Dataengine to get bookmarks

[Eduardo Robles Elvira]

On Mon, Dec 7, 2009 at 12:44 PM, Sebastian Kügler <sebas at kde.org> wrote:
> On Sunday 06 December 2009 18:56:39 Jan Gerrit Marker wrote:
>> I committed a dataengine to playground. This dataengine should get the
>> bookmarks of various browsers like Firefox, Konqueror or Opera. At the
>>  moment  it only supports Firefox.
>>
>> As some of you may know I added Firefox support to the bookmark runner for
>>  KDE  SC 4.4. I noticed that it is easy to get the bookmarks of Konqueror
>>  but hard to get Firefox's. So I thought it would be nice to get this data
>>  easily. The dataengine uses the browsers as source. For every bookmark
>>  there are two entries: num-name and num-url where num is replaced with an
>>  integer value. Two equal integers mean that name and url are the ones of
>>  the same bookmark.
>>
>> This makes it easy to access the bookmarks of different browsers without
>> writing the code for getting them in every application. For example the
>> bookmark runner could use this to get the bookmarks very quick instead of
>> searching for them.
>> I hope you like this idea and I'd be glade if you would comment it. I know
>> that this can't be added to KDE SC 4.4 so I hope I don't annoy you with
>>  this.
>
> It would be good if you got in contact with Eduardo Robles Elvira <edulix at gmail.com>,
> he's working on bookmarks in Akonadi (which will likely be the canonical place to
> store bookmarks).
>
> Cheers,
> --
> sebas

Hello!

Yes, I didn't polish enough the new konqueror bookmarks enough to make
it into 4.4 either, but I'll try hard to move it into KDE SC 4.5. We
can work on this dataengine together ;-). You can see the current
state of the new bookmarks system in http://github.com/edulix/gsoc/,
and we can see how to deal with them in the dataengine.

I don't understand very well what you're trying to do with the
num-name num-bookmark. Is it to avoid bookmark duplication because of
having multiple bookmarks providers? If so, maybe you should also take
into account then also history in that dataengine, because it's quite
likely users actually visit their bookmarks, so even if the bookmarks
are not duplicated in the dataengine, there will duplicated entries:
one for the bookmark runner, the other for the history runner. There
should probably be instead something like a "web browser" runner which
unifies both runners and is configurable to search in history,
bookmarks, or both (?).

Regards,
    Eduardo Robles Elvira.