X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=regexp.h;h=33e7c2091ec7a27c78e79170e88a748aa6a32518;hb=c944cfb99a453bf0989f7e0433237ee9661e575b;hp=d43f05f2d17782d88c0fff006a44c5b2921aad67;hpb=fae667d5a60f37538a5761795f7af2165c7d4fb0;p=p5sagit%2Fp5-mst-13.2.git diff --git a/regexp.h b/regexp.h index d43f05f..33e7c20 100644 --- a/regexp.h +++ b/regexp.h @@ -31,6 +31,7 @@ struct reg_substr_data; struct reg_data; struct regexp_engine; +struct regexp; struct reg_substr_datum { I32 min_offset; @@ -48,11 +49,18 @@ struct reg_substr_data { #else #define SV_SAVED_COPY #endif + +typedef struct regexp_paren_pair { + I32 start; + I32 end; +} regexp_paren_pair; + /* this is ordered such that the most commonly used fields are at the start of the struct */ typedef struct regexp { /* what engine created this regexp? */ const struct regexp_engine* engine; + struct 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 */ @@ -71,17 +79,19 @@ typedef struct regexp { /* Data about the last/current match. These are modified during matching*/ U32 lastparen; /* last open paren matched */ U32 lastcloseparen; /* last close paren matched */ - I32 *startp; /* Array of offsets from start of string (@-) */ - I32 *endp; /* Array of offsets from start of string (@+) */ + regexp_paren_pair *swap; /* Swap copy of *offs */ + regexp_paren_pair *offs; /* Array of offsets for (@-) and (@+) */ + char *subbeg; /* saved or original string so \digit works forever. */ - I32 sublen; /* Length of string pointed by subbeg */ SV_SAVED_COPY /* If non-NULL, SV which is COW from original */ + I32 sublen; /* Length of string pointed by subbeg */ /* Information about the match that isn't often used */ - char *precomp; /* pre-compilation regular expression */ I32 prelen; /* length of precomp */ + const char *precomp; /* pre-compilation regular expression */ + /* wrapped can't be const char*, as it is returned by sv_2pv_flags */ 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 */ @@ -102,7 +112,7 @@ typedef struct re_scream_pos_data_s * 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); + REGEXP* (*comp) (pTHX_ const SV * const pattern, const U32 flags); I32 (*exec) (pTHX_ regexp* prog, char* stringarg, char* strend, char* strbeg, I32 minend, SV* screamer, void* data, U32 flags); @@ -111,6 +121,9 @@ 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); + SV* (*qr_pkg)(pTHX_ const REGEXP * const rx); #ifdef USE_ITHREADS void* (*dupe) (pTHX_ const regexp *r, CLONE_PARAMS *param); #endif @@ -137,6 +150,7 @@ typedef struct regexp_engine { #define RXf_ANCH_SINGLE (RXf_ANCH_SBOL|RXf_ANCH_GPOS) /* Flags indicating special patterns */ +#define RXf_SKIPWHITE 0x00000100 /* Pattern is for a split / / */ #define RXf_START_ONLY 0x00000200 /* Pattern is /^/ */ #define RXf_WHITE 0x00000400 /* Pattern is /\s+/ */ @@ -159,9 +173,13 @@ typedef struct regexp_engine { /* 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 'k' +#define KEEPCOPY_PAT_MOD 'p' #define ONCE_PAT_MOD 'o' #define GLOBAL_PAT_MOD 'g' #define CONTINUE_PAT_MOD 'c' @@ -171,7 +189,7 @@ typedef struct regexp_engine { #define XTENDED_PAT_MOD 'x' #define ONCE_PAT_MODS "o" -#define KEEPCOPY_PAT_MODS "k" +#define KEEPCOPY_PAT_MODS "p" #define EXEC_PAT_MODS "e" #define LOOP_PAT_MODS "gc" @@ -208,7 +226,7 @@ typedef struct regexp_engine { /* Copy and tainted info */ #define RXf_COPY_DONE 0x10000000 #define RXf_TAINTED_SEEN 0x20000000 -/* two bits here */ +#define RXf_TAINTED 0x80000000 /* this pattern is tainted */ #define RX_HAS_CUTGROUP(prog) ((prog)->intflags & PREGf_CUTGROUP_SEEN) @@ -293,7 +311,9 @@ 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; @@ -356,8 +376,8 @@ typedef struct regmatch_state { 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 { @@ -386,12 +406,12 @@ typedef struct regmatch_state { /* 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 */ @@ -451,8 +471,7 @@ 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 @@ -475,28 +494,27 @@ 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 */