X-Git-Url: http://erislabs.net/gitweb/?a=blobdiff_plain;f=m4%2Fregex.m4;h=424ae333a3c756c07664b8e2b3ce1afd03ccecf5;hb=3b07ac4d2688471de4c8979e914a5b688b3b66fe;hp=007b166364f40b402cdf9953f302b0582ff58da4;hpb=44ec9731a3e3e51d6bc1d19b782ab32beb14d94f;p=gnulib.git diff --git a/m4/regex.m4 b/m4/regex.m4 index 007b16636..424ae333a 100644 --- a/m4/regex.m4 +++ b/m4/regex.m4 @@ -1,37 +1,273 @@ -## --------------------------------- ## -## Check if --with-regex was given. ## -## --------------------------------- ## - -# serial 1 - -# The idea is to distribute rx.[hc] and regex.[hc] together, for a while. -# The WITH_REGEX symbol (which should also be documented in acconfig.h) -# is used to decide which of regex.h or rx.h should be included in the -# application. If `./configure --with-regex' is given (the default), the -# package will use gawk's regex. If `./configure --without-regex', a -# check is made to see if rx is already installed, as with newer Linux'es. -# If not found, the package will use the rx from the distribution. -# If found, the package will use the system's rx which, on Linux at least, -# will result in a smaller executable file. - -AC_DEFUN(AM_WITH_REGEX, -[AC_MSG_CHECKING(which of GNU rx or gawk's regex is wanted) -AC_ARG_WITH(regex, -[ --without-regex use GNU rx in lieu of gawk's regex for matching], -[test "$withval" = yes && am_with_regex=1], -[am_with_regex=1]) -if test -n "$am_with_regex"; then - AC_MSG_RESULT(regex) - AC_DEFINE(WITH_REGEX) - AC_CACHE_CHECK([for GNU regex in libc], am_cv_gnu_regex, - AC_TRY_LINK([], [extern int re_max_failures; re_max_failures = 1], - am_cv_gnu_regex=yes, am_cv_gnu_regex=no)) - if test $am_cv_gnu_regex = no; then - LIBOBJS="$LIBOBJS regex.o" +# serial 64 + +# Copyright (C) 1996-2001, 2003-2013 Free Software Foundation, Inc. +# +# This file is free software; the Free Software Foundation +# gives unlimited permission to copy and/or distribute it, +# with or without modifications, as long as this notice is preserved. + +dnl Initially derived from code in GNU grep. +dnl Mostly written by Jim Meyering. + +AC_PREREQ([2.50]) + +AC_DEFUN([gl_REGEX], +[ + AC_ARG_WITH([included-regex], + [AS_HELP_STRING([--without-included-regex], + [don't compile regex; this is the default on systems + with recent-enough versions of the GNU C Library + (use with caution on other systems).])]) + + case $with_included_regex in #( + yes|no) ac_use_included_regex=$with_included_regex + ;; + '') + # If the system regex support is good enough that it passes the + # following run test, then default to *not* using the included regex.c. + # If cross compiling, assume the test would fail and use the included + # regex.c. + AC_CHECK_DECLS_ONCE([alarm]) + AC_CACHE_CHECK([for working re_compile_pattern], + [gl_cv_func_re_compile_pattern_working], + [AC_RUN_IFELSE( + [AC_LANG_PROGRAM( + [[#include + + #include + #include + #include + #if HAVE_DECL_ALARM + # include + # include + #endif + ]], + [[int result = 0; + static struct re_pattern_buffer regex; + unsigned char folded_chars[UCHAR_MAX + 1]; + int i; + const char *s; + struct re_registers regs; + +#if HAVE_DECL_ALARM + /* Some builds of glibc go into an infinite loop on this test. */ + signal (SIGALRM, SIG_DFL); + alarm (2); +#endif + if (setlocale (LC_ALL, "en_US.UTF-8")) + { + { + /* http://sourceware.org/ml/libc-hacker/2006-09/msg00008.html + This test needs valgrind to catch the bug on Debian + GNU/Linux 3.1 x86, but it might catch the bug better + on other platforms