<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN"> <HTML> <HEAD> <META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8"> <META NAME="GENERATOR" CONTENT="GtkHTML/3.26.0"> </HEAD> <BODY> On Wed, 2010-08-11 at 15:41 +0200, Arnaud Fontaine wrote: <BLOCKQUOTE TYPE=CITE> <TT>Would you consider applying the patch adding XORG_CHECK_GNU_M4 macro and</TT> <TT>also  the  version  bump?  Let  me  know  if  you  want  me  to  correct</TT> <TT>anything. Thanks much!</TT> </BLOCKQUOTE> I have tested the xcb proposed <TT>XORG_CHECK_GNU_M4</TT> and I found some problems mainly due to assumptions. <BLOCKQUOTE> <PRE> - When m4 is present without -I support (gm4 absent), result is correct,   but m4 is tested twice. - When m4 is absent and gm4 is present in $PATH (not /usr/gnu/bin)   without -I support, there is no error message when there should be one.   gm4 is tested twice. - If there is no m4 or gm4 present in $PATH, /usr/gnu/bin/gm4 will not be found - If a /usr/gnu/bin/gm4 is found, (because there is an m4 in $PATH), it is not   tested for -I support. - If a gnu processor is named m4 rather than gm4, it will not be found in /usr/gnu/bin. </PRE> </BLOCKQUOTE> I have rewritten my version of the patch and fully tested it. It does not make any assumption other than executable name being m4 or gm4. It should resist changes for a longer period. It would be nice if someone else could test it. Keep in mind this macro will be there for decades and may be used on platforms you may not expect (GNU/Hurd, embedded systems ...) Thanks </BODY> </HTML>