Studio not recognising unchanged numbers on reimporting source

Since several months, I find that upon reimporting my source file (which I do very frequently), Studio does report unchanged numbers as a change and therefore does not import such segments as 100 % matches or content matches. It causes a lot of extra work.

Example segment:

Die wesentlichen Handicaps der antibiotischen Behandlung bei LB sind in Tab. 23.1323.13 dargestellt.

And also, a previously translated 'Tab. 23.13' is proposed by Studio as 'Table 23.18' upon reimporting, whilst it was actually translated as 'Table 23.13'.

Why can't Studio reconise such simple numbers, please?

Best regards,

Willem

  • Good morning everyone,

    I have posted the following on 25 June, but there has not been a response. It is an exceedingly annoying problem, leading to wasting a lot of time.

    Could some one look at it, PLEASE?

    Studio not recognising unchanged numbers on reimporting source

    Since several months, I find that upon reimporting my source file (which I do very frequently), Studio does report unchanged numbers (already translated and therefore present in the TM and SDLXLIFF) as a change and therefore does not import such segments as 100 % matches or content matches. It causes a lot of extra work.

    Example segment:

    'Die wesentlichen Handicaps der antibiotischen Behandlung bei LB sind in Tab. 23.1323.13 dargestellt.'

    And also, a previously translated 'Tab. 23.13' is proposed by Studio as 'Table 23.18' upon reimporting, whilst it was actually translated as 'Table 23.13' and is present in the TM.

    How can I get Studio 2017 to recognise such simple numbers, please?

    Best regards,

  • Hi Willem,

    It looks like your TM may be set to not recognize numbers. If this is the case, checking the Numbers checkbox under TM Settings - Fields and Settings should take care of the problem.

  • Thank you, Nora, but it does not solve the problem. The only box not checked on the settings page is the one about counting as one if containing tags.

    Could the fact that, in this particular case, my file is a pure text file (.txt) have something to do with it?

    Best regards,

  • Hi Willem,

    I don't think the file type would have an effect. I tested your text (in a TXT file) with and without number recognition enabled and saw the match go from 100% (CM) to 98%, with the same highlighted differences you have in your screenshot, here's a short video showing that (please ignore the TM languages, just created a quick project and TM with my default languages):

     

    Are your results different when you enable/disable number recognition?

  • Thank you, Nora. Interesting. I will play with that in the morning, because right now my wife is waiting for me to take her on a maternity visit to a niece of hers.

    I will let you know how I got on.

    Best regards,

  • Nora, disabling and re-enabling number recognition seems to have done the trick. After disabling and again after re-enabling, I have re-indexed the TM. It more than doubled the size of the TM. That is not an issue for me.

    The weird thing is that the TM was created in March, specifically for the project, with all the settings bar one (standard) enabled. So, somehow, they were not activated in spite of being enabled. I always run the latest versions of all my software, fully patched and updated. Perhaps, there is a bug, somewhere.

    Anyway, I am happy that you put me in the right direction. Thank you very much.

  • That's great to hear, Willem. And it does indeed seem strange that the TM wasn't working the way it was supposed to with all the default settings.