From e1e595a649d55a304416c1e8e23f99e0df58a452 Mon Sep 17 00:00:00 2001 From: Corinna Vinschen Date: Thu, 4 Feb 2010 12:35:49 +0000 Subject: [PATCH] Replace regex files with multibyte-aware version from FreeBSD. * Makefile.in (install-headers): Remove extra command to install regex.h. (uninstall-headers): Remove extra command to uninstall regex.h. * nlsfuncs.cc (collate_lcid): Make externally available to allow access to collation internals from regex functions. (collate_charset): Ditto. * wchar.h: Add __cplusplus guards to make C-clean. * include/regex.h: New file, replacing regex/regex.h. Remove UCB advertising clause. * regex/COPYRIGHT: Accommodate BSD license. Remove UCB advertising clause. * regex/cclass.h: Remove. * regex/cname.h: New file from FreeBSD. * regex/engine.c: Ditto. (NONCHAR): Tweak for Cygwin. * regex/engine.ih: Remove. * regex/mkh: Remove. * regex/regcomp.c: New file from FreeBSD. Tweak slightly for Cygwin. Import required collate internals from nlsfunc.cc. (p_ere_exp): Add GNU-specific \< and \> handling for word boundaries. (p_simp_re): Ditto. (__collate_range_cmp): Define. (p_b_term): Use Cygwin-specific collate internals. (findmust): Ditto. * regex/regcomp.ih: Remove. * regex/regerror.c: New file from FreeBSD. Fix a few compiler warnings. * regex/regerror.ih: Remove. * regex/regex.7: New file from FreeBSD. Remove UCB advertising clause. * regex/regex.h: Remove. Replaced by include/regex.h. * regex/regexec.c: New file from FreeBSD. Fix a few compiler warnings. * regex/regfree.c: New file from FreeBSD. * regex/tests: Remove. * regex/utils.h: New file from FreeBSD. --- winsup/cygwin/ChangeLog | 37 + winsup/cygwin/Makefile.in | 2 - winsup/cygwin/include/regex.h | 130 +++ winsup/cygwin/nlsfuncs.cc | 4 +- winsup/cygwin/regex/COPYRIGHT | 34 +- winsup/cygwin/regex/cclass.h | 31 - winsup/cygwin/regex/cname.h | 156 ++-- winsup/cygwin/regex/engine.c | 552 +++++++----- winsup/cygwin/regex/engine.ih | 35 - winsup/cygwin/regex/mkh | 76 -- winsup/cygwin/regex/regcomp.c | 1390 +++++++++++++++++++------------ winsup/cygwin/regex/regcomp.ih | 48 -- winsup/cygwin/regex/regerror.c | 113 ++- winsup/cygwin/regex/regerror.ih | 12 - winsup/cygwin/regex/regex.3 | 790 +++++++++++------- winsup/cygwin/regex/regex.7 | 543 ++++++++---- winsup/cygwin/regex/regex.h | 76 -- winsup/cygwin/regex/regex2.h | 172 ++-- winsup/cygwin/regex/regexec.c | 153 +++- winsup/cygwin/regex/regfree.c | 67 +- winsup/cygwin/regex/tests | 477 ----------- winsup/cygwin/regex/utils.h | 42 +- winsup/cygwin/wchar.h | 2 + 23 files changed, 2832 insertions(+), 2110 deletions(-) create mode 100644 winsup/cygwin/include/regex.h delete mode 100644 winsup/cygwin/regex/cclass.h delete mode 100644 winsup/cygwin/regex/engine.ih delete mode 100755 winsup/cygwin/regex/mkh delete mode 100644 winsup/cygwin/regex/regcomp.ih delete mode 100644 winsup/cygwin/regex/regerror.ih delete mode 100644 winsup/cygwin/regex/regex.h delete mode 100644 winsup/cygwin/regex/tests diff --git a/winsup/cygwin/ChangeLog b/winsup/cygwin/ChangeLog index 506ff09c2..12499e1b9 100644 --- a/winsup/cygwin/ChangeLog +++ b/winsup/cygwin/ChangeLog @@ -1,3 +1,40 @@ +2010-02-04 Corinna Vinschen + + Replace regex files with multibyte-aware version from FreeBSD. + * Makefile.in (install-headers): Remove extra command to install + regex.h. + (uninstall-headers): Remove extra command to uninstall regex.h. + * nlsfuncs.cc (collate_lcid): Make externally available to allow + access to collation internals from regex functions. + (collate_charset): Ditto. + * wchar.h: Add __cplusplus guards to make C-clean. + * include/regex.h: New file, replacing regex/regex.h. Remove UCB + advertising clause. + * regex/COPYRIGHT: Accommodate BSD license. Remove UCB advertising + clause. + * regex/cclass.h: Remove. + * regex/cname.h: New file from FreeBSD. + * regex/engine.c: Ditto. + (NONCHAR): Tweak for Cygwin. + * regex/engine.ih: Remove. + * regex/mkh: Remove. + * regex/regcomp.c: New file from FreeBSD. Tweak slightly for Cygwin. + Import required collate internals from nlsfunc.cc. + (p_ere_exp): Add GNU-specific \< and \> handling for word boundaries. + (p_simp_re): Ditto. + (__collate_range_cmp): Define. + (p_b_term): Use Cygwin-specific collate internals. + (findmust): Ditto. + * regex/regcomp.ih: Remove. + * regex/regerror.c: New file from FreeBSD. Fix a few compiler warnings. + * regex/regerror.ih: Remove. + * regex/regex.7: New file from FreeBSD. Remove UCB advertising clause. + * regex/regex.h: Remove. Replaced by include/regex.h. + * regex/regexec.c: New file from FreeBSD. Fix a few compiler warnings. + * regex/regfree.c: New file from FreeBSD. + * regex/tests: Remove. + * regex/utils.h: New file from FreeBSD. + 2010-02-03 Christopher Faylor * sigproc.cc (get_proc_lock): Fix error message typo. diff --git a/winsup/cygwin/Makefile.in b/winsup/cygwin/Makefile.in index 7273921b1..741dd66ff 100644 --- a/winsup/cygwin/Makefile.in +++ b/winsup/cygwin/Makefile.in @@ -329,7 +329,6 @@ install-headers: $(INSTALL_DATA) $$i $(DESTDIR)$(tooldir)/$$sub/`basename $$i` ; \ done ; \ done ; \ - $(INSTALL_DATA) regex/regex.h $(DESTDIR)$(tooldir)/include/regex.h install-man: @$(MKDIRP) $(DESTDIR)$(mandir)/man2 $(DESTDIR)$(mandir)/man3 $(DESTDIR)$(mandir)/man5 $(DESTDIR)$(mandir)/man7 @@ -364,7 +363,6 @@ uninstall-headers: rm -f $(tooldir)/$$sub/`basename $$i` ; \ done ; \ done ; \ - rm -f $(tooldir)/include/regex.h uninstall-man: cd $(srcdir); \ diff --git a/winsup/cygwin/include/regex.h b/winsup/cygwin/include/regex.h new file mode 100644 index 000000000..d7c673c19 --- /dev/null +++ b/winsup/cygwin/include/regex.h @@ -0,0 +1,130 @@ +/*- + * Copyright (c) 1992 Henry Spencer. + * Copyright (c) 1992, 1993 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer of the University of Toronto. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)regex.h 8.2 (Berkeley) 1/3/94 + * $FreeBSD$ + */ + +#ifndef _REGEX_H_ +#define _REGEX_H_ + +#include +#include + +/* types */ +#ifdef __CYGWIN__ +typedef _off_t regoff_t; + +#define __need_size_t +#include +#else /* !__CYGWIN__ */ +typedef __off_t regoff_t; + +#ifndef _SIZE_T_DECLARED +typedef __size_t size_t; +#define _SIZE_T_DECLARED +#endif +#endif /* !__CYGWIN__ */ + +typedef struct { + int re_magic; + size_t re_nsub; /* number of parenthesized subexpressions */ +#ifdef __CYGWIN__ + const char *re_endp; /* end pointer for REG_PEND */ +#else + __const char *re_endp; /* end pointer for REG_PEND */ +#endif + struct re_guts *re_g; /* none of your business :-) */ +} regex_t; + +typedef struct { + regoff_t rm_so; /* start of match */ + regoff_t rm_eo; /* end of match */ +} regmatch_t; + +/* regcomp() flags */ +#define REG_BASIC 0000 +#define REG_EXTENDED 0001 +#define REG_ICASE 0002 +#define REG_NOSUB 0004 +#define REG_NEWLINE 0010 +#define REG_NOSPEC 0020 +#define REG_PEND 0040 +#define REG_DUMP 0200 + +/* regerror() flags */ +#define REG_ENOSYS (-1) +#ifdef __CYGWIN__ +#define REG_NOERROR 0 /* GNU extension */ +#endif +#define REG_NOMATCH 1 +#define REG_BADPAT 2 +#define REG_ECOLLATE 3 +#define REG_ECTYPE 4 +#define REG_EESCAPE 5 +#define REG_ESUBREG 6 +#define REG_EBRACK 7 +#define REG_EPAREN 8 +#define REG_EBRACE 9 +#define REG_BADBR 10 +#define REG_ERANGE 11 +#define REG_ESPACE 12 +#define REG_BADRPT 13 +#define REG_EMPTY 14 +#define REG_ASSERT 15 +#define REG_INVARG 16 +#define REG_ILLSEQ 17 +#define REG_ATOI 255 /* convert name to number (!) */ +#define REG_ITOA 0400 /* convert number to name (!) */ + +/* regexec() flags */ +#define REG_NOTBOL 00001 +#define REG_NOTEOL 00002 +#define REG_STARTEND 00004 +#define REG_TRACE 00400 /* tracing of execution */ +#define REG_LARGE 01000 /* force large representation */ +#define REG_BACKR 02000 /* force use of backref code */ + +__BEGIN_DECLS +int regcomp(regex_t * __restrict, const char * __restrict, int); +size_t regerror(int, const regex_t * __restrict, char * __restrict, size_t); +/* + * XXX forth parameter should be `regmatch_t [__restrict]', but isn't because + * of a bug in GCC 3.2 (when -std=c99 is specified) which perceives this as a + * syntax error. + */ +int regexec(const regex_t * __restrict, const char * __restrict, size_t, + regmatch_t * __restrict, int); +void regfree(regex_t *); +__END_DECLS + +#endif /* !_REGEX_H_ */ diff --git a/winsup/cygwin/nlsfuncs.cc b/winsup/cygwin/nlsfuncs.cc index f8892533b..3310fb752 100644 --- a/winsup/cygwin/nlsfuncs.cc +++ b/winsup/cygwin/nlsfuncs.cc @@ -638,9 +638,9 @@ __set_lc_monetary_from_win (const char *name, return 1; } -static LCID collate_lcid = 0; +LCID collate_lcid = 0; static mbtowc_p collate_mbtowc = __ascii_mbtowc; -static char collate_charset[ENCODING_LEN + 1] = "ASCII"; +char collate_charset[ENCODING_LEN + 1] = "ASCII"; /* Called from newlib's setlocale() if category is LC_COLLATE. Stores LC_COLLATE locale information. This is subsequently accessed by the diff --git a/winsup/cygwin/regex/COPYRIGHT b/winsup/cygwin/regex/COPYRIGHT index 30c1f7a48..dc823b124 100644 --- a/winsup/cygwin/regex/COPYRIGHT +++ b/winsup/cygwin/regex/COPYRIGHT @@ -1,4 +1,4 @@ -Copyright 1992, 1993, 1994, 1997 Henry Spencer. All rights reserved. +Copyright 1992, 1993, 1994 Henry Spencer. All rights reserved. This software is not subject to any license of the American Telephone and Telegraph Company or of the Regents of the University of California. @@ -18,3 +18,35 @@ to the following restrictions: ever read sources, credits must appear in the documentation. 4. This notice may not be removed or altered. + +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= +/*- + * Copyright (c) 1994 + * The Regents of the University of California. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)COPYRIGHT 8.1 (Berkeley) 3/16/94 + */ diff --git a/winsup/cygwin/regex/cclass.h b/winsup/cygwin/regex/cclass.h deleted file mode 100644 index 7ddb44afb..000000000 --- a/winsup/cygwin/regex/cclass.h +++ /dev/null @@ -1,31 +0,0 @@ -/* character-class table */ -static struct cclass { - const char *name; - const char *chars; - const char *multis; -} cclasses[] = { - {"alnum", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\ -0123456789", ""}, - {"alpha", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz", - ""}, - {"blank", " \t", ""}, - {"cntrl", "\007\b\t\n\v\f\r\1\2\3\4\5\6\16\17\20\21\22\23\24\ - \25\26\27\30\31\32\33\34\35\36\37\177", ""}, - {"digit", "0123456789", ""}, - {"graph", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\ - 0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", - ""}, - {"lower", "abcdefghijklmnopqrstuvwxyz", - ""}, - {"print", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\ - 0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~ ", - ""}, - {"punct", "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", - ""}, - {"space", "\t\n\v\f\r ", ""}, - {"upper", "ABCDEFGHIJKLMNOPQRSTUVWXYZ", - ""}, - {"xdigit", "0123456789ABCDEFabcdef", - ""}, - {NULL, 0, ""} -}; diff --git a/winsup/cygwin/regex/cname.h b/winsup/cygwin/regex/cname.h index 1000f603e..4ad1ea93b 100644 --- a/winsup/cygwin/regex/cname.h +++ b/winsup/cygwin/regex/cname.h @@ -1,82 +1,118 @@ +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)cname.h 8.3 (Berkeley) 3/20/94 + * $FreeBSD: src/lib/libc/regex/cname.h,v 1.4 2007/01/09 00:28:04 imp Exp $ + */ + /* character-name table */ static struct cname { const char *name; char code; } cnames[] = { - {"NUL", '\0'}, - {"SOH", '\001'}, - {"STX", '\002'}, - {"ETX", '\003'}, - {"EOT", '\004'}, - {"ENQ", '\005'}, - {"ACK", '\006'}, - {"BEL", '\007'}, - {"alert", '\007'}, - {"BS", '\010'}, - {"backspace", '\b'}, - {"HT", '\011'}, - {"tab", '\t'}, - {"LF", '\012'}, - {"newline", '\n'}, - {"VT", '\013'}, + {"NUL", '\0'}, + {"SOH", '\001'}, + {"STX", '\002'}, + {"ETX", '\003'}, + {"EOT", '\004'}, + {"ENQ", '\005'}, + {"ACK", '\006'}, + {"BEL", '\007'}, + {"alert", '\007'}, + {"BS", '\010'}, + {"backspace", '\b'}, + {"HT", '\011'}, + {"tab", '\t'}, + {"LF", '\012'}, + {"newline", '\n'}, + {"VT", '\013'}, {"vertical-tab", '\v'}, - {"FF", '\014'}, - {"form-feed", '\f'}, - {"CR", '\015'}, + {"FF", '\014'}, + {"form-feed", '\f'}, + {"CR", '\015'}, {"carriage-return", '\r'}, - {"SO", '\016'}, - {"SI", '\017'}, - {"DLE", '\020'}, - {"DC1", '\021'}, - {"DC2", '\022'}, - {"DC3", '\023'}, - {"DC4", '\024'}, - {"NAK", '\025'}, - {"SYN", '\026'}, - {"ETB", '\027'}, - {"CAN", '\030'}, - {"EM", '\031'}, - {"SUB", '\032'}, - {"ESC", '\033'}, - {"IS4", '\034'}, - {"FS", '\034'}, - {"IS3", '\035'}, - {"GS", '\035'}, - {"IS2", '\036'}, - {"RS", '\036'}, - {"IS1", '\037'}, - {"US", '\037'}, + {"SO", '\016'}, + {"SI", '\017'}, + {"DLE", '\020'}, + {"DC1", '\021'}, + {"DC2", '\022'}, + {"DC3", '\023'}, + {"DC4", '\024'}, + {"NAK", '\025'}, + {"SYN", '\026'}, + {"ETB", '\027'}, + {"CAN", '\030'}, + {"EM", '\031'}, + {"SUB", '\032'}, + {"ESC", '\033'}, + {"IS4", '\034'}, + {"FS", '\034'}, + {"IS3", '\035'}, + {"GS", '\035'}, + {"IS2", '\036'}, + {"RS", '\036'}, + {"IS1", '\037'}, + {"US", '\037'}, {"space", ' '}, {"exclamation-mark", '!'}, {"quotation-mark", '"'}, {"number-sign", '#'}, {"dollar-sign", '$'}, - {"percent-sign", '%'}, + {"percent-sign", '%'}, {"ampersand", '&'}, {"apostrophe", '\''}, {"left-parenthesis", '('}, {"right-parenthesis", ')'}, - {"asterisk", '*'}, - {"plus-sign", '+'}, - {"comma", ','}, - {"hyphen", '-'}, + {"asterisk", '*'}, + {"plus-sign", '+'}, + {"comma", ','}, + {"hyphen", '-'}, {"hyphen-minus", '-'}, - {"period", '.'}, - {"full-stop", '.'}, - {"slash", '/'}, - {"solidus", '/'}, + {"period", '.'}, + {"full-stop", '.'}, + {"slash", '/'}, + {"solidus", '/'}, {"zero", '0'}, - {"one", '1'}, - {"two", '2'}, - {"three", '3'}, + {"one", '1'}, + {"two", '2'}, + {"three", '3'}, {"four", '4'}, {"five", '5'}, - {"six", '6'}, - {"seven", '7'}, - {"eight", '8'}, + {"six", '6'}, + {"seven", '7'}, + {"eight", '8'}, {"nine", '9'}, - {"colon", ':'}, - {"semicolon", ';'}, + {"colon", ':'}, + {"semicolon", ';'}, {"less-than-sign", '<'}, {"equals-sign", '='}, {"greater-than-sign", '>'}, @@ -85,12 +121,12 @@ static struct cname { {"left-square-bracket", '['}, {"backslash", '\\'}, {"reverse-solidus", '\\'}, - {"right-square-bracket", ']'}, + {"right-square-bracket",']'}, {"circumflex", '^'}, {"circumflex-accent", '^'}, {"underscore", '_'}, {"low-line", '_'}, - {"grave-accent", '`'}, + {"grave-accent", '`'}, {"left-brace", '{'}, {"left-curly-bracket", '{'}, {"vertical-line", '|'}, diff --git a/winsup/cygwin/regex/engine.c b/winsup/cygwin/regex/engine.c index 42c91dc0f..ca24cc50c 100644 --- a/winsup/cygwin/regex/engine.c +++ b/winsup/cygwin/regex/engine.c @@ -1,3 +1,41 @@ +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)engine.c 8.5 (Berkeley) 3/20/94 + */ + +#include +__FBSDID("$FreeBSD: src/lib/libc/regex/engine.c,v 1.23 2009/09/16 06:32:23 dds Exp $"); + /* * The matching engine and friends. This file is #included by regexec.c * after suitable #defines of a variety of macros used herein, so that @@ -27,25 +65,77 @@ #define at lat #define match lmat #endif +#ifdef MNAMES +#define matcher mmatcher +#define fast mfast +#define slow mslow +#define dissect mdissect +#define backref mbackref +#define step mstep +#define print mprint +#define at mat +#define match mmat +#endif /* another structure passed up and down to avoid zillions of parameters */ struct match { struct re_guts *g; int eflags; regmatch_t *pmatch; /* [nsub+1] (0 element unused) */ - char *offp; /* offsets work from here */ - char *beginp; /* start of string -- virtual NUL precedes */ - char *endp; /* end of string -- virtual NUL here */ - char *coldp; /* can be no match starting before here */ - char **lastpos; /* [nplus+1] */ + const char *offp; /* offsets work from here */ + const char *beginp; /* start of string -- virtual NUL precedes */ + const char *endp; /* end of string -- virtual NUL here */ + const char *coldp; /* can be no match starting before here */ + const char **lastpos; /* [nplus+1] */ STATEVARS; states st; /* current states */ states fresh; /* states for a fresh start */ states tmp; /* temporary */ states empty; /* empty set of states */ + mbstate_t mbs; /* multibyte conversion state */ }; -#include "engine.ih" +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +/* === engine.c === */ +static int matcher(struct re_guts *g, const char *string, size_t nmatch, regmatch_t pmatch[], int eflags); +static const char *dissect(struct match *m, const char *start, const char *stop, sopno startst, sopno stopst); +static const char *backref(struct match *m, const char *start, const char *stop, sopno startst, sopno stopst, sopno lev, int); +static const char *fast(struct match *m, const char *start, const char *stop, sopno startst, sopno stopst); +static const char *slow(struct match *m, const char *start, const char *stop, sopno startst, sopno stopst); +static states step(struct re_guts *g, sopno start, sopno stop, states bef, wint_t ch, states aft); +#define MAX_RECURSION 100 +#define BOL (OUT-1) +#define EOL (BOL-1) +#define BOLEOL (BOL-2) +#define NOTHING (BOL-3) +#define BOW (BOL-4) +#define EOW (BOL-5) +#define BADCHAR (BOL-6) +#ifdef __CYGWIN__ +/* In contrast to BSD, wint_t on Cygwin is unsigned. This breaks this test, + unless the compared values are casted to signed. */ +#define NONCHAR(c) ((int)(c) <= (int)OUT) +#else +#define NONCHAR(c) ((c) <= OUT) +#endif +#ifdef REDEBUG +static void print(struct match *m, const char *caption, states st, int ch, FILE *d); +#endif +#ifdef REDEBUG +static void at(struct match *m, const char *title, const char *start, const char *stop, sopno startst, sopno stopst); +#endif +#ifdef REDEBUG +static const char *pchar(int ch); +#endif + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ #ifdef REDEBUG #define SP(t, s, c) print(m, t, s, c, stdout) @@ -59,26 +149,32 @@ struct match { /* - matcher - the actual matching engine - == static int matcher(register struct re_guts *g, char *string, \ + == static int matcher(struct re_guts *g, const char *string, \ == size_t nmatch, regmatch_t pmatch[], int eflags); */ static int /* 0 success, REG_NOMATCH failure */ -matcher(g, string, nmatch, pmatch, eflags) -register struct re_guts *g; -char *string; -size_t nmatch; -regmatch_t pmatch[]; -int eflags; +matcher(struct re_guts *g, + const char *string, + size_t nmatch, + regmatch_t pmatch[], + int eflags) { - register char *endp; - register size_t i; + const char *endp; + int i; struct match mv; - register struct match *m = &mv; - register char *dp; - const register sopno gf = g->firststate+1; /* +1 for OEND */ - const register sopno gl = g->laststate; - char *start; - char *stop; + struct match *m = &mv; + const char *dp; + const sopno gf = g->firststate+1; /* +1 for OEND */ + const sopno gl = g->laststate; + const char *start; + const char *stop; + /* Boyer-Moore algorithms variables */ + const char *pp; + int cj, mj; + const char *mustfirst; + const char *mustlast; + int *matchjump; + int *charjump; /* simplify the situation where possible */ if (g->cflags®_NOSUB) @@ -95,12 +191,46 @@ int eflags; /* prescreening; this does wonders for this rather slow code */ if (g->must != NULL) { - for (dp = start; dp < stop; dp++) - if (*dp == g->must[0] && stop - dp >= g->mlen && - memcmp(dp, g->must, (size_t)g->mlen) == 0) - break; - if (dp == stop) /* we didn't find g->must */ - return(REG_NOMATCH); + if (g->charjump != NULL && g->matchjump != NULL) { + mustfirst = g->must; + mustlast = g->must + g->mlen - 1; + charjump = g->charjump; + matchjump = g->matchjump; + pp = mustlast; + for (dp = start+g->mlen-1; dp < stop;) { + /* Fast skip non-matches */ + while (dp < stop && charjump[(int)*dp]) + dp += charjump[(int)*dp]; + + if (dp >= stop) + break; + + /* Greedy matcher */ + /* We depend on not being used for + * for strings of length 1 + */ + while (*--dp == *--pp && pp != mustfirst); + + if (*dp == *pp) + break; + + /* Jump to next possible match */ + mj = matchjump[pp - mustfirst]; + cj = charjump[(int)*dp]; + dp += (cj < mj ? mj : cj); + pp = mustlast; + } + if (pp != mustfirst) + return(REG_NOMATCH); + } else { + for (dp = start; dp < stop; dp++) + if (*dp == g->must[0] && + stop - dp >= g->mlen && + memcmp(dp, g->must, (size_t)g->mlen) == 0) + break; + if (dp == stop) /* we didn't find g->must */ + return(REG_NOMATCH); + } } /* match struct setup */ @@ -117,11 +247,22 @@ int eflags; SETUP(m->tmp); SETUP(m->empty); CLEAR(m->empty); + ZAPSTATE(&m->mbs); + + /* Adjust start according to moffset, to speed things up */ + if (g->moffset > -1) + start = ((dp - g->moffset) < start) ? start : dp - g->moffset; + + SP("mloop", m->st, *start); /* this loop does only one repetition except for backrefs */ for (;;) { endp = fast(m, start, stop, gf, gl); if (endp == NULL) { /* a miss */ + if (m->pmatch != NULL) + free((char *)m->pmatch); + if (m->lastpos != NULL) + free((char *)m->lastpos); STATETEARDOWN(m); return(REG_NOMATCH); } @@ -136,7 +277,8 @@ int eflags; if (endp != NULL) break; assert(m->coldp < m->endp); - m->coldp++; + m->coldp += XMBRTOWC(NULL, m->coldp, + m->endp - m->coldp, &m->mbs, 0); } if (nmatch == 1 && !g->backrefs) break; /* no further info needed */ @@ -156,15 +298,15 @@ int eflags; dp = dissect(m, m->coldp, endp, gf, gl); } else { if (g->nplus > 0 && m->lastpos == NULL) - m->lastpos = (char **)malloc((g->nplus+1) * - sizeof(char *)); + m->lastpos = malloc((g->nplus+1) * + sizeof(const char *)); if (g->nplus > 0 && m->lastpos == NULL) { free(m->pmatch); STATETEARDOWN(m); return(REG_ESPACE); } NOTE("backref dissect"); - dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); + dp = backref(m, m->coldp, endp, gf, gl, (sopno)0, 0); } if (dp != NULL) break; @@ -187,7 +329,7 @@ int eflags; } #endif NOTE("backoff dissect"); - dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); + dp = backref(m, m->coldp, endp, gf, gl, (sopno)0, 0); } assert(dp == NULL || dp == endp); if (dp != NULL) /* found a shorter one */ @@ -195,7 +337,9 @@ int eflags; /* despite initial appearances, there is no match here */ NOTE("false alarm"); - start = m->coldp + 1; /* recycle starting later */ + /* recycle starting later */ + start = m->coldp + XMBRTOWC(NULL, m->coldp, + stop - m->coldp, &m->mbs, 0); assert(start <= stop); } @@ -225,30 +369,29 @@ int eflags; /* - dissect - figure out what matched what, no back references - == static char *dissect(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); + == static const char *dissect(struct match *m, const char *start, \ + == const char *stop, sopno startst, sopno stopst); */ -static char * /* == stop (success) always */ -dissect(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; +static const char * /* == stop (success) always */ +dissect(struct match *m, + const char *start, + const char *stop, + sopno startst, + sopno stopst) { - register int i; - register sopno ss; /* start sop of current subRE */ - register sopno es; /* end sop of current subRE */ - register char *sp; /* start of string matched by it */ - register char *stp; /* string matched by it cannot pass here */ - register char *rest; /* start of rest of string */ - register char *tail; /* string unmatched by rest of RE */ - register sopno ssub; /* start sop of subsubRE */ - register sopno esub; /* end sop of subsubRE */ - register char *ssp; /* start of string matched by subsubRE */ - register char *sep; /* end of string matched by subsubRE */ - register char *oldssp; /* previous ssp */ - register char *dp; + int i; + sopno ss; /* start sop of current subRE */ + sopno es; /* end sop of current subRE */ + const char *sp; /* start of string matched by it */ + const char *stp; /* string matched by it cannot pass here */ + const char *rest; /* start of rest of string */ + const char *tail; /* string unmatched by rest of RE */ + sopno ssub; /* start sop of subsubRE */ + sopno esub; /* end sop of subsubRE */ + const char *ssp; /* start of string matched by subsubRE */ + const char *sep; /* end of string matched by subsubRE */ + const char *oldssp; /* previous ssp */ + const char *dp; AT("diss", start, stop, startst, stopst); sp = start; @@ -273,7 +416,7 @@ sopno stopst; assert(nope); break; case OCHAR: - sp++; + sp += XMBRTOWC(NULL, sp, stop - start, &m->mbs, 0); break; case OBOL: case OEOL: @@ -282,7 +425,7 @@ sopno stopst; break; case OANY: case OANYOF: - sp++; + sp += XMBRTOWC(NULL, sp, stop - start, &m->mbs, 0); break; case OBACK_: case O_BACK: @@ -413,30 +556,31 @@ sopno stopst; /* - backref - figure out what matched what, figuring in back references - == static char *backref(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst, sopno lev); + == static const char *backref(struct match *m, const char *start, \ + == const char *stop, sopno startst, sopno stopst, sopno lev); */ -static char * /* == stop (success) or NULL (failure) */ -backref(m, start, stop, startst, stopst, lev) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; -sopno lev; /* PLUS nesting level */ +static const char * /* == stop (success) or NULL (failure) */ +backref(struct match *m, + const char *start, + const char *stop, + sopno startst, + sopno stopst, + sopno lev, /* PLUS nesting level */ + int rec) { - register int i; - register sopno ss; /* start sop of current subRE */ - register char *sp; /* start of string matched by it */ - register sopno ssub; /* start sop of subsubRE */ - register sopno esub; /* end sop of subsubRE */ - register char *ssp; /* start of string matched by subsubRE */ - register char *dp; - register size_t len; - register int hard; - register sop s; - register regoff_t offsave; - register cset *cs; + int i; + sopno ss; /* start sop of current subRE */ + const char *sp; /* start of string matched by it */ + sopno ssub; /* start sop of subsubRE */ + sopno esub; /* end sop of subsubRE */ + const char *ssp; /* start of string matched by subsubRE */ + const char *dp; + size_t len; + int hard; + sop s; + regoff_t offsave; + cset *cs; + wint_t wc; AT("back", start, stop, startst, stopst); sp = start; @@ -446,17 +590,25 @@ sopno lev; /* PLUS nesting level */ for (ss = startst; !hard && ss < stopst; ss++) switch (OP(s = m->g->strip[ss])) { case OCHAR: - if (sp == stop || *sp++ != (char)OPND(s)) + if (sp == stop) + return(NULL); + sp += XMBRTOWC(&wc, sp, stop - sp, &m->mbs, BADCHAR); + if (wc != OPND(s)) return(NULL); break; case OANY: if (sp == stop) return(NULL); - sp++; + sp += XMBRTOWC(&wc, sp, stop - sp, &m->mbs, BADCHAR); + if (wc == BADCHAR) + return (NULL); break; case OANYOF: + if (sp == stop) + return (NULL); cs = &m->g->sets[OPND(s)]; - if (sp == stop || !CHIN(cs, *sp++)) + sp += XMBRTOWC(&wc, sp, stop - sp, &m->mbs, BADCHAR); + if (wc == BADCHAR || !CHIN(cs, wc)) return(NULL); break; case OBOL: @@ -529,6 +681,8 @@ sopno lev; /* PLUS nesting level */ return(NULL); assert(m->pmatch[i].rm_so != -1); len = m->pmatch[i].rm_eo - m->pmatch[i].rm_so; + if (len == 0 && rec++ > MAX_RECURSION) + return(NULL); assert(stop - m->beginp >= len); if (sp > stop - len) return(NULL); /* not enough left to match */ @@ -537,28 +691,28 @@ sopno lev; /* PLUS nesting level */ return(NULL); while (m->g->strip[ss] != SOP(O_BACK, i)) ss++; - return(backref(m, sp+len, stop, ss+1, stopst, lev)); + return(backref(m, sp+len, stop, ss+1, stopst, lev, rec)); break; case OQUEST_: /* to null or not */ - dp = backref(m, sp, stop, ss+1, stopst, lev); + dp = backref(m, sp, stop, ss+1, stopst, lev, rec); if (dp != NULL) return(dp); /* not */ - return(backref(m, sp, stop, ss+OPND(s)+1, stopst, lev)); + return(backref(m, sp, stop, ss+OPND(s)+1, stopst, lev, rec)); break; case OPLUS_: assert(m->lastpos != NULL); assert(lev+1 <= m->g->nplus); m->lastpos[lev+1] = sp; - return(backref(m, sp, stop, ss+1, stopst, lev+1)); + return(backref(m, sp, stop, ss+1, stopst, lev+1, rec)); break; case O_PLUS: if (sp == m->lastpos[lev]) /* last pass matched null */ - return(backref(m, sp, stop, ss+1, stopst, lev-1)); + return(backref(m, sp, stop, ss+1, stopst, lev-1, rec)); /* try another pass */ m->lastpos[lev] = sp; - dp = backref(m, sp, stop, ss-OPND(s)+1, stopst, lev); + dp = backref(m, sp, stop, ss-OPND(s)+1, stopst, lev, rec); if (dp == NULL) - return(backref(m, sp, stop, ss+1, stopst, lev-1)); + return(backref(m, sp, stop, ss+1, stopst, lev-1, rec)); else return(dp); break; @@ -567,7 +721,7 @@ sopno lev; /* PLUS nesting level */ esub = ss + OPND(s) - 1; assert(OP(m->g->strip[esub]) == OOR1); for (;;) { /* find first matching branch */ - dp = backref(m, sp, stop, ssub, esub, lev); + dp = backref(m, sp, stop, ssub, esub, lev, rec); if (dp != NULL) return(dp); /* that one missed, try next one */ @@ -588,7 +742,7 @@ sopno lev; /* PLUS nesting level */ assert(0 < i && i <= m->g->nsub); offsave = m->pmatch[i].rm_so; m->pmatch[i].rm_so = sp - m->offp; - dp = backref(m, sp, stop, ss+1, stopst, lev); + dp = backref(m, sp, stop, ss+1, stopst, lev, rec); if (dp != NULL) return(dp); m->pmatch[i].rm_so = offsave; @@ -599,7 +753,7 @@ sopno lev; /* PLUS nesting level */ assert(0 < i && i <= m->g->nsub); offsave = m->pmatch[i].rm_eo; m->pmatch[i].rm_eo = sp - m->offp; - dp = backref(m, sp, stop, ss+1, stopst, lev); + dp = backref(m, sp, stop, ss+1, stopst, lev, rec); if (dp != NULL) return(dp); m->pmatch[i].rm_eo = offsave; @@ -613,42 +767,57 @@ sopno lev; /* PLUS nesting level */ /* "can't happen" */ assert(nope); /* NOTREACHED */ - return((char *)NULL); /* dummy */ + return "shut up gcc"; } /* - fast - step through the string at top speed - == static char *fast(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); + == static const char *fast(struct match *m, const char *start, \ + == const char *stop, sopno startst, sopno stopst); */ -static char * /* where tentative match ended, or NULL */ -fast(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; +static const char * /* where tentative match ended, or NULL */ +fast( struct match *m, + const char *start, + const char *stop, + sopno startst, + sopno stopst) { - register states st = m->st; - register states fresh = m->fresh; - register states tmp = m->tmp; - register char *p = start; - register int c = (start == m->beginp) ? OUT : *(start-1); - register int lastc; /* previous c */ - register int flagch; - register int i; - register char *coldp; /* last p after which no match was underway */ + states st = m->st; + states fresh = m->fresh; + states tmp = m->tmp; + const char *p = start; + wint_t c; + wint_t lastc; /* previous c */ + wint_t flagch; + int i; + const char *coldp; /* last p after which no match was underway */ + size_t clen; CLEAR(st); SET1(st, startst); + SP("fast", st, *p); st = step(m->g, startst, stopst, st, NOTHING, st); ASSIGN(fresh, st); SP("start", st, *p); coldp = NULL; + if (start == m->beginp) + c = OUT; + else { + /* + * XXX Wrong if the previous character was multi-byte. + * Newline never is (in encodings supported by FreeBSD), + * so this only breaks the ISWORD tests below. + */ + c = (uch)*(start - 1); + } for (;;) { /* next character */ lastc = c; - c = (p == m->endp) ? OUT : *p; + if (p == m->endp) { + clen = 0; + c = OUT; + } else + clen = XMBRTOWC(&c, p, m->endp - p, &m->mbs, BADCHAR); if (EQ(st, fresh)) coldp = p; @@ -686,7 +855,7 @@ sopno stopst; } /* are we done? */ - if (ISSET(st, stopst) || p == stop) + if (ISSET(st, stopst) || p == stop || clen > stop - p) break; /* NOTE BREAK OUT */ /* no, we must deal with this character */ @@ -696,39 +865,39 @@ sopno stopst; st = step(m->g, startst, stopst, tmp, c, st); SP("aft", st, c); assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); - p++; + p += clen; } assert(coldp != NULL); m->coldp = coldp; if (ISSET(st, stopst)) - return(p+1); + return(p+XMBRTOWC(NULL, p, stop - p, &m->mbs, 0)); else return(NULL); } /* - slow - step through the string more deliberately - == static char *slow(register struct match *m, char *start, \ - == char *stop, sopno startst, sopno stopst); + == static const char *slow(struct match *m, const char *start, \ + == const char *stop, sopno startst, sopno stopst); */ -static char * /* where it ended */ -slow(m, start, stop, startst, stopst) -register struct match *m; -char *start; -char *stop; -sopno startst; -sopno stopst; +static const char * /* where it ended */ +slow( struct match *m, + const char *start, + const char *stop, + sopno startst, + sopno stopst) { - register states st = m->st; - register states empty = m->empty; - register states tmp = m->tmp; - register char *p = start; - register int c = (start == m->beginp) ? OUT : *(start-1); - register int lastc; /* previous c */ - register int flagch; - register int i; - register char *matchp; /* last p at which a match ended */ + states st = m->st; + states empty = m->empty; + states tmp = m->tmp; + const char *p = start; + wint_t c; + wint_t lastc; /* previous c */ + wint_t flagch; + int i; + const char *matchp; /* last p at which a match ended */ + size_t clen; AT("slow", start, stop, startst, stopst); CLEAR(st); @@ -736,10 +905,24 @@ sopno stopst; SP("sstart", st, *p); st = step(m->g, startst, stopst, st, NOTHING, st); matchp = NULL; + if (start == m->beginp) + c = OUT; + else { + /* + * XXX Wrong if the previous character was multi-byte. + * Newline never is (in encodings supported by FreeBSD), + * so this only breaks the ISWORD tests below. + */ + c = (uch)*(start - 1); + } for (;;) { /* next character */ lastc = c; - c = (p == m->endp) ? OUT : *p; + if (p == m->endp) { + c = OUT; + clen = 0; + } else + clen = XMBRTOWC(&c, p, m->endp - p, &m->mbs, BADCHAR); /* is there an EOL and/or BOL between lastc and c? */ flagch = '\0'; @@ -777,7 +960,7 @@ sopno stopst; /* are we done? */ if (ISSET(st, stopst)) matchp = p; - if (EQ(st, empty) || p == stop) + if (EQ(st, empty) || p == stop || clen > stop - p) break; /* NOTE BREAK OUT */ /* no, we must deal with this character */ @@ -787,7 +970,7 @@ sopno stopst; st = step(m->g, startst, stopst, tmp, c, st); SP("saft", st, c); assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); - p++; + p += clen; } return(matchp); @@ -796,33 +979,31 @@ sopno stopst; /* - step - map set of states reachable before char to set reachable after - == static states step(register struct re_guts *g, sopno start, sopno stop, \ - == register states bef, int ch, register states aft); - == #define BOL (OUT+1) - == #define EOL (BOL+1) - == #define BOLEOL (BOL+2) - == #define NOTHING (BOL+3) - == #define BOW (BOL+4) - == #define EOW (BOL+5) - == #define CODEMAX (BOL+5) // highest code used - == #define NONCHAR(c) ((c) > CHAR_MAX) - == #define NNONCHAR (CODEMAX-CHAR_MAX) + == static states step(struct re_guts *g, sopno start, sopno stop, \ + == states bef, int ch, states aft); + == #define BOL (OUT-1) + == #define EOL (BOL-1) + == #define BOLEOL (BOL-2) + == #define NOTHING (BOL-3) + == #define BOW (BOL-4) + == #define EOW (BOL-5) + == #define BADCHAR (BOL-6) + == #define NONCHAR(c) ((c) <= OUT) */ static states -step(g, start, stop, bef, ch, aft) -register struct re_guts *g; -sopno start; /* start state within strip */ -sopno stop; /* state after stop state within strip */ -register states bef; /* states reachable before */ -int ch; /* character or NONCHAR code */ -register states aft; /* states already known reachable after */ +step(struct re_guts *g, + sopno start, /* start state within strip */ + sopno stop, /* state after stop state within strip */ + states bef, /* states reachable before */ + wint_t ch, /* character or NONCHAR code */ + states aft) /* states already known reachable after */ { - register cset *cs; - register sop s; - register sopno pc; - register onestate here; /* note, macros know this name */ - register sopno look; - register long i; + cset *cs; + sop s; + sopno pc; + onestate here; /* note, macros know this name */ + sopno look; + int i; for (pc = start, INIT(here, pc); pc != stop; pc++, INC(here)) { s = g->strip[pc]; @@ -832,8 +1013,8 @@ register states aft; /* states already known reachable after */ break; case OCHAR: /* only characters can match */ - assert(!NONCHAR(ch) || ch != (char)OPND(s)); - if (ch == (char)OPND(s)) + assert(!NONCHAR(ch) || ch != OPND(s)); + if (ch == OPND(s)) FWD(aft, bef, 1); break; case OBOL: @@ -900,7 +1081,7 @@ register states aft; /* states already known reachable after */ OP(s = g->strip[pc+look]) != O_CH; look += OPND(s)) assert(OP(s) == OOR2); - FWD(aft, aft, look); + FWD(aft, aft, look + 1); } break; case OOR2: /* propagate OCH_'s marking */ @@ -926,21 +1107,20 @@ register states aft; /* states already known reachable after */ /* - print - print a set of states == #ifdef REDEBUG - == static void print(struct match *m, char *caption, states st, \ + == static void print(struct match *m, const char *caption, states st, \ == int ch, FILE *d); == #endif */ static void -print(m, caption, st, ch, d) -struct match *m; -char *caption; -states st; -int ch; -FILE *d; +print(struct match *m, + const char *caption, + states st, + int ch, + FILE *d) { - register struct re_guts *g = m->g; - register int i; - register int first = 1; + struct re_guts *g = m->g; + int i; + int first = 1; if (!(m->eflags®_TRACE)) return; @@ -959,18 +1139,17 @@ FILE *d; /* - at - print current situation == #ifdef REDEBUG - == static void at(struct match *m, char *title, char *start, char *stop, \ - == sopno startst, sopno stopst); + == static void at(struct match *m, const char *title, const char *start, \ + == const char *stop, sopno startst, sopno stopst); == #endif */ static void -at(m, title, start, stop, startst, stopst) -struct match *m; -char *title; -char *start; -char *stop; -sopno startst; -sopno stopst; +at( struct match *m, + const char *title, + const char *start, + const char *stop, + sopno startst, + sopno stopst) { if (!(m->eflags®_TRACE)) return; @@ -985,7 +1164,7 @@ sopno stopst; /* - pchar - make a character printable == #ifdef REDEBUG - == static char *pchar(int ch); + == static const char *pchar(int ch); == #endif * * Is this identical to regchar() over in debug.c? Well, yes. But a @@ -993,13 +1172,12 @@ sopno stopst; * a matching debug.o, and this is convenient. It all disappears in * the non-debug compilation anyway, so it doesn't matter much. */ -static char * /* -> representation */ -pchar(ch) -int ch; +static const char * /* -> representation */ +pchar(int ch) { static char pbuf[10]; - if (isprint(ch) || ch == ' ') + if (isprint((uch)ch) || ch == ' ') sprintf(pbuf, "%c", ch); else sprintf(pbuf, "\\%o", ch); diff --git a/winsup/cygwin/regex/engine.ih b/winsup/cygwin/regex/engine.ih deleted file mode 100644 index cc98334e7..000000000 --- a/winsup/cygwin/regex/engine.ih +++ /dev/null @@ -1,35 +0,0 @@ -/* ========= begin header generated by ./mkh ========= */ -#ifdef __cplusplus -extern "C" { -#endif - -/* === engine.c === */ -static int matcher(register struct re_guts *g, char *string, size_t nmatch, regmatch_t pmatch[], int eflags); -static char *dissect(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static char *backref(register struct match *m, char *start, char *stop, sopno startst, sopno stopst, sopno lev); -static char *fast(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static char *slow(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); -static states step(register struct re_guts *g, sopno start, sopno stop, register states bef, int ch, register states aft); -#define BOL (OUT+1) -#define EOL (BOL+1) -#define BOLEOL (BOL+2) -#define NOTHING (BOL+3) -#define BOW (BOL+4) -#define EOW (BOL+5) -#define CODEMAX (BOL+5) /* highest code used */ -#define NONCHAR(c) ((c) > CHAR_MAX) -#define NNONCHAR (CODEMAX-CHAR_MAX) -#ifdef REDEBUG -static void print(struct match *m, char *caption, states st, int ch, FILE *d); -#endif -#ifdef REDEBUG -static void at(struct match *m, char *title, char *start, char *stop, sopno startst, sopno stopst); -#endif -#ifdef REDEBUG -static char *pchar(int ch); -#endif - -#ifdef __cplusplus -} -#endif -/* ========= end header generated by ./mkh ========= */ diff --git a/winsup/cygwin/regex/mkh b/winsup/cygwin/regex/mkh deleted file mode 100755 index 252b246c7..000000000 --- a/winsup/cygwin/regex/mkh +++ /dev/null @@ -1,76 +0,0 @@ -#! /bin/sh -# mkh - pull headers out of C source -PATH=/bin:/usr/bin ; export PATH - -# egrep pattern to pick out marked lines -egrep='^ =([ ]|$)' - -# Sed program to process marked lines into lines for the header file. -# The markers have already been removed. Two things are done here: removal -# of backslashed newlines, and some fudging of comments. The first is done -# because -o needs to have prototypes on one line to strip them down. -# Getting comments into the output is tricky; we turn C++-style // comments -# into /* */ comments, after altering any existing */'s to avoid trouble. -peel=' /\\$/N - /\\\n[ ]*/s///g - /\/\//s;\*/;* /;g - /\/\//s;//\(.*\);/*\1 */;' - -for a -do - case "$a" in - -o) # old (pre-function-prototype) compiler - # add code to comment out argument lists - peel="$peel - "'/^\([^#\/][^\/]*[a-zA-Z0-9_)]\)(\(.*\))/s;;\1(/*\2*/);' - shift - ;; - -b) # funny Berkeley __P macro - peel="$peel - "'/^\([^#\/][^\/]*[a-zA-Z0-9_)]\)(\(.*\))/s;;\1 __P((\2));' - shift - ;; - -s) # compiler doesn't like `static foo();' - # add code to get rid of the `static' - peel="$peel - "'/^static[ ][^\/]*[a-zA-Z0-9_)](.