Pavel Roskin <[EMAIL PROTECTED]> writes:
> Done.
Great, thanks!
> The fix appears to be quite simple. The testsuite passes. Some real
> package using noinst_SCRIPTS is now in a good shape (i.e. `make install'
> works as it should and even `make distcheck' passes).
If you didn't apply it, I wil
Akim Demaille <[EMAIL PROTECTED]> writes:
> (&require_file_internal): Don't warn about installed files, just
> output on STDERR.
I'm fine with making this a warning again some day, but then it should
be something like --warning=installing or something, set by default,
but we need to
Pavel Roskin <[EMAIL PROTECTED]> writes:
> .PHONY uninstall-am: uninstall-%DIR%SCRIPTS
I fixed this typo (missing ?INSTALL?).
> -uninstall-%DIR%SCRIPTS:
> - @$(NORMAL_UNINSTALL)
> - @list='$(%DIR%_SCRIPTS)'; for p in $$list; do \
> - f="`echo $$p|sed '$(transform)'`"; \
> -
On 6 Mar 2001, Akim Demaille wrote:
> Pavel Roskin <[EMAIL PROTECTED]> writes:
>
> > .PHONY uninstall-am: uninstall-%DIR%SCRIPTS
>
> I fixed this typo (missing ?INSTALL?).
Thanks! I've removed noinstdir.test from XFAIL_TESTS.
Regards,
Pavel Roskin
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> `make distcheck' is working again. No more questions from `rm'.
Thanks.
BTW this is the sort of thing that we can check for in
`maintainer-check'. That way we can easily find out if we've
accidentally written a new test that will
Hello, Tom!
> Thanks.
> BTW this is the sort of thing that we can check for in
> `maintainer-check'. That way we can easily find out if we've
> accidentally written a new test that will break.
Libtool has a test (sh.test) that checks for suspicious constructs in
ltmain.sh. But it would be harde
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> Libtool has a test (sh.test) that checks for suspicious
Pavel> constructs in ltmain.sh. But it would be harder to process *.am
Pavel> and automake.in.
We just do it in an ad hoc way in maintainer-check.
Pavel> Speaking of "rm" wit
Akim> from Akim Demaille <[EMAIL PROTECTED]>
Akim> * automake.in (&push_dist_common): Simplify.
Ok.
Tom
>> (&require_file_internal): Don't warn about installed files, just
>> output on STDERR.
Akim> I'm fine with making this a warning again some day, but then it
Akim> should be something like --warning=installing or something, set
Akim> by default, but we need to be able to switch it off when runni
> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
Akim> * automake.in: Backquote is inert in double quotes.
Two problems with this patch:
* You didn't write a ChangeLog entry for your (seemingly unrelated)
Makefile.am change
* You didn't update aclocal.in as requested by the commen
Tom Tromey <[EMAIL PROTECTED]> writes:
> >> (&require_file_internal): Don't warn about installed files, just
> >> output on STDERR.
>
> Akim> I'm fine with making this a warning again some day, but then it
> Akim> should be something like --warning=installing or something, set
> Akim> by default
> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
Akim> FYI, here is the result on the fileutils:
Akim> @@ -307,8 +307,8 @@
Akim>END { for (i in files) print i; }'`; \
Akim> test -z "$(ETAGS_ARGS)config.h.in$$unique$(LISP)$$tags" \
Akim> || etags $(ETAGS
>> Sometimes this error is really an error. That is why we save and
>> optionally restore the exit status around the message printing. Let's
>> make this work correctly instead of breaking it.
Akim> I believe my patch is correct, note the $exit_status = 1.
I was just responding to your comment
I am fairly new to automake/autoconf/etc. so perhaps this question is newbie:
I have searched many docs to no avail...
What is the preferred source directory layout for projects using
automake+autoconf+libtool
when maintaining multiple platform-specific implementations of the same interfaces?
Tom Tromey <[EMAIL PROTECTED]> writes:
> > "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
>
> Akim> * automake.in: Backquote is inert in double quotes.
>
> Two problems with this patch:
>
> * You didn't write a ChangeLog entry for your (seemingly unrelated)
> Makefile.am cha
Tom Tromey <[EMAIL PROTECTED]> writes:
> > "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
>
> Akim> FYI, here is the result on the fileutils:
>
> Akim> @@ -307,8 +307,8 @@
> Akim>END { for (i in files) print i; }'`; \
> Akim> test -z "$(ETAGS_ARGS)config.h.in$$u
Tom Tromey <[EMAIL PROTECTED]> writes:
> >> Sometimes this error is really an error. That is why we save and
> >> optionally restore the exit status around the message printing. Let's
> >> make this work correctly instead of breaking it.
>
> Akim> I believe my patch is correct, note the $exit_
Tom Tromey <[EMAIL PROTECTED]> writes:
> > "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
>
> Akim> What is the general policy wrt `optimizations' in automake vs
> Akim> leaving some job to make?
>
> Let make do it.
>
> Seriously, I've never done any performance measurements to see ho
> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
>> * You didn't update aclocal.in as requested by the comment near
>> %obsolete_macros
Akim> Didn't notice it. That's just for style, or it introduces a
Akim> functional difference?
I think of it as being important for maintenance. I d
I'm sure other people on this list will tell how this can or cannot
be achieved in an autotools-specific way, but you could
get what you want the same way you get your 'generic' sources to
behave. In your 'foo.c' example, you could have a foo.c in your
'common' directory, which simply forwards to
> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
Akim> I just meant to have only
Akim> push (@files, '$(top_srcdir)/acconfig.h');
I guess that's fine. If there are problems we'll hear about it.
Akim> And anyway, when we use $(top_srcdir), we don't depend upon
Akim> VPATH at al
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> The fix appears to be quite simple. The testsuite passes. Some
Pavel> real package using noinst_SCRIPTS is now in a good shape
Pavel> (i.e. `make install' works as it should and even `make
Pavel> distcheck' passes).
Thanks for fixi
> "Tim" == Tim Van Holder <[EMAIL PROTECTED]> writes:
Tim> Alternatively, you could have configure copy /foo.c to
Tim> ./foo.c based on the configured host platform.
FYI, the standard way to do this is to use AC_LINK_FILES.
Tom
> "Kevin" == Kevin Walsh <[EMAIL PROTECTED]> writes:
Kevin> What is the preferred source directory layout for projects
Kevin> using automake+autoconf+libtool when maintaining multiple
Kevin> platform-specific implementations of the same interfaces?
There isn't a single standard, but there ar
Akim> * automake.in (&handle_merge_targets): Ventilate its non `all'
Akim> related content into...
Akim> (&generate_makefile): here.
Akim> (&handle_merge_targets): Rename as...
Akim> (&handle_all): this.
Akim> Remove a useless `if': `@all' is obviously not empty, since it
Akim> conta
> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
Akim> Tom, I'd like to know what you'd prefer: fixing the uses of $2
Akim> etc., or the definition of the groups in the regexps?
Whichever is clearest in the given situation.
Tom
Akim> * automake.in (&handle_all): Use an array instead of a scalar for
Akim> local_headers.
Akim> Get rid of all-redirect: let `all' be that target.
Ok.
Akim> +# FIXME: I believe Tom wants to remove this --akim.
Akim> +push @local_headers, '$(BUILT_SOURCES)'
Akim> + if &varia
27 matches
Mail list logo
