← Back to team overview

gtg-contributors team mailing list archive

Re: Refactorisation progress : an indicible truth?

 

On Sat, Feb 27, 2010 at 9:36 PM, Bryce Harrington <bryce@xxxxxxxxxxxxx> wrote:
> On Fri, Feb 26, 2010 at 01:31:25PM +0100, Lionel Dricot wrote:
>> Le vendredi 26 f??vrier 2010 ?? 13:24 +0100, Bertrand Rousseau a ??crit :
>> > > 1) either they belong to the tagstore and, as such, should be created
>> > > with the tagstore and any view of the tagstore can use/see them.
>> > >
>> > > 2) they are part of the view and, thus, should not modify the tagstore
>> > > at all and be special things completely handled by the tagtreeview.
>> >
>> > My take is that those two pseudo tags are actually UI-objects that
>> > allow the user to display a certain list of tag, but which have no
>> > intrinsic meaning as a data structure.
>> >
>> > That was the reason they were added from the browser in the first
>> > place. I agree they are more specific than the browser, and that they
>> > feel out of place there. Actually I think we should have some sidebar
>> > object, and sidebar tabs object, to decouple the browser from other
>> > task-specific widget and allow to populate the sidebar with other tabs
>> > (much like gedit does). As such, it could be interesting to define
>> > some kind of tag-list "sidebar tab", and that would be where those
>> > pseudo would be handled. Note that it could well be that those pseudo
>> > are not added at all in the tagstore. We could just as well create an
>> > adhoc TreeModel with just those tags.
>>
>> That's indeed a good idea. Make it a UI-only object. But it is then
>> important to not add them to the tagstore at all. Also, we should remove
>> them completely from the requester.
>>
>> Having them only in the TagTreeModel makes sense for me.
>
> Nifty, will this enable plugins to add their own custom pseudo-tags?

I haven't thought about that, but yes, that could be a great idea. I
guess it could work with this, yes.

> Bryce
>
>



-- 
Bertrand Rousseau



References