*)/s;static.;;' - shift - ;; - -p) # private declarations - egrep='^ ==([ ]|$)' - shift - ;; - -i) # wrap in #ifndef, argument is name - ifndef="$2" - shift ; shift - ;; - *) break - ;; - esac -done - -if test " $ifndef" != " " -then - echo "#ifndef $ifndef" - echo "#define $ifndef /* never again */" -fi -echo "/* ========= begin header generated by $0 ========= */" -echo '#ifdef __cplusplus' -echo 'extern "C" {' -echo '#endif' -for f -do - echo - echo "/* === $f === */" - egrep "$egrep" $f | sed 's/^ ==*[ ]//;s/^ ==*$//' | sed "$peel" - echo -done -echo '#ifdef __cplusplus' -echo '}' -echo '#endif' -echo "/* ========= end header generated by $0 ========= */" -if test " $ifndef" != " " -then - echo "#endif" -fi -exit 0 diff --git a/winsup/cygwin/regex/regcomp.c b/winsup/cygwin/regex/regcomp.c index f771a1ac4..5bcaa98c3 100644 --- a/winsup/cygwin/regex/regcomp.c +++ b/winsup/cygwin/regex/regcomp.c @@ -1,18 +1,77 @@ +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)regcomp.c 8.5 (Berkeley) 3/20/94 + */ + +#if defined(LIBC_SCCS) && !defined(lint) +static char sccsid[] = "@(#)regcomp.c 8.5 (Berkeley) 3/20/94"; +#endif /* LIBC_SCCS and not lint */ +#include +__FBSDID("$FreeBSD: src/lib/libc/regex/regcomp.c,v 1.36 2007/06/11 03:05:54 delphij Exp $"); + +#ifdef __CYGWIN__ #include "winsup.h" +#endif #include #include #include #include #include #include -#include "regex.h" +#include +#ifndef __CYGWIN__ +#include +#endif +#include +#include + +#ifndef __CYGWIN__ +#include "collate.h" +#endif #include "utils.h" #include "regex2.h" -#include "cclass.h" #include "cname.h" +#ifdef __CYGWIN__ +/* Don't pull in windows headers just for LCID. */ +typedef unsigned long LCID; +/* These are defined in nlsfuncs.cc. */ +extern LCID collate_lcid; +extern char collate_charset[]; +#endif + /* * parse structure, passed up and down to avoid global variables and * other clumsinesses @@ -31,7 +90,61 @@ struct parse { sopno pend[NPAREN]; /* -> ) ([0] unused) */ }; -#include "regcomp.ih" +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +/* === regcomp.c === */ +#ifdef __CYGWIN__ /* Defined below `int stop'. Our gcc chokes on that. */ +static void p_ere(struct parse *p, int stop); +#else +static void p_ere(struct parse *p, wint_t stop); +#endif +static void p_ere_exp(struct parse *p); +static void p_str(struct parse *p); +#ifdef __CYGWIN__ /* Defined below `int end1/end2'. Our gcc chokes on that. */ +static void p_bre(struct parse *p, int end1, int end2); +#else +static void p_bre(struct parse *p, wint_t end1, wint_t end2); +#endif +static int p_simp_re(struct parse *p, int starordinary); +static int p_count(struct parse *p); +static void p_bracket(struct parse *p); +static void p_b_term(struct parse *p, cset *cs); +static void p_b_cclass(struct parse *p, cset *cs); +static void p_b_eclass(struct parse *p, cset *cs); +static wint_t p_b_symbol(struct parse *p); +static wint_t p_b_coll_elem(struct parse *p, wint_t endc); +static wint_t othercase(wint_t ch); +static void bothcases(struct parse *p, wint_t ch); +static void ordinary(struct parse *p, wint_t ch); +static void nonnewline(struct parse *p); +static void repeat(struct parse *p, sopno start, int from, int to); +static int seterr(struct parse *p, int e); +static cset *allocset(struct parse *p); +static void freeset(struct parse *p, cset *cs); +static void CHadd(struct parse *p, cset *cs, wint_t ch); +static void CHaddrange(struct parse *p, cset *cs, wint_t min, wint_t max); +static void CHaddtype(struct parse *p, cset *cs, wctype_t wct); +static wint_t singleton(cset *cs); +static sopno dupl(struct parse *p, sopno start, sopno finish); +static void doemit(struct parse *p, sop op, size_t opnd); +static void doinsert(struct parse *p, sop op, size_t opnd, sopno pos); +static void dofwd(struct parse *p, sopno pos, sop value); +static void enlarge(struct parse *p, sopno size); +static void stripsnug(struct parse *p, struct re_guts *g); +static void findmust(struct parse *p, struct re_guts *g); +static int altoffset(sop *scan, int offset); +static void computejumps(struct parse *p, struct re_guts *g); +static void computematchjumps(struct parse *p, struct re_guts *g); +static sopno pluscount(struct parse *p, struct re_guts *g); +static wint_t wgetnext(struct parse *p); + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ static char nuls[10]; /* place to point scanner in event of error */ @@ -51,8 +164,9 @@ static char nuls[10]; /* place to point scanner in event of error */ #define NEXT2() (p->next += 2) #define NEXTn(n) (p->next += (n)) #define GETNEXT() (*p->next++) +#define WGETNEXT() wgetnext(p) #define SETERROR(e) seterr(p, (e)) -#define REQUIRE(co, e) (void) ((co) || SETERROR(e)) +#define REQUIRE(co, e) ((co) || SETERROR(e)) #define MUSTSEE(c, e) (REQUIRE(MORE() && PEEK() == (c), e)) #define MUSTEAT(c, e) (REQUIRE(MORE() && GETNEXT() == (c), e)) #define MUSTNOTSEE(c, e) (REQUIRE(!MORE() || PEEK() != (c), e)) @@ -71,6 +185,9 @@ static int never = 0; /* for use in asserts; shuts lint up */ #define never 0 /* some s have bugs too */ #endif +/* Macro used by computejump()/computematchjump() */ +#define MIN(a,b) ((a)<(b)?(a):(b)) + /* - regcomp - interface for parser and compilation = extern int regcomp(regex_t *, const char *, int); @@ -84,16 +201,15 @@ static int never = 0; /* for use in asserts; shuts lint up */ = #define REG_DUMP 0200 */ int /* 0 success, otherwise REG_something */ -regcomp(preg, pattern, cflags) -regex_t *preg; -const char *pattern; -int cflags; +regcomp(regex_t * __restrict preg, + const char * __restrict pattern, + int cflags) { struct parse pa; - register struct re_guts *g; - register struct parse *p = &pa; - register int i; - register size_t len; + struct re_guts *g; + struct parse *p = &pa; + int i; + size_t len; #ifdef REDEBUG # define GOODFLAGS(f) (f) #else @@ -112,8 +228,7 @@ int cflags; len = strlen((char *)pattern); /* do the mallocs early so failure handling is easy */ - g = (struct re_guts *)malloc(sizeof(struct re_guts) + - (NC-1)*sizeof(cat_t)); + g = (struct re_guts *)malloc(sizeof(struct re_guts)); if (g == NULL) return(REG_ESPACE); p->ssize = len/(size_t)2*(size_t)3 + (size_t)1; /* ugh */ @@ -134,20 +249,18 @@ int cflags; p->pbegin[i] = 0; p->pend[i] = 0; } - g->csetsize = NC; g->sets = NULL; - g->setbits = NULL; g->ncsets = 0; g->cflags = cflags; g->iflags = 0; g->nbol = 0; g->neol = 0; g->must = NULL; + g->moffset = -1; + g->charjump = NULL; + g->matchjump = NULL; g->mlen = 0; g->nsub = 0; - g->ncategories = 1; /* category 0 is "everything else" */ - g->categories = &g->catspace[-(CHAR_MIN)]; - (void) memset((char *)g->catspace, 0, NC*sizeof(cat_t)); g->backrefs = 0; /* do it */ @@ -163,9 +276,19 @@ int cflags; g->laststate = THERE(); /* tidy up loose ends and fill things in */ - categorize(p, g); stripsnug(p, g); findmust(p, g); + /* only use Boyer-Moore algorithm if the pattern is bigger + * than three characters + */ + if(g->mlen > 3) { + computejumps(p, g); + computematchjumps(p, g); + if(g->matchjump == NULL && g->charjump != NULL) { + free(g->charjump); + g->charjump = NULL; + } + } g->nplus = pluscount(p, g); g->magic = MAGIC2; preg->re_nsub = g->nsub; @@ -185,25 +308,24 @@ int cflags; /* - p_ere - ERE parser top level, concatenation and alternation - == static void p_ere(register struct parse *p, int stop); + == static void p_ere(struct parse *p, int stop); */ static void -p_ere(p, stop) -register struct parse *p; -int stop; /* character this ERE should end at */ +p_ere(struct parse *p, + int stop) /* character this ERE should end at */ { - register char c; - register sopno prevback = 0; - register sopno prevfwd = 0; - register sopno conc = 0; - register int first = 1; /* is this the first alternative? */ + char c; + sopno prevback; + sopno prevfwd; + sopno conc; + int first = 1; /* is this the first alternative? */ for (;;) { /* do a bunch of concatenated expressions */ conc = HERE(); while (MORE() && (c = PEEK()) != '|' && c != stop) p_ere_exp(p); - REQUIRE(HERE() != conc, REG_EMPTY); /* require nonempty */ + (void)REQUIRE(HERE() != conc, REG_EMPTY); /* require nonempty */ if (!EAT('|')) break; /* NOTE BREAK OUT */ @@ -231,17 +353,17 @@ int stop; /* character this ERE should end at */ /* - p_ere_exp - parse one subERE, an atom possibly followed by a repetition op - == static void p_ere_exp(register struct parse *p); + == static void p_ere_exp(struct parse *p); */ static void -p_ere_exp(p) -register struct parse *p; +p_ere_exp(struct parse *p) { - register char c; - register sopno pos; - register int count; - register int count2; - register sopno subno; + char c; + wint_t wc; + sopno pos; + int count; + int count2; + sopno subno; int wascaret = 0; assert(MORE()); /* caller should have ensured this */ @@ -250,7 +372,7 @@ register struct parse *p; pos = HERE(); switch (c) { case '(': - REQUIRE(MORE(), REG_EPAREN); + (void)REQUIRE(MORE(), REG_EPAREN); p->g->nsub++; subno = p->g->nsub; if (subno < NPAREN) @@ -263,7 +385,7 @@ register struct parse *p; assert(p->pend[subno] != 0); } EMIT(ORPAREN, subno); - MUSTEAT(')', REG_EPAREN); + (void)MUSTEAT(')', REG_EPAREN); break; #ifndef POSIX_MISTAKE case ')': /* happens only if no current unmatched ( */ @@ -306,15 +428,33 @@ register struct parse *p; p_bracket(p); break; case '\\': - REQUIRE(MORE(), REG_EESCAPE); - c = GETNEXT(); - ordinary(p, c); + (void)REQUIRE(MORE(), REG_EESCAPE); + wc = WGETNEXT(); +#ifdef __CYGWIN__ + /* \< and \> are the GNU equivalents to [[:<:]] and [[:>:]] */ + switch (wc) + { + case L'<': + EMIT(OBOW, 0); + break; + case L'>': + EMIT(OEOW, 0); + break; + default: + ordinary(p, wc); + break; + } +#else + ordinary(p, wc); +#endif break; case '{': /* okay as ordinary except if digit follows */ - REQUIRE(!MORE() || !isdigit((unsigned char)PEEK()), REG_BADRPT); + (void)REQUIRE(!MORE() || !isdigit((uch)PEEK()), REG_BADRPT); /* FALLTHROUGH */ default: - ordinary(p, c); + p->next--; + wc = WGETNEXT(); + ordinary(p, wc); break; } @@ -323,11 +463,11 @@ register struct parse *p; c = PEEK(); /* we call { a repetition if followed by a digit */ if (!( c == '*' || c == '+' || c == '?' || - (c == '{' && MORE2() && isdigit((unsigned char)PEEK2())) )) + (c == '{' && MORE2() && isdigit((uch)PEEK2())) )) return; /* no repetition, we're done */ NEXT(); - REQUIRE(!wascaret, REG_BADRPT); + (void)REQUIRE(!wascaret, REG_BADRPT); switch (c) { case '*': /* implemented as +? */ /* this case does not require the (y|) trick, noKLUDGE */ @@ -352,9 +492,9 @@ register struct parse *p; case '{': count = p_count(p); if (EAT(',')) { - if (isdigit((unsigned char)PEEK())) { + if (isdigit((uch)PEEK())) { count2 = p_count(p); - REQUIRE(count <= count2, REG_BADBR); + (void)REQUIRE(count <= count2, REG_BADBR); } else /* single number with comma */ count2 = INFINITY; } else /* just a single number */ @@ -363,7 +503,7 @@ register struct parse *p; if (!EAT('}')) { /* error heuristics */ while (MORE() && PEEK() != '}') NEXT(); - REQUIRE(MORE(), REG_EBRACE); + (void)REQUIRE(MORE(), REG_EBRACE); SETERROR(REG_BADBR); } break; @@ -373,45 +513,41 @@ register struct parse *p; return; c = PEEK(); if (!( c == '*' || c == '+' || c == '?' || - (c == '{' && MORE2() && isdigit((unsigned char)PEEK2())) ) ) + (c == '{' && MORE2() && isdigit((uch)PEEK2())) ) ) return; SETERROR(REG_BADRPT); } /* - p_str - string (no metacharacters) "parser" - == static void p_str(register struct parse *p); + == static void p_str(struct parse *p); */ static void -p_str(p) -register struct parse *p; +p_str(struct parse *p) { - REQUIRE(MORE(), REG_EMPTY); + (void)REQUIRE(MORE(), REG_EMPTY); while (MORE()) - ordinary(p, GETNEXT()); + ordinary(p, WGETNEXT()); } /* - p_bre - BRE parser top level, anchoring and concatenation - == static void p_bre(register struct parse *p, register int end1, \ - == register int end2); + == static void p_bre(struct parse *p, int end1, \ + == int end2); * Giving end1 as OUT essentially eliminates the end1/end2 check. * * This implementation is a bit of a kludge, in that a trailing $ is first - * taken as an ordinary character and then revised to be an anchor. The - * only undesirable side effect is that '$' gets included as a character - * category in such cases. This is fairly harmless; not worth fixing. + * taken as an ordinary character and then revised to be an anchor. * The amount of lookahead needed to avoid this kludge is excessive. */ static void -p_bre(p, end1, end2) -register struct parse *p; -register int end1; /* first terminating character */ -register int end2; /* second terminating character */ +p_bre(struct parse *p, + int end1, /* first terminating character */ + int end2) /* second terminating character */ { - register sopno start = HERE(); - register int first = 1; /* first subexpression? */ - register int wasdollar = 0; + sopno start = HERE(); + int first = 1; /* first subexpression? */ + int wasdollar = 0; if (EAT('^')) { EMIT(OBOL, 0); @@ -429,24 +565,24 @@ register int end2; /* second terminating character */ p->g->neol++; } - REQUIRE(HERE() != start, REG_EMPTY); /* require nonempty */ + (void)REQUIRE(HERE() != start, REG_EMPTY); /* require nonempty */ } /* - p_simp_re - parse a simple RE, an atom possibly followed by a repetition - == static int p_simp_re(register struct parse *p, int starordinary); + == static int p_simp_re(struct parse *p, int starordinary); */ static int /* was the simple RE an unbackslashed $? */ -p_simp_re(p, starordinary) -register struct parse *p; -int starordinary; /* is a leading * an ordinary character? */ +p_simp_re(struct parse *p, + int starordinary) /* is a leading * an ordinary character? */ { - register int c; - register int count; - register int count2; - register sopno pos; - register int i; - register sopno subno; + int c; + int count; + int count2; + sopno pos; + int i; + wint_t wc; + sopno subno; # define BACKSL (1<': + /* \> is the GNU equivalents to [[:>:]] */ + EMIT(OEOW, 0); + break; +#endif case BACKSL|'{': SETERROR(REG_BADRPT); break; @@ -484,7 +630,7 @@ int starordinary; /* is a leading * an ordinary character? */ assert(p->pend[subno] != 0); } EMIT(ORPAREN, subno); - REQUIRE(EATTWO('\\', ')'), REG_EPAREN); + (void)REQUIRE(EATTWO('\\', ')'), REG_EPAREN); break; case BACKSL|')': /* should not get here -- must be user */ case BACKSL|'}': @@ -514,10 +660,12 @@ int starordinary; /* is a leading * an ordinary character? */ p->g->backrefs = 1; break; case '*': - REQUIRE(starordinary, REG_BADRPT); + (void)REQUIRE(starordinary, REG_BADRPT); /* FALLTHROUGH */ default: - ordinary(p, (char)c); /* takes off BACKSL, if any */ + p->next--; + wc = WGETNEXT(); + ordinary(p, wc); break; } @@ -530,9 +678,9 @@ int starordinary; /* is a leading * an ordinary character? */ } else if (EATTWO('\\', '{')) { count = p_count(p); if (EAT(',')) { - if (MORE() && isdigit((unsigned char)PEEK())) { + if (MORE() && isdigit((uch)PEEK())) { count2 = p_count(p); - REQUIRE(count <= count2, REG_BADBR); + (void)REQUIRE(count <= count2, REG_BADBR); } else /* single number with comma */ count2 = INFINITY; } else /* just a single number */ @@ -541,10 +689,10 @@ int starordinary; /* is a leading * an ordinary character? */ if (!EATTWO('\\', '}')) { /* error heuristics */ while (MORE() && !SEETWO('\\', '}')) NEXT(); - REQUIRE(MORE(), REG_EBRACE); + (void)REQUIRE(MORE(), REG_EBRACE); SETERROR(REG_BADBR); } - } else if (c == (unsigned char)'$') /* $ (but not \$) ends it */ + } else if (c == '$') /* $ (but not \$) ends it */ return(1); return(0); @@ -552,37 +700,32 @@ int starordinary; /* is a leading * an ordinary character? */ /* - p_count - parse a repetition count - == static int p_count(register struct parse *p); + == static int p_count(struct parse *p); */ static int /* the value */ -p_count(p) -register struct parse *p; +p_count(struct parse *p) { - register int count = 0; - register int ndigits = 0; + int count = 0; + int ndigits = 0; - while (MORE() && isdigit((unsigned char)PEEK()) && count <= DUPMAX) { + while (MORE() && isdigit((uch)PEEK()) && count <= DUPMAX) { count = count*10 + (GETNEXT() - '0'); ndigits++; } - REQUIRE(ndigits > 0 && count <= DUPMAX, REG_BADBR); + (void)REQUIRE(ndigits > 0 && count <= DUPMAX, REG_BADBR); return(count); } /* - p_bracket - parse a bracketed character list - == static void p_bracket(register struct parse *p); - * - * Note a significant property of this code: if the allocset() did SETERROR, - * no set operations are done. + == static void p_bracket(struct parse *p); */ static void -p_bracket(p) -register struct parse *p; +p_bracket(struct parse *p) { - register cset *cs = allocset(p); - register int invert = 0; + cset *cs; + wint_t ch; /* Dept of Truly Sickening Special-Case Kludges */ if (p->next + 5 < p->end && strncmp(p->next, "[:<:]]", 6) == 0) { @@ -596,69 +739,57 @@ register struct parse *p; return; } + if ((cs = allocset(p)) == NULL) + return; + + if (p->g->cflags®_ICASE) + cs->icase = 1; if (EAT('^')) - invert++; /* make note to invert set at end */ + cs->invert = 1; if (EAT(']')) - CHadd(cs, ']'); + CHadd(p, cs, ']'); else if (EAT('-')) - CHadd(cs, '-'); + CHadd(p, cs, '-'); while (MORE() && PEEK() != ']' && !SEETWO('-', ']')) p_b_term(p, cs); if (EAT('-')) - CHadd(cs, '-'); - MUSTEAT(']', REG_EBRACK); + CHadd(p, cs, '-'); + (void)MUSTEAT(']', REG_EBRACK); if (p->error != 0) /* don't mess things up further */ return; - if (p->g->cflags®_ICASE) { - register int i; - register int ci; + if (cs->invert && p->g->cflags®_NEWLINE) + cs->bmp['\n' >> 3] |= 1 << ('\n' & 7); - for (i = p->g->csetsize - 1; i >= 0; i--) - if (CHIN(cs, i) && isalpha(i)) { - ci = othercase(i); - if (ci != i) - CHadd(cs, ci); - } - if (cs->multis != NULL) - mccase(p, cs); - } - if (invert) { - register int i; - - for (i = p->g->csetsize - 1; i >= 0; i--) - if (CHIN(cs, i)) - CHsub(cs, i); - else - CHadd(cs, i); - if (p->g->cflags®_NEWLINE) - CHsub(cs, '\n'); - if (cs->multis != NULL) - mcinvert(p, cs); - } - - assert(cs->multis == NULL); /* xxx */ - - if (nch(p, cs) == 1) { /* optimize singleton sets */ - ordinary(p, firstch(p, cs)); + if ((ch = singleton(cs)) != OUT) { /* optimize singleton sets */ + ordinary(p, ch); freeset(p, cs); } else - EMIT(OANYOF, freezeset(p, cs)); + EMIT(OANYOF, (int)(cs - p->g->sets)); } +#ifdef __CYGWIN__ +/* This function is usually part of FreeBSD's libc. */ +int +__collate_range_cmp(int c1, int c2) +{ + char s1[2] = { c1, '\0' }; + char s2[2] = { c2, '\0' }; + return strcoll (s1, s2); +} +#endif + /* - p_b_term - parse one term of a bracketed character list - == static void p_b_term(register struct parse *p, register cset *cs); + == static void p_b_term(struct parse *p, cset *cs); */ static void -p_b_term(p, cs) -register struct parse *p; -register cset *cs; +p_b_term(struct parse *p, cset *cs) { - register char c; - register char start, finish; - register int i; + char c; + wint_t start, finish; + wint_t i; /* classify what we've got */ switch ((MORE()) ? PEEK() : '\0') { @@ -677,24 +808,23 @@ register cset *cs; switch (c) { case ':': /* character class */ NEXT2(); - REQUIRE(MORE(), REG_EBRACK); + (void)REQUIRE(MORE(), REG_EBRACK); c = PEEK(); - REQUIRE(c != '-' && c != ']', REG_ECTYPE); + (void)REQUIRE(c != '-' && c != ']', REG_ECTYPE); p_b_cclass(p, cs); - REQUIRE(MORE(), REG_EBRACK); - REQUIRE(EATTWO(':', ']'), REG_ECTYPE); + (void)REQUIRE(MORE(), REG_EBRACK); + (void)REQUIRE(EATTWO(':', ']'), REG_ECTYPE); break; case '=': /* equivalence class */ NEXT2(); - REQUIRE(MORE(), REG_EBRACK); + (void)REQUIRE(MORE(), REG_EBRACK); c = PEEK(); - REQUIRE(c != '-' && c != ']', REG_ECOLLATE); + (void)REQUIRE(c != '-' && c != ']', REG_ECOLLATE); p_b_eclass(p, cs); - REQUIRE(MORE(), REG_EBRACK); - REQUIRE(EATTWO('=', ']'), REG_ECOLLATE); + (void)REQUIRE(MORE(), REG_EBRACK); + (void)REQUIRE(EATTWO('=', ']'), REG_ECOLLATE); break; default: /* symbol, ordinary character, or range */ -/* xxx revision needed for multichar stuff */ start = p_b_symbol(p); if (SEE('-') && MORE2() && PEEK2() != ']') { /* range */ @@ -705,97 +835,106 @@ register cset *cs; finish = p_b_symbol(p); } else finish = start; -/* xxx what about signed chars here... */ - REQUIRE(start <= finish, REG_ERANGE); - for (i = start; i <= finish; i++) - CHadd(cs, i); + if (start == finish) + CHadd(p, cs, start); + else { +#ifdef __CYGWIN__ + if (!collate_lcid) { +#else + if (__collate_load_error) { +#endif + (void)REQUIRE((uch)start <= (uch)finish, REG_ERANGE); + CHaddrange(p, cs, start, finish); + } else { + (void)REQUIRE(__collate_range_cmp(start, finish) <= 0, REG_ERANGE); + for (i = 0; i <= UCHAR_MAX; i++) { + if ( __collate_range_cmp(start, i) <= 0 + && __collate_range_cmp(i, finish) <= 0 + ) + CHadd(p, cs, i); + } + } + } break; } } /* - p_b_cclass - parse a character-class name and deal with it - == static void p_b_cclass(register struct parse *p, register cset *cs); + == static void p_b_cclass(struct parse *p, cset *cs); */ static void -p_b_cclass(p, cs) -register struct parse *p; -register cset *cs; +p_b_cclass(struct parse *p, cset *cs) { - register char *sp = p->next; - register struct cclass *cp; - register size_t len; - register const char *u; - register char c; + char *sp = p->next; + size_t len; + wctype_t wct; + char clname[16]; - while (MORE() && isalpha((unsigned char)PEEK())) + while (MORE() && isalpha((uch)PEEK())) NEXT(); len = p->next - sp; - for (cp = cclasses; cp->name != NULL; cp++) - if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0') - break; - if (cp->name == NULL) { - /* oops, didn't find it */ + if (len >= sizeof(clname) - 1) { SETERROR(REG_ECTYPE); return; } - - u = cp->chars; - while ((c = *u++) != '\0') - CHadd(cs, c); - for (u = cp->multis; *u != '\0'; u += strlen(u) + 1) - MCadd(p, cs, u); + memcpy(clname, sp, len); + clname[len] = '\0'; + if ((wct = wctype(clname)) == 0) { + SETERROR(REG_ECTYPE); + return; + } + CHaddtype(p, cs, wct); } /* - p_b_eclass - parse an equivalence-class name and deal with it - == static void p_b_eclass(register struct parse *p, register cset *cs); + == static void p_b_eclass(struct parse *p, cset *cs); * * This implementation is incomplete. xxx */ static void -p_b_eclass(p, cs) -register struct parse *p; -register cset *cs; +p_b_eclass(struct parse *p, cset *cs) { - register char c; + wint_t c; c = p_b_coll_elem(p, '='); - CHadd(cs, c); + CHadd(p, cs, c); } /* - p_b_symbol - parse a character or [..]ed multicharacter collating symbol - == static char p_b_symbol(register struct parse *p); + == static char p_b_symbol(struct parse *p); */ -static char /* value of symbol */ -p_b_symbol(p) -register struct parse *p; +static wint_t /* value of symbol */ +p_b_symbol(struct parse *p) { - register char value; + wint_t value; - REQUIRE(MORE(), REG_EBRACK); + (void)REQUIRE(MORE(), REG_EBRACK); if (!EATTWO('[', '.')) - return(GETNEXT()); + return(WGETNEXT()); /* collating symbol */ value = p_b_coll_elem(p, '.'); - REQUIRE(EATTWO('.', ']'), REG_ECOLLATE); + (void)REQUIRE(EATTWO('.', ']'), REG_ECOLLATE); return(value); } /* - p_b_coll_elem - parse a collating-element name and look it up - == static char p_b_coll_elem(register struct parse *p, int endc); + == static char p_b_coll_elem(struct parse *p, int endc); */ -static char /* value of collating element */ -p_b_coll_elem(p, endc) -register struct parse *p; -int endc; /* name ended by endc,']' */ +static wint_t /* value of collating element */ +p_b_coll_elem(struct parse *p, + wint_t endc) /* name ended by endc,']' */ { - register char *sp = p->next; - register struct cname *cp; - register int len; + char *sp = p->next; + struct cname *cp; + int len; + mbstate_t mbs; + wchar_t wc; + size_t clen; while (MORE() && !SEETWO(endc, ']')) NEXT(); @@ -807,9 +946,13 @@ int endc; /* name ended by endc,']' */ for (cp = cnames; cp->name != NULL; cp++) if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0') return(cp->code); /* known name */ - if (len == 1) - return(*sp); /* single character */ - SETERROR(REG_ECOLLATE); /* neither */ + memset(&mbs, 0, sizeof(mbs)); + if ((clen = mbrtowc(&wc, sp, len, &mbs)) == len) + return (wc); /* single character */ + else if (clen == (size_t)-1 || clen == (size_t)-2) + SETERROR(REG_ILLSEQ); + else + SETERROR(REG_ECOLLATE); /* neither */ return(0); } @@ -817,78 +960,83 @@ int endc; /* name ended by endc,']' */ - othercase - return the case counterpart of an alphabetic == static char othercase(int ch); */ -static char /* if no counterpart, return ch */ -othercase(ch) -int ch; +static wint_t /* if no counterpart, return ch */ +othercase(wint_t ch) { - assert(isalpha(ch)); - if (isupper(ch)) - return(tolower(ch)); - else if (islower(ch)) - return(toupper(ch)); + assert(iswalpha(ch)); + if (iswupper(ch)) + return(towlower(ch)); + else if (iswlower(ch)) + return(towupper(ch)); else /* peculiar, but could happen */ return(ch); } /* - bothcases - emit a dualcase version of a two-case character - == static void bothcases(register struct parse *p, int ch); + == static void bothcases(struct parse *p, int ch); * * Boy, is this implementation ever a kludge... */ static void -bothcases(p, ch) -register struct parse *p; -int ch; +bothcases(struct parse *p, wint_t ch) { - register char *oldnext = p->next; - register char *oldend = p->end; - char bracket[3]; + char *oldnext = p->next; + char *oldend = p->end; + char bracket[3 + MB_LEN_MAX]; + size_t n; + mbstate_t mbs; assert(othercase(ch) != ch); /* p_bracket() would recurse */ p->next = bracket; - p->end = bracket+2; - bracket[0] = ch; - bracket[1] = ']'; - bracket[2] = '\0'; + memset(&mbs, 0, sizeof(mbs)); + n = wcrtomb(bracket, ch, &mbs); + assert(n != (size_t)-1); + bracket[n] = ']'; + bracket[n + 1] = '\0'; + p->end = bracket+n+1; p_bracket(p); - assert(p->next == bracket+2); + assert(p->next == p->end); p->next = oldnext; p->end = oldend; } /* - ordinary - emit an ordinary character - == static void ordinary(register struct parse *p, register int ch); + == static void ordinary(struct parse *p, int ch); */ static void -ordinary(p, ch) -register struct parse *p; -register int ch; +ordinary(struct parse *p, wint_t ch) { - register cat_t *cap = p->g->categories; + cset *cs; - if ((p->g->cflags®_ICASE) && isalpha(ch) && othercase(ch) != ch) + if ((p->g->cflags®_ICASE) && iswalpha(ch) && othercase(ch) != ch) bothcases(p, ch); + else if ((ch & OPDMASK) == ch) + EMIT(OCHAR, ch); else { - EMIT(OCHAR, (unsigned char)ch); - if (cap[ch] == 0) - cap[ch] = p->g->ncategories++; + /* + * Kludge: character is too big to fit into an OCHAR operand. + * Emit a singleton set. + */ + if ((cs = allocset(p)) == NULL) + return; + CHadd(p, cs, ch); + EMIT(OANYOF, (int)(cs - p->g->sets)); } } /* - nonnewline - emit REG_NEWLINE version of OANY - == static void nonnewline(register struct parse *p); + == static void nonnewline(struct parse *p); * * Boy, is this implementation ever a kludge... */ static void -nonnewline(p) -register struct parse *p; +nonnewline(struct parse *p) { - register char *oldnext = p->next; - register char *oldend = p->end; + char *oldnext = p->next; + char *oldend = p->end; char bracket[4]; p->next = bracket; @@ -905,21 +1053,20 @@ register struct parse *p; /* - repeat - generate code for a bounded repetition, recursively if needed - == static void repeat(register struct parse *p, sopno start, int from, int to); + == static void repeat(struct parse *p, sopno start, int from, int to); */ static void -repeat(p, start, from, to) -register struct parse *p; -sopno start; /* operand from here to end of strip */ -int from; /* repeated from this number */ -int to; /* to this number of times (maybe INFINITY) */ +repeat(struct parse *p, + sopno start, /* operand from here to end of strip */ + int from, /* repeated from this number */ + int to) /* to this number of times (maybe INFINITY) */ { - register sopno finish = HERE(); + sopno finish = HERE(); # define N 2 # define INF 3 # define REP(f, t) ((f)*8 + (t)) # define MAP(n) (((n) <= 1) ? (n) : ((n) == INFINITY) ? INF : N) - register sopno copy; + sopno copy; if (p->error != 0) /* head off possible runaway recursion */ return; @@ -975,14 +1122,36 @@ int to; /* to this number of times (maybe INFINITY) */ } } +/* + - wgetnext - helper function for WGETNEXT() macro. Gets the next wide + - character from the parse struct, signals a REG_ILLSEQ error if the + - character can't be converted. Returns the number of bytes consumed. + */ +static wint_t +wgetnext(struct parse *p) +{ + mbstate_t mbs; + wchar_t wc; + size_t n; + + memset(&mbs, 0, sizeof(mbs)); + n = mbrtowc(&wc, p->next, p->end - p->next, &mbs); + if (n == (size_t)-1 || n == (size_t)-2) { + SETERROR(REG_ILLSEQ); + return (0); + } + if (n == 0) + n = 1; + p->next += n; + return (wc); +} + /* - seterr - set an error condition - == static int seterr(register struct parse *p, int e); + == static int seterr(struct parse *p, int e); */ static int /* useless but makes type checking happy */ -seterr(p, e) -register struct parse *p; -int e; +seterr(struct parse *p, int e) { if (p->error == 0) /* keep earliest error condition */ p->error = e; @@ -993,295 +1162,150 @@ int e; /* - allocset - allocate a set of characters for [] - == static cset *allocset(register struct parse *p); + == static cset *allocset(struct parse *p); */ static cset * -allocset(p) -register struct parse *p; +allocset(struct parse *p) { - register int no = p->g->ncsets++; - register size_t nc; - register size_t nbytes; - register cset *cs; - register size_t css = (size_t)p->g->csetsize; - register int i; + cset *cs, *ncs; - if (no >= p->ncsalloc) { /* need another column of space */ - p->ncsalloc += CHAR_BIT; - nc = p->ncsalloc; - assert(nc % CHAR_BIT == 0); - nbytes = nc / CHAR_BIT * css; - if (p->g->sets == NULL) - p->g->sets = (cset *)malloc(nc * sizeof(cset)); - else - p->g->sets = (cset *)realloc((char *)p->g->sets, - nc * sizeof(cset)); - if (p->g->setbits == NULL) - p->g->setbits = (uch *)malloc(nbytes); - else { - p->g->setbits = (uch *)realloc((char *)p->g->setbits, - nbytes); - /* xxx this isn't right if setbits is now NULL */ - for (i = 0; i < no; i++) - p->g->sets[i].ptr = p->g->setbits + css*(i/CHAR_BIT); - } - if (p->g->sets != NULL && p->g->setbits != NULL) - (void) memset((char *)p->g->setbits + (nbytes - css), - 0, css); - else { - no = 0; - SETERROR(REG_ESPACE); - /* caller's responsibility not to do set ops */ - } + ncs = realloc(p->g->sets, (p->g->ncsets + 1) * sizeof(*ncs)); + if (ncs == NULL) { + SETERROR(REG_ESPACE); + return (NULL); } - - assert(p->g->sets != NULL); /* xxx */ - cs = &p->g->sets[no]; - cs->ptr = p->g->setbits + css*((no)/CHAR_BIT); - cs->mask = 1 << ((no) % CHAR_BIT); - cs->hash = 0; - cs->smultis = 0; - cs->multis = NULL; + p->g->sets = ncs; + cs = &p->g->sets[p->g->ncsets++]; + memset(cs, 0, sizeof(*cs)); return(cs); } /* - freeset - free a now-unused set - == static void freeset(register struct parse *p, register cset *cs); + == static void freeset(struct parse *p, cset *cs); */ static void -freeset(p, cs) -register struct parse *p; -register cset *cs; +freeset(struct parse *p, cset *cs) { - register size_t i; - register cset *top = &p->g->sets[p->g->ncsets]; - register size_t css = (size_t)p->g->csetsize; + cset *top = &p->g->sets[p->g->ncsets]; - for (i = 0; i < css; i++) - CHsub(cs, i); + free(cs->wides); + free(cs->ranges); + free(cs->types); + memset(cs, 0, sizeof(*cs)); if (cs == top-1) /* recover only the easy case */ p->g->ncsets--; } /* - - freezeset - final processing on a set of characters - == static int freezeset(register struct parse *p, register cset *cs); - * - * The main task here is merging identical sets. This is usually a waste - * of time (although the hash code minimizes the overhead), but can win - * big if REG_ICASE is being used. REG_ICASE, by the way, is why the hash - * is done using addition rather than xor -- all ASCII [aA] sets xor to - * the same value! + - singleton - Determine whether a set contains only one character, + - returning it if so, otherwise returning OUT. */ -static int /* set number */ -freezeset(p, cs) -register struct parse *p; -register cset *cs; +static wint_t +singleton(cset *cs) { - register uch h = cs->hash; - register size_t i; - register cset *top = &p->g->sets[p->g->ncsets]; - register cset *cs2; - register size_t css = (size_t)p->g->csetsize; + wint_t i, s, n; - /* look for an earlier one which is the same */ - for (cs2 = &p->g->sets[0]; cs2 < top; cs2++) - if (cs2->hash == h && cs2 != cs) { - /* maybe */ - for (i = 0; i < css; i++) - if (!!CHIN(cs2, i) != !!CHIN(cs, i)) - break; /* no */ - if (i == css) - break; /* yes */ - } - - if (cs2 < top) { /* found one */ - freeset(p, cs); - cs = cs2; - } - - return((int)(cs - p->g->sets)); -} - -/* - - firstch - return first character in a set (which must have at least one) - == static int firstch(register struct parse *p, register cset *cs); - */ -static int /* character; there is no "none" value */ -firstch(p, cs) -register struct parse *p; -register cset *cs; -{ - register size_t i; - register size_t css = (size_t)p->g->csetsize; - - for (i = 0; i < css; i++) - if (CHIN(cs, i)) - return((char)i); - assert(never); - return(0); /* arbitrary */ -} - -/* - - nch - number of characters in a set - == static int nch(register struct parse *p, register cset *cs); - */ -static int -nch(p, cs) -register struct parse *p; -register cset *cs; -{ - register size_t i; - register size_t css = (size_t)p->g->csetsize; - register int n = 0; - - for (i = 0; i < css; i++) - if (CHIN(cs, i)) + for (i = n = 0; i < NC; i++) + if (CHIN(cs, i)) { n++; - return(n); + s = i; + } + if (n == 1) + return (s); + if (cs->nwides == 1 && cs->nranges == 0 && cs->ntypes == 0 && + cs->icase == 0) + return (cs->wides[0]); + /* Don't bother handling the other cases. */ + return (OUT); } /* - - mcadd - add a collating element to a cset - == static void mcadd(register struct parse *p, register cset *cs, \ - == register char *cp); + - CHadd - add character to character set. */ static void -mcadd(p, cs, cp) -register struct parse *p; -register cset *cs; -register const char *cp; +CHadd(struct parse *p, cset *cs, wint_t ch) { - register size_t oldend = cs->smultis; + wint_t nch, *newwides; + assert(ch >= 0); + if (ch < NC) + cs->bmp[ch >> 3] |= 1 << (ch & 7); + else { + newwides = realloc(cs->wides, (cs->nwides + 1) * + sizeof(*cs->wides)); + if (newwides == NULL) { + SETERROR(REG_ESPACE); + return; + } + cs->wides = newwides; + cs->wides[cs->nwides++] = ch; + } + if (cs->icase) { + if ((nch = towlower(ch)) < NC) + cs->bmp[nch >> 3] |= 1 << (nch & 7); + if ((nch = towupper(ch)) < NC) + cs->bmp[nch >> 3] |= 1 << (nch & 7); + } +} - cs->smultis += strlen(cp) + 1; - if (cs->multis == NULL) - cs->multis = malloc(cs->smultis); - else - cs->multis = realloc(cs->multis, cs->smultis); - if (cs->multis == NULL) { +/* + - CHaddrange - add all characters in the range [min,max] to a character set. + */ +static void +CHaddrange(struct parse *p, cset *cs, wint_t min, wint_t max) +{ + crange *newranges; + + for (; min < NC && min <= max; min++) + CHadd(p, cs, min); + if (min >= max) + return; + newranges = realloc(cs->ranges, (cs->nranges + 1) * + sizeof(*cs->ranges)); + if (newranges == NULL) { SETERROR(REG_ESPACE); return; } - - (void) strcpy(cs->multis + oldend - 1, cp); - cs->multis[cs->smultis - 1] = '\0'; + cs->ranges = newranges; + cs->ranges[cs->nranges].min = min; + cs->ranges[cs->nranges].min = max; + cs->nranges++; } /* - - mcinvert - invert the list of collating elements in a cset - == static void mcinvert(register struct parse *p, register cset *cs); - * - * This would have to know the set of possibilities. Implementation - * is deferred. + - CHaddtype - add all characters of a certain type to a character set. */ static void -mcinvert(p, cs) -register struct parse *p; -register cset *cs; +CHaddtype(struct parse *p, cset *cs, wctype_t wct) { - assert(cs->multis == NULL); /* xxx */ -} + wint_t i; + wctype_t *newtypes; -/* - - mccase - add case counterparts of the list of collating elements in a cset - == static void mccase(register struct parse *p, register cset *cs); - * - * This would have to know the set of possibilities. Implementation - * is deferred. - */ -static void -mccase(p, cs) -register struct parse *p; -register cset *cs; -{ - assert(cs->multis == NULL); /* xxx */ -} - -/* - - isinsets - is this character in any sets? - == static int isinsets(register struct re_guts *g, int c); - */ -static int /* predicate */ -isinsets(g, c) -register struct re_guts *g; -int c; -{ - register uch *col; - register int i; - register int ncols = (g->ncsets+(CHAR_BIT-1)) / CHAR_BIT; - register unsigned uc = (unsigned char)c; - - for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) - if (col[uc] != 0) - return(1); - return(0); -} - -/* - - samesets - are these two characters in exactly the same sets? - == static int samesets(register struct re_guts *g, int c1, int c2); - */ -static int /* predicate */ -samesets(g, c1, c2) -register struct re_guts *g; -int c1; -int c2; -{ - register uch *col; - register int i; - register int ncols = (g->ncsets+(CHAR_BIT-1)) / CHAR_BIT; - register unsigned uc1 = (unsigned char)c1; - register unsigned uc2 = (unsigned char)c2; - - for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) - if (col[uc1] != col[uc2]) - return(0); - return(1); -} - -/* - - categorize - sort out character categories - == static void categorize(struct parse *p, register struct re_guts *g); - */ -static void -categorize(p, g) -struct parse *p; -register struct re_guts *g; -{ - register cat_t *cats = g->categories; - register int c; - register int c2; - register cat_t cat; - - /* avoid making error situations worse */ - if (p->error != 0) + for (i = 0; i < NC; i++) + if (iswctype(i, wct)) + CHadd(p, cs, i); + newtypes = realloc(cs->types, (cs->ntypes + 1) * + sizeof(*cs->types)); + if (newtypes == NULL) { + SETERROR(REG_ESPACE); return; - - for (c = CHAR_MIN; c <= CHAR_MAX; c++) - if (cats[c] == 0 && isinsets(g, c)) { - cat = g->ncategories++; - cats[c] = cat; - for (c2 = c+1; c2 <= CHAR_MAX; c2++) - if (cats[c2] == 0 && samesets(g, c, c2)) - cats[c2] = cat; - } + } + cs->types = newtypes; + cs->types[cs->ntypes++] = wct; } /* - dupl - emit a duplicate of a bunch of sops - == static sopno dupl(register struct parse *p, sopno start, sopno finish); + == static sopno dupl(struct parse *p, sopno start, sopno finish); */ static sopno /* start of duplicate */ -dupl(p, start, finish) -register struct parse *p; -sopno start; /* from here */ -sopno finish; /* to this less one */ +dupl(struct parse *p, + sopno start, /* from here */ + sopno finish) /* to this less one */ { - register sopno ret = HERE(); - register sopno len = finish - start; + sopno ret = HERE(); + sopno len = finish - start; assert(finish >= start); if (len == 0) @@ -1296,17 +1320,14 @@ sopno finish; /* to this less one */ /* - doemit - emit a strip operator - == static void doemit(register struct parse *p, sop op, size_t opnd); + == static void doemit(struct parse *p, sop op, size_t opnd); * * It might seem better to implement this as a macro with a function as * hard-case backup, but it's just too big and messy unless there are * some changes to the data structures. Maybe later. */ static void -doemit(p, op, opnd) -register struct parse *p; -sop op; -size_t opnd; +doemit(struct parse *p, sop op, size_t opnd) { /* avoid making error situations worse */ if (p->error != 0) @@ -1326,18 +1347,14 @@ size_t opnd; /* - doinsert - insert a sop into the strip - == static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos); + == static void doinsert(struct parse *p, sop op, size_t opnd, sopno pos); */ static void -doinsert(p, op, opnd, pos) -register struct parse *p; -sop op; -size_t opnd; -sopno pos; +doinsert(struct parse *p, sop op, size_t opnd, sopno pos) { - register sopno sn; - register sop s; - register int i; + sopno sn; + sop s; + int i; /* avoid making error situations worse */ if (p->error != 0) @@ -1366,13 +1383,10 @@ sopno pos; /* - dofwd - complete a forward reference - == static void dofwd(register struct parse *p, sopno pos, sop value); + == static void dofwd(struct parse *p, sopno pos, sop value); */ static void -dofwd(p, pos, value) -register struct parse *p; -register sopno pos; -sop value; +dofwd(struct parse *p, sopno pos, sop value) { /* avoid making error situations worse */ if (p->error != 0) @@ -1384,14 +1398,12 @@ sop value; /* - enlarge - enlarge the strip - == static void enlarge(register struct parse *p, sopno size); + == static void enlarge(struct parse *p, sopno size); */ static void -enlarge(p, size) -register struct parse *p; -register sopno size; +enlarge(struct parse *p, sopno size) { - register sop *sp; + sop *sp; if (p->ssize >= size) return; @@ -1407,12 +1419,10 @@ register sopno size; /* - stripsnug - compact the strip - == static void stripsnug(register struct parse *p, register struct re_guts *g); + == static void stripsnug(struct parse *p, struct re_guts *g); */ static void -stripsnug(p, g) -register struct parse *p; -register struct re_guts *g; +stripsnug(struct parse *p, struct re_guts *g) { g->nstates = p->slen; g->strip = (sop *)realloc((char *)p->strip, p->slen * sizeof(sop)); @@ -1424,7 +1434,7 @@ register struct re_guts *g; /* - findmust - fill in must and mlen with longest mandatory literal string - == static void findmust(register struct parse *p, register struct re_guts *g); + == static void findmust(struct parse *p, struct re_guts *g); * * This algorithm could do fancy things like analyzing the operands of | * for common subsequences. Someday. This code is simple and finds most @@ -1433,32 +1443,53 @@ register struct re_guts *g; * Note that must and mlen got initialized during setup. */ static void -findmust(p, g) -struct parse *p; -register struct re_guts *g; +findmust(struct parse *p, struct re_guts *g) { - register sop *scan; - sop *start = NULL; - register sop *newstart = NULL; - register sopno newlen; - register sop s; - register char *cp; - register sopno i; + sop *scan; + sop *start; + sop *newstart; + sopno newlen; + sop s; + char *cp; + int offset; + char buf[MB_LEN_MAX]; + size_t clen; + mbstate_t mbs; /* avoid making error situations worse */ if (p->error != 0) return; + /* + * It's not generally safe to do a ``char'' substring search on + * multibyte character strings, but it's safe for at least + * UTF-8 (see RFC 3629). + */ + if (MB_CUR_MAX > 1 && +#ifdef __CYGWIN__ + strcmp(collate_charset, "UTF-8") != 0) +#else + strcmp(_CurrentRuneLocale->__encoding, "UTF-8") != 0) +#endif + return; + /* find the longest OCHAR sequence in strip */ newlen = 0; + offset = 0; + g->moffset = 0; scan = g->strip + 1; do { s = *scan++; switch (OP(s)) { case OCHAR: /* sequence member */ - if (newlen == 0) /* new sequence */ + if (newlen == 0) { /* new sequence */ + memset(&mbs, 0, sizeof(mbs)); newstart = scan - 1; - newlen++; + } + clen = wcrtomb(buf, OPND(s), &mbs); + if (clen == (size_t)-1) + goto toohard; + newlen += clen; break; case OPLUS_: /* things that don't break one */ case OLPAREN: @@ -1466,6 +1497,7 @@ register struct re_guts *g; break; case OQUEST_: /* things that must be skipped */ case OCH_: + offset = altoffset(scan, offset); scan--; do { scan += OPND(s); @@ -1477,51 +1509,317 @@ register struct re_guts *g; return; } } while (OP(s) != O_QUEST && OP(s) != O_CH); - /* fallthrough */ - default: /* things that break a sequence */ + /* FALLTHROUGH */ + case OBOW: /* things that break a sequence */ + case OEOW: + case OBOL: + case OEOL: + case O_QUEST: + case O_CH: + case OEND: if (newlen > g->mlen) { /* ends one */ start = newstart; g->mlen = newlen; + if (offset > -1) { + g->moffset += offset; + offset = newlen; + } else + g->moffset = offset; + } else { + if (offset > -1) + offset += newlen; } newlen = 0; break; + case OANY: + if (newlen > g->mlen) { /* ends one */ + start = newstart; + g->mlen = newlen; + if (offset > -1) { + g->moffset += offset; + offset = newlen; + } else + g->moffset = offset; + } else { + if (offset > -1) + offset += newlen; + } + if (offset > -1) + offset++; + newlen = 0; + break; + case OANYOF: /* may or may not invalidate offset */ + /* First, everything as OANY */ + if (newlen > g->mlen) { /* ends one */ + start = newstart; + g->mlen = newlen; + if (offset > -1) { + g->moffset += offset; + offset = newlen; + } else + g->moffset = offset; + } else { + if (offset > -1) + offset += newlen; + } + if (offset > -1) + offset++; + newlen = 0; + break; + toohard: + default: + /* Anything here makes it impossible or too hard + * to calculate the offset -- so we give up; + * save the last known good offset, in case the + * must sequence doesn't occur later. + */ + if (newlen > g->mlen) { /* ends one */ + start = newstart; + g->mlen = newlen; + if (offset > -1) + g->moffset += offset; + else + g->moffset = offset; + } + offset = -1; + newlen = 0; + break; } } while (OP(s) != OEND); - if (g->mlen == 0) /* there isn't one */ + if (g->mlen == 0) { /* there isn't one */ + g->moffset = -1; return; + } /* turn it into a character string */ g->must = malloc((size_t)g->mlen + 1); if (g->must == NULL) { /* argh; just forget it */ g->mlen = 0; + g->moffset = -1; return; } cp = g->must; scan = start; - for (i = g->mlen; i > 0; i--) { + memset(&mbs, 0, sizeof(mbs)); + while (cp < g->must + g->mlen) { while (OP(s = *scan++) != OCHAR) continue; - assert(cp < g->must + g->mlen); - *cp++ = (char)OPND(s); + clen = wcrtomb(cp, OPND(s), &mbs); + assert(clen != (size_t)-1); + cp += clen; } assert(cp == g->must + g->mlen); *cp++ = '\0'; /* just on general principles */ } +/* + - altoffset - choose biggest offset among multiple choices + == static int altoffset(sop *scan, int offset); + * + * Compute, recursively if necessary, the largest offset among multiple + * re paths. + */ +static int +altoffset(sop *scan, int offset) +{ + int largest; + int try; + sop s; + + /* If we gave up already on offsets, return */ + if (offset == -1) + return -1; + + largest = 0; + try = 0; + s = *scan++; + while (OP(s) != O_QUEST && OP(s) != O_CH) { + switch (OP(s)) { + case OOR1: + if (try > largest) + largest = try; + try = 0; + break; + case OQUEST_: + case OCH_: + try = altoffset(scan, try); + if (try == -1) + return -1; + scan--; + do { + scan += OPND(s); + s = *scan; + if (OP(s) != O_QUEST && OP(s) != O_CH && + OP(s) != OOR2) + return -1; + } while (OP(s) != O_QUEST && OP(s) != O_CH); + /* We must skip to the next position, or we'll + * leave altoffset() too early. + */ + scan++; + break; + case OANYOF: + case OCHAR: + case OANY: + try++; + case OBOW: + case OEOW: + case OLPAREN: + case ORPAREN: + case OOR2: + break; + default: + try = -1; + break; + } + if (try == -1) + return -1; + s = *scan++; + } + + if (try > largest) + largest = try; + + return largest+offset; +} + +/* + - computejumps - compute char jumps for BM scan + == static void computejumps(struct parse *p, struct re_guts *g); + * + * This algorithm assumes g->must exists and is has size greater than + * zero. It's based on the algorithm found on Computer Algorithms by + * Sara Baase. + * + * A char jump is the number of characters one needs to jump based on + * the value of the character from the text that was mismatched. + */ +static void +computejumps(struct parse *p, struct re_guts *g) +{ + int ch; + int mindex; + + /* Avoid making errors worse */ + if (p->error != 0) + return; + + g->charjump = (int*) malloc((NC + 1) * sizeof(int)); + if (g->charjump == NULL) /* Not a fatal error */ + return; + /* Adjust for signed chars, if necessary */ + g->charjump = &g->charjump[-(CHAR_MIN)]; + + /* If the character does not exist in the pattern, the jump + * is equal to the number of characters in the pattern. + */ + for (ch = CHAR_MIN; ch < (CHAR_MAX + 1); ch++) + g->charjump[ch] = g->mlen; + + /* If the character does exist, compute the jump that would + * take us to the last character in the pattern equal to it + * (notice that we match right to left, so that last character + * is the first one that would be matched). + */ + for (mindex = 0; mindex < g->mlen; mindex++) + g->charjump[(int)g->must[mindex]] = g->mlen - mindex - 1; +} + +/* + - computematchjumps - compute match jumps for BM scan + == static void computematchjumps(struct parse *p, struct re_guts *g); + * + * This algorithm assumes g->must exists and is has size greater than + * zero. It's based on the algorithm found on Computer Algorithms by + * Sara Baase. + * + * A match jump is the number of characters one needs to advance based + * on the already-matched suffix. + * Notice that all values here are minus (g->mlen-1), because of the way + * the search algorithm works. + */ +static void +computematchjumps(struct parse *p, struct re_guts *g) +{ + int mindex; /* General "must" iterator */ + int suffix; /* Keeps track of matching suffix */ + int ssuffix; /* Keeps track of suffixes' suffix */ + int* pmatches; /* pmatches[k] points to the next i + * such that i+1...mlen is a substring + * of k+1...k+mlen-i-1 + */ + + /* Avoid making errors worse */ + if (p->error != 0) + return; + + pmatches = (int*) malloc(g->mlen * sizeof(unsigned int)); + if (pmatches == NULL) { + g->matchjump = NULL; + return; + } + + g->matchjump = (int*) malloc(g->mlen * sizeof(unsigned int)); + if (g->matchjump == NULL) /* Not a fatal error */ + return; + + /* Set maximum possible jump for each character in the pattern */ + for (mindex = 0; mindex < g->mlen; mindex++) + g->matchjump[mindex] = 2*g->mlen - mindex - 1; + + /* Compute pmatches[] */ + for (mindex = g->mlen - 1, suffix = g->mlen; mindex >= 0; + mindex--, suffix--) { + pmatches[mindex] = suffix; + + /* If a mismatch is found, interrupting the substring, + * compute the matchjump for that position. If no + * mismatch is found, then a text substring mismatched + * against the suffix will also mismatch against the + * substring. + */ + while (suffix < g->mlen + && g->must[mindex] != g->must[suffix]) { + g->matchjump[suffix] = MIN(g->matchjump[suffix], + g->mlen - mindex - 1); + suffix = pmatches[suffix]; + } + } + + /* Compute the matchjump up to the last substring found to jump + * to the beginning of the largest must pattern prefix matching + * it's own suffix. + */ + for (mindex = 0; mindex <= suffix; mindex++) + g->matchjump[mindex] = MIN(g->matchjump[mindex], + g->mlen + suffix - mindex); + + ssuffix = pmatches[suffix]; + while (suffix < g->mlen) { + while (suffix <= ssuffix && suffix < g->mlen) { + g->matchjump[suffix] = MIN(g->matchjump[suffix], + g->mlen + ssuffix - suffix); + suffix++; + } + if (suffix < g->mlen) + ssuffix = pmatches[ssuffix]; + } + + free(pmatches); +} + /* - pluscount - count + nesting - == static sopno pluscount(register struct parse *p, register struct re_guts *g); + == static sopno pluscount(struct parse *p, struct re_guts *g); */ static sopno /* nesting depth */ -pluscount(p, g) -struct parse *p; -register struct re_guts *g; +pluscount(struct parse *p, struct re_guts *g) { - register sop *scan; - register sop s; - register sopno plusnest = 0; - register sopno maxnest = 0; + sop *scan; + sop s; + sopno plusnest = 0; + sopno maxnest = 0; if (p->error != 0) return(0); /* there may not be an OEND */ diff --git a/winsup/cygwin/regex/regcomp.ih b/winsup/cygwin/regex/regcomp.ih deleted file mode 100644 index e16d5ab2f..000000000 --- a/winsup/cygwin/regex/regcomp.ih +++ /dev/null @@ -1,48 +0,0 @@ -/* ========= begin header generated by ./mkh ========= */ -#ifdef __cplusplus -extern "C" { -#endif - -/* === regcomp.c === */ -static void p_ere(register struct parse *p, int stop); -static void p_ere_exp(register struct parse *p); -static void p_str(register struct parse *p); -static void p_bre(register struct parse *p, register int end1, register int end2); -static int p_simp_re(register struct parse *p, int starordinary); -static int p_count(register struct parse *p); -static void p_bracket(register struct parse *p); -static void p_b_term(register struct parse *p, register cset *cs); -static void p_b_cclass(register struct parse *p, register cset *cs); -static void p_b_eclass(register struct parse *p, register cset *cs); -static char p_b_symbol(register struct parse *p); -static char p_b_coll_elem(register struct parse *p, int endc); -static char othercase(int ch); -static void bothcases(register struct parse *p, int ch); -static void ordinary(register struct parse *p, register int ch); -static void nonnewline(register struct parse *p); -static void repeat(register struct parse *p, sopno start, int from, int to); -static int seterr(register struct parse *p, int e); -static cset *allocset(register struct parse *p); -static void freeset(register struct parse *p, register cset *cs); -static int freezeset(register struct parse *p, register cset *cs); -static int firstch(register struct parse *p, register cset *cs); -static int nch(register struct parse *p, register cset *cs); -static void mcadd(register struct parse *p, register cset *cs, register const char *cp); -static void mcinvert(register struct parse *p, register cset *cs); -static void mccase(register struct parse *p, register cset *cs); -static int isinsets(register struct re_guts *g, int c); -static int samesets(register struct re_guts *g, int c1, int c2); -static void categorize(struct parse *p, register struct re_guts *g); -static sopno dupl(register struct parse *p, sopno start, sopno finish); -static void doemit(register struct parse *p, sop op, size_t opnd); -static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos); -static void dofwd(register struct parse *p, sopno pos, sop value); -static void enlarge(register struct parse *p, sopno size); -static void stripsnug(register struct parse *p, register struct re_guts *g); -static void findmust(register struct parse *p, register struct re_guts *g); -static sopno pluscount(register struct parse *p, register struct re_guts *g); - -#ifdef __cplusplus -} -#endif -/* ========= end header generated by ./mkh ========= */ diff --git a/winsup/cygwin/regex/regerror.c b/winsup/cygwin/regex/regerror.c index 42b045bae..7e88be5cb 100644 --- a/winsup/cygwin/regex/regerror.c +++ b/winsup/cygwin/regex/regerror.c @@ -1,17 +1,66 @@ -#include "winsup.h" +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)regerror.c 8.4 (Berkeley) 3/20/94 + */ + +#if defined(LIBC_SCCS) && !defined(lint) +static char sccsid[] = "@(#)regerror.c 8.4 (Berkeley) 3/20/94"; +#endif /* LIBC_SCCS and not lint */ +#include +__FBSDID("$FreeBSD: src/lib/libc/regex/regerror.c,v 1.11 2007/06/11 03:05:54 delphij Exp $"); + #include #include #include -#include #include #include -#include "regex.h" +#include #include "utils.h" -#include "regerror.ih" +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +/* === regerror.c === */ +static char *regatoi(const regex_t *preg, char *localbuf); + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ /* - = #define REG_OKAY 0 = #define REG_NOMATCH 1 = #define REG_BADPAT 2 = #define REG_ECOLLATE 3 @@ -28,15 +77,20 @@ = #define REG_EMPTY 14 = #define REG_ASSERT 15 = #define REG_INVARG 16 + = #define REG_ILLSEQ 17 = #define REG_ATOI 255 // convert name to number (!) = #define REG_ITOA 0400 // convert number to name (!) */ static struct rerr { int code; +#ifdef __CYGWIN__ /* Avoid whining compiler */ const char *name; const char *explain; +#else + char *name; + char *explain; +#endif } rerrs[] = { - {REG_OKAY, "REG_OKAY", "no errors detected"}, {REG_NOMATCH, "REG_NOMATCH", "regexec() failed to match"}, {REG_BADPAT, "REG_BADPAT", "invalid regular expression"}, {REG_ECOLLATE, "REG_ECOLLATE", "invalid collating element"}, @@ -53,7 +107,8 @@ static struct rerr { {REG_EMPTY, "REG_EMPTY", "empty (sub)expression"}, {REG_ASSERT, "REG_ASSERT", "\"can't happen\" -- you found a bug"}, {REG_INVARG, "REG_INVARG", "invalid argument to regex routine"}, - {-1, "", "*** unknown regexp error code ***"}, + {REG_ILLSEQ, "REG_ILLSEQ", "illegal byte sequence"}, + {0, "", "*** unknown regexp error code ***"} }; /* @@ -62,27 +117,30 @@ static struct rerr { */ /* ARGSUSED */ size_t -regerror(errcode, preg, errbuf, errbuf_size) -int errcode; -const regex_t *preg; -char *errbuf; -size_t errbuf_size; +regerror(int errcode, + const regex_t * __restrict preg, + char * __restrict errbuf, + size_t errbuf_size) { - register struct rerr *r; - register size_t len; - register int target = errcode &~ REG_ITOA; - register const char *s; + struct rerr *r; + size_t len; + int target = errcode &~ REG_ITOA; +#ifdef __CYGWIN__ /* Avoid whining compiler */ + const char *s; +#else + char *s; +#endif char convbuf[50]; if (errcode == REG_ATOI) s = regatoi(preg, convbuf); else { - for (r = rerrs; r->code >= 0; r++) + for (r = rerrs; r->code != 0; r++) if (r->code == target) break; if (errcode®_ITOA) { - if (r->code >= 0) + if (r->code != 0) (void) strcpy(convbuf, r->name); else sprintf(convbuf, "REG_0x%x", target); @@ -109,18 +167,23 @@ size_t errbuf_size; - regatoi - internal routine to implement REG_ATOI == static char *regatoi(const regex_t *preg, char *localbuf); */ -static const char * -regatoi(preg, localbuf) -const regex_t *preg; -char *localbuf; +static char * +regatoi(const regex_t *preg, char *localbuf) { - register struct rerr *r; + struct rerr *r; - for (r = rerrs; r->code >= 0; r++) + for (r = rerrs; r->code != 0; r++) if (strcmp(r->name, preg->re_endp) == 0) break; - if (r->code < 0) + if (r->code == 0) +#ifdef __CYGWIN__ /* Avoid whining compiler */ + { + static const char null[] = "0"; + return null; + } +#else return("0"); +#endif sprintf(localbuf, "%d", r->code); return(localbuf); diff --git a/winsup/cygwin/regex/regerror.ih b/winsup/cygwin/regex/regerror.ih deleted file mode 100644 index d428544e3..000000000 --- a/winsup/cygwin/regex/regerror.ih +++ /dev/null @@ -1,12 +0,0 @@ -/* ========= begin header generated by ./mkh ========= */ -#ifdef __cplusplus -extern "C" { -#endif - -/* === regerror.c === */ -static const char *regatoi(const regex_t *preg, char *localbuf); - -#ifdef __cplusplus -} -#endif -/* ========= end header generated by ./mkh ========= */ diff --git a/winsup/cygwin/regex/regex.3 b/winsup/cygwin/regex/regex.3 index bc747096d..f848d66c3 100644 --- a/winsup/cygwin/regex/regex.3 +++ b/winsup/cygwin/regex/regex.3 @@ -1,509 +1,727 @@ -.TH REGEX 3 "25 Sept 1997" -.BY "Henry Spencer" -.de ZR -.\" one other place knows this name: the SEE ALSO section -.IR regex (7) \\$1 -.. -.SH NAME -regcomp, regexec, regerror, regfree \- regular-expression library -.SH SYNOPSIS -.ft B -.\".na -#include -.br -#include -.HP 10 -int regcomp(regex_t\ *preg, const\ char\ *pattern, int\ cflags); -.HP -int\ regexec(const\ regex_t\ *preg, const\ char\ *string, -size_t\ nmatch, regmatch_t\ pmatch[], int\ eflags); -.HP -size_t\ regerror(int\ errcode, const\ regex_t\ *preg, -char\ *errbuf, size_t\ errbuf_size); -.HP -void\ regfree(regex_t\ *preg); -.\".ad -.ft -.SH DESCRIPTION -These routines implement POSIX 1003.2 regular expressions (``RE''s); +.\" Copyright (c) 1992, 1993, 1994 Henry Spencer. +.\" Copyright (c) 1992, 1993, 1994 +.\" The Regents of the University of California. All rights reserved. +.\" +.\" This code is derived from software contributed to Berkeley by +.\" Henry Spencer. +.\" +.\" Redistribution and use in source and binary forms, with or without +.\" modification, are permitted provided that the following conditions +.\" are met: +.\" 1. Redistributions of source code must retain the above copyright +.\" notice, this list of conditions and the following disclaimer. +.\" 2. Redistributions in binary form must reproduce the above copyright +.\" notice, this list of conditions and the following disclaimer in the +.\" documentation and/or other materials provided with the distribution. +.\" 4. Neither the name of the University nor the names of its contributors +.\" may be used to endorse or promote products derived from this software +.\" without specific prior written permission. +.\" +.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND +.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE +.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS +.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) +.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT +.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY +.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF +.\" SUCH DAMAGE. +.\" +.\" @(#)regex.3 8.4 (Berkeley) 3/20/94 +.\" $FreeBSD: src/lib/libc/regex/regex.3,v 1.21 2007/01/09 00:28:04 imp Exp $ +.\" +.Dd August 17, 2005 +.Dt REGEX 3 +.Os +.Sh NAME +.Nm regcomp , +.Nm regexec , +.Nm regerror , +.Nm regfree +.Nd regular-expression library +.Sh LIBRARY +.Lb libc +.Sh SYNOPSIS +.In regex.h +.Ft int +.Fo regcomp +.Fa "regex_t * restrict preg" "const char * restrict pattern" "int cflags" +.Fc +.Ft int +.Fo regexec +.Fa "const regex_t * restrict preg" "const char * restrict string" +.Fa "size_t nmatch" "regmatch_t pmatch[restrict]" "int eflags" +.Fc +.Ft size_t +.Fo regerror +.Fa "int errcode" "const regex_t * restrict preg" +.Fa "char * restrict errbuf" "size_t errbuf_size" +.Fc +.Ft void +.Fn regfree "regex_t *preg" +.Sh DESCRIPTION +These routines implement +.St -p1003.2 +regular expressions +.Pq Do RE Dc Ns s ; see -.ZR . -.I Regcomp +.Xr re_format 7 . +The +.Fn regcomp +function compiles an RE written as a string into an internal form, -.I regexec +.Fn regexec matches that internal form against a string and reports results, -.I regerror +.Fn regerror transforms error codes from either into human-readable messages, and -.I regfree +.Fn regfree frees any dynamically-allocated storage used by the internal form of an RE. -.PP +.Pp The header -.I +.In regex.h declares two structure types, -.I regex_t +.Ft regex_t and -.IR regmatch_t , +.Ft regmatch_t , the former for compiled internal forms and the latter for match reporting. It also declares the four functions, a type -.IR regoff_t , -and a number of constants with names starting with ``REG_''. -.PP -.I Regcomp +.Ft regoff_t , +and a number of constants with names starting with +.Dq Dv REG_ . +.Pp +The +.Fn regcomp +function compiles the regular expression contained in the -.I pattern +.Fa pattern string, subject to the flags in -.IR cflags , +.Fa cflags , and places the results in the -.I regex_t +.Ft regex_t structure pointed to by -.IR preg . -.I Cflags +.Fa preg . +The +.Fa cflags +argument is the bitwise OR of zero or more of the following flags: -.IP REG_EXTENDED \w'REG_EXTENDED'u+2n -Compile modern (``extended'') REs, -rather than the obsolete (``basic'') REs that +.Bl -tag -width REG_EXTENDED +.It Dv REG_EXTENDED +Compile modern +.Pq Dq extended +REs, +rather than the obsolete +.Pq Dq basic +REs that are the default. -.IP REG_BASIC +.It Dv REG_BASIC This is a synonym for 0, -provided as a counterpart to REG_EXTENDED to improve readability. -This is an extension, -compatible with but not specified by POSIX 1003.2, -and should be used with -caution in software intended to be portable to other systems. -.IP REG_NOSPEC +provided as a counterpart to +.Dv REG_EXTENDED +to improve readability. +.It Dv REG_NOSPEC Compile with recognition of all special characters turned off. All characters are thus considered ordinary, -so the ``RE'' is a literal string. +so the +.Dq RE +is a literal string. This is an extension, -compatible with but not specified by POSIX 1003.2, +compatible with but not specified by +.St -p1003.2 , and should be used with caution in software intended to be portable to other systems. -REG_EXTENDED and REG_NOSPEC may not be used +.Dv REG_EXTENDED +and +.Dv REG_NOSPEC +may not be used in the same call to -.IR regcomp . -.IP REG_ICASE +.Fn regcomp . +.It Dv REG_ICASE Compile for matching that ignores upper/lower case distinctions. See -.ZR . -.IP REG_NOSUB +.Xr re_format 7 . +.It Dv REG_NOSUB Compile for matching that need only report success or failure, not what was matched. -.IP REG_NEWLINE +.It Dv REG_NEWLINE Compile for newline-sensitive matching. By default, newline is a completely ordinary character with no special meaning in either REs or strings. With this flag, -`[^' bracket expressions and `.' never match newline, -a `^' anchor matches the null string after any newline in the string +.Ql [^ +bracket expressions and +.Ql .