X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/785a26d510947a2b97507d7acf9b8c13bd59b310..d56b30146911a1c03e38fbb1e59b6f4edf749c91:/regexp.h diff --git a/regexp.h b/regexp.h index 248090a..f6d3c7b 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 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. @@ -82,6 +82,8 @@ typedef struct regexp { /* Information about the match that isn't often used */ char *precomp; /* pre-compilation regular expression */ I32 prelen; /* length of precomp */ + char *wrapped; /* wrapped version of the pattern */ + I32 wraplen; /* length of wrapped */ I32 seen_evals; /* number of eval groups in the pattern - for security checks */ HV *paren_names; /* Optional hash of paren names */ @@ -109,6 +111,8 @@ typedef struct regexp_engine { struct re_scream_pos_data_s *data); SV* (*checkstr) (pTHX_ regexp *prog); void (*free) (pTHX_ struct regexp* r); + SV* (*numbered_buff_get) (pTHX_ const REGEXP * const rx, I32 paren, SV* usesv); + SV* (*named_buff_get)(pTHX_ const REGEXP * const rx, SV* namesv, U32 flags); #ifdef USE_ITHREADS void* (*dupe) (pTHX_ const regexp *r, CLONE_PARAMS *param); #endif @@ -116,6 +120,10 @@ typedef struct regexp_engine { /* 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. */ /* Anchor and GPOS related stuff */ @@ -125,24 +133,65 @@ typedef struct regexp_engine { #define RXf_ANCH_GPOS 0x00000008 #define RXf_GPOS_SEEN 0x00000010 #define RXf_GPOS_FLOAT 0x00000020 -/* five bits here */ +/* 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_ANCH_SINGLE (RXf_ANCH_SBOL|RXf_ANCH_GPOS) + +/* Flags indicating special patterns */ +#define RXf_START_ONLY 0x00000200 /* Pattern is /^/ */ +#define RXf_WHITE 0x00000400 /* Pattern is /\s+/ */ + +/* 0x1F800 of extflags is used by (RXf_)PMf_COMPILETIME */ +#define RXf_PMf_LOCALE 0x00000800 /* use locale */ +#define RXf_PMf_MULTILINE 0x00001000 /* /m */ +#define RXf_PMf_SINGLELINE 0x00002000 /* /s */ +#define RXf_PMf_FOLD 0x00004000 /* /i */ +#define RXf_PMf_EXTENDED 0x00008000 /* /x */ +#define RXf_PMf_KEEPCOPY 0x00010000 /* /k */ +/* these flags are transfered from the PMOP->op_pmflags member during compilation */ +#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 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 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 + /* What we have seen */ -/* one bit here */ #define RXf_LOOKBEHIND_SEEN 0x00020000 #define RXf_EVAL_SEEN 0x00040000 #define RXf_CANY_SEEN 0x00080000 @@ -250,7 +299,9 @@ typedef struct { /* structures for holding and saving the state maintained by regmatch() */ +#ifndef MAX_RECURSE_EVAL_NOCHANGE_DEPTH #define MAX_RECURSE_EVAL_NOCHANGE_DEPTH 50 +#endif typedef I32 CHECKPOINT; @@ -267,16 +318,41 @@ 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; @@ -295,6 +371,28 @@ typedef struct regmatch_state { 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 */ CHECKPOINT cp; /* remember current savestack index */ bool minmod; @@ -322,14 +420,6 @@ 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 */ @@ -350,21 +440,6 @@ 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; @@ -438,6 +513,7 @@ 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