Correct recognition of languages during import of XML

I try to import Glossary from Excell with 30 langugages. First I convert the Excell-File with Multiterm 2022 Convert to an Multiterm XML File, then I try to import in Multitermbase. All languages are imported correct, only the languages Bosnian, Croatian and Serbian are imported to one of them with 3 entries. I want to have the terms in TDB with correct languages, but I didn´t find the way, how to do it.

Screenshot showing language entries for Czech, Bosnian (Latin, Bosnia and Herzegovina), and Bulgarian with the term 'Terminus Hinzufugen' and a placeholder '%s mesic' for Czech and three '%s mesec' entries for Bosnian.Screenshot of a language selection list including Bosnian (Latin, Bosnia and Herzegovina), Bulgarian, Croatian (Croatia), Czech, and other languages in alphabetical order.Close-up screenshot of a language code list showing SH-B2 for Bosnian (Latin, Bosnia and Herzegovina), SH-HR for Croatian (Croatia), and SH-LATN-RS for Serbian (Latin, Serbia).



Generated Image Alt-Text
[edited by: RWS Community AI at 1:22 PM (GMT 1) on 23 May 2024]
emoji
  •  

    It sounds as though you need to map them to the appropriate sub-languages as opposed to simply mapping to Bosnian.  So always show the fields in the UI settings so you get the list of languages when you convert.  Then you can set the specific variants you want:

    Screenshot of MultiTerm product showing a list of languages. Bosnian variants are highlighted with a red dashed border, including Bosnian, Bosnian (Cyrillic), and Bosnian (Latin).

    Paul Filkin | RWS Group

    ________________________
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

    emoji


    Generated Image Alt-Text
    [edited by: RWS Community AI at 9:01 PM (GMT 1) on 23 May 2024]
  • This don´t solve the problem. I set In MultiTerm Convert the languages correct, identically with Header in Excell and definition in Termbase, but after import it is always under one language. I think there is a bug, the code of all this languages (bosnian, croatian, serbian) start with SH, so Multiterm Import procedure do not recognize the difference. 

    Screenshot of MultiTerm Convert dialog box step 59 showing available fields for column headers with languages Bosnian, Croatian, and Serbian listed, and a language field dropdown menu with Croatian selected.

    emoji


    Generated Image Alt-Text
    [edited by: RWS Community AI at 7:48 AM (GMT 1) on 24 May 2024]
  •  

    I see what you mean.  I misread your original post and actually showed you a screenshot from the Glossary Converter,  So I just tested with MultiTerm Convert and see your problem now.  I checked and this is already a logged bug under our reference CRQ-30837 (you can't access that but it's helpful for following up if I note it here).

    I then tested with the Glossary Converter and this handles things correctly, so I believe the developer has most likely fixed this issue in the code for the Glossary Converter.  See video:

    Paul Filkin | RWS Group

    ________________________
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

    emoji
  • Hi Paul,

    thanks for advice, with Glossary Converter it´s ok, I create local termbase, the language recognition is correct.

    Then I create an termbase on server, with the same definition and try in Multiterm to export the local termbase to xml and then import in server termbase, and the language recognition was incorrect and the items from hr-HR and sr-Latn-RS are merge under bs-Latn-BA.

    I want continuously update the local termbase and then in an easy way - export-import - periodically update the server termbase, but this way in Multiterm is therefore impassable :-(

    Screenshot of a language selection list in MultiTerm with various languages, including bs-Latn-BA, hr-HR, and sr-Latn-RS highlighted. Screenshot of MultiTerm showing language entries for BS-LATN-BA with placeholder text '%s mjesec' repeated three times.

    emoji


    Generated Image Alt-Text
    [edited by: RWS Community AI at 8:23 AM (GMT 1) on 10 Jun 2024]
  •  

    Have you tried these things?

    • create the termbase definition using the Glossary Converter
    • export to XML using the Glossary Converter

    Maybe that will help if you can avoid MultiTerm desktop altogether?  Although I'm not sure and have not tested to see if the Server part also carries the same issue would still merge the entries on import.  But perhaps worth testing as you're already set up.

    Paul Filkin | RWS Group

    ________________________
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

    emoji