\& +never match newline, +a +.Ql ^\& +anchor matches the null string after any newline in the string in addition to its normal function, -and the `$' anchor matches the null string before any newline in the +and the +.Ql $\& +anchor matches the null string before any newline in the string in addition to its normal function. -.IP REG_PEND +.It Dv REG_PEND The regular expression ends, not at the first NUL, but just before the character pointed to by the -.I re_endp +.Va re_endp member of the structure pointed to by -.IR preg . +.Fa preg . The -.I re_endp +.Va re_endp member is of type -.IR const\ char\ * . +.Ft "const char *" . This flag permits inclusion of NULs in the RE; they are considered ordinary characters. This is an extension, -compatible with but not specified by POSIX 1003.2, +compatible with but not specified by +.St -p1003.2 , and should be used with caution in software intended to be portable to other systems. -.PP +.El +.Pp When successful, -.I regcomp +.Fn regcomp returns 0 and fills in the structure pointed to by -.IR preg . +.Fa preg . One member of that structure (other than -.IR re_endp ) +.Va re_endp ) is publicized: -.IR re_nsub , +.Va re_nsub , of type -.IR size_t , +.Ft size_t , contains the number of parenthesized subexpressions within the RE (except that the value of this member is undefined if the -REG_NOSUB flag was used). +.Dv REG_NOSUB +flag was used). If -.I regcomp +.Fn regcomp fails, it returns a non-zero error code; -see DIAGNOSTICS. -.PP -.I Regexec +see +.Sx DIAGNOSTICS . +.Pp +The +.Fn regexec +function matches the compiled RE pointed to by -.I preg +.Fa preg against the -.IR string , +.Fa string , subject to the flags in -.IR eflags , +.Fa eflags , and reports results using -.IR nmatch , -.IR pmatch , +.Fa nmatch , +.Fa pmatch , and the returned value. The RE must have been compiled by a previous invocation of -.IR regcomp . +.Fn regcomp . The compiled form is not altered during execution of -.IR regexec , +.Fn regexec , so a single compiled RE can be used simultaneously by multiple threads. -.PP +.Pp By default, the NUL-terminated string pointed to by -.I string -is considered to be the text of an entire line, -with the NUL indicating the end of the line. -(That is, -any other end-of-line marker is considered to have been removed -and replaced by the NUL.) +.Fa string +is considered to be the text of an entire line, minus any terminating +newline. The -.I eflags +.Fa eflags argument is the bitwise OR of zero or more of the following flags: -.IP REG_NOTBOL \w'REG_STARTEND'u+2n +.Bl -tag -width REG_STARTEND +.It Dv REG_NOTBOL The first character of the string -is not the beginning of a line, so the `^' anchor should not match before it. -This does not affect the behavior of newlines under REG_NEWLINE. -.IP REG_NOTEOL +is not the beginning of a line, so the +.Ql ^\& +anchor should not match before it. +This does not affect the behavior of newlines under +.Dv REG_NEWLINE . +.It Dv REG_NOTEOL The NUL terminating the string -does not end a line, so the `$' anchor should not match before it. -This does not affect the behavior of newlines under REG_NEWLINE. -.IP REG_STARTEND +does not end a line, so the +.Ql $\& +anchor should not match before it. +This does not affect the behavior of newlines under +.Dv REG_NEWLINE . +.It Dv REG_STARTEND The string is considered to start at -\fIstring\fR\ + \fIpmatch\fR[0].\fIrm_so\fR +.Fa string ++ +.Fa pmatch Ns [0]. Ns Va rm_so and to have a terminating NUL located at -\fIstring\fR\ + \fIpmatch\fR[0].\fIrm_eo\fR +.Fa string ++ +.Fa pmatch Ns [0]. Ns Va rm_eo (there need not actually be a NUL at that location), regardless of the value of -.IR nmatch . +.Fa nmatch . See below for the definition of -.IR pmatch +.Fa pmatch and -.IR nmatch . +.Fa nmatch . This is an extension, -compatible with but not specified by POSIX 1003.2, +compatible with but not specified by +.St -p1003.2 , and should be used with caution in software intended to be portable to other systems. -Note that a non-zero \fIrm_so\fR does not imply REG_NOTBOL; -REG_STARTEND affects only the location of the string, +Note that a non-zero +.Va rm_so +does not imply +.Dv REG_NOTBOL ; +.Dv REG_STARTEND +affects only the location of the string, not how it is matched. -.PP +.El +.Pp See -.ZR +.Xr re_format 7 for a discussion of what is matched in situations where an RE or a portion thereof could match any of several substrings of -.IR string . -.PP +.Fa string . +.Pp Normally, -.I regexec -returns 0 for success and the non-zero code REG_NOMATCH for failure. +.Fn regexec +returns 0 for success and the non-zero code +.Dv REG_NOMATCH +for failure. Other non-zero error codes may be returned in exceptional situations; -see DIAGNOSTICS. -.PP -If REG_NOSUB was specified in the compilation of the RE, +see +.Sx DIAGNOSTICS . +.Pp +If +.Dv REG_NOSUB +was specified in the compilation of the RE, or if -.I nmatch +.Fa nmatch is 0, -.I regexec +.Fn regexec ignores the -.I pmatch -argument (but see below for the case where REG_STARTEND is specified). +.Fa pmatch +argument (but see below for the case where +.Dv REG_STARTEND +is specified). Otherwise, -.I pmatch +.Fa pmatch points to an array of -.I nmatch +.Fa nmatch structures of type -.IR regmatch_t . +.Ft regmatch_t . Such a structure has at least the members -.I rm_so +.Va rm_so and -.IR rm_eo , +.Va rm_eo , both of type -.I regoff_t +.Ft regoff_t (a signed arithmetic type at least as large as an -.I off_t +.Ft off_t and a -.IR ssize_t ), +.Ft ssize_t ) , containing respectively the offset of the first character of a substring and the offset of the first character after the end of the substring. Offsets are measured from the beginning of the -.I string +.Fa string argument given to -.IR regexec . +.Fn regexec . An empty substring is denoted by equal offsets, both indicating the character following the empty substring. -.PP +.Pp The 0th member of the -.I pmatch +.Fa pmatch array is filled in to indicate what substring of -.I string +.Fa string was matched by the entire RE. Remaining members report what substring was matched by parenthesized subexpressions within the RE; member -.I i +.Va i reports subexpression -.IR i , +.Va i , with subexpressions counted (starting at 1) by the order of their opening parentheses in the RE, left to right. -Unused entries in the array\(emcorresponding either to subexpressions that +Unused entries in the array (corresponding either to subexpressions that did not participate in the match at all, or to subexpressions that do not -exist in the RE (that is, \fIi\fR\ > \fIpreg\fR\->\fIre_nsub\fR)\(emhave both -.I rm_so +exist in the RE (that is, +.Va i +> +.Fa preg Ns -> Ns Va re_nsub ) ) +have both +.Va rm_so and -.I rm_eo -set to \-1. +.Va rm_eo +set to -1. If a subexpression participated in the match several times, the reported substring is the last one it matched. -(Note, as an example in particular, that when the RE `(b*)+' matches `bbb', -the parenthesized subexpression matches the three `b's and then -an infinite number of empty strings following the last `b', +(Note, as an example in particular, that when the RE +.Ql "(b*)+" +matches +.Ql bbb , +the parenthesized subexpression matches each of the three +.So Li b Sc Ns s +and then +an infinite number of empty strings following the last +.Ql b , so the reported substring is one of the empties.) -.PP -If REG_STARTEND is specified, -.I pmatch +.Pp +If +.Dv REG_STARTEND +is specified, +.Fa pmatch must point to at least one -.I regmatch_t +.Ft regmatch_t (even if -.I nmatch -is 0 or REG_NOSUB was specified), -to hold the input offsets for REG_STARTEND. +.Fa nmatch +is 0 or +.Dv REG_NOSUB +was specified), +to hold the input offsets for +.Dv REG_STARTEND . Use for output is still entirely controlled by -.IR nmatch ; +.Fa nmatch ; if -.I nmatch -is 0 or REG_NOSUB was specified, +.Fa nmatch +is 0 or +.Dv REG_NOSUB +was specified, the value of -.IR pmatch [0] +.Fa pmatch Ns [0] will not be changed by a successful -.IR regexec . -.PP -.I Regerror +.Fn regexec . +.Pp +The +.Fn regerror +function maps a non-zero -.I errcode +.Fa errcode from either -.I regcomp +.Fn regcomp or -.I regexec +.Fn regexec to a human-readable, printable message. If -.I preg -is non-NULL, +.Fa preg +is +.No non\- Ns Dv NULL , the error code should have arisen from use of the -.I regex_t +.Ft regex_t pointed to by -.IR preg , +.Fa preg , and if the error code came from -.IR regcomp , +.Fn regcomp , it should have been the result from the most recent -.I regcomp +.Fn regcomp using that -.IR regex_t . -.RI ( Regerror +.Ft regex_t . +The +.Fn ( regerror may be able to supply a more detailed message using information from the -.IR regex_t .) -.I Regerror +.Ft regex_t . ) +The +.Fn regerror +function places the NUL-terminated message into the buffer pointed to by -.IR errbuf , +.Fa errbuf , limiting the length (including the NUL) to at most -.I errbuf_size +.Fa errbuf_size bytes. -If the whole message won't fit, +If the whole message will not fit, as much of it as will fit before the terminating NUL is supplied. In any case, the returned value is the size of buffer needed to hold the whole message (including terminating NUL). If -.I errbuf_size +.Fa errbuf_size is 0, -.I errbuf +.Fa errbuf is ignored but the return value is still correct. -.PP +.Pp If the -.I errcode +.Fa errcode given to -.I regerror -is first ORed with REG_ITOA, -the ``message'' that results is the printable name of the error code, -e.g. ``REG_NOMATCH'', +.Fn regerror +is first ORed with +.Dv REG_ITOA , +the +.Dq message +that results is the printable name of the error code, +e.g.\& +.Dq Dv REG_NOMATCH , rather than an explanation thereof. If -.I errcode -is REG_ATOI, +.Fa errcode +is +.Dv REG_ATOI , then -.I preg -shall be non-NULL and the -.I re_endp +.Fa preg +shall be +.No non\- Ns Dv NULL +and the +.Va re_endp member of the structure it points to must point to the printable name of an error code; in this case, the result in -.I errbuf +.Fa errbuf is the decimal digits of the numeric value of the error code (0 if the name is not recognized). -REG_ITOA and REG_ATOI are intended primarily as debugging facilities; +.Dv REG_ITOA +and +.Dv REG_ATOI +are intended primarily as debugging facilities; they are extensions, -compatible with but not specified by POSIX 1003.2, +compatible with but not specified by +.St -p1003.2 , and should be used with caution in software intended to be portable to other systems. Be warned also that they are considered experimental and changes are possible. -.PP -.I Regfree +.Pp +The +.Fn regfree +function frees any dynamically-allocated storage associated with the compiled RE pointed to by -.IR preg . +.Fa preg . The remaining -.I regex_t +.Ft regex_t is no longer a valid compiled RE and the effect of supplying it to -.I regexec +.Fn regexec or -.I regerror +.Fn regerror is undefined. -.PP +.Pp None of these functions references global variables except for tables of constants; all are safe for use from multiple threads if the arguments are safe. -.SH IMPLEMENTATION CHOICES -There are a number of decisions that 1003.2 leaves up to the implementor, -either by explicitly saying ``undefined'' or by virtue of them being +.Sh IMPLEMENTATION CHOICES +There are a number of decisions that +.St -p1003.2 +leaves up to the implementor, +either by explicitly saying +.Dq undefined +or by virtue of them being forbidden by the RE grammar. This implementation treats them as follows. -.PP +.Pp See -.ZR +.Xr re_format 7 for a discussion of the definition of case-independent matching. -.PP +.Pp There is no particular limit on the length of REs, except insofar as memory is limited. Memory usage is approximately linear in RE size, and largely insensitive to RE complexity, except for bounded repetitions. -See BUGS for one short RE using them +See +.Sx BUGS +for one short RE using them that will run almost any system out of memory. -.PP +.Pp A backslashed character other than one specifically given a magic meaning -by 1003.2 (such magic meanings occur only in obsolete [``basic''] REs) +by +.St -p1003.2 +(such magic meanings occur only in obsolete +.Bq Dq basic +REs) is taken as an ordinary character. -.PP -Any unmatched [ is a REG_EBRACK error. -.PP +.Pp +Any unmatched +.Ql [\& +is a +.Dv REG_EBRACK +error. +.Pp Equivalence classes cannot begin or end bracket-expression ranges. The endpoint of one range cannot begin another. -.PP -RE_DUP_MAX, the limit on repetition counts in bounded repetitions, is 255. -.PP -A repetition operator (?, *, +, or bounds) cannot follow another +.Pp +.Dv RE_DUP_MAX , +the limit on repetition counts in bounded repetitions, is 255. +.Pp +A repetition operator +.Ql ( ?\& , +.Ql *\& , +.Ql +\& , +or bounds) +cannot follow another repetition operator. A repetition operator cannot begin an expression or subexpression -or follow `^' or `|'. -.PP -`|' cannot appear first or last in a (sub)expression or after another `|', -i.e. an operand of `|' cannot be an empty subexpression. -An empty parenthesized subexpression, `()', is legal and matches an +or follow +.Ql ^\& +or +.Ql |\& . +.Pp +.Ql |\& +cannot appear first or last in a (sub)expression or after another +.Ql |\& , +i.e., an operand of +.Ql |\& +cannot be an empty subexpression. +An empty parenthesized subexpression, +.Ql "()" , +is legal and matches an empty (sub)string. An empty string is not a legal RE. -.PP -A `{' followed by a digit is considered the beginning of bounds for a +.Pp +A +.Ql {\& +followed by a digit is considered the beginning of bounds for a bounded repetition, which must then follow the syntax for bounds. -A `{' \fInot\fR followed by a digit is considered an ordinary character. -.PP -`^' and `$' beginning and ending subexpressions in obsolete (``basic'') +A +.Ql {\& +.Em not +followed by a digit is considered an ordinary character. +.Pp +.Ql ^\& +and +.Ql $\& +beginning and ending subexpressions in obsolete +.Pq Dq basic REs are anchors, not ordinary characters. -.SH SEE ALSO -grep(1), regex(7) -.PP -POSIX 1003.2, sections 2.8 (Regular Expression Notation) +.Sh DIAGNOSTICS +Non-zero error codes from +.Fn regcomp +and +.Fn regexec +include the following: +.Pp +.Bl -tag -width REG_ECOLLATE -compact +.It Dv REG_NOMATCH +The +.Fn regexec +function +failed to match +.It Dv REG_BADPAT +invalid regular expression +.It Dv REG_ECOLLATE +invalid collating element +.It Dv REG_ECTYPE +invalid character class +.It Dv REG_EESCAPE +.Ql \e +applied to unescapable character +.It Dv REG_ESUBREG +invalid backreference number +.It Dv REG_EBRACK +brackets +.Ql "[ ]" +not balanced +.It Dv REG_EPAREN +parentheses +.Ql "( )" +not balanced +.It Dv REG_EBRACE +braces +.Ql "{ }" +not balanced +.It Dv REG_BADBR +invalid repetition count(s) in +.Ql "{ }" +.It Dv REG_ERANGE +invalid character range in +.Ql "[ ]" +.It Dv REG_ESPACE +ran out of memory +.It Dv REG_BADRPT +.Ql ?\& , +.Ql *\& , +or +.Ql +\& +operand invalid +.It Dv REG_EMPTY +empty (sub)expression +.It Dv REG_ASSERT +cannot happen - you found a bug +.It Dv REG_INVARG +invalid argument, e.g.\& negative-length string +.It Dv REG_ILLSEQ +illegal byte sequence (bad multibyte character) +.El +.Sh SEE ALSO +.Xr grep 1 , +.Xr re_format 7 +.Pp +.St -p1003.2 , +sections 2.8 (Regular Expression Notation) and B.5 (C Binding for Regular Expression Matching). -.SH DIAGNOSTICS -Non-zero error codes from -.I regcomp -and -.I regexec -include the following: -.PP -.nf -.ta \w'REG_ECOLLATE'u+3n -REG_NOMATCH regexec() failed to match -REG_BADPAT invalid regular expression -REG_ECOLLATE invalid collating element -REG_ECTYPE invalid character class -REG_EESCAPE \e applied to unescapable character -REG_ESUBREG invalid backreference number -REG_EBRACK brackets [ ] not balanced -REG_EPAREN parentheses ( ) not balanced -REG_EBRACE braces { } not balanced -REG_BADBR invalid repetition count(s) in { } -REG_ERANGE invalid character range in [ ] -REG_ESPACE ran out of memory -REG_BADRPT ?, *, or + operand invalid -REG_EMPTY empty (sub)expression -REG_ASSERT ``can't happen''\(emyou found a bug -REG_INVARG invalid argument, e.g. negative-length string -.fi -.SH HISTORY -Written by Henry Spencer, -henry@zoo.toronto.edu. -.SH BUGS +.Sh HISTORY +Originally written by +.An Henry Spencer . +Altered for inclusion in the +.Bx 4.4 +distribution. +.Sh BUGS This is an alpha release with known defects. Please report problems. -.PP -There is one known functionality bug. -The implementation of internationalization is incomplete: -the locale is always assumed to be the default one of 1003.2, -and only the collating elements etc. of that locale are available. -.PP +.Pp The back-reference code is subtle and doubts linger about its correctness in complex cases. -.PP -.I Regexec +.Pp +The +.Fn regexec +function performance is poor. This will improve with later releases. -.I Nmatch +The +.Fa nmatch +argument exceeding 0 is expensive; -.I nmatch +.Fa nmatch exceeding 1 is worse. -.I Regexec -is largely insensitive to RE complexity \fIexcept\fR that back +The +.Fn regexec +function +is largely insensitive to RE complexity +.Em except +that back references are massively expensive. RE length does matter; in particular, there is a strong speed bonus for keeping RE length under about 30 characters, with most special characters counting roughly double. -.PP -.I Regcomp +.Pp +The +.Fn regcomp +function implements bounded repetitions by macro expansion, which is costly in time and space if counts are large or bounded repetitions are nested. An RE like, say, -`((((a{1,100}){1,100}){1,100}){1,100}){1,100}' +.Ql "((((a{1,100}){1,100}){1,100}){1,100}){1,100}" will (eventually) run almost any existing machine out of swap space. -.PP +.Pp There are suspected problems with response to obscure error conditions. Notably, certain kinds of internal overflow, produced only by truly enormous REs or by multiply nested bounded repetitions, are probably not handled well. -.PP -Due to a mistake in 1003.2, things like `a)b' are legal REs because `)' is -a special character only in the presence of a previous unmatched `('. -This can't be fixed until the spec is fixed. -.PP +.Pp +Due to a mistake in +.St -p1003.2 , +things like +.Ql "a)b" +are legal REs because +.Ql )\& +is +a special character only in the presence of a previous unmatched +.Ql (\& . +This cannot be fixed until the spec is fixed. +.Pp The standard's definition of back references is vague. For example, does -`a\e(\e(b\e)*\e2\e)*d' match `abbbd'? +.Ql "a\e(\e(b\e)*\e2\e)*d" +match +.Ql "abbbd" ? Until the standard is clarified, behavior in such cases should not be relied on. -.PP +.Pp The implementation of word-boundary matching is a bit of a kludge, and bugs may lurk in combinations of word-boundary matching and anchoring. +.Pp +Word-boundary matching does not work properly in multibyte locales. diff --git a/winsup/cygwin/regex/regex.7 b/winsup/cygwin/regex/regex.7 index dc53e8c47..79fecc197 100644 --- a/winsup/cygwin/regex/regex.7 +++ b/winsup/cygwin/regex/regex.7 @@ -1,151 +1,317 @@ -.TH REGEX 7 "25 Oct 1995" -.BY "Henry Spencer" -.SH NAME -regex \- POSIX 1003.2 regular expressions -.SH DESCRIPTION -Regular expressions (``RE''s), -as defined in POSIX 1003.2, come in two forms: +.\" Copyright (c) 1992, 1993, 1994 Henry Spencer. +.\" Copyright (c) 1992, 1993, 1994 +.\" The Regents of the University of California. All rights reserved. +.\" +.\" This code is derived from software contributed to Berkeley by +.\" Henry Spencer. +.\" +.\" Redistribution and use in source and binary forms, with or without +.\" modification, are permitted provided that the following conditions +.\" are met: +.\" 1. Redistributions of source code must retain the above copyright +.\" notice, this list of conditions and the following disclaimer. +.\" 2. Redistributions in binary form must reproduce the above copyright +.\" notice, this list of conditions and the following disclaimer in the +.\" documentation and/or other materials provided with the distribution. +.\" 4. Neither the name of the University nor the names of its contributors +.\" may be used to endorse or promote products derived from this software +.\" without specific prior written permission. +.\" +.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND +.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE +.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS +.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) +.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT +.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY +.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF +.\" SUCH DAMAGE. +.\" +.\" @(#)re_format.7 8.3 (Berkeley) 3/20/94 +.\" $FreeBSD: src/lib/libc/regex/re_format.7,v 1.12 2008/09/05 17:41:20 keramida Exp $ +.\" +.Dd March 20, 1994 +.Dt RE_FORMAT 7 +.Os +.Sh NAME +.Nm re_format +.Nd POSIX 1003.2 regular expressions +.Sh DESCRIPTION +Regular expressions +.Pq Dq RE Ns s , +as defined in +.St -p1003.2 , +come in two forms: modern REs (roughly those of -.IR egrep ; -1003.2 calls these ``extended'' REs) +.Xr egrep 1 ; +1003.2 calls these +.Dq extended +REs) and obsolete REs (roughly those of -.IR ed ; -1003.2 ``basic'' REs). +.Xr ed 1 ; +1003.2 +.Dq basic +REs). Obsolete REs mostly exist for backward compatibility in some old programs; they will be discussed at the end. -1003.2 leaves some aspects of RE syntax and semantics open; -`\(dg' marks decisions on these aspects that -may not be fully portable to other 1003.2 implementations. -.PP -A (modern) RE is one\(dg or more non-empty\(dg \fIbranches\fR, -separated by `|'. +.St -p1003.2 +leaves some aspects of RE syntax and semantics open; +`\(dd' marks decisions on these aspects that +may not be fully portable to other +.St -p1003.2 +implementations. +.Pp +A (modern) RE is one\(dd or more non-empty\(dd +.Em branches , +separated by +.Ql \&| . It matches anything that matches one of the branches. -.PP -A branch is one\(dg or more \fIpieces\fR, concatenated. +.Pp +A branch is one\(dd or more +.Em pieces , +concatenated. It matches a match for the first, followed by a match for the second, etc. -.PP -A piece is an \fIatom\fR possibly followed -by a single\(dg `*', `+', `?', or \fIbound\fR. -An atom followed by `*' matches a sequence of 0 or more matches of the atom. -An atom followed by `+' matches a sequence of 1 or more matches of the atom. -An atom followed by `?' matches a sequence of 0 or 1 matches of the atom. -.PP -A \fIbound\fR is `{' followed by an unsigned decimal integer, -possibly followed by `,' +.Pp +A piece is an +.Em atom +possibly followed +by a single\(dd +.Ql \&* , +.Ql \&+ , +.Ql \&? , +or +.Em bound . +An atom followed by +.Ql \&* +matches a sequence of 0 or more matches of the atom. +An atom followed by +.Ql \&+ +matches a sequence of 1 or more matches of the atom. +An atom followed by +.Ql ?