"Paul Smith" <[EMAIL PROTECTED]> writes: > %% Dan Espen <[EMAIL PROTECTED]> writes: > > de> Guess I've been fooled by configure again. > > Oh that rude autoconf, behaving exactly as described in its manual yet > again!! How tiresome!
> :) I'm going to guess you had something to do with the manual. :) I just read the description at: http://www.gnu.org/manual/autoconf/html_mono/autoconf.html and I understand it now, but I'm sure it didn't sink in the first time. Unlike the gnumake manual, (which is great), there's no example. > In this case you actually got fooled by GCC, which has this > super-advanced extension that made what is unquestionably a syntax error > in any standard compiler, only a warning in GCC. I didn't know about GCC, but my bigger problem was I thought the code was coming from autoconf. When I went looking for the code in configure.in I was searching for the f77 part of the code. Since its not in stuff generated by autoconf 2.13, I guessed the whole test was coming from autoconf. I know, my understanding of autoconf macros is abysmal, but I think I'm improving. > The moral of the story > is that when it comes to portability, autoconf is no panacea and no > substitute for actually trying it on lots of different systems and with > lots of different compilers. At least two of each, anyway :). > > I know that can be difficult these days where GCC is so prevalent and > most anything else costs $$... I think different versions of autoconf is an issue too. 2.57 is complaining about our input files, but it seems to work. I just updated the notes I keep about preparing releases to tell myself to do a compile with Sun's compiler before release. Seems like a good idea with gcc having weird extensions. Thanks. -- Dan Espen E-mail: [EMAIL PROTECTED] -- Visit the official FVWM web page at <URL: http://www.fvwm.org/>. To unsubscribe from the list, send "unsubscribe fvwm" in the body of a message to [EMAIL PROTECTED] To report problems, send mail to [EMAIL PROTECTED]
