From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 19383 invoked by alias); 5 Feb 2007 15:14:53 -0000 Received: (qmail 19366 invoked by uid 22791); 5 Feb 2007 15:14:52 -0000 X-Spam-Check-By: sourceware.org Received: from sunsite.ms.mff.cuni.cz (HELO sunsite.mff.cuni.cz) (195.113.15.26) by sourceware.org (qpsmtpd/0.31) with ESMTP; Mon, 05 Feb 2007 15:14:44 +0000 Received: from sunsite.mff.cuni.cz (localhost.localdomain [127.0.0.1]) by sunsite.mff.cuni.cz (8.13.8/8.13.8) with ESMTP id l15FFCVY007353; Mon, 5 Feb 2007 16:15:12 +0100 Received: (from jakub@localhost) by sunsite.mff.cuni.cz (8.13.8/8.13.8/Submit) id l15FFCb8007351; Mon, 5 Feb 2007 16:15:12 +0100 Date: Mon, 05 Feb 2007 15:14:00 -0000 From: Jakub Jelinek To: Ulrich Drepper Cc: Glibc hackers Subject: [PATCH] Fix regex REG_NEWLINE/RE_HAT_LISTS_NOT_NEWLINE (BZ #3957) Message-ID: <20070205151511.GA4219@sunsite.mff.cuni.cz> Reply-To: Jakub Jelinek Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.2i Mailing-List: contact libc-hacker-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libc-hacker-owner@sourceware.org X-SW-Source: 2007-02/txt/msg00001.txt.bz2 Hi! RE_HAT_LISTS_NOT_NEWLINE should add \n to [^...] lists, not \0. For build_charclass_op (i.e. \s, \S, \w, \W) only \W might make a difference, but I checked the pre-2002 regex code and \W matched there '\n' no matter whether RE_HAT_LISTS_NOT_NEWLINE was set or not. Both testcases pass with pre-2002 regex code. 2007-02-05 Jakub Jelinek [BZ #3957] * posix/regcomp.c (parse_bracket_exp): Set '\n' bit rather than '\0' bit for RE_HAT_LISTS_NOT_NEWLINE. (build_charclass_op): Remove bogus comment. * posix/Makefile (tests): Add bug-regex27 and bug-regex28. * posix/bug-regex27.c: New test. * posix/bug-regex28.c: New test. --- libc/posix/regcomp.c.jj 2006-01-31 20:17:14.000000000 +0100 +++ libc/posix/regcomp.c 2007-02-05 15:50:37.000000000 +0100 @@ -1,5 +1,5 @@ /* Extended regular expression matching and search library. - Copyright (C) 2002,2003,2004,2005,2006 Free Software Foundation, Inc. + Copyright (C) 2002,2003,2004,2005,2006,2007 Free Software Foundation, Inc. This file is part of the GNU C Library. Contributed by Isamu Hasegawa . @@ -3019,7 +3019,7 @@ parse_bracket_exp (re_string_t *regexp, #endif /* not RE_ENABLE_I18N */ non_match = 1; if (syntax & RE_HAT_LISTS_NOT_NEWLINE) - bitset_set (sbcset, '\0'); + bitset_set (sbcset, '\n'); re_string_skip_bytes (regexp, token_len); /* Skip a token. */ token_len = peek_token_bracket (token, regexp, syntax); if (BE (token->type == END_OF_RE, 0)) @@ -3549,10 +3549,6 @@ build_charclass_op (re_dfa_t *dfa, RE_TR if (non_match) { #ifdef RE_ENABLE_I18N - /* - if (syntax & RE_HAT_LISTS_NOT_NEWLINE) - bitset_set(cset->sbcset, '\0'); - */ mbcset->non_match = 1; #endif /* not RE_ENABLE_I18N */ } --- libc/posix/Makefile.jj 2007-01-10 17:15:15.000000000 +0100 +++ libc/posix/Makefile 2007-02-05 15:50:18.000000000 +0100 @@ -81,7 +81,8 @@ tests := tstgetopt testfnm runtests run bug-regex13 bug-regex14 bug-regex15 bug-regex16 \ bug-regex17 bug-regex18 bug-regex19 bug-regex20 \ bug-regex21 bug-regex22 bug-regex23 bug-regex24 \ - bug-regex25 bug-regex26 