X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/bbe252da68db3f0a4d00844fdb5e30bf18a0828f..1af6200d185d478f416345eedc9641a043650e1b:/regexp.h diff --git a/regexp.h b/regexp.h index 1355e1e..ad59cc7 100644 --- a/regexp.h +++ b/regexp.h @@ -1,7 +1,7 @@ /* regexp.h * * Copyright (C) 1993, 1994, 1996, 1997, 1999, 2000, 2001, 2003, - * by Larry Wall and others + * 2005, 2006, 2007, by Larry Wall and others * * You may distribute under the terms of either the GNU General Public * License or the Artistic License, as specified in the README file. @@ -15,8 +15,8 @@ * not the System V one. */ #ifndef PLUGGABLE_RE_EXTENSION -/* we don't want to include this stuff if we are inside Nicholas' - * pluggable regex engine code */ +/* we don't want to include this stuff if we are inside of + an external regex engine based on the core one - like re 'debug'*/ struct regnode { U8 flags; @@ -31,11 +31,18 @@ struct reg_substr_data; struct reg_data; struct regexp_engine; - -typedef struct regexp_paren_ofs { - I32 *startp; - I32 *endp; -} regexp_paren_ofs; +struct regexp; + +struct reg_substr_datum { + I32 min_offset; + I32 max_offset; + SV *substr; /* non-utf8 variant */ + SV *utf8_substr; /* utf8 variant */ + I32 end_shift; +}; +struct reg_substr_data { + struct reg_substr_datum data[3]; /* Actual array */ +}; #ifdef PERL_OLD_COPY_ON_WRITE #define SV_SAVED_COPY SV *saved_copy; /* If non-NULL, SV which is COW from original */ @@ -43,188 +50,372 @@ typedef struct regexp_paren_ofs { #define SV_SAVED_COPY #endif +typedef struct regexp_paren_pair { + I32 start; + I32 end; +} regexp_paren_pair; + +/* + The regexp/REGEXP struct, see L for further documentation + on the individual fields. The struct is ordered so that the most + commonly used fields are placed at the start. + + Any patch that adds items to this struct will need to include + changes to F (C) and F + (C). This involves freeing or cloning items in the + regexp's data array based on the data item's type. +*/ + typedef struct regexp { - /* Generic details */ - const struct regexp_engine* engine; /* what created this regexp? */ - I32 refcnt; /* Refcount of this regexp */ - - /* The original string as passed to the compilation routine */ - char *precomp; /* pre-compilation regular expression */ - I32 prelen; /* length of precomp */ - - /* Used for generic optimisations by the perl core. - All engines are expected to provide this information. */ + _XPV_HEAD; + _XPVMG_HEAD; + /* what engine created this regexp? */ + const struct regexp_engine* engine; + REGEXP *mother_re; /* what re is this a lightweight copy of? */ + + /* Information about the match that the perl core uses to manage things */ U32 extflags; /* Flags used both externally and internally */ I32 minlen; /* mininum possible length of string to match */ I32 minlenret; /* mininum possible length of $& */ U32 gofs; /* chars left of pos that we search from */ - U32 nparens; /* number of capture buffers */ - HV *paren_names; /* Optional hash of paren names */ - struct reg_substr_data *substrs; /* substring data about strings that must appear + struct reg_substr_data *substrs; /* substring data about strings that must appear in the final match, used for optimisations */ + U32 nparens; /* number of capture buffers */ + + /* private engine specific data */ + U32 intflags; /* Engine Specific Internal flags */ + void *pprivate; /* Data private to the regex engine which + created this object. */ + + /* Data about the last/current match. These