ICU + HTML strings are not counting properly for memory

We’re attempting to use Translation Memory with strings that have both ICU and HTML in them. This is supported as far as I can tell. The system appears to be counting ICU’s syntax as translatable words:

For example:

Sometime in {date, date, monthYear}

Counts as five words, not two. Is that expected behavior?

@ruppel this is the expected behavior. By default, only the HTML tags can be skipped from counting in case of the Project Settings > Word count is set to “Skip Tags”.

See the Word Counter article in Crowdin’s KB for more details.

1 Like

Thanks for clearing that up! Obviously this is less than ideal since variables inside ICU syntax don’t need to be translated. Is there another recommended substitution format besides ICU that doesn’t count against the word count?

Hi ruppel,

Currently, this behavior can’t be omitted. But the improvement is on our roadmap :slightly_smiling_face:


Awesome, glad to hear it. Thank you!

Hello @ruppel!

Getting back to you regarding your request! We have released an improvement that should fix your issue.

You’ve welcome to check how everything works on your side,