tst-nice tst-nanosleep tst-regex2 \ + bug-regex25 bug-regex26 bug-regex27 bug-regex28 \ + tst-nice tst-nanosleep tst-regex2 \ transbug tst-rxspencer tst-pcre tst-boost \ bug-ga1 tst-vfork1 tst-vfork2 tst-vfork3 tst-waitid \ tst-getaddrinfo2 bug-glob1 bug-glob2 tst-sysconf \ --- libc/posix/bug-regex27.c.jj 2007-02-05 15:18:18.000000000 +0100 +++ libc/posix/bug-regex27.c 2007-02-05 14:52:35.000000000 +0100 @@ -0,0 +1,64 @@ +/* Test REG_NEWLINE. + Copyright (C) 2007 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Jakub Jelinek , 2007. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. */ + +#include +#include +#include + +struct tests +{ + const char *regex; + const char *string; + int cflags; + int retval; +} tests[] = { + { "a.b", "a\nb", REG_EXTENDED | REG_NEWLINE, REG_NOMATCH }, + { "a.b", "a\nb", REG_EXTENDED, 0 }, + { "a[^x]b", "a\nb", REG_EXTENDED | REG_NEWLINE, REG_NOMATCH }, + { "a[^x]b", "a\nb", REG_EXTENDED, 0 } +}; + +int +main (void) +{ + regex_t r; + size_t i; + int ret = 0; + + for (i = 0; i < sizeof (tests) / sizeof (tests[i]); ++i) + { + memset (&r, 0, sizeof (r)); + if (regcomp (&r, tests[i].regex, tests[i].cflags)) + { + printf ("regcomp %zd failed\n", i); + ret = 1; + continue; + } + int rv = regexec (&r, tests[i].string, 0, NULL, 0); + if (rv != tests[i].retval) + { + printf ("regexec %zd unexpected value %d != %d\n", + i, rv, tests[i].retval); + ret = 1; + } + regfree (&r); + } + return ret; +} --- libc/posix/bug-regex28.c.jj 2007-02-05 15:18:30.000000000 +0100 +++ libc/posix/bug-regex28.c 2007-02-05 16:04:41.000000000 +0100 @@ -0,0 +1,75 @@ +/* Test RE_HAT_LISTS_NOT_NEWLINE and RE_DOT_NEWLINE. + Copyright (C) 2007 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Jakub Jelinek , 2007. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. */ + +#include +#include +#include + +struct tests +{ + const char *regex; + const char *string; + reg_syntax_t syntax; + int retval; +} tests[] = { +#define EGREP RE_SYNTAX_EGREP +#define EGREP_NL (RE_SYNTAX_EGREP | RE_DOT_NEWLINE) & ~RE_HAT_LISTS_NOT_NEWLINE + { "a.b", "a\nb", EGREP, -1 }, + { "a.b", "a\nb", EGREP_NL, 0 }, + { "a[^x]b", "a\nb", EGREP, -1 }, + { "a[^x]b", "a\nb", EGREP_NL, 0 }, + /* While \S and \W are internally handled as [^[:space:]] and [^[:alnum:]_], + RE_HAT_LISTS_NOT_NEWLINE did not make any difference, so ensure + it doesn't change. */ + { "a\\Sb", "a\nb", EGREP, -1 }, + { "a\\Sb", "a\nb", EGREP_NL, -1 }, + { "a\\Wb", "a\nb", EGREP, 0 }, + { "a\\Wb", "a\nb", EGREP_NL, 0 } +}; + +int +main (void) +{ + struct re_pattern_buffer r; + size_t i; + int ret = 0; + + for (i = 0; i < sizeof (tests) / sizeof (tests[i]); ++i) + { + re_set_syntax (tests[i].syntax); + memset (&r, 0, sizeof (r)); + if (re_compile_pattern (tests[i].regex, strlen (tests[i].regex), &r)) + { + printf ("re_compile_pattern %zd failed\n", i); + ret = 1; + continue; + } + size_t len = strlen (tests[i].string); + int rv = re_search (&r, tests[i].string, len, 0, len, NULL); + if (rv != tests[i].retval) + { + printf ("re_search %zd unexpected value %d != %d\n", + i, rv, tests[i].retval); + ret = 1; + } + regfree (&r); + } + return ret; +} Jakub