Hi

Could you please try to switch off the entities conversion in your embedded content processor? Or check, if these entities are there properly recognized and processed?

Regards, Jerzy

No, that unfortunately doesn't change anything... I even tried to create my own embedded content processor (copy from html 4)...

So I'm afraid I do not see any better solution than creating a customized file type with legacy embedded content, where the entities will be masked. These entities seem to be quite unusual. One can be identified as non breaking space, the other two however I can't tell what they are.

Hi Torben,

This seems to be a bug of some kind. We have logged this with development after they had a quick look at your files and couldn't see why it behaved this way.  So I think using the legacy embedded content is one way to go with this one, or maybe use the regex autosuggest?  That would allow the translator to handle any unrecognised entities with a single keyboard shortcut which might at least reduce the likelihood of errors.  I guess you could also add them as variables given there only a couple.

Not great, but hopefully development will resolve this in a future patch.

I will try to create a new filetyp using legacy embedded content... Not sure about the Regex autosuggest, because you don't really see the correct entity in the sdlxliff file...? And how would you distinguish between the different unknown character entities in the file... I only included two examples of these custom character entities in the sample, but there are a lot more in the actual source file... :-)

That is tricky I agree.  But I thought that as you could probably come up with a single expression to find any entity then it might not be too hard.  Find the entity and then place it with Ctrl+Shift+F12.

The ones in Chinese pose a different problem I agree...but maybe you can recognise them as variables?

I haven't tried any of this Torben... just thinking out loud.  This is a problem for you, an unusual one for sure, but maybe something like this will help you to minimise the risk of error at least.

Using the legacy embedded content kind of solves it... One downside are very long segments with tags/character entities in between...

Hi Torben, just fyi really so you know what to quote when looking for fixes in a future release.  Logged with development under ID 47672.