How to use OmegaT TMX files?

2008-09-21
2012-09-14
  • Serhiy Kuznyetsov

    Hi!

    I failed to use OmegaT TMX files.
    Using Linux, I opened TMX files in Kate editor, saved as UTF16 file, it loads, but counts 0 TU, even file has about 500 kb.

    Is it possible somehowe to convert OmegaT TMX files for using with Anaphraseus?

    Thank you!

    Serhiy

     
    • Gabix

      Gabix - 2008-09-21

      So far, Anaphraseus cannot import TMX files — only export to TMX is possible. You may try out Wf2TMX utility developed by Wordfast community to convert a TMX file to a Wordfast/Anaphraseus translation memory (and vice versa). It's a Windows program, but should run with the latest wine. You can download it from the Files section of the Wordfast group in Yahoo! (http://tech.groups.yahoo.com/group/wordfast/files).

      Note that there are some issues about compatibility of Anaphraseus and Wordfast TMs. If you encounter errors/crashes when loading a file produced with the above utility, try creating a new TM and copying translation units (all strings except the header) into it from that file.

       
    • Gabix

      Gabix - 2008-09-22

      Still, for TMX files produced by OmegaT you'll need the converter, as OmegaT's output is UTF-8, not UTF-16 (that's a little follow-up to our private talk with Oleg).

       
    • Serhiy Kuznyetsov

      Thank you for responses!

      Till now i tried the tool Wf2TMX and I got garbage symbols instead of Cirillyc. I tried both options in Wf2TMX, Unicode and UTF-8, and result looked the same. I couldn't get Russian text, even I tried to change encoding in Kate editor.

      Regards, Serhiy

       
    • Gabix

      Gabix - 2008-09-23

      Serhiy, are you sure you use the latest build of Wf2tmx? That problem with Cyrillic existed in the previous builds, but I see it rectified in the latest one. I've just checked out 1.0.11.37 with an OmegaT TMX file (on Windows XP though), and it converted that file into a WF TM without any problem (although the output was UTF-8, whilst genuine WF TMs must be ASCII or UTF-16).

       
    • Serhiy Kuznyetsov

      Hi again!

      Thank you for your help.
      Yes, I used the latest Wf2tmx. And I tried to convert under Windows and I found that only few first dozens TUs were spoiled, but all others were correct. Probably that was under Linux also, Just that I didn't look further down into document. Perhaps my TMX was somehow broken.

      Then I copied TUs from the converted document and Anaphraseus does count them, only that I needed to change RU-RU into RU-01.

      Hm, First time the dictionary loaded well, but now on 60% of loading it gives Basic Error in ShowTranslation: Reading exceeds EOF. Line 43.

      Can it be something wrong with TM file? Can it be that there is too big TU?

      Regards,
      Serhiy

       
    • Gabix

      Gabix - 2008-09-29

      Serhiy, there can be something wrong with Anaphraseus :) Don't forget — it's still beta. Could you send your dictionary (glossary) file to Oleg to check out? Or to me to verify, whether it's really problematic?

       
    • Ole Yansen

      Ole Yansen - 2008-10-03

      Well, Anaphraseus can take previously converted OmegaT files.
      All you need to do is convert it from UTF-8 to UTF-16 (with OpenOffice's "Text encoded (*.txt)" filter for example), and load as TMX.

      Regards, Oleg.

       
    • Ole Yansen

      Ole Yansen - 2008-09-22

      Anaphraseus can get TMX version 1.0 only as the Wordfast.

      Regards, Oleg.

       

Log in to post a comment.

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:

JavaScript is required for this form.





No, thanks