[Digikam-users] Importing F-Spot tags to DigiKam

Caulier Gilles caulier.gilles at kdemail.net
Wed Jan 17 15:18:29 GMT 2007


Le mercredi 17 janvier 2007 16:04, Dotan Cohen a écrit :
> On 17/01/07, Caulier Gilles <caulier.gilles at kdemail.net> wrote:> Le mardi
> 16 janvier 2007 17:32, Dotan Cohen a écrit:> > I'm seriously considering
> converting to DigiKam 0.9.0 from the> > wonderful F-Spot. It is paramount
> that my 500+ tags on 8000+ images be> > imported from F-Spot. The tags are
> currently stored in XMP format in> > the photos themselves, and in the
> F-Spot sqlite database. What must I> > do to import the tags? The heiarchy
> is not so important, as I can> > reproduce that fairly easily. Thanks in
> advance.> >>> Well, digiKAm do not support yet XMP. All pictures metadata
> are supported by> digiKam core using Exiv2 library. This library need to be
> updated for that.> There is a wish in Exiv2 bugzilla about this subject :>>
> http://dev.robotbattle.com/bugs/view.php?id=463>> Personally, i think than
> XMP _must_ be supported by Exiv2 and digiKam. It very> important.>>
> Actually digiKam/Exiv2 support IPTC/IIM metadata, witch are the old IPTC>
> schema. XMP is the new IPTC schema designed by Adobe...>> If you want to
> import your XMP info to digiKam, a solution is to write a> little script
> witch use ExifTool to copy XMP informations to IPTC/IIM in you> pictures.>>
> When you will start the first digiKam instance with your pictures, IPTC>
> metadata will be parsed and the right digiKAm tags will be automaticly add
> to> digiKam database.>> Regards>> Gilles Caulier 

> Thanks Gilles. Actually, 
> most of the pictures haveIPTC data as well,but it is not showing up as
> tags. It does show up inMetadata->IPTC->Contact but in the wrong charset.
> The IPTC data isHebrew stored in UTF-8 (as IPTC always is UTF-8), however
> Digikamtries to display it as ISO-8859-8, which leads to me
> seeingX-giberish-X-giberish-X. See the the 8th example on this
> page:http://dotancohen.com/heb/encoding.phpIt's in Hebrew, but it's an
> example of UTF-8 text being displayed as ISO-8859-8. I went looking for the
> Digikam database but it's not in~/.kde/share/apps/digikam. Where is it? I
> might just run a few querieson the sqlite database to inject the tags.
> Dotan Cohen

Yes, i know the problem. In fact the UTF-8 problem is know.

IPTC UTF-8 support is not clear in IPTC/IIM spec. We have found a solution in 
this B.K.O report :

http://bugs.kde.org/show_bug.cgi?id=132244

It's not yet corrected in current implementation. I will try to do it before 
0.9.1 release date.

I will very interressed if you can post comments in this B.K.O file, for 
example, to attach some pictures with UTF8 IPTC char-set.

Gilles




More information about the Digikam-users mailing list