Multiterm ignores sublanguages when importing



I have some problems with Multiterm 2017 when importing data wich was originally converted from Excel into .mtf.xml with Muliterm Convert 2017.

The source Excel file is a very simple one: 1 column for source, 1 column for target, 1 column for information and 1 column for abbreviation source. All languages have separate Excel files as the terms differ.


When I tried to make a termbase with several languages and sublanguages, Multiterm does ignore the next sublanguage each time. For example, if I import de-AT, de-CH and de-DE, only the first of the sublanguages remains shown in Multiterm, altough there is an empty entry for the next sublanguage.

I checked the settings several times, I tried numerous approaches, like having the source language as sublanguage (English United Kingdom), clicking the checkbox "Ignore sublanguages", literally everything.




As you can see, the Swiss entry remains empty even if I select Swiss German from the dropdown, and the Swiss term becomes Austrian even if I selected the right sublanguage when importing.


Do you know if I missed a step or is this a bug?


Best regards,

Adam Biczo

5 Replies Latest Replies: 21 Jan 2019 12:13 PM by Steven Whale
  • Hello ,

    Maybe this is an issue with the (excel) import file not matching the definition of your Termbase correctly.
    Have you tried to convert using the Glossary Converter app-
    This tool (Glossary Converter) overcomes many of the technical issues in MultiTerm

    Please let me know if this (the app) helps
  • In reply to Steven Whale:

    Hi Steven,

    thank you for your answer. I'm going to be a little painful with this one.

    Unfortunately working with one xlsx for all languages (20+ with several sublanguages) is not an option as the number of source terms differs in many cases.

    I have installed Multiterm 2017 SR1, and here does Glossary Converter not support sdltb as output.

    To make things even more weird, I'm currently testing on a VM Multiterm 2019 where Glossary Converter works fine for sdltb-output.

    I found this on the matter after doing some digging:


    So, after this I made the good old xlsx->xml conversion with Glossary Converter, and imported these xml's into the termbase set up with several sublanguages. Unfortunately, the problem still remains, I import for Swiss German, everything is set up for Swiss German and it overwrites the Swiss entries to Austrian and leaves Swiss entries empty. I have not selected anything for Austrian, altough it the termbase contains this sublanguage, too - this should not be a problem for a multilingual database.

    As you can see, the XML is set up properly for DE-CH...

    but after importing it becomes Austrian.


    Maybe I should try to write some rules in Excel to get every entry for all languages into one sheet which I wanted to avoid as the source terms differ in many cases.


    Best regards,


  • In reply to Ádám Biczó:

    Hi Ádám,

    I work with large excels with +30 languages including variants, like en-US, en-GB, es-MX, es-ES and I always have some empty translations in some languages for a given term... Just to clarify, termbases doesn't work like translation memories, therefore there is not "source" as such. All you need is to organize correspondences, you can leave a term blank let's say in en-GB, and then have the term translated in de-CH and de-AT. I highly recommend coding all your terms together in an only excel divided in columns marked with language-country, even for English (not just en but en-GB or en-US).You may have something like


    and then you can open your excel in glossary converter and tell the program which language is which (if mistakenly recognized) in settings -> fields:


    I hope it helps,



  • In reply to Ádám Biczó:

    Hello  ,

    Further to  reply I would like to direct you to the following thread- which should help you if the GC app is not working properly.

    Also if you still have problems- could you provide a screen shot of your TB definition please and maybe a small copy of your Excel-

    all we would need is just a few lines (and notthe entire content) for testing (


  • In reply to Almudena Ballester:

    Hi Almudena,

    thank you for your answer. Yes, I'm aware of the fact that TB's doesn't not work like TM's, altough you can set for example in Multiterm that a selected language should be only source or target. Sorry it's German, but I hope you know this option in Multiterm.

    Regarding the Excel, yepp, that's what I wanted to avoid, because like I mentioned, the terms differ and there are 20+ languages with 500+ terms. To write a function in Excel requires some time and I wanted to avoid it as I thougt Multiterm was a multilingual database that handles such functionalities.


    Best regards,