Uploaded image for project: 'translate5'
  1. translate5
  2. TRANSLATE-2834

Change repetition behaviour in pre-translation

    XMLWordPrintable

Details

    • Critical
    • The calculation of the match-rate of repeated and pre-translated fuzzy segments has been changed. More details in the concrete change log entry of that issue.
    • Hide
      On pre-translations with using fuzzy matches, repeated segments may be filled with different tags / amount of tags as there are tags in the source content. Then the repetition algorithm could not process such segments as repetitions and finally the analysis was not counting them as repetitions.
      Now such segments always count as repetition in the analysis, but it does not get the 102% matchrate (since this may lead the translator to jump over the segment and ignore its content). Therefore such a repeated segment is filled with the fuzzy match content and the fuzzy match-rate. If the translator then edits and fix the fuzzy to be the correct translation , and then uses the repetition editor to fill the repetitions, then it is set to 102% matchrate.
      Show
      On pre-translations with using fuzzy matches, repeated segments may be filled with different tags / amount of tags as there are tags in the source content. Then the repetition algorithm could not process such segments as repetitions and finally the analysis was not counting them as repetitions. Now such segments always count as repetition in the analysis, but it does not get the 102% matchrate (since this may lead the translator to jump over the segment and ignore its content). Therefore such a repeated segment is filled with the fuzzy match content and the fuzzy match-rate. If the translator then edits and fix the fuzzy to be the correct translation , and then uses the repetition editor to fill the repetitions, then it is set to 102% matchrate.

    Description

      problem

      On pre-translation repeated segments where treated as repetitions. If pre-translation is used with fuzzy matches, the segment may be filled with different tags / amount of tags as there are tags in the source content.

      Then the repetition algorithm could not process such segments as repetitions and finally the analysis was not counting them as repetitions.

      solution

      If a segment was pre-translated with a fuzzy match, and this segment is a repetition, it is counted as repetition in the analysis, but it does not get the 102% matchrate (since this may lead the translator to jump over the segment and ignore its content). Therefore such a repeated segment is filled with the fuzzy match content and the fuzzy match-rate. If the translator then edits and fix the fuzzy to be the correct translation , and then uses the repetition editor to fill the repetitions, then it is set to 102% matchrate.

      Attachments

        Issue Links

          is cloned by

          Bug - A problem which impairs or prevents the functions of the product. TRANSLATE-3068 Fix repetition behaviour in pre-translation with MT only

          • Done
          relates to

          Bug - A problem which impairs or prevents the functions of the product. TRANSLATE-2756 Segments were locked after pre-translation but no translation content was set

          • Done

          Bug - A problem which impairs or prevents the functions of the product. TRANSLATE-2764 Refactor MatchAnalysis code and reduce memory usage

          • Selected for dev

          Activity

            People

              tlauria Thomas Lauria
              marcmittag Marc Mittag [Administrator]
              Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: