blog

commit: ab79f131ed8e0624e712747d9e9291bc71a539b5
parent 5353e204adc5144c5409eb725599e6e1f33c38ab
Author: Haelwenn (lanodan) Monnier <contact@hacktivis.me>
Date:   Mon, 11 Feb 2019 02:08:09 +0100

articles/My git server setup: link to hook code repository, dumb http can be done with post-receive too

Diffstat:
M
articles/My git server setup.html
29
++++-------------------------

1 file changed, 4 insertions(+), 25 deletions(-)
diff --git a/articles/My git server setup.html b/articles/My git server setup.html
@@ -1,7 +1,7 @@
 <h1 class="p-name"><a class="u-url" href="/articles/My%20git%20server%20setup">My git server setup</a></h1>
 <p>So after having problems with gitlab.com (not being able to push to your own repository for example). I decided to have a very simple git setup, inspired by git.linkmauve.fr. I putted all my git repos into <code>/git</code>.</p>
 <h2>started with just nginx and ssh</h2>
-<p>This one is dead-simple when you know the trick, simply put <code>git update-server-info</code> into hooks/post-update of your git repo (they have to be bare repo, that’s done with --bare)</p>
+<p>This one is dead-simple when you know the trick, simply put <code>git update-server-info</code> into <code>hooks/post-update</code> or <code>hooks/post-receive</code> of your git repo (they have to be bare repo, that’s done with --bare)</p>
 <h2>Added git-daemon</h2>
 <p>That one was even more simple, just had to point to where the git repositories are stored</p>
 <h2>Wanted a better interface</h2>
@@ -9,30 +9,9 @@
 <pre><code>repo=$(pwd)
 cd "$(pwd | sed s/.git$//)" &amp;&amp; stagit -c "$repo.cache" "$repo"
 cd /git &amp;&amp; stagit-index *.git &gt; /git/index.html</code></pre></p>
-<h2>Deployement to blog</h2>
-<p>This one output an error(probably because the GIT_INDEX_FILE should not be empty) but it works so who cares :P</p>
-<code>GIT_INDEX_FILE='' git --work-tree=/srv/web/hacktivis.me --git-dir=/git/blog.git checkout -f</code>
 <h2>Garbage Collector</h2>
 <p>It’s not like my repos were getting big, git is supposed to do that itself but it seems like it doesn’t, so I’m doing <code>git gc</code> each time, which isn’t very optimised.</p>
-<h2>Final post-update hook</h2>
-<pre><code>
-#!/bin/sh
-repo=$(pwd)
-update_f=' * Updating %s…'
-is_ok(){ echo ' [OK]'; }
-
-printf "$update_f" 'Garbage Collector'
-git gc &amp;&amp; is_ok
-
-printf "$update_f" info
-git update-server-info &amp;&amp; is_ok
-
-printf "$update_f" stagit
-cd "$(pwd | sed s/.git$//)" &amp;&amp; stagit -c "$repo.cache" "$repo" &amp;&amp; is_ok
-
-printf "$update_f" stagit-index
-cd /git &amp;&amp; stagit-index *.git &gt; /git/index.html &amp;&amp; is_ok
-
-# Errors but works
-grep blog &lt;&lt;&lt;$repo &amp;&amp; printf "$update_f" blog &amp;&amp; GIT_INDEX_FILE='' git --work-tree=/srv/web/hacktivis.me --git-dir=/git/blog.git checkout -f &amp;&amp; is_ok
+<h2>Final Hook code (deploy, stagit, …)</h2>
+<p>This can be seen in my <a href="/git/utils">/git/utils</a> repo, in the <code>git-hooks</code> folder.</p>
+<p>Also the blog is a symlink to <code>/git/blog.work</code> which are the raw files done by that hook.</p>
 </code></pre>