and it shouldn't hurt to try the + test here. */ + static char const pat[] = "insert into"; + static char const data[] = + "\xFF\0\x12\xA2\xAA\xC4\xB1,K\x12\xC4\xB1*\xACK"; + re_set_syntax (RE_SYNTAX_GREP | RE_HAT_LISTS_NOT_NEWLINE + | RE_ICASE); + memset (®ex, 0, sizeof regex); + s = re_compile_pattern (pat, sizeof pat - 1, ®ex); + if (s) + result |= 1; + else if (re_search (®ex, data, sizeof data - 1, + 0, sizeof data - 1, ®s) + != -1) + result |= 1; + } + + { + /* This test is from glibc bug 15078. + The test case is from Andreas Schwab in + . + */ + static char const pat[] = "[^x]x"; + static char const data[] = + /* */ + "\xe1\x80\x80" + "\xe1\x80\xbb" + "\xe1\x80\xbd" + "\xe1\x80\x94" + "\xe1\x80\xba" + "\xe1\x80\xaf" + "\xe1\x80\x95" + "\xe1\x80\xba" + "x"; + re_set_syntax (0); + memset (®ex, 0, sizeof regex); + s = re_compile_pattern (pat, sizeof pat - 1, ®ex); + if (s) + result |= 1; + else + { + i = re_search (®ex, data, sizeof data - 1, + 0, sizeof data - 1, 0); + if (i != 0 && i != 21) + result |= 1; + } + } + + if (! setlocale (LC_ALL, "C")) + return 1; + } + + /* This test is from glibc bug 3957, reported by Andrew Mackey. */ + re_set_syntax (RE_SYNTAX_EGREP | RE_HAT_LISTS_NOT_NEWLINE); + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("a[^x]b", 6, ®ex); + if (s) + result |= 2; + /* This should fail, but succeeds for glibc-2.5. */ + else if (re_search (®ex, "a\nb", 3, 0, 3, ®s) != -1) + result |= 2; + + /* This regular expression is from Spencer ere test number 75 + in grep-2.3. */ + re_set_syntax (RE_SYNTAX_POSIX_EGREP); + memset (®ex, 0, sizeof regex); + for (i = 0; i <= UCHAR_MAX; i++) + folded_chars[i] = i; + regex.translate = folded_chars; + s = re_compile_pattern ("a[[:@:>@:]]b\n", 11, ®ex); + /* This should fail with _Invalid character class name_ error. */ + if (!s) + result |= 4; + + /* Ensure that [b-a] is diagnosed as invalid, when + using RE_NO_EMPTY_RANGES. */ + re_set_syntax (RE_SYNTAX_POSIX_EGREP | RE_NO_EMPTY_RANGES); + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("a[b-a]", 6, ®ex); + if (s == 0) + result |= 8; + + /* This should succeed, but does not for glibc-2.1.3. */ + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("{1", 2, ®ex); + if (s) + result |= 8; + + /* The following example is derived from a problem report + against gawk from Jorge Stolfi . */ + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("[an\371]*n", 7, ®ex); + if (s) + result |= 8; + /* This should match, but does not for glibc-2.2.1. */ + else if (re_match (®ex, "an", 2, 0, ®s) != 2) + result |= 8; + + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("x", 1, ®ex); + if (s) + result |= 8; + /* glibc-2.2.93 does not work with a negative RANGE argument. */ + else if (re_search (®ex, "wxy", 3, 2, -2, ®s) != 1) + result |= 8; + + /* The version of regex.c in older versions of gnulib + ignored RE_ICASE. Detect that problem too. */ + re_set_syntax (RE_SYNTAX_EMACS | RE_ICASE); + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("x", 1, ®ex); + if (s) + result |= 16; + else if (re_search (®ex, "WXY", 3, 0, 3, ®s) < 0) + result |= 16; + + /* Catch a bug reported by Vin Shelton in + http://lists.gnu.org/archive/html/bug-coreutils/2007-06/msg00089.html + */ + re_set_syntax (RE_SYNTAX_POSIX_BASIC + & ~RE_CONTEXT_INVALID_DUP + & ~RE_NO_EMPTY_RANGES); + memset (®ex, 0, sizeof regex); + s = re_compile_pattern ("[[:alnum:]_-]\\\\+$", 16, ®ex); + if (s) + result |= 32; + + /* REG_STARTEND was added to glibc on 2004-01-15. + Reject