Importing SKOS duplicating concepts

28 views
Skip to first unread message

Divino Júnio Batista Lopes

unread,
Aug 12, 2020, 3:54:09 PM8/12/20
to AtoM Users

Hello there. I'm trying import some subjects using a SKOS file by CLI (I'm using the flags index and verbose), but this action is making many clones of the originals terms and i don't know why. I suspected that it could be something about the hierarchical tree of the subjects, since turning all relationships in Related Terms fix the problem. I tried using the AtoM 2.5.3 and 2.6.0 and I had the same issue. There's a image showing the result below and the file used in annex.

Thanks!

Junio.
AtoM_Error.png

Divino Júnio Batista Lopes

unread,
Aug 12, 2020, 4:03:51 PM8/12/20
to AtoM Users

I tried to annex the file, but it's bigger than google allow.
Then I uploaded on my Google Drive and this is the link: https://drive.google.com/file/d/11oVY0MLVTbtqzbpmWLnjow6OmMMnUCVm/view?usp=sharing

Dan Gillean

unread,
Aug 13, 2020, 6:15:06 PM8/13/20
to ICA-AtoM Users
Hi Divino, 

I've managed to reproduce the issue that you are seeing locally. I haven't entirely isolated the cause yet either, but it appears to have something to do with the way that broader/narrower relationships are represented in the SKOS file. For example, take the term "Abacate" shown in your screenshot. When you find the original term that is properly linked to all ancestor terms, it appears to have 8 ancestors - and you end up with 8 total versions of the term. I suspect that something about how these relations are being created is causing the duplicate, though I'm not yet sure what exactly. 

In the meantime, something you could try: running the taxonomy normalization task: 
I do suggest you create a backup first! This task will identify terms with an exact duplicate authorized form of name, and merge them. However, it does so by moving description relations from duplicate terms to the oldest / original version of the term. I'm not sure in this case what order these duplicate terms were created, so it's possible it will delete the wrong term, breaking the hierarchy. However, at this point, if you back up first, I think it's worth a shot. 

I will try to investigate this further in the coming days to see if I can determine the cause, and hopefully file a bug report if needed. 

Cheers, 

Dan Gillean, MAS, MLIS
AtoM Program Manager
Artefactual Systems, Inc.
604-527-2056
@accesstomemory
he / him


On Wed, Aug 12, 2020 at 4:03 PM Divino Júnio Batista Lopes <junio.b...@gmail.com> wrote:

I tried to annex the file, but it's bigger than google allow.
Then I uploaded on my Google Drive and this is the link: https://drive.google.com/file/d/11oVY0MLVTbtqzbpmWLnjow6OmMMnUCVm/view?usp=sharing

--
You received this message because you are subscribed to the Google Groups "AtoM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ica-atom-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ica-atom-users/af0d8595-e352-41df-86f0-f0c8e5f21793n%40googlegroups.com.

Divino Júnio Batista Lopes

unread,
Aug 14, 2020, 4:48:43 PM8/14/20
to AtoM Users
Hi Dan. I tried use the taxonomy:normalize, but i was receiving error messages about the PHP trying delete a null object. I was wondering about the SKOS structure and I'll try remake that file removing any broader relationship and just let the narrowers. Thanks for the tip, this command seems like exactly what I wanted.
Reply all
Reply to author
Forward
0 new messages