Can I merge two .xlf translation files into one?

I have two .xlf translation files: source-language="en" target-language="fr"

The source language is identical in both, but the target language in one has many updated segments. I need to merge the two .xlf files because the one with the updated French segments is based upon an older source file.

Does SDL Trados Studio offer a way to merge the two .xlf files and remove the replaced French segments?

Thank you

Parents
  • Hello ,

    If I have understood your question correctly, you should be able to do this by using the “Translate to Fuzzy Match” feature in Trados Studio.

    Create a new project in Studio, adding both files (copied to a new folder) ensuring they have a different name to make it easier to see (OLD and NEW). Don’t use a TM to create the project, but do create a NEW TM in the Project Settings once the project is created.

    Selecting the original OLD file only in the Files View, then update Main Translation Memory In your Batch Tasks.
    In the screenshot below I have highlighted the original translation for comparison

    Once you have populated your TM with the contents of the OLD file open the 2nd (NEW) file for translation.  You will see the changes made in the second file where the translations in the file are different to the OLD file:

    Now select “Confirm and Translate until Next Fuzzy match”:

    Studio will now run through the translation and stop at the first segment where there are changes and warn you.  These will be the segments in the NEW file that have translations older than the ones in the OLD file.  This gives you an opportunity to verify it’s correct and also replace the translation with the one from your TM:

    Once complete “Generate the target file”, or just “Save target As”, and the XLIFF file you get will be the one containing a “merge” of your two files.


    If you feel confident that everything in the OLD XLIFF (now in your TM) should always replace the translations in your NEW XLIFF then you could also do this faster and automatically by just pre-translating the NEW XLIFF instead of using “Translate to Fuzzy” and use the option to always overwrite from the TM.

  • Don’t use a TM to create the project, but do create a NEW TM in the Project Settings once the project is created.

    Could you please explain why not to create?

    emoji
  •  

    Because you don't want to change what's in the files at all.  So create the project first without a TM, then create a new TM so it's empty.  Then you can go through the process above to achieve the aim.

    emoji
  • sorry Paul but I am not following

    How does creation of a TM affects already made xliff files?

    If that is related to consistency of segmentation, than it was made during creation of the xliffs to be merged later.

    Could you please clarify in some more details the 

    Because you don't want to change what's in the files at all.

    ========

    btw, I usually utilize Perfect Match function to get segments from other translators xliffs (in existing project, simply making PM and importing others xliffs to my TM).

    But now I realize that the Translate until next Fuzzy Match helps to see difference in proposed translation at the segment level (not phrase level unfortunately) by mean of penalties over the differently translated segments.

    However in my PM approach I simply prioritize my current xliff content over other PMed content by playing with Segment Actions.

    Anyway, I find both methods, and especially the proposed above, rather many-clicking. I believe there may be a more convenient method

    emoji
Reply
  • sorry Paul but I am not following

    How does creation of a TM affects already made xliff files?

    If that is related to consistency of segmentation, than it was made during creation of the xliffs to be merged later.

    Could you please clarify in some more details the 

    Because you don't want to change what's in the files at all.

    ========

    btw, I usually utilize Perfect Match function to get segments from other translators xliffs (in existing project, simply making PM and importing others xliffs to my TM).

    But now I realize that the Translate until next Fuzzy Match helps to see difference in proposed translation at the segment level (not phrase level unfortunately) by mean of penalties over the differently translated segments.

    However in my PM approach I simply prioritize my current xliff content over other PMed content by playing with Segment Actions.

    Anyway, I find both methods, and especially the proposed above, rather many-clicking. I believe there may be a more convenient method

    emoji
Children
  •   

    How does creation of a TM affects already made xliff files?

    It doesn't.  But not all translators have the professional version and the ability to use Perfect Match.  So the answer provided was simply a way to use the Translate to Fuzzy feature to allow a translator to take the contents of the existing xliff file (by putting them into a TM all of their own) and then comparing the contents of the updated xliff files with that TM.

    Could you please clarify in some more details the 

    Because you don't want to change what's in the files at all.

    Sure... I was answering what you asked here:

    Don’t use a TM to create the project, but do create a NEW TM in the Project Settings once the project is created.

    Could you please explain why not to create?

    If you used an existing TM there is a chance you might introduce something into the xliff you don't want to have to think about because it may be a match not related to these two files at all.  So for this small and simple exercise it makes sense to just create a new TM.

    If you wanted to then you could of course use an existing TM.  But the task is so trivial I would just create a new one for this exercise, and then update the main TM afterwards.  There are always many ways to skin a cat!

    btw, I usually utilize Perfect Match function to get segments from other translators xliffs

    Absolutely perfect when you have the Professional version.  And these days you could probably work with the update file feature as well.

    How successful this whole exercise was we'll never know since  didn't bother to come back and tell us if the solution provided was helpful or not!

    emoji