Creating a TB from two tables EN-GE and EN-Gr

I have two tables, one containing English and German and the other containing English and Greek.

Both were converted to SDLTB via Glossary Converter. I then exported the TBs to xml, created a new one EN-GE-GR, imported EN-GE and then created an import model to synchronize on English term, adding new and merging existing entries. The result is bad–I have two glossaries in one. English terms are all duplicated and show either GE or GR as target, but not both. In previous versions this method worked and I received a glossary with EN-GE-GR. What am I doing wrong now?

I attacEPC.ziph the Excel files here.

Parents Reply Children
  • ok - I've had a quick play with this.  Must admit I would never do it this way... much much longer and more painful... but now that I have I can reproduce the problem you found in 2021.

    I then tested in 2017 (didn't bother with 2019.. will take your word for that) and had the exact same problem.  I don't have an older version than this installed... but that's probably irrelevant now anyway.

    Having said all of this I can still make it work.  And actually it does a much better job than the Glossary Converter so I may have to rethink my original asessment.  For example, the Glossary Converter does this with nearly duplicates:

    Screenshot of Trados Studio showing a list of terms with the English term 'Analyse root cause and continue stabilising the patient' and its German translation.

    Screenshot of Trados Studio displaying a list of terms including the English term 'Analyse the root cause and continue stabilising the patient' with a warning for a duplicate entry in German.

    Technically correct I guess because it creates new entries for the difference in case.  However, MultiTerm import does this:

    Screenshot of Trados Studio with a list of terms, highlighting the English term 'Analyse root cause and continue stabilising the patient' with a warning for potential duplicate German translation.

    And I think this is a neater solution.  Termbase attached:

    en-de-gr.zip

    The secret to making this work is this (adapting your workflow):

    1. GC to convert the two Excel files into separate SDLTB termbases
    2. Export both termbases as multiterm XML
    3. Create a new en-de-gr termbase in multiterm
    4. create an import definition synchronising on the English term
    5. import the en-de
    6. Now, I deleted the import definition and created it again, and I also ticked the box to run the import at the end
    7. this works

    The reason it seems to work is that I don't think the import definition is being saved properly.  So when I run it the advanced settings to synchronise on English are omitted.  You can see this in the final screen where you see a summary of what will happen.  When you create the import definition and run it the first time you'll see that summary does mention the synchronisation setting.

    So I think the bug is that the definition doesn't save so it can be run as intended afterwards.  I'll log this... but at least you should have a workaround that seems to work.  You should check the result I attached however as I didn't check it thoroughly.

    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: Trados AI at 1:50 PM (GMT 0) on 5 Mar 2024]
  • Dear Paul

    Many thanks for all your work and confirming the existence of the problem. I have already delivered the results from GC and the customer is OK with that. Nonetheless, I would again go that way.

    My process was similar, but after having created a en-ge-gr termbase, I used default import definition, then created a new one to synchronize on English term. And this did not work.