From b988e673a1322982f92dcce5bc82b973c29216ed Mon Sep 17 00:00:00 2001 From: Karl Williamson Date: Sun, 26 Dec 2010 10:39:12 -0700 Subject: [PATCH] regex: Convert regnode FLAGS fields to charset enum The FLAGS fields of certain regnodes were encoded with USE_UNI if unicode semantics were to be used. This patch instead sets them to the character set used, one of the possibilities of which is to use unicode semantics. This shortens the code somewhat, and always puts the character set in the flags field, which can allow use of switch statements on it for efficiency, especially as new values are added. --- regcomp.c | 26 +++++++++++++------------- regcomp.h | 5 +++-- regexec.c | 32 ++++++++++++++++---------------- 3 files changed, 32 insertions(+), 31 deletions(-) diff --git a/regcomp.c b/regcomp.c index 7d9e573..327abde 100644 --- a/regcomp.c +++ b/regcomp.c @@ -3632,7 +3632,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, if (flags & SCF_DO_STCLASS_AND) { if (!(data->start_class->flags & ANYOF_LOCALE)) { ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NALNUM); - if (FLAGS(scan) & USE_UNI) { + if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (!isWORDCHAR_L1(value)) { ANYOF_BITMAP_CLEAR(data->start_class, value); @@ -3650,7 +3650,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, else { if (data->start_class->flags & ANYOF_LOCALE) ANYOF_CLASS_SET(data->start_class,ANYOF_ALNUM); - else if (FLAGS(scan) & USE_UNI) { + else if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (isWORDCHAR_L1(value)) { ANYOF_BITMAP_SET(data->start_class, value); @@ -3679,7 +3679,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, if (flags & SCF_DO_STCLASS_AND) { if (!(data->start_class->flags & ANYOF_LOCALE)) { ANYOF_CLASS_CLEAR(data->start_class,ANYOF_ALNUM); - if (FLAGS(scan) & USE_UNI) { + if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (isWORDCHAR_L1(value)) { ANYOF_BITMAP_CLEAR(data->start_class, value); @@ -3718,7 +3718,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, if (flags & SCF_DO_STCLASS_AND) { if (!(data->start_class->flags & ANYOF_LOCALE)) { ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NSPACE); - if (FLAGS(scan) & USE_UNI) { + if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (!isSPACE_L1(value)) { ANYOF_BITMAP_CLEAR(data->start_class, value); @@ -3737,7 +3737,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, if (data->start_class->flags & ANYOF_LOCALE) { ANYOF_CLASS_SET(data->start_class,ANYOF_SPACE); } - else if (FLAGS(scan) & USE_UNI) { + else if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (isSPACE_L1(value)) { ANYOF_BITMAP_SET(data->start_class, value); @@ -3766,7 +3766,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, if (flags & SCF_DO_STCLASS_AND) { if (!(data->start_class->flags & ANYOF_LOCALE)) { ANYOF_CLASS_CLEAR(data->start_class,ANYOF_SPACE); - if (FLAGS(scan) & USE_UNI) { + if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (isSPACE_L1(value)) { ANYOF_BITMAP_CLEAR(data->start_class, value); @@ -3784,7 +3784,7 @@ S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp, else { if (data->start_class->flags & ANYOF_LOCALE) ANYOF_CLASS_SET(data->start_class,ANYOF_NSPACE); - else if (FLAGS(scan) & USE_UNI) { + else if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { for (value = 0; value < 256; value++) { if (!isSPACE_L1(value)) { ANYOF_BITMAP_SET(data->start_class, value); @@ -7400,8 +7400,8 @@ tryagain: ret = reg_node(pRExC_state, (U8)(ALNUML)); } else { ret = reg_node(pRExC_state, (U8)(ALNUM)); - FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0; } + FLAGS(ret) = get_regex_charset(RExC_flags); *flagp |= HASWIDTH|SIMPLE; goto finish_meta_pat; case 'W': @@ -7409,8 +7409,8 @@ tryagain: ret = reg_node(pRExC_state, (U8)(NALNUML)); } else { ret = reg_node(pRExC_state, (U8)(NALNUM)); - FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0; } + FLAGS(ret) = get_regex_charset(RExC_flags); *flagp |= HASWIDTH|SIMPLE; goto finish_meta_pat; case 'b': @@ -7420,8 +7420,8 @@ tryagain: ret = reg_node(pRExC_state, (U8)(BOUNDL)); } else { ret = reg_node(pRExC_state, (U8)(BOUND)); - FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0; } + FLAGS(ret) = get_regex_charset(RExC_flags); *flagp |= SIMPLE; goto finish_meta_pat; case 'B': @@ -7431,8 +7431,8 @@ tryagain: ret = reg_node(pRExC_state, (U8)(NBOUNDL)); } else { ret = reg_node(pRExC_state, (U8)(NBOUND)); - FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0; } + FLAGS(ret) = get_regex_charset(RExC_flags); *flagp |= SIMPLE; goto finish_meta_pat; case 's': @@ -7440,8 +7440,8 @@ tryagain: ret = reg_node(pRExC_state, (U8)(SPACEL)); } else { ret = reg_node(pRExC_state, (U8)(SPACE)); - FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0; } + FLAGS(ret) = get_regex_charset(RExC_flags); *flagp |= HASWIDTH|SIMPLE; goto finish_meta_pat; case 'S': @@ -7449,8 +7449,8 @@ tryagain: ret = reg_node(pRExC_state, (U8)(NSPACEL)); } else { ret = reg_node(pRExC_state, (U8)(NSPACE)); - FLAGS(ret) = (UNI_SEMANTICS) ? USE_UNI : 0; } + FLAGS(ret) = get_regex_charset(RExC_flags); *flagp |= HASWIDTH|SIMPLE; goto finish_meta_pat; case 'd': diff --git a/regcomp.h b/regcomp.h index 96e7ae1..48e270f 100644 --- a/regcomp.h +++ b/regcomp.h @@ -272,8 +272,9 @@ struct regnode_charclass_class { #undef STRING #define OP(p) ((p)->type) -#define FLAGS(p) ((p)->flags) /* Caution: Doesn't apply to all \ - regnode types */ +#define FLAGS(p) ((p)->flags) /* Caution: Doesn't apply to all \ + regnode types. For some, it's the \ + character set of the regnode */ #define OPERAND(p) (((struct regnode_string *)p)->string) #define MASK(p) ((char*)OPERAND(p)) #define STR_LEN(p) (((struct regnode_string *)p)->str_len) diff --git a/regexec.c b/regexec.c index be0feeb..3644302 100644 --- a/regexec.c +++ b/regexec.c @@ -219,7 +219,7 @@ * between 128 and 255 using Unicode (latin1) semantics. */ #define CCC_TRY_AFF_U(NAME,NAMEL,CLASS,STR,LCFUNC_utf8,FUNCU,LCFUNC) \ _CCC_TRY_AFF_COMMON(NAME,NAMEL,CLASS,STR,LCFUNC_utf8,FUNC) \ - if (!(OP(scan) == NAMEL ? LCFUNC(nextchr) : (FUNCU(nextchr) && (isASCII(nextchr) || (FLAGS(scan) & USE_UNI))))) \ + if (!(OP(scan) == NAMEL ? LCFUNC(nextchr) : (FUNCU(nextchr) && (isASCII(nextchr) || (FLAGS(scan) == REGEX_UNICODE_CHARSET))))) \ sayNO; \ nextchr = UCHARAT(++locinput); \ break @@ -261,7 +261,7 @@ #define CCC_TRY_NEG_U(NAME,NAMEL,CLASS,STR,LCFUNC_utf8,FUNCU,LCFUNC) \ _CCC_TRY_NEG_COMMON(NAME,NAMEL,CLASS,STR,LCFUNC_utf8,FUNCU) \ - if ((OP(scan) == NAMEL ? LCFUNC(nextchr) : (FUNCU(nextchr) && (isASCII(nextchr) || (FLAGS(scan) & USE_UNI))))) \ + if ((OP(scan) == NAMEL ? LCFUNC(nextchr) : (FUNCU(nextchr) && (isASCII(nextchr) || (FLAGS(scan) == REGEX_UNICODE_CHARSET))))) \ sayNO; \ nextchr = UCHARAT(++locinput); \ break @@ -1557,13 +1557,13 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, tmp = cBOOL((OP(c) == BOUNDL) ? isALNUM_LC(tmp) : (isWORDCHAR_L1(tmp) - && (isASCII(tmp) || (FLAGS(c) & USE_UNI)))); + && (isASCII(tmp) || (FLAGS(c) == REGEX_UNICODE_CHARSET)))); REXEC_FBC_SCAN( if (tmp == !