[Tracker] two diffenernt tags from one keyword.

[Alexey Fisher <bug-track fisher-privat net>]

Hallo all,

i have some issue with tracker caused by digiKam. Just curious how
should it actually work. 

My digiKam write metadata back to files, so it produce xmp part like
this:

dc:subject[1] Hermannsburg
dc:subject[2] Studenten
dc:subject[3] Missionsseminar
dc:subject[4] BW
dc:subject[5] Archive
dc:subject[6] Alte fotos
....
digiKam:TagList[1] Hermannsburg
digiKam:TagList[2] Studenten
digiKam:TagList[3] Missionsseminar
digiKam:TagList[4] BW
digiKam:TagList[5] Archive
digiKam:TagList[6] Alte fotos
....
lr:hierarchicalSubject[1] Hermannsburg
lr:hierarchicalSubject[2] Studenten
lr:hierarchicalSubject[3] Missionsseminar
lr:hierarchicalSubject[4] BW
lr:hierarchicalSubject[5] Archive
lr:hierarchicalSubject[6] Alte fotos

tracker extracted it and it make me some problems. First of all, most
tags have double entries.

For example keyword "Heramnnsburg":
====================================================
tracker-tag -t -s "Hermannsburg"
Tag: 1 (shown by name)
   Hermannsburg 
    urn:uuid:78b2fd14-7d2f-7206-8944-5719b64f5150
    153 files
file:///home/lex/Bilder/Fotos/2010/07/01/38._Kursus_(1946-1953).tif.png.jpg
....

tracker-tag -t -s " Hermannsburg"
Tag: 1 (shown by name)
  Hermannsburg 
    urn:uuid:d71396b4-6fdb-32ef-3823-02601ef5c964
    60 files
file:///home/lex/Bilder/Fotos/2010/07/01/38._Kursus_(1946-1953).tif.png.jpg
....
===================================================

I assume tracker read dc:subject and lr:hierarchicalSubject as two
different tags and create empty space before second similar tag. But if
this assumption true, why it read not all tags from this file.
Theoretically it should be 12 "nao:hasTag" but there is only 8

Here is part of tracker-info, combined with my comments:
  'http://purl.org/dc/elements/1.1/source' =
'urn:nepomuk:datasource:9291a450-1d49-11de-8c30-0800200c9a66'
  'nao:hasTag' = 'urn:uuid:e17bbb5c-39ad-ac0d-1157-80800d8e14b9'
---------> Alte fotos
  'nao:hasTag' = 'urn:uuid:9b59d4d2-aa6f-e69d-ed27-d229b92a7850'
---------> Archive
  'nao:hasTag' = 'urn:uuid:19ba8abe-b238-50f3-4d29-4b11876e27f1'
---------> BW
  'nao:hasTag' = 'urn:uuid:4796abf9-3e03-ddb8-f457-fa5401d14c87'
---------> Missionsseminar
  'nao:hasTag' = 'urn:uuid:71c181a7-1d0c-6660-59b6-3fc456cf1f5b'
---------> Studenten
  'nao:hasTag' = 'urn:uuid:78b2fd14-7d2f-7206-8944-5719b64f5150'
---------> Hermannsburg
  'nao:hasTag' = 'urn:uuid:d71396b4-6fdb-32ef-3823-02601ef5c964'
---------> Hermannsburg
  'nao:hasTag' = 'urn:uuid:f3bda3b4-40c1-ea61-9460-399b6d7e4c15'
---------> Alte fotos
  'nie:byteSize' = '500940'

So my question is, how it actually should work and is it possible to
ignore doubletags?

In attachment is one affected file.

Regards,
Alexey

Attachment: 38._Kursus_(1946-1953).tif.png.jpg
Description: JPEG image