• State Verified Answer
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 189 subscribers
  • Views 5440 views
  • Users 0 members are here
Options
Related

Multiterm failing to import/distinguish language variants

Almudena Ballester

I have an excel glossary for several languages including variants for Spanish (Mexico, Argentina) and Portuguese (Brazil, Portugal). I treated each variant as a new language and therefore I create separate columns for them like this:

Then, I use Multiterm Convert to create Multiterm Database and carefully asign each column to language variant like this:

Same for ptPT for Portuguese (Portugal). Then I get my structure:

which distinguish between both language variants. But when I create my Termbase with Multiterm Desktop or Multiterm Server, I just get all portuguese terms under Portuguese (Brazil) and all spanish terms under Spanish (Argentina), no terms for Spanish (Spain), Spanish (Mexico), nor for Portuguese (Portugal), like this:

I've been using Multiterm, Multiterm convert and all Multiterm products since ever, so this time I think it is not a matter of not doing things right... I think it's a bug :(

I like to use Multiterm Convert instead other solutions like Glossary Converter because I want to keep .xcd, .xdt files for further Termbase updates and I also like to have control over structure and fields, etc.

Any suggestions?

Thanks in advance,

Almudena
Rate translation
×
Suggest better translation
×
Moderator UI
×
  • Get AI Suggestion
Parents Reply Children
  • 0 in reply to Almudena Ballester

    Hi Almudena,

    I have done this using the Glossary Converter (just like this tool better), but had to structure your spreadsheet a little differently to achieve it.  So I get this now for example:

    I have attached the spreadsheet I used and the converted termbase.  maybe this will help you as you could also use the termbase to edit the definition to suit what you really wanted and then base your proper termbase on that.

    gc_no_picklists.xlsx

    https://community.rws.com/cfs-file/__key/communityserver-discussions-components-files/91/gc_5F00_no_5F00_picklists.sdltb

    Hope this helps... also hope it's right!!

    Regards

    Paul

    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

  • 0 in reply to Paul

    Hi Paul,

    Many thanks for this solution. I see that you only leave one cell for each language whereas I keep three: a cell for main term, a cell for a synonym and another cell for a special form of term, which I also categorise with a field (cat).

    I know that your option, all terms in one cell separated through "|" is an easier way but I like to keep translations separate so I can add info for each translation variant. For example, it is a forbidden translation, it is an acronym, it is preferred... If I don't do this way, my cat field goes under an (unique) term level and categorises terms wrongly like here:

    It should read under term form DELE. I do this to warn my translators in case a special project or document doesn't allow a particular translation, etc.

    Also I'd like to show language names in .sdltb with full original name and not with those codes I use in excel for practical reasons...

    So, what about Multiterm Convert? Is there really a bug with language variants or am I doing something wrong...? I will give a try to Glossary Converter and re-build my termbase the way you did, but I'd really like to know if I need to discard my old method if termbase involves language variants.

    Many thanks again,

    Almudena
  • 0 in reply to Almudena Ballester
    Hi Almudena,

    You can also do what you want this way too. You just separate the notes with the same pipe separator. I do have one set up this way ready to go, and also with a different way to categorise the variants, but when I tested it found a bug with the Glossary Converter so the developer will work on this for his next update. As explained to me by the developer... "The problem is that the fields are assigned to the language left of them. But if that language has no term, the field "falls through" to the next one. For example, the English column is followed by empty Spanish and Finnish columns, and thus ends up with the fields for all three. "

    Once this is fixed it should work.

    I'm going to break a golden rule and have a look at MT Convert :-(

    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

  • 0 in reply to Paul
    Thanks a lot for that, Paul! We the big Multiterm lovers and sufferers really appreciate your efforts :)
  • 0 in reply to Almudena Ballester

    Hi Almudena,

    It's starting to look like a bug to me too... so I have asked the team to take a look in case I'm doing something silly.  I can get the structure you want in MultiTerm using the Glossary Converter but only if I fill all the spreadsheet so there are no gaps (that's the bug that needs fixing).  So at least I was able to try two ways of doing it to make sure I understood what you wanted.

    I attach them both here just for interest, but will respond again after the team have had a look... how I love working with MultiTerm Convert ;-)

    https://community.rws.com/cfs-file/__key/communityserver-discussions-components-files/91/gc_5F00_no_5F00_picklists_5F00_full.sdltb

    https://community.rws.com/cfs-file/__key/communityserver-discussions-components-files/91/gc_5F00_picklists.sdltb

    I don't think either of these will help you right now but you never know.

    Regards

    Paul

    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

  • 0 in reply to Paul

    Hi Almudena,

    ok the only bug here is in my head!  At least as far as MT Convert goes.,.. was a good exercise to find the problem with the Glossary Converter!!

    What you have to do is make sure you uncheck the option to ignore sublanguages in the import definition.  Obvious really!!  So the advice I had, that works for me was this:

    1. Make a copy of the default definition
    2. Edit it to ensure this is unchecked:
    3. Save it, activate it and process the import

    When I do this I get exactly what you need.

    Hope this helps and many apologies for my messing around... I'm not sure who's teaching who here!!!

    Regards

    Paul 

    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

  • 0 in reply to Paul
    Hi Paul,

    Yes! That little and significant checkbox -- so easy to overlook at the end of import process! I finally got my Termbase with all I need :)

    Many thanks for all and I hope this issue would help somebody else too.

    Almudena