older versions. */ + if (! REG_STARTEND) + result |= 64; + +#if 0 + /* It would be nice to reject hosts whose regoff_t values are too + narrow (including glibc on hosts with 64-bit ptrdiff_t and + 32-bit int), but we should wait until glibc implements this + feature. Otherwise, support for equivalence classes and + multibyte collation symbols would always be broken except + when compiling --without-included-regex. */ + if (sizeof (regoff_t) < sizeof (ptrdiff_t) + || sizeof (regoff_t) < sizeof (ssize_t)) + result |= 64; +#endif + + return result; + ]])], + [gl_cv_func_re_compile_pattern_working=yes], + [gl_cv_func_re_compile_pattern_working=no], + dnl When crosscompiling, assume it is not working. + [gl_cv_func_re_compile_pattern_working=no])]) + case $gl_cv_func_re_compile_pattern_working in #( + yes) ac_use_included_regex=no;; #( + no) ac_use_included_regex=yes;; + esac + ;; + *) AC_MSG_ERROR([Invalid value for --with-included-regex: $with_included_regex]) + ;; + esac + + if test $ac_use_included_regex = yes; then + AC_DEFINE([_REGEX_INCLUDE_LIMITS_H], [1], + [Define if you want to include , so that it + consistently overrides 's RE_DUP_MAX.]) + AC_DEFINE([_REGEX_LARGE_OFFSETS], [1], + [Define if you want regoff_t to be at least as wide POSIX requires.]) + AC_DEFINE([re_syntax_options], [rpl_re_syntax_options], + [Define to rpl_re_syntax_options if the replacement should be used.]) + AC_DEFINE([re_set_syntax], [rpl_re_set_syntax], + [Define to rpl_re_set_syntax if the replacement should be used.]) + AC_DEFINE([re_compile_pattern], [rpl_re_compile_pattern], + [Define to rpl_re_compile_pattern if the replacement should be used.]) + AC_DEFINE([re_compile_fastmap], [rpl_re_compile_fastmap], + [Define to rpl_re_compile_fastmap if the replacement should be used.]) + AC_DEFINE([re_search], [rpl_re_search], + [Define to rpl_re_search if the replacement should be used.]) + AC_DEFINE([re_search_2], [rpl_re_search_2], + [Define to rpl_re_search_2 if the replacement should be used.]) + AC_DEFINE([re_match], [rpl_re_match], + [Define to rpl_re_match if the replacement should be used.]) + AC_DEFINE([re_match_2], [rpl_re_match_2], + [Define to rpl_re_match_2 if the replacement should be used.]) + AC_DEFINE([re_set_registers], [rpl_re_set_registers], + [Define to rpl_re_set_registers if the replacement should be used.]) + AC_DEFINE([re_comp], [rpl_re_comp], + [Define to rpl_re_comp if the replacement should be used.]) + AC_DEFINE([re_exec], [rpl_re_exec], + [Define to rpl_re_exec if the replacement should be used.]) + AC_DEFINE([regcomp], [rpl_regcomp], + [Define to rpl_regcomp if the replacement should be used.]) + AC_DEFINE([regexec], [rpl_regexec], + [Define to rpl_regexec if the replacement should be used.]) + AC_DEFINE([regerror], [rpl_regerror], + [Define to rpl_regerror if the replacement should be used.]) + AC_DEFINE([regfree], [rpl_regfree], + [Define to rpl_regfree if the replacement should be used.]) fi -else - AC_MSG_RESULT(rx) - AC_CHECK_FUNC(re_rx_search, , [LIBOBJS="$LIBOBJS rx.o"]) -fi -AC_SUBST(LIBOBJS)dnl +]) + +# Prerequisites of lib/regex.c and lib/regex_internal.c. +AC_DEFUN([gl_PREREQ_REGEX], +[ + AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS]) + AC_REQUIRE([AC_C_INLINE]) + AC_REQUIRE([AC_C_RESTRICT]) + AC_REQUIRE([AC_TYPE_MBSTATE_T]) + AC_REQUIRE([gl_EEMALLOC]) + AC_REQUIRE([gl_GLIBC21]) + AC_CHECK_HEADERS([libintl.h]) + AC_CHECK_FUNCS_ONCE([isblank iswctype]) + AC_CHECK_DECLS([isblank], [], [], [[#include ]]) ])