are modified during matching*/ + U32 lastparen; /* last open paren matched */ + U32 lastcloseparen; /* last close paren matched */ + regexp_paren_pair *swap; /* Swap copy of *offs */ + regexp_paren_pair *offs; /* Array of offsets for (@-) and (@+) */ - /* Data about the last/current match. Used by the core and therefore - must be populated by all engines. */ char *subbeg; /* saved or original string so \digit works forever. */ - I32 sublen; /* Length of string pointed by subbeg */ - I32 *startp; /* Array of offsets from start of string (@-) */ - I32 *endp; /* Array of offsets from start of string (@+) */ - SV_SAVED_COPY /* If non-NULL, SV which is COW from original */ - U32 lastparen; /* last open paren matched */ - U32 lastcloseparen; /* last close paren matched */ - - /* Perl Regex Engine specific data. Other engines shouldn't need - to touch this. Should be refactored out into a different structure - and accessed via the *pprivate field. (except intflags) */ - U32 intflags; /* Internal flags */ - void *pprivate; /* Data private to the regex engine which - created this object. Perl will never mess with - this member at all. */ - regexp_paren_ofs *swap; /* Swap copy of *startp / *endp */ - U32 *offsets; /* offset annotations 20001228 MJD - data about mapping the program to the - string*/ - regnode *regstclass; /* Optional startclass as identified or constructed - by the optimiser */ - struct reg_data *data; /* Additional miscellaneous data used by the program. - Used to make it easier to clone and free arbitrary - data that the regops need. Often the ARG field of - a regop is an index into this structure */ - regnode program[1]; /* Unwarranted chumminess with compiler. */ + I32 sublen; /* Length of string pointed by subbeg */ + + + /* Information about the match that isn't often used */ + unsigned pre_prefix:4; /* offset from wrapped to the start of precomp */ + unsigned seen_evals:28; /* number of eval groups in the pattern - for security checks */ } regexp; +/* HV *paren_names; Optional hash of paren names + now stored in the IV union */ + +#define RXp_PAREN_NAMES(rx) ((rx)->xiv_u.xivu_hv) +/* used for high speed searches */ typedef struct re_scream_pos_data_s { char **scream_olds; /* match pos */ I32 *scream_pos; /* Internal iterator of scream. */ } re_scream_pos_data; +/* regexp_engine structure. This is the dispatch table for regexes. + * Any regex engine implementation must be able to build one of these. + */ typedef struct regexp_engine { - regexp* (*comp) (pTHX_ char* exp, char* xend, PMOP* pm); - I32 (*exec) (pTHX_ regexp* prog, char* stringarg, char* strend, - char* strbeg, I32 minend, SV* screamer, - void* data, U32 flags); - char* (*intuit) (pTHX_ regexp *prog, SV *sv, char *strpos, - char *strend, U32 flags, - struct re_scream_pos_data_s *data); - SV* (*checkstr) (pTHX_ regexp *prog); - void (*free) (pTHX_ struct regexp* r); - char* (*as_str) (pTHX_ MAGIC *mg, STRLEN *lp, U32 *flags, I32 *haseval); + REGEXP* (*comp) (pTHX_ const SV * const pattern, const U32 flags); + I32 (*exec) (pTHX_ REGEXP * const rx, char* stringarg, char* strend, + char* strbeg, I32 minend, SV* screamer, + void* data, U32 flags); + char* (*intuit) (pTHX_ REGEXP * const rx, SV *sv, char *strpos, + char *strend, const U32 flags, + re_scream_pos_data *data); + SV* (*checkstr) (pTHX_ REGEXP * const rx); + void (*free) (pTHX_ REGEXP * const rx); + void (*numbered_buff_FETCH) (pTHX_ REGEXP * const rx, const I32 paren, + SV * const sv); + void (*numbered_buff_STORE) (pTHX_ REGEXP * const rx, const I32 paren, + SV const * const value); + I32 (*numbered_buff_LENGTH) (pTHX_ REGEXP * const rx, const SV * const sv, + const I32 paren); + SV* (*named_buff) (pTHX_ REGEXP * const rx, SV * const key, + SV * const value, const U32 flags); + SV* (*named_buff_iter) (pTHX_ REGEXP * const rx, const SV * const lastkey, + const U32 flags); + SV* (*qr_package)(pTHX_ REGEXP * const rx); #ifdef USE_ITHREADS - regexp* (*dupe) (pTHX_ const regexp *r, CLONE_PARAMS *param); -#endif + void* (*dupe) (pTHX_ REGEXP * const rx, CLONE_PARAMS *param); +#endif } regexp_engine; -/* - * Flags stored in regexp->intflags - * These are used only internally to the regexp engine - */ -#define PREGf_SKIP 0x00000001 -#define PREGf_IMPLICIT 0x00000002 /* Converted .* to ^.* */ -#define PREGf_NAUGHTY 0x00000004 /* how exponential is this pattern? */ -#define PREGf_VERBARG_SEEN 0x00000008 -#define PREGf_CUTGROUP_SEEN 0x00000010 +/* + These are passed to the numbered capture variable callbacks as the + paren name. >= 1 is reserved for actual numbered captures, i.e. $1, + $2 etc. +*/ +#define RX_BUFF_IDX_PREMATCH -2 /* $` / ${^PREMATCH} */ +#define RX_BUFF_IDX_POSTMATCH -1 /* $' / ${^POSTMATCH} */ +#define RX_BUFF_IDX_FULLMATCH 0 /* $& / ${^MATCH} */ + +/* + Flags that are passed to the named_buff and named_buff_iter + callbacks above. Those routines are called from universal.c via the + Tie::Hash::NamedCapture interface for %+ and %- and the re:: + functions in the same file. +*/ + +/* The Tie::Hash::NamedCapture operation this is part of, if any */ +#define RXapif_FETCH 0x0001 +#define RXapif_STORE 0x0002 +#define RXapif_DELETE 0x0004 +#define RXapif_CLEAR 0x0008 +#define RXapif_EXISTS 0x0010 +#define RXapif_SCALAR 0x0020 +#define RXapif_FIRSTKEY 0x0040 +#define RXapif_NEXTKEY 0x0080 + +/* Whether %+ or %- is being operated on */ +#define RXapif_ONE 0x0100 /* %+ */ +#define RXapif_ALL 0x0200 /* %- */ + +/* Whether this is being called from a re:: function */ +#define RXapif_REGNAME 0x0400 +#define RXapif_REGNAMES 0x0800 +#define RXapif_REGNAMES_COUNT 0x1000 + +/* +=head1 REGEXP Functions + +=for apidoc Am|REGEXP *|SvRX|SV *sv + +Convenience macro to get the REGEXP from a SV. This is approximately +equivalent to the following snippet: + + if (SvMAGICAL(sv)) + mg_get(sv); + if (SvROK(sv) && + (tmpsv = (SV*)SvRV(sv)) && + SvTYPE(tmpsv) == SVt_PVMG && + (tmpmg = mg_find(tmpsv, PERL_MAGIC_qr))) + { + return (REGEXP *)tmpmg->mg_obj; + } + +NULL will be returned if a REGEXP* is not found. + +=for apidoc Am|bool|SvRXOK|SV* sv + +Returns a boolean indicating whether the SV contains qr magic +(PERL_MAGIC_qr). + +If you want to do something with the REGEXP* later use SvRX instead +and check for NULL. + +=cut +*/ + +#define SvRX(sv) (Perl_get_re_arg(aTHX_ sv)) +#define SvRXOK(sv) (Perl_get_re_arg(aTHX_ sv) ? TRUE : FALSE) /* Flags stored in regexp->extflags * These are used by code external to the regexp engine + * + * Note that flags starting with RXf_PMf_ have exact equivalents + * stored in op_pmflags and which are defined in op.h, they are defined + * numerically here only for clarity. + * + * NOTE: if you modify any RXf flags you should run regen.pl or regcomp.pl + * so that regnodes.h is updated with the changes. + * + */ + +/* 0x3F of extflags is used by (RXf_)PMf_COMPILETIME + * If you change these you need to change the equivalent flags in op.h, and + * vice versa. */ +#define RXf_PMf_MULTILINE 0x00000001 /* /m */ +#define RXf_PMf_SINGLELINE 0x00000002 /* /s */ +#define RXf_PMf_FOLD 0x00000004 /* /i */ +#define RXf_PMf_EXTENDED 0x00000008 /* /x */ +#define RXf_PMf_KEEPCOPY 0x00000010 /* /p */ +#define RXf_PMf_LOCALE 0x00000020 /* use locale */ +/* these flags are transfered from the PMOP->op_pmflags member during compilation */ +#define RXf_PMf_STD_PMMOD_SHIFT 0 +#define RXf_PMf_STD_PMMOD (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_FOLD|RXf_PMf_EXTENDED) +#define RXf_PMf_COMPILETIME (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_LOCALE|RXf_PMf_FOLD|RXf_PMf_EXTENDED|RXf_PMf_KEEPCOPY) + +#define CASE_STD_PMMOD_FLAGS_PARSE_SET(pmfl) \ + case IGNORE_PAT_MOD: *(pmfl) |= RXf_PMf_FOLD; break; \ + case MULTILINE_PAT_MOD: *(pmfl) |= RXf_PMf_MULTILINE; break; \ + case SINGLE_PAT_MOD: *(pmfl) |= RXf_PMf_SINGLELINE; break; \ + case XTENDED_PAT_MOD: *(pmfl) |= RXf_PMf_EXTENDED; break + +/* chars and strings used as regex pattern modifiers + * Singlular is a 'c'har, plural is a "string" + * + * NOTE, KEEPCOPY was originally 'k', but was changed to 'p' for preserve + * for compatibility reasons with Regexp::Common which highjacked (?k:...) + * for its own uses. So 'k' is out as well. + */ +#define EXEC_PAT_MOD 'e' +#define KEEPCOPY_PAT_MOD 'p' +#define ONCE_PAT_MOD 'o' +#define GLOBAL_PAT_MOD 'g' +#define CONTINUE_PAT_MOD 'c' +#define MULTILINE_PAT_MOD 'm' +#define SINGLE_PAT_MOD 's' +#define IGNORE_PAT_MOD 'i' +#define XTENDED_PAT_MOD 'x' + +#define ONCE_PAT_MODS "o" +#define KEEPCOPY_PAT_MODS "p" +#define EXEC_PAT_MODS "e" +#define LOOP_PAT_MODS "gc" + +#define STD_PAT_MODS "msix" + +#define INT_PAT_MODS STD_PAT_MODS KEEPCOPY_PAT_MODS + +#define EXT_PAT_MODS ONCE_PAT_MODS KEEPCOPY_PAT_MODS +#define QR_PAT_MODS STD_PAT_MODS EXT_PAT_MODS +#define M_PAT_MODS QR_PAT_MODS LOOP_PAT_MODS +#define S_PAT_MODS M_PAT_MODS EXEC_PAT_MODS + +/* + * NOTE: if you modify any RXf flags you should run regen.pl or regcomp.pl + * so that regnodes.h is updated with the changes. + * */ /* Anchor and GPOS related stuff */ -#define RXf_ANCH_BOL 0x00000001 -#define RXf_ANCH_MBOL 0x00000002 -#define RXf_ANCH_SBOL 0x00000004 -#define RXf_ANCH_GPOS 0x00000008 -#define RXf_GPOS_SEEN 0x00000010 -#define RXf_GPOS_FLOAT 0x00000020 -/* five bits here */ +#define RXf_ANCH_BOL 0x00000100 +#define RXf_ANCH_MBOL 0x00000200 +#define RXf_ANCH_SBOL 0x00000400 +#define RXf_ANCH_GPOS 0x00000800 +#define RXf_GPOS_SEEN 0x00001000 +#define RXf_GPOS_FLOAT 0x00002000 +/* two bits here */ #define RXf_ANCH (RXf_ANCH_BOL|RXf_ANCH_MBOL|RXf_ANCH_GPOS|RXf_ANCH_SBOL) #define RXf_GPOS_CHECK (RXf_GPOS_SEEN|RXf_ANCH_GPOS) -#define RXf_ANCH_SINGLE (RXf_ANCH_SBOL|RXf_ANCH_GPOS) -/* - * 0xF800 of extflags is used by PMf_COMPILETIME - * These are the regex equivelent of the PMf_xyz stuff defined - * in op.h - */ -#define RXf_PMf_LOCALE 0x00000800 -#define RXf_PMf_MULTILINE 0x00001000 -#define RXf_PMf_SINGLELINE 0x00002000 -#define RXf_PMf_FOLD 0x00004000 -#define RXf_PMf_EXTENDED 0x00008000 -#define RXf_PMf_COMPILETIME (RXf_PMf_MULTILINE|RXf_PMf_SINGLELINE|RXf_PMf_LOCALE|RXf_PMf_FOLD|RXf_PMf_EXTENDED) +#define RXf_ANCH_SINGLE (RXf_ANCH_SBOL|RXf_ANCH_GPOS) /* What we have seen */ -/* one bit here */ -#define RXf_LOOKBEHIND_SEEN 0x00020000 -#define RXf_EVAL_SEEN 0x00040000 -#define RXf_CANY_SEEN 0x00080000 +#define RXf_LOOKBEHIND_SEEN 0x00004000 +#define RXf_EVAL_SEEN 0x00008000 +#define RXf_CANY_SEEN 0x00010000 /* Special */ -#define RXf_NOSCAN 0x00100000 -#define RXf_CHECK_ALL 0x00200000 +#define