Copy link to clipboard
Copied
I have used RoboHelp (v 2022.3.93) to export XLIFFs of our web content for language translation. After testing out the translation with a couple different independent tools then importing the XLIFFs back into RoboHelp, I'm finding that the translated content is missing (blank). One vendor has told me they are unable to parse the sample XLIFF files I provided (in the format generated by RoboHelp).
Wondering if others have encountered this problem - the process I'm using doesn't seem outrageously complex. We'd like to use machine translation as a starting point and then have native speakers within our organization perfect this. However if the translations aren't visible after re-importing the XLIFFs then the process is obviously not usable. I would appreciate any advice. Thank you.
Copy link to clipboard
Copied
Copy link to clipboard
Copied
That's actually a different issue, which I also submitted. It was before I tried to incorporate a translation app into the mix. Something about the interaction between RoboHelp and the app is causing the translated strings to come back blank when imported as XLIFFs.
Copy link to clipboard
Copied
I was thinking it was the issue with the second solution (something about the MRK things) might have been your issue. Are you using one of the translation tools that Stefan talks about in https://community.adobe.com/t5/robohelp-discussions/searching-for-a-translation-tool-that-works-with... ?
Copy link to clipboard
Copied
I have not tried any of the tools he references. I can take a look at these. If they work and aren't terribly pricey, I'm hoping we might be in business.
Copy link to clipboard
Copied
That might be the issue - something about supporting pre-segmenting or something?
Copy link to clipboard
Copied
Something went wrong during processing the XLIFF files in translation, when importing them, or both.
1) something went wrong during processing
Your XLIFF file should have the following structure:
XLIFF file after generating it (from base project)
XLIFF file after translation
If you get a different XLIFF structure, your translation vendor probably did not use pre-segmenting.
2) something went wrong during import
The import functionality is [censored] and you may need the stars to align correctly to have everything work.
On the top of my head, I can think of the following issues:
*Variables, glossaries, TOCs, and template files remain untranslatable for our translation vendor for reasons. I had to build a custom conversion-to-XLIFF tool to remedy that problem.
***
I also have doubts about your proposed workflow. If I read your post correctly, you want to have users post-edit machine translations in the translation project after importing the XLIFFs. If that's the case, I foresee issues.