automake-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: bug#8168: macros directory not created automatically


From: Stefano Lattarini
Subject: Re: bug#8168: macros directory not created automatically
Date: Wed, 30 Mar 2011 10:37:15 +0200
User-agent: KMail/1.13.3 (Linux/2.6.30-2-686; KDE/4.4.4; i686; ; )

On Monday 14 March 2011, Stefano Lattarini wrote:
> [dropping automake list, adding automake-patches]
> 
> References:
>   <http://lists.gnu.org/archive/html/automake/2011-03/msg00000.html>
>   <http://lists.gnu.org/archive/html/bug-automake/2011-03/msg00004.html>
> 
> -*-*-
> 
> Hello Javier and all automakes, and sorry for the delay.
> 
> On Thursday 03 March 2011, Stefano Lattarini wrote:
> > 
> > On Thursday 03 March 2011, Javier Jardón wrote:
> > > Hi,
> > >
> > Hi Javier, and thanks for the report.
> > 
> > > I'm using
> > > 
> > > ACLOCAL_AMFLAGS = -I m4 ${ACLOCAL_FLAGS}
> > > 
> > > in my main Makefile.am, but I have to add the m4 directory to the git
> > > tree, or create it in my autogen script (and this doesnt work if I do
> > > not use a autogen file and only autoreconf instead)
> > > 
> > > Shouldnt be automake smart enough to create that directory automatically?
> > >
> > Probably yes -- but IMHO it should then also be smart enough to create
> > the directory *only* when it makes sense, i.e. only when the `--install'
> > flag is used too.
> > 
> > Also, it might be useful if aclocal would not complain (or at least would
> > not error out) in case it is passed a non-existent or unreadable directory
> > through the `-I' option.  This "laxer" behaviour is implemented in many
> > other gnu tools already (e.g., make, m4. gcc).
> > 
> > I might take a look at this issues in the next(ish) days.
> > 
> > Thanks,
> >   Stefano
> > 
> Attached is a patch series (2 patches) that should address the issue.
> The series is based off of maint -- as I'm not yet sure whether it
> would be better to merge it to master only, or to maint too.
> 
> Javier, are you ok with your name being added to THANKS?
> 
> Ralf, OK to apply?  If yes, where (maint, or master only)?
> 
> Regards,
>   Stefano
> 
> -*-*-
> 
> PATCH 1/2] aclocal: `-I' does not bail out on invalid directories.
> 
> Related to automake bug#8168.
> 
> * aclocal.in (scan_m4_dirs): If a user-specified "include
> directory" is unreadable or non-existent, do not issue a
> fatal error anymore, but simply issue a warning, and only
> when running in verbose mode.
> * NEWS: Update.
> * tests/aclocal-bad-dirlist-include-dir.test: New test.
> * tests/aclocal-bad-local-include-dir.test: Likewise.
> * tests/aclocal-bad-system-include-dir.test: Likewise.
> * tests/Makefile.am (TESTS): Update.
> * tests/.gitignore: Update.
> ---
>  ChangeLog                                  |   15 +++++
>  NEWS                                       |    5 ++
>  aclocal.in                                 |   10 +++-
>  tests/.gitignore                           |    4 +-
>  tests/Makefile.am                          |    3 +
>  tests/Makefile.in                          |    3 +
>  tests/aclocal-bad-dirlist-include-dir.test |   36 +++++++++++
>  tests/aclocal-bad-local-include-dir.test   |   90 
> ++++++++++++++++++++++++++++
>  tests/aclocal-bad-system-include-dir.test  |   36 +++++++++++
>  9 files changed, 199 insertions(+), 3 deletions(-)
>  create mode 100755 tests/aclocal-bad-dirlist-include-dir.test
>  create mode 100755 tests/aclocal-bad-local-include-dir.test
>  create mode 100755 tests/aclocal-bad-system-include-dir.test
> 
> -*-*-
> 
> [PATCH 2/2] aclocal: create local directory where to install m4 files
> 
> Before this change, a call like `aclocal -I m4 --install' would
> fail if the `m4' directory wasn't pre-existing.  This could be
> particularly annoying when running in a checked-out version
> from a VCS like git, which doesn't allow empty directories to
> be tracked.
> 
> Closes automake bug#8168.
> 
> * aclocal.in (install_file): Change signature: take as second
> argument the destination directory rather than the destination
> file.  Crate the destination directory if it doesn't already
> exist.  In verbose mode, tell what is being copied where.
> (write_aclocal: Update.
> * NEWS: Update.
> * THANKS: Update.
> * tests/aclocal-install-fail.test: New test.
> * tests/aclocal-install-mkdir.test: Likewise.
> * tests/aclocal-no-install-no-mkdir.test: Likewise.
> * tests/aclocal-verbose-install.test: Likewise.
> * tests/Makefile.am (TESTS): Update.
> 
> From a report by Javier Jardón.
> ---
>  ChangeLog                              |   23 ++++++++++++
>  NEWS                                   |    4 ++
>  THANKS                                 |    1 +
>  aclocal.in                             |   13 ++++---
>  tests/Makefile.am                      |    4 ++
>  tests/Makefile.in                      |    4 ++
>  tests/aclocal-install-fail.test        |   52 ++++++++++++++++++++++++++
>  tests/aclocal-install-mkdir.test       |   62 
> ++++++++++++++++++++++++++++++++
>  tests/aclocal-no-install-no-mkdir.test |   39 ++++++++++++++++++++
>  tests/aclocal-verbose-install.test     |   53 +++++++++++++++++++++++++++
>  10 files changed, 250 insertions(+), 5 deletions(-)
>  create mode 100755 tests/aclocal-install-fail.test
>  create mode 100755 tests/aclocal-install-mkdir.test
>  create mode 100755 tests/aclocal-no-install-no-mkdir.test
>  create mode 100755 tests/aclocal-verbose-install.test
> 

Ping on this?  I'll push in 72 hours if there is no objection.

Regards,
  Stefano



reply via email to

[Prev in Thread] Current Thread [Next in Thread]