[pacman-dev] sourcing /etc/profile and set -e bug in makepkg
Xavier
shiningxc at gmail.com
Mon Aug 11 14:28:35 EDT 2008
On Mon, Aug 11, 2008 at 6:41 PM, Allan McRae <allan at archlinux.org> wrote:
>
> Wow, I am very, very impressed that you found this! It does seem
> bash-completion is broken but it does not appear to be developed upstream
> any longer. I wonder if there is a replacement project somewhere?
>
Well, it just took me 1 hour of adding debug statements everywhere :D
We don't even need to care about upstream, because the
/etc/profile.d/bash-completion.sh script in fault was made for Arch,
as far as I can tell.
So let's just fix that for now.
But on that note, I just noticed a very weird thing. The
bash-completion package is newer in the old cvs repo than in the new
svn repo.
http://repos.archlinux.org/viewvc.cgi/extra/system/bash-completion/?root=extra
http://repos.archlinux.org/viewvc.cgi/bash-completion/repos/extra-i686/
I know this is offtopic here, but since this was made by both Dan and
Aaron, maybe one of them could take care of it?
Otherwise, I am willing to take the time to write a bug tracker if necessary
About upstream, there was apparently an attempt of Debian to take it over :
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=472468
But the alioth page seems still empty, the code was not even imported :
http://alioth.debian.org/projects/bash-completion/
I also found this, but well, no comment :
http://bash-completion.alioth.debian.org/
I can't even find the 20080705 version than Debian is using anywhere
else than on the debian package page :
http://packages.debian.org/fr/sid/bash-completion
In the bug report above, the last comment mentions this new project :
http://fvue.nl/wiki/Bash_completion_lib
http://code.google.com/p/bash-completion-lib/
That could be a better replacement.
But anyway, as I said, this is not related to our problem.
> Despite locating the actual cause, I am still in favour of including the
> work-around in makepkg for robustness. Even though it was not makepkg's
> fault, it should not fail without any obvious cause like it currently is.
>
Hm yeah, that looks like a good idea indeed.
More information about the pacman-dev
mailing list