Mailinglist Archive: yast-devel (177 mails)

< Previous Next >
Re: [yast-devel] Direction of YaST Architecture?
  • From: "Duncan Mac-Vicar P." <dmacvicar@xxxxxxx>
  • Date: Wed, 02 Feb 2011 10:36:18 +0100
  • Message-id: <4D492592.8010408@suse.de>
On 01/31/2011 03:56 PM, Jiri Srain wrote:
> In the past, we have made attempts with various languages. We have enhanced
the infrastructure so that it should allow a python-only module (take Python
as an example here), we have made some prototypes, but from some reason I
cannot remember a single non-YCP module which was finished. Maybe we should
collect the issues found during these experiments, there can be easy solutions
of them.

You mean the attempts using the ycp language bindings?

I think if you want to provide a real replacement a module needs to be done and iterated until the framework below matures to a point where it is more useful.

e.g.: ruby bindings never supported UI except for a very weird way: there was no support for UI terms in ruby, and the real way to do it, using blocks and DSLs was never finished. There was no support for translations, etc etc etc (the rest of the 20%).

Also, YaST has strong conventions due to having lot of developers in the same room coding ycp for years. If a new way of doing a module exists, it needs to get conventions and good practices, as a language like python or ruby allow for more creativity when it comes to do weird things, which is cool in the exception case, but not as the default.

--
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 >
List Navigation
This Thread