Re: [Tracker] Tracker testing environment

[Ivan Frade <ivan frade gmail com>]

Hi,

 I usually work installing on /usr but your sandbox script is a great idea.

 A couple of comments on the script:

* I think it should also set the TRACKER_ONTOLOGIES_DIR variable
* It should also report if the test directory exists (to know if
tracker is writing a new DB or reusing some old version)

 I guess we assume that if you modify something in tracker, you
already know tracker-control and what processes are running and how.
It doesn't harm to repeat that information in some Hacking note.

 Thanks for the script, once it is tested a bit, we could add it in master.

 Regards,

Ivan

On Mon, Mar 12, 2012 at 6:14 PM, Sam Thursfield <ssssam gmail com> wrote:
> Hi everyone
>
> Is there a recommended method for doing Tracker development and
> testing it on a desktop where you also have a 'production' Tracker
> installation? I wrote a quick script this weekend that gives you a
> shell with a test environment where you can play with a build of
> Tracker installed in /opt (find it here:
> http://afuera.me.uk/junk/tracker-sandbox ), similar to how the
> functional tests work. I'm interested if all Tracker developers do
> something like this, or use a separate user account, or jhbuild, or a
> make install directly into /usr and hope for the best.
>
> https://live.gnome.org/Tracker/Documentation/Debugging doesn't have
> too much useful info, I was thinking of adding whatever comes out of
> this discussion up there and maybe into a HACKING / README in the git
> tree as well. I worry that if it's not trivial for me to work out the
> best way to test out changes without messing up my real tracker-store,
> it could be quite a big barrier to new contributors who'll find
> themselves in a sort of "make, make install, what now??" situation.
>
> Thanks
> Sam
> _______________________________________________
> tracker-list mailing list
> tracker-list gnome org
> http://mail.gnome.org/mailman/listinfo/tracker-list