git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH] Add a git-completion rpm subpackage to the spec

From
Ben Walton <bwalton@artsci.utoronto.ca>
Date
Mar 21, 2010, 20:06 UTC
Message-ID
<1269199866-sup-8342@pinkfloyd.chass.utoronto.ca>
In-Reply-To
<20100321155156.GH27676@inocybe.localdomain>
Excerpts from Todd Zullinger's message of Sun Mar 21 11:51:56 -0400 2010:
Hi Todd,
Show 6 quoted lines
> The paths haven't changed currently, though I thought I read that
> eventually they'd like to move the default location for completion
> scripts to /usr/share/bash-completion/ (many are already installed
> there, but are symlinked to /etc/bash_completion.d by
> bash-completion packages (in the Fedora/EPEL packages via %triggers
> when the applications they complete are installed).
Should this be taken as:
* Eventually, /etc/bash_completion will source files directly from
  /usr/share/...
OR
* To play nice with the distros, triggers and symlinks are the way to
  go after installing to /usr/share/...?
> I also haven't made time to see what, if any, of the changes we've
> made in the Fedora packages might be worthwhile to integrate into
> the git.git rpm spec file. :/

I just took a look at the spec from 1.7.0.1-1 in rawhide. Some of the stuff it includes is Fedora/RHEL specific, but a few items that I wouldn't mind seeing merged into the stock .spec file are the splitting of the daemon package with the addition of an xinetd.d/git file and the emacs support subpackage.

I also like the gitweb sub-package, but it doesn't meet the (site specific) needs I have for deployment...In general, I think support in the standard .spec for it is a good thing though.

> simple to source /etc/bash_completion.d/git.  In that light, is
> there a compelling reason not to install the completion by default?

Well, that's a good point. I think my personal inclination for a separate package comes from having it separated in OpenCSW. Most of the reasons for a separate package there likely don't apply here though, so maybe it should be part of git-core? As I said when I sent the alternate (separate rpm) version, I'm happy with either solution as long as the completion support is more easily available when I build rpms in the future.

Thanks -Ben

-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302
Previous: Todd Zullinger
Message 8 of 8 in “RPM spec: optionally include bash completion support”
  1. RPM spec: optionally include bash completion supportIan Ward Comfort, Mar 20, 2010
  2. Add a git-completion rpm subpackage to the specBen Walton, Mar 20, 2010
  3. Junio C HamanoMar 20, 2010
  4. Ian Ward ComfortMar 20, 2010
  5. Ben WaltonMar 20, 2010
  6. Junio C HamanoMar 21, 2010
  7. Todd ZullingerMar 21, 2010
  8. Ben WaltonMar 21, 2010

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.