RXf_NOSCAN 0x00020000 +#define RXf_CHECK_ALL 0x00040000 /* UTF8 related */ -#define RXf_UTF8 0x00400000 -#define RXf_MATCH_UTF8 0x00800000 +#define RXf_MATCH_UTF8 0x00100000 /* Intuit related */ -#define RXf_USE_INTUIT_NOML 0x01000000 -#define RXf_USE_INTUIT_ML 0x02000000 -#define RXf_INTUIT_TAIL 0x04000000 -/* one bit here */ +#define RXf_USE_INTUIT_NOML 0x00200000 +#define RXf_USE_INTUIT_ML 0x00400000 +#define RXf_INTUIT_TAIL 0x00800000 + +/* + Set in Perl_pmruntime if op_flags & OPf_SPECIAL, i.e. split. Will + be used by regex engines to check whether they should set + RXf_SKIPWHITE +*/ +#define RXf_SPLIT 0x01000000 + #define RXf_USE_INTUIT (RXf_USE_INTUIT_NOML|RXf_USE_INTUIT_ML) /* Copy and tainted info */ -#define RXf_COPY_DONE 0x10000000 -#define RXf_TAINTED_SEEN 0x20000000 -/* two bits here */ +#define RXf_COPY_DONE 0x02000000 +#define RXf_TAINTED_SEEN 0x04000000 +#define RXf_TAINTED 0x08000000 /* this pattern is tainted */ +/* Flags indicating special patterns */ +#define RXf_START_ONLY 0x10000000 /* Pattern is /^/ */ +#define RXf_SKIPWHITE 0x20000000 /* Pattern is for a split / / */ +#define RXf_WHITE 0x40000000 /* Pattern is /\s+/ */ +#define RXf_NULL 0x80000000 /* Pattern is // */ + +/* + * NOTE: if you modify any RXf flags you should run regen.pl or regcomp.pl + * so that regnodes.h is updated with the changes. + * + */ #define RX_HAS_CUTGROUP(prog) ((prog)->intflags & PREGf_CUTGROUP_SEEN) -#define RX_MATCH_TAINTED(prog) ((prog)->extflags & RXf_TAINTED_SEEN) -#define RX_MATCH_TAINTED_on(prog) ((prog)->extflags |= RXf_TAINTED_SEEN) -#define RX_MATCH_TAINTED_off(prog) ((prog)->extflags &= ~RXf_TAINTED_SEEN) +#define RXp_MATCH_TAINTED(prog) (RXp_EXTFLAGS(prog) & RXf_TAINTED_SEEN) +#define RX_MATCH_TAINTED(prog) (RX_EXTFLAGS(prog) & RXf_TAINTED_SEEN) +#define RX_MATCH_TAINTED_on(prog) (RX_EXTFLAGS(prog) |= RXf_TAINTED_SEEN) +#define RX_MATCH_TAINTED_off(prog) (RX_EXTFLAGS(prog) &= ~RXf_TAINTED_SEEN) #define RX_MATCH_TAINTED_set(prog, t) ((t) \ ? RX_MATCH_TAINTED_on(prog) \ : RX_MATCH_TAINTED_off(prog)) -#define RX_MATCH_COPIED(prog) ((prog)->extflags & RXf_COPY_DONE) -#define RX_MATCH_COPIED_on(prog) ((prog)->extflags |= RXf_COPY_DONE) -#define RX_MATCH_COPIED_off(prog) ((prog)->extflags &= ~RXf_COPY_DONE) +#define RXp_MATCH_COPIED(prog) (RXp_EXTFLAGS(prog) & RXf_COPY_DONE) +#define RX_MATCH_COPIED(prog) (RX_EXTFLAGS(prog) & RXf_COPY_DONE) +#define RXp_MATCH_COPIED_on(prog) (RXp_EXTFLAGS(prog) |= RXf_COPY_DONE) +#define RX_MATCH_COPIED_on(prog) (RX_EXTFLAGS(prog) |= RXf_COPY_DONE) +#define RXp_MATCH_COPIED_off(prog) (RXp_EXTFLAGS(prog) &= ~RXf_COPY_DONE) +#define RX_MATCH_COPIED_off(prog) (RX_EXTFLAGS(prog) &= ~RXf_COPY_DONE) #define RX_MATCH_COPIED_set(prog,t) ((t) \ ? RX_MATCH_COPIED_on(prog) \ : RX_MATCH_COPIED_off(prog)) +#define RXp_EXTFLAGS(rx) ((rx)->extflags) + +/* For source compatibility. We used to store these explicitly. */ +#define RX_PRECOMP(prog) (RX_WRAPPED(prog) + ((struct regexp *)SvANY(prog))->pre_prefix) +/* FIXME? Are we hardcoding too much here and constraining plugin extension + writers? Specifically, the value 1 assumes that the wrapped version always + has exactly one character at the end, a ')'. Will that always be true? */ +#define RX_PRELEN(prog) (RX_WRAPLEN(prog) - ((struct regexp *)SvANY(prog))->pre_prefix - 1) +#define RX_WRAPPED(prog) SvPVX(prog) +#define RX_WRAPLEN(prog) SvCUR(prog) +#define RX_CHECK_SUBSTR(prog) (((struct regexp *)SvANY(prog))->check_substr) +#define