> * Eric Blake <roy...@erqung.pbz> [2011-05-03 16:03:16 -0600]: > > On 05/03/2011 03:52 PM, Sam Steingold wrote: >> when I regenerate config.h with the current gnulib, I get: >> >> @@ -871,12 +979,6 @@ >> /* `struct sockaddr_un' from <sys/un.h> has a `sun_len' field */ >> #undef HAVE_SOCKADDR_UN_LEN >> >> -/* Define to 1 if stdbool.h conforms to C99. */ >> -#undef HAVE_STDBOOL_H >> - >> -/* Define to 1 if you have the <stddef.h> header file. */ >> -#undef HAVE_STDDEF_H >> - >> /* Define to 1 if you have the <stdint.h> header file. */ >> #undef HAVE_STDINT_H >> >> >> what happenned? are we assuming that all platforms now have these files? > > <stddef.h> is required by C89, so yes, it was a redundant declaration. > > <stdbool.h> is not required by C89. However, nothing else in gnulib > used the results of the test, and the idea is that if you are using > gnulib's stdbool module, you don't care about a fully > standards-compliant <stdbool.h>, rather you care about the subset of > <stdbool.h> guaranteed by gnulib. So we trimmed the size of config.h > for the sake of emacs: > http://lists.gnu.org/archive/html/bug-gnulib/2011-01/msg00571.html
So, how do I know if gnulib created the file or not? (if you must know, in clisp we create a header clisp.h for external modules to include and it either includes the generated "stdbool.h" literally or does "#include <stdbool.h>"; we used HAVE_STDBOOL_H to decide what to do). Thanks. -- Sam Steingold (http://sds.podval.org/) on CentOS release 5.6 (Final) X 11.0.60900031 http://truepeace.org http://pmw.org.il http://iris.org.il http://jihadwatch.org http://www.PetitionOnline.com/tap12009/ http://openvotingconsortium.org A computer scientist is someone who fixes things that aren't broken.
