Hi, It was pointed out to me that the re_compile_pattern check was failing for some reason. On investigation, it turns out to be a broken test.
Tim. */ --- coreutils-4.5.3/aclocal.m4.regex 2003-01-16 14:29:30.000000000 +0000 +++ coreutils-4.5.3/aclocal.m4 2003-01-16 14:32:20.000000000 +0000 @@ -1214,6 +1214,7 @@ jm_cv_func_working_re_compile_pattern, AC_TRY_RUN( [#include <stdio.h> +#include <string.h> #include <regex.h> int main () @@ -1222,12 +1223,14 @@ const char *s; struct re_registers regs; re_set_syntax (RE_SYNTAX_POSIX_EGREP); + memset (®ex, 0, sizeof (regex)); [s = re_compile_pattern ("a[[:@:>@:]]b\n", 9, ®ex);] /* This should fail with _Invalid character class name_ error. */ if (!s) exit (1); /* This should succeed, but doesn't for e.g. glibc-2.1.3. */ + memset (®ex, 0, sizeof (regex)); s = re_compile_pattern ("{1", 2, ®ex); if (s) @@ -1235,6 +1238,7 @@ /* The following example is derived from a problem report against gawk from Jorge Stolfi <[EMAIL PROTECTED]>. */ + memset (®ex, 0, sizeof (regex)); s = re_compile_pattern ("[[an�]]*n", 7, ®ex); if (s) exit (1); _______________________________________________ Bug-coreutils mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-coreutils