RX_EXTFLAGS(prog) RXp_EXTFLAGS((struct regexp *)SvANY(prog)) +#define RX_REFCNT(prog) SvREFCNT(prog) +#define RX_ENGINE(prog) (((struct regexp *)SvANY(prog))->engine) +#define RX_SUBBEG(prog) (((struct regexp *)SvANY(prog))->subbeg) +#define RX_OFFS(prog) (((struct regexp *)SvANY(prog))->offs) +#define RX_NPARENS(prog) (((struct regexp *)SvANY(prog))->nparens) +#define RX_SUBLEN(prog) (((struct regexp *)SvANY(prog))->sublen) +#define RX_SUBBEG(prog) (((struct regexp *)SvANY(prog))->subbeg) +#define RX_MINLEN(prog) (((struct regexp *)SvANY(prog))->minlen) +#define RX_MINLENRET(prog) (((struct regexp *)SvANY(prog))->minlenret) +#define RX_GOFS(prog) (((struct regexp *)SvANY(prog))->gofs) +#define RX_LASTPAREN(prog) (((struct regexp *)SvANY(prog))->lastparen) +#define RX_LASTCLOSEPAREN(prog) (((struct regexp *)SvANY(prog))->lastcloseparen) +#define RX_SEEN_EVALS(prog) (((struct regexp *)SvANY(prog))->seen_evals) +#define RX_SAVED_COPY(prog) (((struct regexp *)SvANY(prog))->saved_copy) + #endif /* PLUGGABLE_RE_EXTENSION */ /* Stuff that needs to be included in the plugable extension goes below here */ #ifdef PERL_OLD_COPY_ON_WRITE #define RX_MATCH_COPY_FREE(rx) \ - STMT_START {if (rx->saved_copy) { \ - SV_CHECK_THINKFIRST_COW_DROP(rx->saved_copy); \ + STMT_START {if (RX_SAVED_COPY(rx)) { \ + SV_CHECK_THINKFIRST_COW_DROP(RX_SAVED_COPY(rx)); \ } \ if (RX_MATCH_COPIED(rx)) { \ - Safefree(rx->subbeg); \ + Safefree(RX_SUBBEG(rx)); \ RX_MATCH_COPIED_off(rx); \ }} STMT_END #else #define RX_MATCH_COPY_FREE(rx) \ STMT_START {if (RX_MATCH_COPIED(rx)) { \ - Safefree(rx->subbeg); \ + Safefree(RX_SUBBEG(rx)); \ RX_MATCH_COPIED_off(rx); \ }} STMT_END #endif -#define RX_MATCH_UTF8(prog) ((prog)->extflags & RXf_MATCH_UTF8) -#define RX_MATCH_UTF8_on(prog) ((prog)->extflags |= RXf_MATCH_UTF8) -#define RX_MATCH_UTF8_off(prog) ((prog)->extflags &= ~RXf_MATCH_UTF8) +#define RXp_MATCH_UTF8(prog) (RXp_EXTFLAGS(prog) & RXf_MATCH_UTF8) +#define RX_MATCH_UTF8(prog) (RX_EXTFLAGS(prog) & RXf_MATCH_UTF8) +#define RX_MATCH_UTF8_on(prog) (RX_EXTFLAGS(prog) |= RXf_MATCH_UTF8) +#define RX_MATCH_UTF8_off(prog) (RX_EXTFLAGS(prog) &= ~RXf_MATCH_UTF8) #define RX_MATCH_UTF8_set(prog, t) ((t) \ ? (RX_MATCH_UTF8_on(prog), (PL_reg_match_utf8 = 1)) \ : (RX_MATCH_UTF8_off(prog), (PL_reg_match_utf8 = 0))) + +/* Whether the pattern stored at RX_WRAPPED is in UTF-8 */ +#define RX_UTF8(prog) SvUTF8(prog) #define REXEC_COPY_STR 0x01 /* Need to copy the string. */ #define REXEC_CHECKED 0x02 /* check_substr already checked. */ @@ -232,8 +423,25 @@ typedef struct regexp_engine { #define REXEC_IGNOREPOS 0x08 /* \G matches at start. */ #define REXEC_NOT_FIRST 0x10 /* This is another iteration of //g. */ -#define ReREFCNT_inc(re) ((void)(re && re->refcnt++), re) -#define ReREFCNT_dec(re) CALLREGFREE(re) +#if defined(__GNUC__) && !defined(__STRICT_ANSI__) && !defined(PERL_GCC_PEDANTIC) +# define ReREFCNT_inc(re) \ + ({ \ + /* This is here to generate a casting warning if incorrect. */ \ + REGEXP *const zwapp = (re); \ + SvREFCNT_inc(zwapp); \ + }) +# define ReREFCNT_dec(re) \ + ({ \ + /* This is here to generate a casting warning if incorrect. */ \ + REGEXP *const boff = (re); \ + SvREFCNT_dec(boff); \ + }) +#else +# define ReREFCNT_dec(re) SvREFCNT_dec(re) +# define ReREFCNT_inc(re) SvREFCNT_inc(re) +#endif + +/* FIXME for plugins. */ #define FBMcf_TAIL_DOLLAR 1 #define FBMcf_TAIL_DOLLARM 2 @@ -254,7 +462,7 @@ typedef struct _reg_trie_accepted reg_trie_accepted; * Perl_regexec_flags and then passed to