From: Michael Widenius Date: September 11 2001 2:03pm Subject: Re: two identical header files? List-Archive: http://lists.mysql.com/internals/1754 Message-Id: <15262.6593.426898.640418@tik.mysql.fi> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi! >>>>> "Dana" == Dana Powers writes: Dana> I believe this is a product of running 'autoheaders'. The source repository Dana> does not contain duplicates, but the source distribution ( .tar.gz ) is Dana> created Dana> after running 'autolocal;autoheaders;autolocal;autoconf;automake;' - so any Dana> files that get created here you'll get in the tar ball ( like the configure Dana> script, or all the Makefile.in files ). The above is one part of the story. Dana> From: "Dmitry Mityugov" >> > mySQL 3.23.41. Files global.h and my_global.h in the include directory Dana> are >> > identical. Could somebody explain me why you need both? The reason for the above is that we did at one point needed to make global.h available in the exported include directory and we got a name conflict with 'global.h'. Becasue of this we export global.h as my_global.h. In MySQL 4.0 we will change all source files to use my_global.h directly, after which we will not need global.h anymore. >> Another example: zlib/zlib.h and include/zlib.h. The file in include/zlib.h is there to not have to add the path to zlib when compiling the client directories. I didn't want to remove zlib/zlib.h because I wanted to keep this directory similar to the original zlib distribution. Regards, Monty