Improve TM with respect of text with xml tags

[ohallot]

Hi

I'm translating docs with tons of XML. Some strings are clean (no XML) and others have XML intermixed. weblate translation memory fails to offer a suggestion in the following example case:

"Hello World!"

and

"<mytag a="1234">Hello Word!</mytag>.

The first string is not offered as suggestion to the (later) second string.

Is there a setting to put in TM so it can ignore/bypass xml tags?

Thank you.

[nijel]

This sounds like#11179 which was fixed in d53bce1.

[ohallot]

Bug #11179 seems to address machine translation issue.
Here, my issue is internal translation memory used in weblate. Machine translation is not available in the installation I use.

[nijel]

I see. There is currently no such option and Weblate suggests only very similar strings (using pg_trgm).