[Digikam-users] Using digikam on two computers?

Arnd Baecker arnd.baecker at web.de
Fri Jun 16 07:21:35 BST 2006

On Thu, 15 Jun 2006, Turgut Durduran wrote:

> Hello,
> I apologize for not checking the archives/bug reports
> carefully before asking the question. Thank you very
> much for answering my question nevertheless.

Don't worry - I did not know about BKO before posting
here and also asked a registered complaint (aeh feature wish ;-).

I think this discussion is very helpful as it
highlights the possible problems of merging databases.
Somehow I get the feeling that this is doable though!


> The other idea of keeping an identical copy of the
> whole album structure but with tiny images is
> interesting too but don't you get into complications
> when trying to keep track of what is new/what is
> old/what is large/what is small?  I imagine this one
> would be able to fool digikam into working state on
> both computers?

Yes, but maybe if one does an "intelligent" syncing procedure it
might work. Let's try the following workflow:
(Notation: "main computer" has the full database, all figures
in large size. "laptop": the one which is taken into the field,
to this one new figures are being added)

(1) create a full copy of all Pictures on the main computer,
    with all images rescaled to some prescribed resolution.
(2) copy this (including the corresponding digikam3.db) onto
    the laptop in some directory.
    Working with digikam on those files is no problem (tested).

    *All these files are considered to be unchangeable, including
     their tags etc. - all changes will be ignored*
(3) In the field: add new pictures to new folders,
    tag them, rate them, and also add tags if necessary.
(4) Adding the changes back to the main computer:

    (a) First add all additional tags into the main digikam3.db
    (b) For each *new* folder on the laptop: create that
        folder on the main computer and go over all
        files (and if needed recursively subfolders) and
        add them to the database, copy them into the corresponding
        In addition, for each files the corresponding tags are added.

So with some organization on the users side, this might work ...

Some problems/points:
ad (1):
   * Presently I only implemented something like this for jpg
     - what about raw files?
     - what about tiff/png/...
   * If one has changed Pictures on the main machine,
     one does not want to re-create small-size copies of
     all images again. So it should be possible to
     create the small size copies only for newer files.
     (However, files could have become deleted,
     changed or moved - so this is not completely trivial...)
ad (4):
   (a) this should be possible with some sqlite magic, or?
   (b) presumably we need some "tag" to id association
       to make sure that deletion of tags does not screw up things

I think all this should be doable with a python script
of less than 100 lines. Unfortunately, I know close to
nothing about sql, so I cannot go for a proof-of-concept
straight away.
(Are there any good quick references about sqlite,
in particular in connection with digikam?)

More importantly before coding is, whether the above approach
is robust enough - so any comments/criticism is wellcome!



P.S.: I just installed http://sourceforge.net/projects/sqlitebrowser/
      and it seems very helpful to explore/understand the digikam
      database ...

More information about the Digikam-users mailing list