What we want to achieve

If Okapi configurations for resource files (resx etc.) or for json files are set correctly, the ID of the to be translated string will be extracted by Okapi as resname of the trans-unit.

For HTML files it is possible to extract the ID of a segment as segment ID of the xliff.

So far translate5 saves the database segmentid of a segment as context in t5memory.

Currently if the value of context is the same for a match and the document name is also the same, we are getting a 103% match in translate5.

In the future we want to get a 103% match, if:

• document name is the same and
  • resname value of the segment is the same
  • OR if resname is not defined segment number in xliff is the same

Solution

The value resname attribute of the trans-unit in xliff should be read and saved as meta-data of a segment (if it exists).

Everywhere where context is evaluated for t5memory or saved to it in the future the resname value should be used for a segment, if it exists. If it does not exist, the segment number in task should be used instead.