Re: $ac_aux_dir issue (Was on automake@gnu.org: Re: amtraces)

2001-02-03 Thread Alexandre Oliva
On Feb 3, 2001, Alexandre Duret-Lutz <[EMAIL PROTECTED]> wrote: > OTOH, autoconf defines @INSTALL@ using $ac_aux_dir, and $(INSTALL) is > usable anywhere. That's because this variable has a special treatment > in config.status (it's relative path is adjusted with regard to the > depth of the ge

$ac_aux_dir issue (Was on automake@gnu.org: Re: amtraces)

2001-02-03 Thread Alexandre Duret-Lutz
>>> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes: Akim> "Derek R. Price" <[EMAIL PROTECTED]> writes: [...] >> +AC_CONFIG_AUX_DIR => sub { @config_aux_path = $_[2] }, Akim> This macro gives too many problem. Alexandre D. knows what I'm Akim> referring to, I'd like him to start a

Re: autoconf 2.49c fails if '.' is in PATH

2001-02-03 Thread Alexandre Oliva
On Feb 3, 2001, Akim Demaille <[EMAIL PROTECTED]> wrote: > The question is `is $FILE an executable in the common sense'. I think the best thing to do is to just ignore the issue of whether the found executable is a directory while testing -x or -f, and test for -d later on, notifying the user a

Re: autoconf 2.49c fails if '.' is in PATH

2001-02-03 Thread Akim Demaille
This is becoming way too confusing. I'd like to ask a single answer from each person who feel s/he knows well the architecture concerned. Don't try to answer for all the architectures, just yours. Try to avoid as much as possible using $exeext, as we need to find the compiler before trying to c

Re: `configure --help' overwrites existing config.log

2001-02-03 Thread Akim Demaille
"Lars J. Aas" <[EMAIL PROTECTED]> writes: > As long as the log stays the same it's great (but it looks to me like > moving the LOG_FD initialization will cause messages to be dropped from > the log). Are you referring to the logging of options errors? I agree it's good to have them. But I also