Is it really such a hard work to parse PKGBUILD and simulate a bash behaviour only on variables preceding the build() function ?<br>The parser needs to be able to :<br>- affect and replace variables<br>- support arrays<br>
- support the simple ${x//find/replace} syntax<br>- what else ?<br><br>If this can be tedious with PHP, is it that difficult to realize in python ?<br><br>Cilyan<br><br><div class="gmail_quote">2008/6/10 Loui &lt;<a href="mailto:louipc.ist@gmail.com">louipc.ist@gmail.com</a>&gt;:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">On Tue, 10 Jun 2008 15:16:29 +0800<br>
<div class="Ih2E3d">&quot;Callan Barrett&quot; &lt;<a href="mailto:wizzomafizzo@gmail.com">wizzomafizzo@gmail.com</a>&gt; wrote:<br>
<br>
</div><div class="Ih2E3d">&gt; Note to self: restricted mode is apparently not as restricted as it<br>
&gt; sounds, need to look into that and probably how namcap deals with it.<br>
&gt;<br>
&gt; Possibly a chroot?<br>
&gt;<br>
&gt; Also I&#39;m assuming everyone is fine and dandy with this behavior so<br>
&gt; when it&#39;s not terribly insecure anymore I&#39;ll apply this.<br>
<br>
</div>Yeah I&#39;m concerned about how secure it is to actually source the<br>
PKGBUILD in bash. I don&#39;t know if AUR really needs to cover every<br>
exotic PKGBUILD out there. Having to set up a chroot seems like a bit<br>
much. It would be nice to have all variables resolved though, and bash<br>
could make it a lot simpler.<br>
<br>
</blockquote></div><br>