Mailinglist Archive: yast-devel (17 mails)

< Previous Next >
Re: [yast-devel] ini-agent vs Augeas, Re: [yast-commit] r62714
  • From: "Duncan Mac-Vicar P." <dmacvicar@xxxxxxx>
  • Date: Fri, 12 Nov 2010 10:08:25 +0100
  • Message-id: <4CDD0409.4090605@xxxxxxx>
On 11/09/2010 04:57 PM, Martin Vidner wrote:
On Tue, Nov 09, 2010 at 02:51:21PM +0100, Martin Kudlvasr wrote:
I read the "Creating a lens" tutorial and I realize that augeas is an
actual parsing tool (and not only a heap of regexps). Using it will
actually improve yast2. If zypper already has a dependency on Augeas,
that's all the more reason for doing this change.

Anyone interested in playing with it?

I'm all in. Can you suggest a place in yast2, where I can start?

One thing is enabling the use of Augeas. I think we should use the
existing Perl or Ruby bindings (we don't have RPMs yet) and write
some helper/wrapper for the YCP modules.

Also we can look at the current agent usage:
grep -h '^`ag_' /usr/share/YaST2/scrconf/*.scr | sort | uniq -c
(spoiler alert: Ini is the most popular, with Any the second)
and decide where the low hanging fruits are for replacement.

Is the current INI agent is plain C++ right now?

I saw more naturally a design where the basic augeas agent is built as a C++ agent linking to libaugeas, and then the derived agents are specified as an .scr referencing ag_augeas and adding additional information like inlining the lense description.

Duncan Mac-Vicar P. - Novell® Making IT Work As One™
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nuernberg)

To unsubscribe, e-mail: yast-devel+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: yast-devel+help@xxxxxxxxxxxx

< Previous Next >