On Oct 31, 2007, at 5:05 PM, Ralf Wildenhues wrote:
Hello Benoit,Thanks for the patch. * Benoit Sigoure wrote on Wed, Oct 31, 2007 at 04:43:27PM CET:+2007-10-31 Benoit Sigoure <[EMAIL PROTECTED]> + + Adjust the documentation of autotest WRT atlocal. + * doc/autoconf.texi (Making testsuite Scripts): It is not necessary + to write a rule to produce atlocal since it's an AC_CONFIG_FILES. + Mention that atlocal.in needs to be distributed, not atconfig.in.[...]--- a/doc/autoconf.texi +++ b/doc/autoconf.texi @@ -20244,15 +20244,10 @@ subdir = tests atconfig: $(top_builddir)/config.status cd $(top_builddir) && \ $(SHELL) ./config.status $(subdir)/$@@ - -atlocal: $(srcdir)/atlocal.in $(top_builddir)/config.status - cd $(top_builddir) && \ - $(SHELL) ./config.status $(subdir)/$@@ @end exampleThis paragraph starts with | With strict Autoconf, you might need to add lines inspired from the | following: meaning that you would need those lines only when not using Automake. Only Automake adds the atlocal rebuilding rule for you, given AC_CONFIG_FILES(atlocal...).@noindent -and manage to have @file{atconfig.in} and @code{$(EXTRA_DIST)} -distributed. +and manage to have @file{atlocal.in} distributed.Again, without using Automake, the developer must ensure that the stufflisted further up in EXTRA_DIST is being distributed. Is the section too difficult to understand? If yes, how can it be improved?
OK I misunderstood what "strict" autoconf meant (actually I wasn't quite sure about what it was :D), maybe something along the lines of "if you don't use automake" or "without automake" would've been clearer. Anyways, the section is still unclear because even if it's true that this must be done when one doesn't use automake, it's also true that *with* automake you must add the rule to refresh `atconfig' anyways.
I'll post another patch. -- Benoit Sigoure aka Tsuna EPITA Research and Development Laboratory
PGP.sig
Description: This is a digitally signed message part
