Hello Paul, > Debian string.h #defines some of the symbols that string.h therefore > attempts to #define to foo_is_unportable, and the #defines collide.
Oh well. Thanks for the fix. > A program that uses mountlist shouldn't > need to use the gnulib strstr if it doesn't want to. I think it's time for me to report a glibc bug on strstr and strcasestr, then... so that gnulib's strstr.c be no longer needed on glibc systems. > But with gnulib string_.h as it is, it will have to use gnulib strstr. Or it can use gnulib's c_strstr. Or it can define GNULIB_STRSTR itself: dnl Tell gnulib that the non-i18n implementation of strstr() is enough dnl for our purposes. AC_DEFINE([GNULIB_STRSTR], 1) > How about if we remove the > strstr_is_unportable__use_gnulib_module_strcasestr_for_portability > symbols? They serve the purpose of detecting portability problems. I think it would be actually good to introduce a module called, say, 'posixcheck' (or a gnulib-tool option) that would generalize this technique from string.h to all other headers that declare POSIX functions for which gnulib has a replacement. Bruno
