Uwe Pezoldt 

Not really a solution, but a comment.

I’ve seen those pm tags with XLIFFs coming from an alignment step and then running a Perfect Match. I guess it’s your case here as well, right

These tags look different in Editor, but they won’t prevent you from creating the target document.

Also, if you change the tag view from the actual Tag ID to Full Tag Text mode, and if I recall correctly, you’ll see the same tags in Editor.

Hi Jesús,

Thnaks for your reply. It's correct, these segments come from alignment.

The problem is not so much the export as that the index entries are all wrong afterwards.

We had to correct them all in the exported files manually.

Regards,

Uwe

Hi Uwe Pezoldt 

Uwe Pezoldt said:

We had to correct them all in the exported files manually.

Do you mean that you need to correct tags in Word, in the Polish DOCX translated file? I don’t remember this step…

Hi Uwe Pezoldt 

Uwe Pezoldt said:

We had to correct them all in the exported files manually.

Do you mean that you need to correct tags in Word, in the Polish DOCX translated file? I don’t remember this step…

Yes, that's correct. In the exported target Word file we found not the actual keywords anymore but instead the XE boxes were filled with the wrong tag codes from Trados Studio.

It looks like this:

Uwe Pezoldt 

There is another topic about this:  RE: Problems with PerfectMatch and picture tags

Looks like the developers found this is a bug and will be corrected.