((OP(c) == BOUNDL) ? isALNUM_LC(*s) : (isWORDCHAR_L1((U8) *s) - && (isASCII((U8) *s) || (FLAGS(c) & USE_UNI))))) + && (isASCII((U8) *s) || (FLAGS(c) == REGEX_UNICODE_CHARSET))))) { tmp = !tmp; REXEC_FBC_TRYIT; @@ -1600,13 +1600,13 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, tmp = cBOOL((OP(c) == NBOUNDL) ? isALNUM_LC(tmp) : (isWORDCHAR_L1(tmp) - && (isASCII(tmp) || (FLAGS(c) & USE_UNI)))); + && (isASCII(tmp) || (FLAGS(c) == REGEX_UNICODE_CHARSET)))); REXEC_FBC_SCAN( if (tmp == ! cBOOL( (OP(c) == NBOUNDL) ? isALNUM_LC(*s) : (isWORDCHAR_L1((U8) *s) - && (isASCII((U8) *s) || (FLAGS(c) & USE_UNI))))) + && (isASCII((U8) *s) || (FLAGS(c) == REGEX_UNICODE_CHARSET))))) { tmp = !tmp; } @@ -1620,7 +1620,7 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, REXEC_FBC_CSCAN_PRELOAD( LOAD_UTF8_CHARCLASS_PERL_WORD(), swash_fetch(RE_utf8_perl_word, (U8*)s, utf8_target), - (FLAGS(c) & USE_UNI) ? isWORDCHAR_L1((U8) *s) : isALNUM(*s) + (FLAGS(c) == REGEX_UNICODE_CHARSET) ? isWORDCHAR_L1((U8) *s) : isALNUM(*s) ); case ALNUML: REXEC_FBC_CSCAN_TAINT( @@ -1631,7 +1631,7 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, REXEC_FBC_CSCAN_PRELOAD( LOAD_UTF8_CHARCLASS_PERL_WORD(), !swash_fetch(RE_utf8_perl_word, (U8*)s, utf8_target), - ! ((FLAGS(c) & USE_UNI) ? isWORDCHAR_L1((U8) *s) : isALNUM(*s)) + ! ((FLAGS(c) == REGEX_UNICODE_CHARSET) ? isWORDCHAR_L1((U8) *s) : isALNUM(*s)) ); case NALNUML: REXEC_FBC_CSCAN_TAINT( @@ -1642,7 +1642,7 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, REXEC_FBC_CSCAN_PRELOAD( LOAD_UTF8_CHARCLASS_PERL_SPACE(), *s == ' ' || swash_fetch(RE_utf8_perl_space,(U8*)s, utf8_target), - isSPACE_L1((U8) *s) && (isASCII((U8) *s) || (FLAGS(c) & USE_UNI)) + isSPACE_L1((U8) *s) && (isASCII((U8) *s) || (FLAGS(c) == REGEX_UNICODE_CHARSET)) ); case SPACEL: REXEC_FBC_CSCAN_TAINT( @@ -1653,7 +1653,7 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, REXEC_FBC_CSCAN_PRELOAD( LOAD_UTF8_CHARCLASS_PERL_SPACE(), !(*s == ' ' || swash_fetch(RE_utf8_perl_space,(U8*)s, utf8_target)), - !(isSPACE_L1((U8) *s) && (isASCII((U8) *s) || (FLAGS(c) & USE_UNI))) + !(isSPACE_L1((U8) *s) && (isASCII((U8) *s) || (FLAGS(c) == REGEX_UNICODE_CHARSET))) ); case NSPACEL: REXEC_FBC_CSCAN_TAINT( @@ -3651,10 +3651,10 @@ S_regmatch(pTHX_ regmatch_info *reginfo, regnode *prog) else { ln = (locinput != PL_bostr) ? UCHARAT(locinput - 1) : '\n'; - if (FLAGS(scan) & USE_UNI) { + if (FLAGS(scan) == REGEX_UNICODE_CHARSET) { /* Here, can't be BOUNDL or NBOUNDL because they never set - * the flags to USE_UNI */ + * the flags to REGEX_UNICODE_CHARSET */ ln = isWORDCHAR_L1(ln); n = isWORDCHAR_L1(nextchr); } @@ -5942,7 +5942,7 @@ S_regrepeat(pTHX_ const regexp *prog, const regnode *p, I32 max, int depth) scan += UTF8SKIP(scan); hardcount++; } - } else if (FLAGS(p) & USE_UNI) { + } else if (FLAGS(p) == REGEX_UNICODE_CHARSET) { while (scan < loceol && isWORDCHAR_L1((U8) *scan)) { scan++; } @@ -5976,7 +5976,7 @@ S_regrepeat(pTHX_ const regexp *prog, const regnode *p, I32 max, int depth) scan += UTF8SKIP(scan); hardcount++; } - } else if (FLAGS(p) & USE_UNI) { + } else if (FLAGS(p) == REGEX_UNICODE_CHARSET) { while (scan < loceol && ! isWORDCHAR_L1((U8) *scan)) { scan++; } @@ -6011,7 +6011,7 @@ S_regrepeat(pTHX_ const regexp *prog, const regnode *p, I32 max, int depth) scan += UTF8SKIP(scan); hardcount++; } - } else if (FLAGS(p) & USE_UNI) { + } else if (FLAGS(p) == REGEX_UNICODE_CHARSET) { while (scan < loceol && isSPACE_L1((U8) *scan)) { scan++; } @@ -6045,7 +6045,7 @@ S_regrepeat(pTHX_ const regexp *prog, const regnode *p, I32 max, int depth) scan += UTF8SKIP(scan); hardcount++; } - } else if (FLAGS(p) & USE_UNI) { + } else if (FLAGS(p) == REGEX_UNICODE_CHARSET) { while (scan < loceol && ! isSPACE_L1((U8) *scan)) { scan++; } -- 2.7.4