFH[0102-010B]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{10C}]' => 'ANYOFH[0102-010C]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}]' => 'ANYOFH[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{104}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{105}]' => 'ANYOFHb[0102-0105 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{106}]' => 'ANYOFHb[0102-0106 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{107}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{108}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{109}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{10A}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{10B}]' => 'ANYOFHb[0102-010B]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{103}-\x{10C}]' => 'ANYOFHb[0102-010C]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}]' => 'ANYOFHb[0102-0104 0108-010A]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{105}]' => 'ANYOFH[0102-0105 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{106}]' => 'ANYOFH[0102-0106 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{107}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{108}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{109}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{10A}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{10B}]' => 'ANYOFH[0102-010B]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{10C}]' => 'ANYOFH[0102-010C]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}]' => 'ANYOFH[0102-0105 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{105}]' => 'ANYOFHb[0102-0105 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{106}]' => 'ANYOFHb[0102-0106 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{107}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{108}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{109}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{10A}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{10B}]' => 'ANYOFHb[0102-010B]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{104}-\x{10C}]' => 'ANYOFHb[0102-010C]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}]' => 'ANYOFHb[0102-0105 0108-010A]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-{INFTY}]' => 'ANYOFH[0102-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-{INFTY_minus_1}]' => 'ANYOFH[0102-INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{106}]' => 'ANYOFH[0102-0106 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{107}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{108}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{109}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{10A}]' => 'ANYOFH[0102-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{10B}]' => 'ANYOFH[0102-010B]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{10C}]' => 'ANYOFH[0102-010C]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}]' => 'ANYOFH[0102-0104 0106 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{106}]' => 'ANYOFHb[0102-0106 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{107}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{108}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{109}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{10A}]' => 'ANYOFHb[0102-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{10B}]' => 'ANYOFHb[0102-010B]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{105}-\x{10C}]' => 'ANYOFHb[0102-010C]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}]' => 'ANYOFHb[0102-0104 0106 0108-010A]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-{INFTY}]' => 'ANYOFH[0102-0104 0106-INFTY]', '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-{INFTY_minus_1}]' => 'ANYOFH[0102-0104 0106-INFTY_minus_1]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{107}]' => 'ANYOFH[0102-0104 0106-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{108}]' => 'ANYOFH[0102-0104 0106-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{109}]' => 'ANYOFH[0102-0104 0106-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{10A}]' => 'ANYOFH[0102-0104 0106-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{10B}]' => 'ANYOFH[0102-0104 0106-010B]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{10C}]' => 'ANYOFH[0102-0104 0106-010C]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{107}]' => 'ANYOFH[0102-0104 0107-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{108}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{109}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{10A}]' => 'ANYOFH[0102-0104 0108-010A]', - '[\x{102}-\x{104}\x{108}-\x{10A}\x{10B}]' => 'ANYOFH[0102-0104 0108-010B]', - '[\x{103}\x{102}]' => 'ANYOFH[0102-0103]', - '[\x{104}\x{102}]' => 'ANYOFH[0102 0104]', - '[\x{104}\x{102}\x{103}]' => 'ANYOFH[0102-0104]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{107}]' => 'ANYOFHb[0102-0104 0106-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{108}]' => 'ANYOFHb[0102-0104 0106-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{109}]' => 'ANYOFHb[0102-0104 0106-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{10A}]' => 'ANYOFHb[0102-0104 0106-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{10B}]' => 'ANYOFHb[0102-0104 0106-010B]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{106}-\x{10C}]' => 'ANYOFHb[0102-0104 0106-010C]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{107}]' => 'ANYOFHb[0102-0104 0107-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{108}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{109}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{10A}]' => 'ANYOFHb[0102-0104 0108-010A]', + '[\x{102}-\x{104}\x{108}-\x{10A}\x{10B}]' => 'ANYOFHb[0102-0104 0108-010B]', + '[\x{103}\x{102}]' => 'ANYOFHb[0102-0103]', + '[\x{104}\x{102}]' => 'ANYOFHb[0102 0104]', + '[\x{104}\x{102}\x{103}]' => 'ANYOFHb[0102-0104]', '[\x{106}-{INFTY}\x{104}]' => 'ANYOFH[0104 0106-INFTY]', '[\x{106}-{INFTY}\x{104}-{INFTY}]' => 'ANYOFH[0104-INFTY]', '[\x{106}-{INFTY}\x{104}-{INFTY_minus_1}]' => 'ANYOFH[0104-INFTY]', @@ -713,7 +713,7 @@ my @tests = ( '(?8)(?i)[\x{100}]' => 'EXACTFU_ONLY8 <\x{101}>', '(?8)(?i)[\x{399}]' => 'EXACTFU_ONLY8 <\x{3b9}>', '(?8)(?i)[\x{345}\x{399}\x{3B9}\x{1FBE}]' => 'EXACTFU_ONLY8 <\x{3b9}>', - '(?i)[\x{2b9}]' => 'ANYOFH[02B9]', # Doesn't participate in a fold + '(?i)[\x{2b9}]' => 'ANYOFHb[02B9]', # Doesn't participate in a fold '(?8)(?i)[\x{2b9}]' => 'EXACT_ONLY8 <\x{2b9}>', '(?i)[\x{2bc}]' => 'EXACTFU_ONLY8 <\x{2bc}>', # Part of a multi-char fold, ASCII component '(?i)[\x{390}]' => 'EXACTFU_ONLY8 <\x{3b9}\x{308}\x{301}>', # Part of a multi-char fold, no ASCII component @@ -721,7 +721,7 @@ my @tests = ( '(?i)[\x{1E9E}]' => 'EXACTFU <ss>', '(?iaa)[\x{1E9E}]' => 'EXACTFAA <\x{17f}\x{17f}>', '(?i)[\x{FB00}]' => 'EXACTFU <ff>', - '(?iaa)[\x{FB00}]' => 'ANYOFH[FB00]', + '(?iaa)[\x{FB00}]' => 'ANYOFHb[FB00]', '(?i)[\x{FB00}]' => 'EXACTFU <ff>', '(?i)[\x{FB01}]' => 'EXACTFU <fi>', '(?i)[\x{FB02}]' => 'EXACTFU <fl>', @@ -841,7 +841,7 @@ for my $char (@single_chars_to_test) { else { die join ", ", @list; } - my $anyof = ($charset eq "l") ? "ANYOFL" : "ANYOFH"; + my $anyof = ($charset eq "l") ? "ANYOFL" : "ANYOFHb"; push @single_tests, "$anyof\[$interior\]"; } } -- Perl5 Master Repository
