X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=regexp.h;h=156e60b4e51ba43be6078f7215fb8049ee625038;hb=523f125d4a71aa467fc6a9acfe6c304944f5a5f5;hp=2960665bb184ad4541465861bb7058d75b4ebf2d;hpb=52d81aa88ad6014ec0ca7c381776c58dd68e428a;p=p5sagit%2Fp5-mst-13.2.git diff --git a/regexp.h b/regexp.h index 2960665..156e60b 100644 --- a/regexp.h +++ b/regexp.h @@ -66,49 +66,58 @@ typedef struct regexp_paren_pair { regexp's data array based on the data item's type. */ +#define _REGEXP_COMMON \ + /* 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 */ \ + /* substring data about strings that must appear in the */ \ + /* final match, used for optimisations */ \ + struct reg_substr_data *substrs; \ + 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 */ \ + /* Array of offsets for (@-) and (@+) */ \ + regexp_paren_pair *offs; \ + /* saved or original string so \digit works forever. */ \ + char *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 */ \ + /* offset from wrapped to the start of precomp */ \ + PERL_BITFIELD32 pre_prefix:4; \ + /* number of eval groups in the pattern - for security checks */\ + PERL_BITFIELD32 seen_evals:28 + 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 */ - 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 */ - 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 (@+) */ - - char *subbeg; /* saved or original string - so \digit works forever. */ - 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 */ - /* 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 */ - 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 */ - HV *paren_names; /* Optional hash of paren names */ - - /* Refcount of this regexp */ - I32 refcnt; /* Refcount of this regexp */ + _XPV_HEAD; + _XPVMG_HEAD; + _REGEXP_COMMON; } regexp; +struct regexp_allocated { + _XPV_ALLOCATED_HEAD; + _XPVMG_HEAD; + _REGEXP_COMMON; +}; + +/* 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 { @@ -120,7 +129,7 @@ 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_ const SV * const pattern, const U32 flags); + REGEXP* (*comp) (pTHX_ const SV * const pattern, U32 flags); I32 (*exec) (pTHX_ REGEXP * const rx, char* stringarg, char* strend, char* strbeg, I32 minend, SV* screamer, void* data, U32 flags); @@ -306,7 +315,6 @@ and check for NULL. #define RXf_CHECK_ALL 0x00040000 /* UTF8 related */ -#define RXf_UTF8 0x00080000 #define RXf_MATCH_UTF8 0x00100000 /* Intuit related */ @@ -341,29 +349,82 @@ and check for NULL. */ #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) ((prog)->wrapped + (prog)->pre_prefix) +#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) ((prog)->wraplen - (prog)->pre_prefix - 1) -#define RX_WRAPPED(prog) ((prog)->wrapped) -#define RX_WRAPLEN(prog) ((prog)->wraplen) - +#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_REFCNT(prog) SvREFCNT(prog) +#if defined(__GNUC__) && !defined(PERL_GCC_BRACE_GROUPS_FORBIDDEN) +# define RX_EXTFLAGS(prog) \ + (*({ \ + const REGEXP *const thwape = (prog); \ + assert(SvTYPE(thwape) == SVt_REGEXP); \ + &RXp_EXTFLAGS(SvANY(thwape)); \ + })) +# define RX_ENGINE(prog) \ + (*({ \ + const REGEXP *const thwape = (prog); \ + assert(SvTYPE(thwape) == SVt_REGEXP); \ + &SvANY(thwape)->engine; \ + })) +# define RX_SUBBEG(prog) \ + (*({ \ + const REGEXP *const thwape = (prog); \ + assert(SvTYPE(thwape) == SVt_REGEXP); \ + &SvANY(thwape)->subbeg; \ + })) +# define RX_OFFS(prog) \ + (*({ \ + const REGEXP *const thwape = (prog); \ + assert(SvTYPE(thwape) == SVt_REGEXP); \ + &SvANY(thwape)->offs; \ + })) +# define RX_NPARENS(prog) \ + (*({ \ + const REGEXP *const thwape = (prog); \ + assert(SvTYPE(thwape) == SVt_REGEXP); \ + &SvANY(thwape)->nparens; \ + })) +#else +# define RX_EXTFLAGS(prog) RXp_EXTFLAGS((struct regexp *)SvANY(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) +#endif +#define RX_SUBLEN(prog) (((struct regexp *)SvANY(prog))->sublen) +#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 */ @@ -371,27 +432,31 @@ and check for NULL. #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. */ @@ -399,8 +464,27 @@ and check for NULL. #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(PERL_GCC_BRACE_GROUPS_FORBIDDEN) +# define ReREFCNT_inc(re) \ + ({ \ + /* This is here to generate a casting warning if incorrect. */ \ + REGEXP *const zwapp = (re); \ + assert(SvTYPE(zwapp) == SVt_REGEXP); \ + SvREFCNT_inc(zwapp); \ + 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) ((REGEXP *) SvREFCNT_inc(re)) +#endif + +/* FIXME for plugins. */ #define FBMcf_TAIL_DOLLAR 1 #define FBMcf_TAIL_DOLLARM 2 @@ -421,7 +505,7 @@ 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; @@ -491,7 +575,7 @@ typedef struct regmatch_state { 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 */