\& +matches a sequence of 0 or 1 matches of the atom. +.Pp +A +.Em bound +is +.Ql \&{ +followed by an unsigned decimal integer, +possibly followed by +.Ql \&, possibly followed by another unsigned decimal integer, -always followed by `}'. -The integers must lie between 0 and RE_DUP_MAX (255\(dg) inclusive, +always followed by +.Ql \&} . +The integers must lie between 0 and +.Dv RE_DUP_MAX +(255\(dd) inclusive, and if there are two of them, the first may not exceed the second. -An atom followed by a bound containing one integer \fIi\fR +An atom followed by a bound containing one integer +.Em i and no comma matches -a sequence of exactly \fIi\fR matches of the atom. +a sequence of exactly +.Em i +matches of the atom. An atom followed by a bound -containing one integer \fIi\fR and a comma matches -a sequence of \fIi\fR or more matches of the atom. +containing one integer +.Em i +and a comma matches +a sequence of +.Em i +or more matches of the atom. An atom followed by a bound -containing two integers \fIi\fR and \fIj\fR matches -a sequence of \fIi\fR through \fIj\fR (inclusive) matches of the atom. -.PP -An atom is a regular expression enclosed in `()' (matching a match for the +containing two integers +.Em i +and +.Em j +matches +a sequence of +.Em i +through +.Em j +(inclusive) matches of the atom. +.Pp +An atom is a regular expression enclosed in +.Ql () +(matching a match for the regular expression), -an empty set of `()' (matching the null string)\(dg, -a \fIbracket expression\fR (see below), `.' -(matching any single character), `^' (matching the null string at the -beginning of a line), `$' (matching the null string at the -end of a line), a `\e' followed by one of the characters -`^.[$()|*+?{\e' +an empty set of +.Ql () +(matching the null string)\(dd, +a +.Em bracket expression +(see below), +.Ql .\& +(matching any single character), +.Ql \&^ +(matching the null string at the beginning of a line), +.Ql \&$ +(matching the null string at the end of a line), a +.Ql \e +followed by one of the characters +.Ql ^.[$()|*+?{\e (matching that character taken as an ordinary character), -a `\e' followed by any other character\(dg +a +.Ql \e +followed by any other character\(dd (matching that character taken as an ordinary character, -as if the `\e' had not been present\(dg), +as if the +.Ql \e +had not been present\(dd), or a single character with no other significance (matching that character). -A `{' followed by a character other than a digit is an ordinary -character, not the beginning of a bound\(dg. -It is illegal to end an RE with `\e'. -.PP -A \fIbracket expression\fR is a list of characters enclosed in `[]'. +A +.Ql \&{ +followed by a character other than a digit is an ordinary +character, not the beginning of a bound\(dd. +It is illegal to end an RE with +.Ql \e . +.Pp +A +.Em bracket expression +is a list of characters enclosed in +.Ql [] . It normally matches any single character from the list (but see below). -If the list begins with `^', +If the list begins with +.Ql \&^ , it matches any single character -(but see below) \fInot\fR from the rest of the list. -If two characters in the list are separated by `\-', this is shorthand -for the full \fIrange\fR of characters between those two (inclusive) in the +(but see below) +.Em not +from the rest of the list. +If two characters in the list are separated by +.Ql \&- , +this is shorthand +for the full +.Em range +of characters between those two (inclusive) in the collating sequence, -e.g. `[0\-9]' in ASCII matches any decimal digit. -It is illegal\(dg for two ranges to share an -endpoint, e.g. `a\-c\-e'. +.No e.g. Ql [0-9] +in ASCII matches any decimal digit. +It is illegal\(dd for two ranges to share an +endpoint, +.No e.g. Ql a-c-e . Ranges are very collating-sequence-dependent, and portable programs should avoid relying on them. -.PP -To include a literal `]' in the list, make it the first character -(following a possible `^'). -To include a literal `\-', make it the first or last character, +.Pp +To include a literal +.Ql \&] +in the list, make it the first character +(following a possible +.Ql \&^ ) . +To include a literal +.Ql \&- , +make it the first or last character, or the second endpoint of a range. -To use a literal `\-' as the first endpoint of a range, -enclose it in `[.' and `.]' to make it a collating element (see below). -With the exception of these and some combinations using `[' (see next -paragraphs), all other special characters, including `\e', lose their -special significance within a bracket expression. -.PP +To use a literal +.Ql \&- +as the first endpoint of a range, +enclose it in +.Ql [.\& +and +.Ql .]\& +to make it a collating element (see below). +With the exception of these and some combinations using +.Ql \&[ +(see next paragraphs), all other special characters, including +.Ql \e , +lose their special significance within a bracket expression. +.Pp Within a bracket expression, a collating element (a character, a multi-character sequence that collates as if it were a single character, or a collating-sequence name for either) -enclosed in `[.' and `.]' stands for the +enclosed in +.Ql [.\& +and +.Ql .]\& +stands for the sequence of characters of that collating element. The sequence is a single element of the bracket expression's list. A bracket expression containing a multi-character collating element can thus match more than one character, -e.g. if the collating sequence includes a `ch' collating element, -then the RE `[[.ch.]]*c' matches the first five characters -of `chchcc'. -.PP -Within a bracket expression, a collating element enclosed in `[=' and -`=]' is an equivalence class, standing for the sequences of characters +e.g.\& if the collating sequence includes a +.Ql ch +collating element, +then the RE +.Ql [[.ch.]]*c +matches the first five characters +of +.Ql chchcc . +.Pp +Within a bracket expression, a collating element enclosed in +.Ql [= +and +.Ql =] +is an equivalence class, standing for the sequences of characters of all collating elements equivalent to that one, including itself. (If there are no other equivalent collating elements, -the treatment is as if the enclosing delimiters were `[.' and `.]'.) -For example, if o and \o'o^' are the members of an equivalence class, -then `[[=o=]]', `[[=\o'o^'=]]', and `[o\o'o^']' are all synonymous. -An equivalence class may not\(dg be an endpoint +the treatment is as if the enclosing delimiters were +.Ql [.\& +and +.Ql .] . ) +For example, if +.Ql x +and +.Ql y +are the members of an equivalence class, +then +.Ql [[=x=]] , +.Ql [[=y=]] , +and +.Ql [xy] +are all synonymous. +An equivalence class may not\(dd be an endpoint of a range. -.PP -Within a bracket expression, the name of a \fIcharacter class\fR enclosed -in `[:' and `:]' stands for the list of all characters belonging to that +.Pp +Within a bracket expression, the name of a +.Em character class +enclosed in +.Ql [: +and +.Ql :] +stands for the list of all characters belonging to that class. Standard character class names are: -.PP -.RS -.nf -.ta 3c 6c 9c -alnum digit punct -alpha graph space -blank lower upper -cntrl print xdigit -.fi -.RE -.PP +.Pp +.Bl -column "alnum" "digit" "xdigit" -offset indent +.It Em "alnum digit punct" +.It Em "alpha graph space" +.It Em "blank lower upper" +.It Em "cntrl print xdigit" +.El +.Pp These stand for the character classes defined in -.IR ctype (3). +.Xr ctype 3 . A locale may provide others. A character class may not be used as an endpoint of a range. -.PP -There are two special cases\(dg of bracket expressions: -the bracket expressions `[[:<:]]' and `[[:>:]]' match the null string at -the beginning and end of a word respectively. -A word is defined as a sequence of -word characters +.Pp +A bracketed expression like +.Ql [[:class:]] +can be used to match a single character that belongs to a character +class. +The reverse, matching any character that does not belong to a specific +class, the negation operator of bracket expressions may be used: +.Ql [^[:class:]] . +.Pp +There are two special cases\(dd of bracket expressions: +the bracket expressions +.Ql [[:<:]] +and +.Ql [[:>:]] +match the null string at the beginning and end of a word respectively. +A word is defined as a sequence of word characters which is neither preceded nor followed by word characters. A word character is an -.I alnum +.Em alnum character (as defined by -.IR ctype (3)) +.Xr ctype 3 ) or an underscore. This is an extension, -compatible with but not specified by POSIX 1003.2, +compatible with but not specified by +.St -p1003.2 , and should be used with caution in software intended to be portable to other systems. -.PP +.Pp In the event that an RE could match more than one substring of a given string, the RE matches the one starting earliest in the string. @@ -157,79 +323,158 @@ with subexpressions starting earlier in the RE taking priority over ones starting later. Note that higher-level subexpressions thus take priority over their lower-level component subexpressions. -.PP +.Pp Match lengths are measured in characters, not collating elements. A null string is considered longer than no match at all. For example, -`bb*' matches the three middle characters of `abbbc', -`(wee|week)(knights|nights)' matches all ten characters of `weeknights', -when `(.*).*' is matched against `abc' the parenthesized subexpression +.Ql bb* +matches the three middle characters of +.Ql abbbc , +.Ql (wee|week)(knights|nights) +matches all ten characters of +.Ql weeknights , +when +.Ql (.*).*\& +is matched against +.Ql abc +the parenthesized subexpression matches all three characters, and -when `(a*)*' is matched against `bc' both the whole RE and the parenthesized +when +.Ql (a*)* +is matched against +.Ql bc +both the whole RE and the parenthesized subexpression match the null string. -.PP +.Pp If case-independent matching is specified, the effect is much as if all case distinctions had vanished from the alphabet. When an alphabetic that exists in multiple cases appears as an ordinary character outside a bracket expression, it is effectively transformed into a bracket expression containing both cases, -e.g. `x' becomes `[xX]'. +.No e.g. Ql x +becomes +.Ql [xX] . When it appears inside a bracket expression, all case counterparts -of it are added to the bracket expression, so that (e.g.) `[x]' -becomes `[xX]' and `[^x]' becomes `[^xX]'. -.PP -No particular limit is imposed on the length of REs\(dg. +of it are added to the bracket expression, so that (e.g.) +.Ql [x] +becomes +.Ql [xX] +and +.Ql [^x] +becomes +.Ql [^xX] . +.Pp +No particular limit is imposed on the length of REs\(dd. Programs intended to be portable should not employ REs longer than 256 bytes, as an implementation can refuse to accept such REs and remain POSIX-compliant. -.PP -Obsolete (``basic'') regular expressions differ in several respects. -`|', `+', and `?' are ordinary characters and there is no equivalent -for their functionality. -The delimiters for bounds are `\e{' and `\e}', -with `{' and `}' by themselves ordinary characters. -The parentheses for nested subexpressions are `\e(' and `\e)', -with `(' and `)' by themselves ordinary characters. -`^' is an ordinary character except at the beginning of the -RE or\(dg the beginning of a parenthesized subexpression, -`$' is an ordinary character except at the end of the -RE or\(dg the end of a parenthesized subexpression, -and `*' is an ordinary character if it appears at the beginning of the +.Pp +Obsolete +.Pq Dq basic +regular expressions differ in several respects. +.Ql \&| +is an ordinary character and there is no equivalent +for its functionality. +.Ql \&+ +and +.Ql ?\& +are ordinary characters, and their functionality +can be expressed using bounds +.No ( Ql {1,} +or +.Ql {0,1} +respectively). +Also note that +.Ql x+ +in modern REs is equivalent to +.Ql xx* . +The delimiters for bounds are +.Ql \e{ +and +.Ql \e} , +with +.Ql \&{ +and +.Ql \&} +by themselves ordinary characters. +The parentheses for nested subexpressions are +.Ql \e( +and +.Ql \e) , +with +.Ql \&( +and +.Ql \&) +by themselves ordinary characters. +.Ql \&^ +is an ordinary character except at the beginning of the +RE or\(dd the beginning of a parenthesized subexpression, +.Ql \&$ +is an ordinary character except at the end of the +RE or\(dd the end of a parenthesized subexpression, +and +.Ql \&* +is an ordinary character if it appears at the beginning of the RE or the beginning of a parenthesized subexpression -(after a possible leading `^'). -Finally, there is one new type of atom, a \fIback reference\fR: -`\e' followed by a non-zero decimal digit \fId\fR +(after a possible leading +.Ql \&^ ) . +Finally, there is one new type of atom, a +.Em back reference : +.Ql \e +followed by a non-zero decimal digit +.Em d matches the same sequence of characters -matched by the \fId\fRth parenthesized subexpression +matched by the +.Em d Ns th +parenthesized subexpression (numbering subexpressions by the positions of their opening parentheses, left to right), -so that (e.g.) `\e([bc]\e)\e1' matches `bb' or `cc' but not `bc'. -.SH SEE ALSO -regex(3) -.PP -POSIX 1003.2, section 2.8 (Regular Expression Notation). -.SH HISTORY -Written by Henry Spencer, based on the 1003.2 spec. -.SH BUGS +so that (e.g.) +.Ql \e([bc]\e)\e1 +matches +.Ql bb +or +.Ql cc +but not +.Ql bc . +.Sh SEE ALSO +.Xr regex 3 +.Rs +.%T Regular Expression Notation +.%R IEEE Std +.%N 1003.2 +.%P section 2.8 +.Re +.Sh BUGS Having two kinds of REs is a botch. -.PP -The current 1003.2 spec says that `)' is an ordinary character in -the absence of an unmatched `('; +.Pp +The current +.St -p1003.2 +spec says that +.Ql \&) +is an ordinary character in +the absence of an unmatched +.Ql \&( ; this was an unintentional result of a wording error, and change is likely. Avoid relying on it. -.PP +.Pp Back references are a dreadful botch, posing major problems for efficient implementations. They are also somewhat vaguely defined (does -`a\e(\e(b\e)*\e2\e)*d' match `abbbd'?). +.Ql a\e(\e(b\e)*\e2\e)*d +match +.Ql abbbd ? ) . Avoid using them. -.PP -1003.2's specification of case-independent matching is vague. -The ``one case implies all cases'' definition given above +.Pp +.St -p1003.2 +specification of case-independent matching is vague. +The +.Dq one case implies all cases +definition given above is current consensus among implementors as to the right interpretation. -.PP +.Pp The syntax for word boundaries is incredibly ugly. diff --git a/winsup/cygwin/regex/regex.h b/winsup/cygwin/regex/regex.h deleted file mode 100644 index 03a182555..000000000 --- a/winsup/cygwin/regex/regex.h +++ /dev/null @@ -1,76 +0,0 @@ -#ifndef _REGEX_H_ -#define _REGEX_H_ /* never again */ -#include -/* ========= begin header generated by ./mkh ========= */ -#ifdef __cplusplus -extern "C" { -#endif - -/* === regex2.h === */ -typedef _off_t regoff_t; -typedef struct { - int re_magic; - size_t re_nsub; /* number of parenthesized subexpressions */ - const char *re_endp; /* end pointer for REG_PEND */ - struct re_guts *re_g; /* none of your business :-) */ -} regex_t; -typedef struct { - regoff_t rm_so; /* start of match */ - regoff_t rm_eo; /* end of match */ -} regmatch_t; - - -/* === regcomp.c === */ -extern int regcomp(regex_t *, const char *, int); -#define REG_BASIC 0000 -#define REG_EXTENDED 0001 -#define REG_ICASE 0002 -#define REG_NOSUB 0004 -#define REG_NEWLINE 0010 -#define REG_NOSPEC 0020 -#define REG_PEND 0040 -#define REG_DUMP 0200 - - -/* === regerror.c === */ -#define REG_OKAY 0 -#define REG_NOMATCH 1 -#define REG_BADPAT 2 -#define REG_ECOLLATE 3 -#define REG_ECTYPE 4 -#define REG_EESCAPE 5 -#define REG_ESUBREG 6 -#define REG_EBRACK 7 -#define REG_EPAREN 8 -#define REG_EBRACE 9 -#define REG_BADBR 10 -#define REG_ERANGE 11 -#define REG_ESPACE 12 -#define REG_BADRPT 13 -#define REG_EMPTY 14 -#define REG_ASSERT 15 -#define REG_INVARG 16 -#define REG_ATOI 255 /* convert name to number (!) */ -#define REG_ITOA 0400 /* convert number to name (!) */ -extern size_t regerror(int, const regex_t *, char *, size_t); - - -/* === regexec.c === */ -extern int regexec(const regex_t *, const char *, size_t, regmatch_t [], int); -#define REG_NOTBOL 00001 -#define REG_NOTEOL 00002 -#define REG_STARTEND 00004 -#define REG_TRACE 00400 /* tracing of execution */ -#define REG_LARGE 01000 /* force large representation */ -#define REG_BACKR 02000 /* force use of backref code */ - - -/* === regfree.c === */ -extern void regfree(regex_t *); - - -#ifdef __cplusplus -} -#endif -/* ========= end header generated by ./mkh ========= */ -#endif diff --git a/winsup/cygwin/regex/regex2.h b/winsup/cygwin/regex/regex2.h index 176038ab0..13bbf64a7 100644 --- a/winsup/cygwin/regex/regex2.h +++ b/winsup/cygwin/regex/regex2.h @@ -1,6 +1,42 @@ +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)regex2.h 8.4 (Berkeley) 3/20/94 + * $FreeBSD: src/lib/libc/regex/regex2.h,v 1.11 2007/01/09 00:28:04 imp Exp $ + */ + /* * First, the stuff that ends up in the outside-world include file - = typedef _off_t regoff_t; + = typedef off_t regoff_t; = typedef struct { = int re_magic; = size_t re_nsub; // number of parenthesized subexpressions @@ -36,66 +72,91 @@ * In state representations, an operator's bit is on to signify a state * immediately *preceding* "execution" of that operator. */ -typedef long sop; /* strip operator */ +typedef unsigned long sop; /* strip operator */ typedef long sopno; -#define OPRMASK 0x7c000000 -#define OPDMASK 0x03ffffff -#define OPSHIFT (26) +#define OPRMASK 0xf8000000L +#define OPDMASK 0x07ffffffL +#define OPSHIFT ((unsigned)27) #define OP(n) ((n)&OPRMASK) #define OPND(n) ((n)&OPDMASK) #define SOP(op, opnd) ((op)|(opnd)) /* operators meaning operand */ /* (back, fwd are offsets) */ -#define OEND (1< uch [csetsize] */ - uch mask; /* bit within array */ - uch hash; /* hash code */ - size_t smultis; - char *multis; /* -> char[smulti] ab\0cd\0ef\0\0 */ + wint_t min; + wint_t max; +} crange; +typedef struct { + unsigned char bmp[NC / 8]; + wctype_t *types; + int ntypes; + wint_t *wides; + int nwides; + crange *ranges; + int nranges; + int invert; + int icase; } cset; -/* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */ -#define CHadd(cs, c) ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c)) -#define CHsub(cs, c) ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c)) -#define CHIN(cs, c) ((cs)->ptr[(uch)(c)] & (cs)->mask) -#define MCadd(p, cs, cp) mcadd(p, cs, cp) /* regcomp() internal fns */ -#define MCsub(p, cs, cp) mcsub(p, cs, cp) -#define MCin(p, cs, cp) mcin(p, cs, cp) -/* stuff for character categories */ -typedef unsigned char cat_t; +static int +CHIN1(cset *cs, wint_t ch) +{ + int i; + + assert(ch >= 0); + if (ch < NC) + return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^ + cs->invert); + for (i = 0; i < cs->nwides; i++) + if (ch == cs->wides[i]) + return (!cs->invert); + for (i = 0; i < cs->nranges; i++) + if (cs->ranges[i].min <= ch && ch <= cs->ranges[i].max) + return (!cs->invert); + for (i = 0; i < cs->ntypes; i++) + if (iswctype(ch, cs->types[i])) + return (!cs->invert); + return (cs->invert); +} + +static __inline int +CHIN(cset *cs, wint_t ch) +{ + + assert(ch >= 0); + if (ch < NC) + return (((cs->bmp[ch >> 3] & (1 << (ch & 7))) != 0) ^ + cs->invert); + else if (cs->icase) + return (CHIN1(cs, ch) || CHIN1(cs, towlower(ch)) || + CHIN1(cs, towupper(ch))); + else + return (CHIN1(cs, ch)); +} /* * main compiled-expression structure @@ -104,10 +165,8 @@ struct re_guts { int magic; # define MAGIC2 ((('R'^0200)<<8)|'E') sop *strip; /* malloced area for strip */ - int csetsize; /* number of bits in a cset vector */ int ncsets; /* number of csets in use */ cset *sets; /* -> cset [ncsets] */ - uch *setbits; /* -> uch[csetsize][ncsets/CHAR_BIT] */ int cflags; /* copy of regcomp() cflags argument */ sopno nstates; /* = number of sops */ sopno firststate; /* the initial OEND (normally 0) */ @@ -118,17 +177,16 @@ struct re_guts { # define BAD 04 /* something wrong */ int nbol; /* number of ^ used */ int neol; /* number of $ used */ - int ncategories; /* how many character categories */ - cat_t *categories; /* ->catspace[-CHAR_MIN] */ char *must; /* match must contain this string */ + int moffset; /* latest point at which must may be located */ + int *charjump; /* Boyer-Moore char jump table */ + int *matchjump; /* Boyer-Moore match jump table */ int mlen; /* length of must */ size_t nsub; /* copy of re_nsub */ int backrefs; /* does it use back references? */ sopno nplus; /* how deep does it nest +s? */ - /* catspace must be last */ - cat_t catspace[1]; /* actually [NC] */ }; /* misc utilities */ -#define OUT (CHAR_MAX+1) /* a non-character value */ -#define ISWORD(c) (isalnum((unsigned char)c) || (c) == '_') +#define OUT (CHAR_MIN - 1) /* a non-character value */ +#define ISWORD(c) (iswalnum((uch)(c)) || (c) == '_') diff --git a/winsup/cygwin/regex/regexec.c b/winsup/cygwin/regex/regexec.c index 35b99c272..6195e508c 100644 --- a/winsup/cygwin/regex/regexec.c +++ b/winsup/cygwin/regex/regexec.c @@ -1,48 +1,131 @@ +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)regexec.c 8.3 (Berkeley) 3/20/94 + */ + +#if defined(LIBC_SCCS) && !defined(lint) +static char sccsid[] = "@(#)regexec.c 8.3 (Berkeley) 3/20/94"; +#endif /* LIBC_SCCS and not lint */ +#include +__FBSDID("$FreeBSD: src/lib/libc/regex/regexec.c,v 1.8 2007/06/11 03:05:54 delphij Exp $"); + /* * the outer shell of regexec() * - * This file includes