How to preserve segment status from XTM file when importing to Trados?

I basically have larger files that I export from XTM and work on in Trados. The issue is that the files contain many fuzzy and out of scope 100% matches as well as unedited MT from the client but Trados seems to randomly set segments to Confirmed that were not confirmed in XTM. This way I can't tell which is which and don't know what segments to work on - is there any way to import the files with the exact same segment status?

emoji
Parents
  •   

    Thanks for sending me the file.  I looked through it and there are only three states used within the entire file:

    no state at all
    translated
    signed-off

    These are mapped as these respectively:

    not translated
    Trados Studio screenshot showing Translation Details with Status: Not Translated and Score: 0%.

    translated
    Trados Studio screenshot showing Translation Details with Status: Translated, Origin: Translation Memory, and Score: 80%.

    translation approved
    Trados Studio screenshot showing Translation Details with Status: Translation Approved, Origin: Translation Memory, and Score: 100%.

    These are the only mappings that are possible with this file.  The reason you perhaps think it's random is because of this sort of thing that is actually in the source file itself:

    Code snippet from Trados Studio with target state 'translated' and state-qualifier 'exact-match' showing match-quality '100%'.

    or this:

    Code snippet from Trados Studio with target state 'translated' and state-qualifier 'fuzzy-match' showing match-quality '80%'.

    The alt-trans match-quality attribute is tool specific and we don't use this.  If you do, by using the option to map the state qualifiers, then you'll get a value in there as I have shown in the screenshots above.  In this case for every "exact-match" you'll get a 100% match, and for every "fuzzy-match" you'll get an 80% match as this is the hardcoded value that Studio will place on it.  There is no mechanism to use the alt-trans "match-quality" value at all.  So you may see different percentages in the file but they won't be used.

    So... in relation to your question here:

    is there any way to import the files with the exact same segment status?

    It already does.

    The issue is that the files contain many fuzzy and out of scope 100% matches as well as unedited MT from the client

    There is nothing in the XLIFF that Trados Studio can use to see this.  There are more state qualifiers available to XTM that might be helpful, for example:

    http://docs.oasis-open.org/xliff/v1.2/os/xliff-core.html#state-qualifier

    But the only ones used in your file are these:

    leveraged-tm
    fuzzy-match
    exact-match
    leveraged-inherited
    x-fuzzy-forward
    x-alphanum
    x-numeric
    x-measure

    The last four are custom extensions so we would not recognise these anyway, but none of them relate to being able to identify the things you have mentioned here (fuzzy, out of scope 100%, unedited MT).

    So, that's your problem as far as I can see.  Nothing we can address without two things:

    1. a new filetype based on supporting XTM specifically with all their custom extensions, and
    2. information being available in the file that we can then use to identify what you want to use it for

    I hope that helps to explain things a little... but maybe not quite as straightforward as you'd hoped?

    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 10:56 AM (GMT 0) on 29 Feb 2024]
Reply
  •   

    Thanks for sending me the file.  I looked through it and there are only three states used within the entire file:

    no state at all
    translated
    signed-off

    These are mapped as these respectively:

    not translated
    Trados Studio screenshot showing Translation Details with Status: Not Translated and Score: 0%.

    translated
    Trados Studio screenshot showing Translation Details with Status: Translated, Origin: Translation Memory, and Score: 80%.

    translation approved
    Trados Studio screenshot showing Translation Details with Status: Translation Approved, Origin: Translation Memory, and Score: 100%.

    These are the only mappings that are possible with this file.  The reason you perhaps think it's random is because of this sort of thing that is actually in the source file itself:

    Code snippet from Trados Studio with target state 'translated' and state-qualifier 'exact-match' showing match-quality '100%'.

    or this:

    Code snippet from Trados Studio with target state 'translated' and state-qualifier 'fuzzy-match' showing match-quality '80%'.

    The alt-trans match-quality attribute is tool specific and we don't use this.  If you do, by using the option to map the state qualifiers, then you'll get a value in there as I have shown in the screenshots above.  In this case for every "exact-match" you'll get a 100% match, and for every "fuzzy-match" you'll get an 80% match as this is the hardcoded value that Studio will place on it.  There is no mechanism to use the alt-trans "match-quality" value at all.  So you may see different percentages in the file but they won't be used.

    So... in relation to your question here:

    is there any way to import the files with the exact same segment status?

    It already does.

    The issue is that the files contain many fuzzy and out of scope 100% matches as well as unedited MT from the client

    There is nothing in the XLIFF that Trados Studio can use to see this.  There are more state qualifiers available to XTM that might be helpful, for example:

    http://docs.oasis-open.org/xliff/v1.2/os/xliff-core.html#state-qualifier

    But the only ones used in your file are these:

    leveraged-tm
    fuzzy-match
    exact-match
    leveraged-inherited
    x-fuzzy-forward
    x-alphanum
    x-numeric
    x-measure

    The last four are custom extensions so we would not recognise these anyway, but none of them relate to being able to identify the things you have mentioned here (fuzzy, out of scope 100%, unedited MT).

    So, that's your problem as far as I can see.  Nothing we can address without two things:

    1. a new filetype based on supporting XTM specifically with all their custom extensions, and
    2. information being available in the file that we can then use to identify what you want to use it for

    I hope that helps to explain things a little... but maybe not quite as straightforward as you'd hoped?

    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 10:56 AM (GMT 0) on 29 Feb 2024]
Children