Re: should an empty "pkgdata_DATA" cause creation of $(pkgdatadir) by "make install"?

2012-03-21 Thread Bruno Haible
Hi Stefano, > We finally decided that the older behaviour was a bug. We've thus fixed it > (in the automake maint branch, fix to appear in automake 1.11.4), Cool! Many thanks for your considerations and perseverance. > and I now propose the attached patch against gnulib-tool. OK to push? > +

Re: GSoC

2012-03-21 Thread Bruno Haible
Hello Vinay, > I was thinking of applying for the implementation of gnulib-tool in Python. It will be exciting to see you apply! > I've been coding in Python for the last few years. This is a good starting position for this project, indeed. > I wanted to know what the mentors expect out of ap

Re: submodules again

2012-03-21 Thread Jim Meyering
Eric Blake wrote: > On 03/21/2012 01:44 PM, Jim Meyering wrote: >> git-submodule's --reference option looks like it does what you want. >> However, balancing the repo duplication cost against the risk of error >> described in git-clone's --reference and --shared descriptions, so far, >> I have pref

Re: submodules again

2012-03-21 Thread Eric Blake
On 03/21/2012 01:44 PM, Jim Meyering wrote: > git-submodule's --reference option looks like it does what you want. > However, balancing the repo duplication cost against the risk of error > described in git-clone's --reference and --shared descriptions, so far, > I have preferred to use more disk s

Re: submodules again

2012-03-21 Thread Jim Meyering
Werner LEMBERG wrote: > Folks, > > the default operation (a) of `bootstrap' is clear: > > 1. The `gnulib' submodule URL is in file `.gitmodules'. > 2. No special submodule entry in `.git/config'. > 3. A (submodule) clone of `gnulib' is created. > 4. All files of `gnulib' are checked out. >

Re: should an empty "pkgdata_DATA" cause creation of $(pkgdatadir) by "make install"?

2012-03-21 Thread Stefano Lattarini
[dropping automake list] On 03/13/2012 01:30 PM, Stefano Lattarini wrote: > [CC:ing Ralf, as I'd like to hear his opinion here] > > Reference: > > > On 03/13/2012 01:14 PM, Stefano Lattarini wrote: >> >> Now that I think about

GSoC

2012-03-21 Thread vinay rao
Hi. I'm a student trying my chances at GSoC, and thought the GNU project was a very interesting opportunity. I was thinking of applying for the implementation of gnulib-tool in Python. I've been coding in Python for the last few years. I cloned the gnulib git and checked the source of gnulib-tool a

announce-gen: require perl-5.9.3

2012-03-21 Thread Jim Meyering
Eric Wong wrote: > How about using the generic Digest module instead? > > Something like this (totally untested) patch: > > diff --git a/build-aux/announce-gen b/build-aux/announce-gen > index 3ca90a9..c453d44 100755 > --- a/build-aux/announce-gen > +++ b/build-aux/announce-gen > @@ -29,8 +29,7 @@

Re: Revised announce-gen patch

2012-03-21 Thread Jim Meyering
Reuben Thomas wrote: > Implementing Jim's suggestions. > >>From 3e6c32b5834341ee8d1870f7b85db546b1a1f2d8 Mon Sep 17 00:00:00 2001 > From: Reuben Thomas > Date: Tue, 20 Mar 2012 20:17:22 + > Subject: [PATCH] announce-gen: make it use Digest::SHA where available. > > * build-aux/announce-gen: Us

submodules again

2012-03-21 Thread Werner LEMBERG
Folks, the default operation (a) of `bootstrap' is clear: 1. The `gnulib' submodule URL is in file `.gitmodules'. 2. No special submodule entry in `.git/config'. 3. A (submodule) clone of `gnulib' is created. 4. All files of `gnulib' are checked out. If I already have a local `gnulib'