engine.c *twice*, after muchos fiddling with the + * This file includes engine.c three times, after muchos fiddling with the * macros that code uses. This lets the same code operate on two different - * representations for state sets. + * representations for state sets and characters. */ +#ifdef __CYGWIN__ #include "winsup.h" +#endif #include #include #include #include #include #include -#include "regex.h" +#include +#include +#include #include "utils.h" #include "regex2.h" -#ifdef lint -static int nope = 0; /* for use in asserts; shuts lint up */ +#ifdef __CYGWIN__ +#define __unused __attribute__ ((unused)) #endif +static int nope __unused = 0; /* for use in asserts; shuts lint up */ + +static __inline size_t +xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy) +{ + size_t nr; + wchar_t wc; + + nr = mbrtowc(&wc, s, n, mbs); + if (wi != NULL) + *wi = wc; + if (nr == 0) + return (1); + else if (nr == (size_t)-1 || nr == (size_t)-2) { + memset(mbs, 0, sizeof(*mbs)); + if (wi != NULL) + *wi = dummy; + return (1); + } else + return (nr); +} + +static __inline size_t +xmbrtowc_dummy(wint_t *wi, + const char *s, + size_t n __unused, + mbstate_t *mbs __unused, + wint_t dummy __unused) +{ + + if (wi != NULL) + *wi = (unsigned char)*s; + return (1); +} + /* macros for manipulating states, small version */ -#define states unsigned -#define states1 unsigned /* for later use in regexec() decision */ +#define states long +#define states1 states /* for later use in regexec() decision */ #define CLEAR(v) ((v) = 0) -#define SET0(v, n) ((v) &= ~((unsigned)1 << (n))) -#define SET1(v, n) ((v) |= (unsigned)1 << (n)) -#define ISSET(v, n) ((v) & ((unsigned)1 << (n))) +#define SET0(v, n) ((v) &= ~((unsigned long)1 << (n))) +#define SET1(v, n) ((v) |= (unsigned long)1 << (n)) +#define ISSET(v, n) (((v) & ((unsigned long)1 << (n))) != 0) #define ASSIGN(d, s) ((d) = (s)) #define EQ(a, b) ((a) == (b)) -#define STATEVARS int dummy /* dummy version */ +#define STATEVARS long dummy /* dummy version */ #define STATESETUP(m, n) /* nothing */ #define STATETEARDOWN(m) /* nothing */ #define SETUP(v) ((v) = 0) -#define onestate unsigned -#define INIT(o, n) ((o) = (unsigned)1 << (n)) +#define onestate long +#define INIT(o, n) ((o) = (unsigned long)1 << (n)) #define INC(o) ((o) <<= 1) -#define ISSTATEIN(v, o) ((v) & (o)) +#define ISSTATEIN(v, o) (((v) & (o)) != 0) /* some abbreviations; note that some of these know variable names! */ /* do "if I'm here, I can also be there" etc without branches */ -#define FWD(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) << (n)) -#define BACK(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) >> (n)) -#define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n))) +#define FWD(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) << (n)) +#define BACK(dst, src, n) ((dst) |= ((unsigned long)(src)&(here)) >> (n)) +#define ISSETBACK(v, n) (((v) & ((unsigned long)here >> (n))) != 0) +/* no multibyte support */ +#define XMBRTOWC xmbrtowc_dummy +#define ZAPSTATE(mbs) ((void)(mbs)) /* function names */ #define SNAMES /* engine.c looks after details */ @@ -68,6 +151,8 @@ static int nope = 0; /* for use in asserts; shuts lint up */ #undef BACK #undef ISSETBACK #undef SNAMES +#undef XMBRTOWC +#undef ZAPSTATE /* macros for manipulating states, large version */ #define states char * @@ -77,13 +162,13 @@ static int nope = 0; /* for use in asserts; shuts lint up */ #define ISSET(v, n) ((v)[n]) #define ASSIGN(d, s) memcpy(d, s, m->g->nstates) #define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) -#define STATEVARS int vn; char *space +#define STATEVARS long vn; char *space #define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ if ((m)->space == NULL) return(REG_ESPACE); \ (m)->vn = 0; } #define STATETEARDOWN(m) { free((m)->space); } #define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) -#define onestate int +#define onestate long #define INIT(o, n) ((o) = (n)) #define INC(o) ((o)++) #define ISSTATEIN(v, o) ((v)[o]) @@ -92,11 +177,24 @@ static int nope = 0; /* for use in asserts; shuts lint up */ #define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) #define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) #define ISSETBACK(v, n) ((v)[here - (n)]) +/* no multibyte support */ +#define XMBRTOWC xmbrtowc_dummy +#define ZAPSTATE(mbs) ((void)(mbs)) /* function names */ #define LNAMES /* flag */ #include "engine.c" +/* multibyte character & large states version */ +#undef LNAMES +#undef XMBRTOWC +#undef ZAPSTATE +#define XMBRTOWC xmbrtowc +#define ZAPSTATE(mbs) memset((mbs), 0, sizeof(*(mbs))) +#define MNAMES + +#include "engine.c" + /* - regexec - interface for matching = extern int regexec(const regex_t *, const char *, size_t, \ @@ -113,14 +211,13 @@ static int nope = 0; /* for use in asserts; shuts lint up */ * have been prototyped. */ int /* 0 success, REG_NOMATCH failure */ -regexec(preg, string, nmatch, pmatch, eflags) -const regex_t *preg; -const char *string; -size_t nmatch; -regmatch_t pmatch[]; -int eflags; +regexec(const regex_t * __restrict preg, + const char * __restrict string, + size_t nmatch, + regmatch_t pmatch[__restrict], + int eflags) { - register struct re_guts *g = preg->re_g; + struct re_guts *g = preg->re_g; #ifdef REDEBUG # define GOODFLAGS(f) (f) #else @@ -134,7 +231,9 @@ int eflags; return(REG_BADPAT); eflags = GOODFLAGS(eflags); - if ((unsigned) g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) + if (MB_CUR_MAX > 1) + return(mmatcher(g, (char *)string, nmatch, pmatch, eflags)); + else if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) return(smatcher(g, (char *)string, nmatch, pmatch, eflags)); else return(lmatcher(g, (char *)string, nmatch, pmatch, eflags)); diff --git a/winsup/cygwin/regex/regfree.c b/winsup/cygwin/regex/regfree.c index b8be143d8..aa795fa78 100644 --- a/winsup/cygwin/regex/regfree.c +++ b/winsup/cygwin/regex/regfree.c @@ -1,8 +1,51 @@ -#include "winsup.h" +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)regfree.c 8.3 (Berkeley) 3/20/94 + */ + +#if defined(LIBC_SCCS) && !defined(lint) +static char sccsid[] = "@(#)regfree.c 8.3 (Berkeley) 3/20/94"; +#endif /* LIBC_SCCS and not lint */ +#include +__FBSDID("$FreeBSD: src/lib/libc/regex/regfree.c,v 1.8 2007/06/11 03:05:54 delphij Exp $"); + #include #include #include -#include "regex.h" +#include +#include +#include +#include #include "utils.h" #include "regex2.h" @@ -12,10 +55,10 @@ = extern void regfree(regex_t *); */ void -regfree(preg) -regex_t *preg; +regfree(regex_t *preg) { - register struct re_guts *g; + struct re_guts *g; + int i; if (preg->re_magic != MAGIC1) /* oops */ return; /* nice to complain, but hard */ @@ -28,11 +71,19 @@ regex_t *preg; if (g->strip != NULL) free((char *)g->strip); - if (g->sets != NULL) + if (g->sets != NULL) { + for (i = 0; i < g->ncsets; i++) { + free(g->sets[i].ranges); + free(g->sets[i].wides); + free(g->sets[i].types); + } free((char *)g->sets); - if (g->setbits != NULL) - free((char *)g->setbits); + } if (g->must != NULL) free(g->must); + if (g->charjump != NULL) + free(&g->charjump[CHAR_MIN]); + if (g->matchjump != NULL) + free(g->matchjump); free((char *)g); } diff --git a/winsup/cygwin/regex/tests b/winsup/cygwin/regex/tests deleted file mode 100644 index e4d928dad..000000000 --- a/winsup/cygwin/regex/tests +++ /dev/null @@ -1,477 +0,0 @@ -# regular expression test set -# Lines are at least three fields, separated by one or more tabs. "" stands -# for an empty field. First field is an RE. Second field is flags. If -# C flag given, regcomp() is expected to fail, and the third field is the -# error name (minus the leading REG_). -# -# Otherwise it is expected to succeed, and the third field is the string to -# try matching it against. If there is no fourth field, the match is -# expected to fail. If there is a fourth field, it is the substring that -# the RE is expected to match. If there is a fifth field, it is a comma- -# separated list of what the subexpressions should match, with - indicating -# no match for that one. In both the fourth and fifth fields, a (sub)field -# starting with @ indicates that the (sub)expression is expected to match -# a null string followed by the stuff after the @; this provides a way to -# test where null strings match. The character `N' in REs and strings -# is newline, `S' is space, `T' is tab, `Z' is NUL. -# -# The full list of flags: -# - placeholder, does nothing -# b RE is a BRE, not an ERE -# & try it as both an ERE and a BRE -# C regcomp() error expected, third field is error name -# i REG_ICASE -# m ("mundane") REG_NOSPEC -# s REG_NOSUB (not really testable) -# n REG_NEWLINE -# ^ REG_NOTBOL -# $ REG_NOTEOL -# # REG_STARTEND (see below) -# p REG_PEND -# -# For REG_STARTEND, the start/end offsets are those of the substring -# enclosed in (). - -# basics -a & a a -abc & abc abc -abc|de - abc abc -a|b|c - abc a - -# parentheses and perversions thereof -a(b)c - abc abc -a\(b\)c b abc abc -a( C EPAREN -a( b a( a( -a\( - a( a( -a\( bC EPAREN -a\(b bC EPAREN -a(b C EPAREN -a(b b a(b a(b -# gag me with a right parenthesis -- 1003.2 goofed here (my fault, partly) -a) - a) a) -) - ) ) -# end gagging (in a just world, those *should* give EPAREN) -a) b a) a) -a\) bC EPAREN -\) bC EPAREN -a()b - ab ab -a\(\)b b ab ab - -# anchoring and REG_NEWLINE -^abc$ & abc abc -a^b - a^b -a^b b a^b a^b -a$b - a$b -a$b b a$b a$b -^ & abc @abc -$ & abc @ -^$ & "" @ -$^ - "" @ -\($\)\(^\) b "" @ -# stop retching, those are legitimate (although disgusting) -^^ - "" @ -$$ - "" @ -b$ & abNc -b$ &n abNc b -^b$ & aNbNc -^b$ &n aNbNc b -^$ &n aNNb @Nb -^$ n abc -^$ n abcN @ -$^ n aNNb @Nb -\($\)\(^\) bn aNNb @Nb -^^ n^ aNNb @Nb -$$ n aNNb @NN -^a ^ a -a$ $ a -^a ^n aNb -^b ^n aNb b -a$ $n bNa -b$ $n bNa b -a*(^b$)c* - b b -a*\(^b$\)c* b b b - -# certain syntax errors and non-errors -| C EMPTY -| b | | -* C BADRPT -* b * * -+ C BADRPT -? C BADRPT -"" &C EMPTY -() - abc @abc -\(\) b abc @abc -a||b C EMPTY -|ab C EMPTY -ab| C EMPTY -(|a)b C EMPTY -(a|)b C EMPTY -(*a) C BADRPT -(+a) C BADRPT -(?a) C BADRPT -({1}a) C BADRPT -\(\{1\}a\) bC BADRPT -(a|*b) C BADRPT -(a|+b) C BADRPT -(a|?b) C BADRPT -(a|{1}b) C BADRPT -^* C BADRPT -^* b * * -^+ C BADRPT -^? C BADRPT -^{1} C BADRPT -^\{1\} bC BADRPT - -# metacharacters, backslashes -a.c & abc abc -a[bc]d & abd abd -a\*c & a*c a*c -a\\b & a\b a\b -a\\\*b & a\*b a\*b -a\bc & abc abc -a\ &C EESCAPE -a\\bc & a\bc a\bc -\{ bC BADRPT -a\[b & a[b a[b -a[b &C EBRACK -# trailing $ is a peculiar special case for the BRE code -a$ & a a -a$ & a$ -a\$ & a -a\$ & a$ a$ -a\\$ & a -a\\$ & a$ -a\\$ & a\$ -a\\$ & a\ a\ - -# back references, ugh -a\(b\)\2c bC ESUBREG -a\(b\1\)c bC ESUBREG -a\(b*\)c\1d b abbcbbd abbcbbd bb -a\(b*\)c\1d b abbcbd -a\(b*\)c\1d b abbcbbbd -^\(.\)\1 b abc -a\([bc]\)\1d b abcdabbd abbd b -a\(\([bc]\)\2\)*d b abbccd abbccd -a\(\([bc]\)\2\)*d b abbcbd -# actually, this next one probably ought to fail, but the spec is unclear -a\(\(b\)*\2\)*d b abbbd abbbd -# here is a case that no NFA implementation does right -\(ab*\)[ab]*\1 b ababaaa ababaaa a -# check out normal matching in the presence of back refs -\(a\)\1bcd b aabcd aabcd -\(a\)\1bc*d b aabcd aabcd -\(a\)\1bc*d b aabd aabd -\(a\)\1bc*d b aabcccd aabcccd -\(a\)\1bc*[ce]d b aabcccd aabcccd -^\(a\)\1b\(c\)*cd$ b aabcccd aabcccd - -# ordinary repetitions -ab*c & abc abc -ab+c - abc abc -ab?c - abc abc -a\(*\)b b a*b a*b -a\(**\)b b ab ab -a\(***\)b bC BADRPT -*a b *a *a -**a b a a -***a bC BADRPT - -# the dreaded bounded repetitions -{ & { { -{abc & {abc {abc -{1 C BADRPT -{1} C BADRPT -a{b & a{b a{b -a{1}b - ab ab -a\{1\}b b ab ab -a{1,}b - ab ab -a\{1,\}b b ab ab -a{1,2}b - aab aab -a\{1,2\}b b aab aab -a{1 C EBRACE -a\{1 bC EBRACE -a{1a C EBRACE -a\{1a bC EBRACE -a{1a} C BADBR -a\{1a\} bC BADBR -a{,2} - a{,2} a{,2} -a\{,2\} bC BADBR -a{,} - a{,} a{,} -a\{,\} bC BADBR -a{1,x} C BADBR -a\{1,x\} bC BADBR -a{1,x C EBRACE -a\{1,x bC EBRACE -a{300} C BADBR -a\{300\} bC BADBR -a{1,0} C BADBR -a\{1,0\} bC BADBR -ab{0,0}c - abcac ac -ab\{0,0\}c b abcac ac -ab{0,1}c - abcac abc -ab\{0,1\}c b abcac abc -ab{0,3}c - abbcac abbc -ab\{0,3\}c b abbcac abbc -ab{1,1}c - acabc abc -ab\{1,1\}c b acabc abc -ab{1,3}c - acabc abc -ab\{1,3\}c b acabc abc -ab{2,2}c - abcabbc abbc -ab\{2,2\}c b abcabbc abbc -ab{2,4}c - abcabbc abbc -ab\{2,4\}c b abcabbc abbc -((a{1,10}){1,10}){1,10} - a a a,a - -# multiple repetitions -a** &C BADRPT -a++ C BADRPT -a?? C BADRPT -a*+ C BADRPT -a*? C BADRPT -a+* C BADRPT -a+? C BADRPT -a?* C BADRPT -a?+ C BADRPT -a{1}{1} C BADRPT -a*{1} C BADRPT -a+{1} C BADRPT -a?{1} C BADRPT -a{1}* C BADRPT -a{1}+ C BADRPT -a{1}? C BADRPT -a*{b} - a{b} a{b} -a\{1\}\{1\} bC BADRPT -a*\{1\} bC BADRPT -a\{1\}* bC BADRPT - -# brackets, and numerous perversions thereof -a[b]c & abc abc -a[ab]c & abc abc -a[^ab]c & adc adc -a[]b]c & a]c a]c -a[[b]c & a[c a[c -a[-b]c & a-c a-c -a[^]b]c & adc adc -a[^-b]c & adc adc -a[b-]c & a-c a-c -a[b &C EBRACK -a[] &C EBRACK -a[1-3]c & a2c a2c -a[3-1]c &C ERANGE -a[1-3-5]c &C ERANGE -a[[.-.]--]c & a-c a-c -a[1- &C ERANGE -a[[. &C EBRACK -a[[.x &C EBRACK -a[[.x. &C EBRACK -a[[.x.] &C EBRACK -a[[.x.]] & ax ax -a[[.x,.]] &C ECOLLATE -a[[.one.]]b & a1b a1b -a[[.notdef.]]b &C ECOLLATE -a[[.].]]b & a]b a]b -a[[:alpha:]]c & abc abc -a[[:notdef:]]c &C ECTYPE -a[[: &C EBRACK -a[[:alpha &C EBRACK -a[[:alpha:] &C EBRACK -a[[:alpha,:] &C ECTYPE -a[[:]:]]b &C ECTYPE -a[[:-:]]b &C ECTYPE -a[[:alph:]] &C ECTYPE -a[[:alphabet:]] &C ECTYPE -[[:alnum:]]+ - -%@a0X- a0X -[[:alpha:]]+ - -%@aX0- aX -[[:blank:]]+ - aSSTb SST -[[:cntrl:]]+ - aNTb NT -[[:digit:]]+ - a019b 019 -[[:graph:]]+ - Sa%bS a%b -[[:lower:]]+ - AabC ab -[[:print:]]+ - NaSbN aSb -[[:punct:]]+ - S%-&T %-& -[[:space:]]+ - aSNTb SNT -[[:upper:]]+ - aBCd BC -[[:xdigit:]]+ - p0f3Cq 0f3C -a[[=b=]]c & abc abc -a[[= &C EBRACK -a[[=b &C EBRACK -a[[=b= &C EBRACK -a[[=b=] &C EBRACK -a[[=b,=]] &C ECOLLATE -a[[=one=]]b & a1b a1b - -# complexities -a(((b)))c - abc abc -a(b|(c))d - abd abd -a(b*|c)d - abbd abbd -# just gotta have one DFA-buster, of course -a[ab]{20} - aaaaabaaaabaaaabaaaab aaaaabaaaabaaaabaaaab -# and an inline expansion in case somebody gets tricky -a[ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab] - aaaaabaaaabaaaabaaaab aaaaabaaaabaaaabaaaab -# and in case somebody just slips in an NFA... -a[ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab](wee|week)(knights|night) - aaaaabaaaabaaaabaaaabweeknights aaaaabaaaabaaaabaaaabweeknights -# fish for anomalies as the number of states passes 32 -12345678901234567890123456789 - a12345678901234567890123456789b 12345678901234567890123456789 -123456789012345678901234567890 - a123456789012345678901234567890b 123456789012345678901234567890 -1234567890123456789012345678901 - a1234567890123456789012345678901b 1234567890123456789012345678901 -12345678901234567890123456789012 - a12345678901234567890123456789012b 12345678901234567890123456789012 -123456789012345678901234567890123 - a123456789012345678901234567890123b 123456789012345678901234567890123 -# and one really big one, beyond any plausible word width -1234567890123456789012345678901234567890123456789012345678901234567890 - a1234567890123456789012345678901234567890123456789012345678901234567890b 1234567890123456789012345678901234567890123456789012345678901234567890 -# fish for problems as brackets go past 8 -[ab][cd][ef][gh][ij][kl][mn] - xacegikmoq acegikm -[ab][cd][ef][gh][ij][kl][mn][op] - xacegikmoq acegikmo -[ab][cd][ef][gh][ij][kl][mn][op][qr] - xacegikmoqy acegikmoq -[ab][cd][ef][gh][ij][kl][mn][op][q] - xacegikmoqy acegikmoq - -# subtleties of matching -abc & xabcy abc -a\(b\)?c\1d b acd -aBc i Abc Abc -a[Bc]*d i abBCcd abBCcd -0[[:upper:]]1 &i 0a1 0a1 -0[[:lower:]]1 &i 0A1 0A1 -a[^b]c &i abc -a[^b]c &i aBc -a[^b]c &i adc adc -[a]b[c] - abc abc -[a]b[a] - aba aba -[abc]b[abc] - abc abc -[abc]b[abd] - abd abd -a(b?c)+d - accd accd -(wee|week)(knights|night) - weeknights weeknights -(we|wee|week|frob)(knights|night|day) - weeknights weeknights -a[bc]d - xyzaaabcaababdacd abd -a[ab]c - aaabc abc -abc s abc abc -a* & b @b - -# Let's have some fun -- try to match a C comment. -# first the obvious, which looks okay at first glance... -/\*.*\*/ - /*x*/ /*x*/ -# but... -/\*.*\*/ - /*x*/y/*z*/ /*x*/y/*z*/ -# okay, we must not match */ inside; try to do that... -/\*([^*]|\*[^/])*\*/ - /*x*/ /*x*/ -/\*([^*]|\*[^/])*\*/ - /*x*/y/*z*/ /*x*/ -# but... -/\*([^*]|\*[^/])*\*/ - /*x**/y/*z*/ /*x**/y/*z*/ -# and a still fancier version, which does it right (I think)... -/\*([^*]|\*+[^*/])*\*+/ - /*x*/ /*x*/ -/\*([^*]|\*+[^*/])*\*+/ - /*x*/y/*z*/ /*x*/ -/\*([^*]|\*+[^*/])*\*+/ - /*x**/y/*z*/ /*x**/ -/\*([^*]|\*+[^*/])*\*+/ - /*x****/y/*z*/ /*x****/ -/\*([^*]|\*+[^*/])*\*+/ - /*x**x*/y/*z*/ /*x**x*/ -/\*([^*]|\*+[^*/])*\*+/ - /*x***x/y/*z*/ /*x***x/y/*z*/ - -# subexpressions -.* - abc abc - -a(b)(c)d - abcd abcd b,c -a(((b)))c - abc abc b,b,b -a(b|(c))d - abd abd b,- -a(b*|c|e)d - abbd abbd bb -a(b*|c|e)d - acd acd c -a(b*|c|e)d - ad ad @d -a(b?)c - abc abc b -a(b?)c - ac ac @c -a(b+)c - abc abc b -a(b+)c - abbbc abbbc bbb -a(b*)c - ac ac @c -(a|ab)(bc([de]+)f|cde) - abcdef abcdef a,bcdef,de -# the regression tester only asks for 9 subexpressions -a(b)(c)(d)(e)(f)(g)(h)(i)(j)k - abcdefghijk abcdefghijk b,c,d,e,f,g,h,i,j -a(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)l - abcdefghijkl abcdefghijkl b,c,d,e,f,g,h,i,j,k -a([bc]?)c - abc abc b -a([bc]?)c - ac ac @c -a([bc]+)c - abc abc b -a([bc]+)c - abcc abcc bc -a([bc]+)bc - abcbc abcbc bc -a(bb+|b)b - abb abb b -a(bbb+|bb+|b)b - abb abb b -a(bbb+|bb+|b)b - abbb abbb bb -a(bbb+|bb+|b)bb - abbb abbb b -(.*).* - abcdef abcdef abcdef -(a*)* - bc @b @b - -# do we get the right subexpression when it is used more than once? -a(b|c)*d - ad ad - -a(b|c)*d - abcd abcd c -a(b|c)+d - abd abd b -a(b|c)+d - abcd abcd c -a(b|c?)+d - ad ad @d -a(b|c?)+d - abcd abcd @d -a(b|c){0,0}d - ad ad - -a(b|c){0,1}d - ad ad - -a(b|c){0,1}d - abd abd b -a(b|c){0,2}d - ad ad - -a(b|c){0,2}d - abcd abcd c -a(b|c){0,}d - ad ad - -a(b|c){0,}d - abcd abcd c -a(b|c){1,1}d - abd abd b -a(b|c){1,1}d - acd acd c -a(b|c){1,2}d - abd abd b -a(b|c){1,2}d - abcd abcd c -a(b|c){1,}d - abd abd b -a(b|c){1,}d - abcd abcd c -a(b|c){2,2}d - acbd acbd b -a(b|c){2,2}d - abcd abcd c -a(b|c){2,4}d - abcd abcd c -a(b|c){2,4}d - abcbd abcbd b -a(b|c){2,4}d - abcbcd abcbcd c -a(b|c){2,}d - abcd abcd c -a(b|c){2,}d - abcbd abcbd b -a(b+|((c)*))+d - abd abd @d,@d,- -a(b+|((c)*))+d - abcd abcd @d,@d,- - -# check out the STARTEND option -[abc] &# a(b)c b -[abc] &# a(d)c -[abc] &# a(bc)d b -[abc] &# a(dc)d c -. &# a()c -b.*c &# b(bc)c bc -b.* &# b(bc)c bc -.*c &# b(bc)c bc - -# plain strings, with the NOSPEC flag -abc m abc abc -abc m xabcy abc -abc m xyz -a*b m aba*b a*b -a*b m ab -"" mC EMPTY - -# cases involving NULs -aZb & a a -aZb &p a -aZb &p# (aZb) aZb -aZ*b &p# (ab) ab -a.b &# (aZb) aZb -a.* &# (aZb)c aZb - -# word boundaries (ick) -[[:<:]]a & a a -[[:<:]]a & ba -[[:<:]]a & -a a -a[[:>:]] & a a -a[[:>:]] & ab -a[[:>:]] & a- a -[[:<:]]a.c[[:>:]] & axcd-dayc-dazce-abc abc -[[:<:]]a.c[[:>:]] & axcd-dayc-dazce-abc-q abc -[[:<:]]a.c[[:>:]] & axc-dayc-dazce-abc axc -[[:<:]]b.c[[:>:]] & a_bxc-byc_d-bzc-q bzc -[[:<:]].x..[[:>:]] & y_xa_-_xb_y-_xc_-axdc _xc_ -[[:<:]]a_b[[:>:]] & x_a_b - -# past problems, and suspected problems -(A[1])|(A[2])|(A[3])|(A[4])|(A[5])|(A[6])|(A[7])|(A[8])|(A[9])|(A[A]) - A1 A1 -abcdefghijklmnop i abcdefghijklmnop abcdefghijklmnop -abcdefghijklmnopqrstuv i abcdefghijklmnopqrstuv abcdefghijklmnopqrstuv -(ALAK)|(ALT[AB])|(CC[123]1)|(CM[123]1)|(GAMC)|(LC[23][EO ])|(SEM[1234])|(SL[ES][12])|(SLWW)|(SLF )|(SLDT)|(VWH[12])|(WH[34][EW])|(WP1[ESN]) - CC11 CC11 -CC[13]1|a{21}[23][EO][123][Es][12]a{15}aa[34][EW]aaaaaaa[X]a - CC11 CC11 -Char \([a-z0-9_]*\)\[.* b Char xyz[k Char xyz[k xyz -a?b - ab ab --\{0,1\}[0-9]*$ b -5 -5 -a*a*a*a*a*a*a* & aaaaaa aaaaaa diff --git a/winsup/cygwin/regex/utils.h b/winsup/cygwin/regex/utils.h index 1a997ac8f..2a2ed9694 100644 --- a/winsup/cygwin/regex/utils.h +++ b/winsup/cygwin/regex/utils.h @@ -1,9 +1,41 @@ +/*- + * Copyright (c) 1992, 1993, 1994 Henry Spencer. + * Copyright (c) 1992, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Henry Spencer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)utils.h 8.3 (Berkeley) 3/20/94 + * $FreeBSD: src/lib/libc/regex/utils.h,v 1.3 2007/01/09 00:28:04 imp Exp $ + */ + /* utility definitions */ -#ifdef _POSIX2_RE_DUP_MAX -#define DUPMAX _POSIX2_RE_DUP_MAX -#else -#define DUPMAX 255 -#endif +#define DUPMAX _POSIX2_RE_DUP_MAX /* xxx is this right? */ #define INFINITY (DUPMAX + 1) #define NC (CHAR_MAX - CHAR_MIN + 1) typedef unsigned char uch; diff --git a/winsup/cygwin/wchar.h b/winsup/cygwin/wchar.h index 820519af5..fbeb04c70 100644 --- a/winsup/cygwin/wchar.h +++ b/winsup/cygwin/wchar.h @@ -49,6 +49,7 @@ extern char *__locale_charset (); #endif #ifdef __INSIDE_CYGWIN__ +#ifdef __cplusplus size_t __stdcall sys_cp_wcstombs (wctomb_p, const char *, char *, size_t, const wchar_t *, size_t = (size_t) -1) __attribute__ ((regparm(3))); @@ -68,6 +69,7 @@ size_t __stdcall sys_mbstowcs (wchar_t * dst, size_t dlen, const char *src, size_t __stdcall sys_mbstowcs_alloc (wchar_t **, int, const char *, size_t = (size_t) -1) __attribute__ ((regparm(3))); +#endif /* __cplusplus */ #endif /* __INSIDE_CYGWIN__ */ #endif /* _CYGWIN_WCHAR_H */