regtry(), regmatch() etc */ typedef struct { - regexp *prog; + REGEXP *prog; char *bol; char *till; SV *sv; @@ -265,7 +473,9 @@ typedef struct { /* structures for holding and saving the state maintained by regmatch() */ -#define MAX_RECURSE_EVAL_NOCHANGE_DEPTH 50 +#ifndef MAX_RECURSE_EVAL_NOCHANGE_DEPTH +#define MAX_RECURSE_EVAL_NOCHANGE_DEPTH 1000 +#endif typedef I32 CHECKPOINT; @@ -282,41 +492,88 @@ typedef struct regmatch_state { struct regmatch_state *prev_yes_state; } yes; - struct { + /* branchlike members */ + /* this is a fake union member that matches the first elements + * of each member that needs to behave like a branch */ + struct { /* this first element must match u.yes */ struct regmatch_state *prev_yes_state; - reg_trie_accepted *accept_buff; + U32 lastparen; + CHECKPOINT cp; + + } branchlike; + + struct { + /* the first elements must match u.branchlike */ + struct regmatch_state *prev_yes_state; + U32 lastparen; + CHECKPOINT cp; + + regnode *next_branch; /* next branch node */ + } branch; + + struct { + /* the first elements must match u.branchlike */ + struct regmatch_state *prev_yes_state; + U32 lastparen; + CHECKPOINT cp; + + reg_trie_accepted *accept_buff; /* accepting states we have seen */ U32 accepted; /* how many accepting states we have seen */ U16 *jump; /* positive offsets from me */ regnode *B; /* node following the trie */ regnode *me; /* Which node am I - needed for jump tries*/ } trie; + /* special types - these members are used to store state for special + regops like eval, if/then, lookaround and the markpoint state */ struct { /* this first element must match u.yes */ struct regmatch_state *prev_yes_state; struct regmatch_state *prev_eval; struct regmatch_state *prev_curlyx; - regexp *prev_rex; + REGEXP *prev_rex; U32 toggle_reg_flags; /* what bits in PL_reg_flags to flip when transitioning between inner and outer rexen */ CHECKPOINT cp; /* remember current savestack indexes */ CHECKPOINT lastcp; - regnode *B; /* the node following us */ U32 close_paren; /* which close bracket is our end */ + regnode *B; /* the node following us */ } eval; struct { /* this first element must match u.yes */ struct regmatch_state *prev_yes_state; + I32 wanted; + I32 logical; /* saved copy of 'logical' var */ + regnode *me; /* the IFMATCH/SUSPEND/UNLESSM node */ + } ifmatch; /* and SUSPEND/UNLESSM */ + + struct { + /* this first element must match u.yes */ + struct regmatch_state *prev_yes_state; + struct regmatch_state *prev_mark; + SV* mark_name; + char *mark_loc; + } mark; + + struct { + int val; + } keeper; + + /* quantifiers - these members are used for storing state for + for the regops used to implement quantifiers */ + struct { + /* this first element must match u.yes */ + struct regmatch_state *prev_yes_state; struct regmatch_state *prev_curlyx; /* previous cur_curlyx */ + regnode *A, *B; /* the nodes corresponding to /A*B/ */ CHECKPOINT cp; /* remember current savestack index */ bool minmod; int parenfloor;/* how far back to strip paren data */ int min; /* the minimal number of A's to match */ int max; /* the maximal number of A's to match */ - regnode *A, *B; /* the nodes corresponding to /A*B/ */ /* these two are modified by WHILEM */ int count; /* how many instances of A we've matched */ @@ -337,14 +594,6 @@ typedef struct regmatch_state { struct { /* this first element must match u.yes */ struct regmatch_state *prev_yes_state; - U32 lastparen; - regnode *next_branch; /* next branch node */ - CHECKPOINT