X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=regexp.h;h=d12df9209a0bf609aea054bc0af31eb099543eb9;hb=25ff0154ccf606eb5512a8cde622caf50e20fba3;hp=a2e0fc1a31cf76d191846b659c84adb3c0f2a0aa;hpb=a20207d7c00bd54d6d2718961a96cd5c7ed1e37b;p=p5sagit%2Fp5-mst-13.2.git diff --git a/regexp.h b/regexp.h index a2e0fc1..d12df92 100644 --- a/regexp.h +++ b/regexp.h @@ -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 @@ -138,7 +142,7 @@ typedef struct regexp_engine { #define RXf_START_ONLY 0x00000200 /* Pattern is /^/ */ #define RXf_WHITE 0x00000400 /* Pattern is /\s+/ */ -/* 0xF800 of extflags is used by (RXf_)PMf_COMPILETIME */ +/* 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 */ @@ -155,6 +159,38 @@ typedef struct regexp_engine { 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 + + /* What we have seen */ #define RXf_LOOKBEHIND_SEEN 0x00020000 #define RXf_EVAL_SEEN 0x00040000 @@ -263,7 +299,9 @@ 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; @@ -280,16 +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; @@ -308,6 +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; @@ -335,14 +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 */ @@ -363,25 +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; - - struct { - int val; - } keeper; } u; } regmatch_state;