Empty segments filled with source text during conversion

Hi there,

When I leave a segment empty during translation, Studio fills it with the source text when converting into the original document format. But sometimes, neither source nor target text is needed, e.g. when a number is given in inches and centimeters, I can completely skip the inches line. I don't really want my target document to have bits of source text in several places that I have to delete manually after conversion. What can I do to avoid this? 

Many thanks,

Birgit

  • Hi Birgit,

    The answer to your question is threefold though you may already know the first two things I'll explain here...

    1: You can prevent source being copied to target when pretranslating via this setting:

      in Project settings or under File>Options if you wish this to always be the case.

    2: When you're actually translating, you can prevent source being automatically saved to target by unchecking this box:

      under File>Options

    3: As far as preventing anything from being saved automatically to the final target document, you need to find a way of filling the target segment with something that will not matter in the target document, then lock it as 'Draft' so it does not get auto-updated by any process such as autopropagation.

    See here:

    https://community.sdl.com/solutions/language/translationproductivity/f/90/p/10067/38529?Redirected=true#38529

    That should give you an idea of how to achieve what you wish.

    If not, or you wish to ask me further questions about this, I'm out from now till about 6 hours' time which is why I've given you the first 2 above, in case that's part of the problem for you.

    Hope this helps,

    All the best,

    Ali

  • Hi Alison,

    Many thanks for your suggestions! I'm familiar with 1) and 2), but this alone doesn't to the trick. Regarding 3), the thing is that I'd already tried inserting a space before converting, and while I thought I remembered it had worked once, last time I tried it wouldn't. But you know what? Now I tried with a different file (with a space and, just to try it out, with a left to right mark as well) - and it worked! I have no idea why it sometimes does and at other times doesn't, but I'll just hope it'll work whenever I try it in the future.

    Thanks again :-)

    Best,

    Birgit

  • Hi Birgit,

    Yes, 1) and 2) won't prevent the autofill of Source text to the Target document unless you also do 3).

    Studio does sometimes do inconsistent things but as far as I know, a space alone is never recognised by Studio as a translation. Studio thus autofills the Source to the Target document if it perceives the Target in the SDLXLIFF bilingual translation file as untranslated (because when there is content in the Source that needs to flow untranslated into the Target document, it is better to keep unnecessary identical source to target TUs out of the TM).

    The LRM mark works because, like other symbols, it is recognised as a character and thus a translation but the advantage is that it is invisible in the final document. As you'll have seen in the other post, if you actually want a space in the Target document you need to surround it with two recognisable characters, LRM's with a space between them for example. If you don't want this saved to your memory, keep the segment as Draft (right-click > Change segment status) and lock it (right-click > Lock segment).

    Hope this all makes sense,
    All the best,
    Ali
  • Hi Alison,

    OK, then I'll use the LRM mark in future files rather than just a space if you say this will always work, after all I'm not very interested in playing the lottery here and see what Studio will do with the space next time. Many thanks for your explanation, it does make sense (though I do think it would be useful to include an option for leaving segments untranslated in future versions).

    All the best,

    Birgit
  • Hi Birgit,

    I agree, it would be useful to have a 'leave blank' option sometimes. Especially if it is impossible to merge segments and the final segment is just a word that has already been translated in the previous segment because of word order differences.

    Nice chatting,
    Ali :))