cp; - } branch; - - struct { - /* this first element must match u.yes */ - struct regmatch_state *prev_yes_state; I32 c1, c2; /* case fold search */ CHECKPOINT cp; I32 alen; /* length of first-matched A string */ @@ -365,21 +614,6 @@ typedef struct regmatch_state { regnode *A, *B; /* the nodes corresponding to /A*B/ */ } curly; /* and CURLYN/PLUS/STAR */ - struct { - /* this first element must match u.yes */ - struct regmatch_state *prev_yes_state; - I32 wanted; - I32 logical; /* saved copy of 'logical' var */ - regnode *me; /* the IFMATCH/SUSPEND/UNLESSM node */ - } ifmatch; /* and SUSPEND/UNLESSM */ - - struct { - /* this first element must match u.yes */ - struct regmatch_state *prev_yes_state; - struct regmatch_state *prev_mark; - SV* mark_name; - char *mark_loc; - } mark; } u; } regmatch_state; @@ -399,8 +633,7 @@ typedef struct regmatch_slab { #define PL_bostr PL_reg_state.re_state_bostr #define PL_reginput PL_reg_state.re_state_reginput #define PL_regeol PL_reg_state.re_state_regeol -#define PL_regstartp PL_reg_state.re_state_regstartp -#define PL_regendp PL_reg_state.re_state_regendp +#define PL_regoffs PL_reg_state.re_state_regoffs #define PL_reglastparen PL_reg_state.re_state_reglastparen #define PL_reglastcloseparen PL_reg_state.re_state_reglastcloseparen #define PL_reg_start_tmp PL_reg_state.re_state_reg_start_tmp @@ -423,28 +656,27 @@ typedef struct regmatch_slab { struct re_save_state { U32 re_state_reg_flags; /* from regexec.c */ + U32 re_state_reg_start_tmpl; /* from regexec.c */ + I32 re_state_reg_eval_set; /* from regexec.c */ + bool re_state_reg_match_utf8; /* from regexec.c */ char *re_state_bostr; char *re_state_reginput; /* String-input pointer. */ char *re_state_regeol; /* End of input, for $ check. */ - I32 *re_state_regstartp; /* Pointer to startp array. */ - I32 *re_state_regendp; /* Ditto for endp. */ + regexp_paren_pair *re_state_regoffs; /* Pointer to start/end pairs */ U32 *re_state_reglastparen; /* Similarly for lastparen. */ U32 *re_state_reglastcloseparen; /* Similarly for lastcloseparen. */ char **re_state_reg_start_tmp; /* from regexec.c */ - U32 re_state_reg_start_tmpl; /* from regexec.c */ - I32 re_state_reg_eval_set; /* from regexec.c */ - bool re_state_reg_match_utf8; /* from regexec.c */ MAGIC *re_state_reg_magic; /* from regexec.c */ - I32 re_state_reg_oldpos; /* from regexec.c */ PMOP *re_state_reg_oldcurpm; /* from regexec.c */ PMOP *re_state_reg_curpm; /* from regexec.c */ char *re_state_reg_oldsaved; /* old saved substr during match */ STRLEN re_state_reg_oldsavedlen; /* old length of saved substr during match */ + STRLEN re_state_reg_poscache_size; /* size of pos cache of WHILEM */ + I32 re_state_reg_oldpos; /* from regexec.c */ I32 re_state_reg_maxiter; /* max wait until caching pos */ I32 re_state_reg_leftiter; /* wait until caching pos */ - char *re_state_reg_poscache; /* cache of pos of WHILEM */ - STRLEN re_state_reg_poscache_size; /* size of pos cache of WHILEM */ U32 re_state_regsize; /* from regexec.c */ + char *re_state_reg_poscache; /* cache of pos of WHILEM */ char *re_state_reg_starttry; /* from regexec.c */ #ifdef PERL_OLD_COPY_ON_WRITE SV *re_state_nrs; /* was placeholder: unused since 5.8.0 (5.7.2 patch #12027 for bug ID 20010815.012). Used to save rx->saved_copy */ @@ -453,6 +685,7 @@ struct re_save_state { #define SAVESTACK_ALLOC_FOR_RE_SAVE_STATE \ (1 + ((sizeof(struct re_save_state) - 1) / sizeof(*PL_savestack))) + /* * Local variables: * c-indentation-style: bsd