Automatic association of screenshots with string ids

[mikedawson]

Describe the problem

I really appreciate the screenshot feature in weblate - it is very helpful to provide critical context for translators. I screenshot new strings in context as I add them to my project.

What is a bit frustrating is that afterwards I have to onto weblate and add the screenshots one by one.

Describe the solution you would like

If I add a new screenshot to the repository under the path specified screenshot path (e.g. as per project settings / Screenshot file mask ), where the name matches the string id e.g. string_id.png , then automatically associate this screenshot with the string id.

Describe alternatives you have considered

Adding screenshots manually one by one.

Screenshots

No response

Additional context

No response

[nijel]

There is OCR feature that might be helpful to associate strings, see also https://github.com/WeblateOrg/weblate/issues/9467

But yes, some kind of automatic association makes sense, the problem I see is that it's typically one screenshot for several strings, so there should be more than just a single string ID in the filename...

[mikedawson]

In our case, we are uploading one screenshot per string (even where it's a single screen, I am highlighting strings). I can see why some would use a single screenshot for multiple strings.

Simple workaround; if someone wants to link a screenshot to more than one string id (or a string id other than the file name)- one could have a small text file e.g. filename.png.stringids

string_id1 [lang1] [lang2]
string_id2 [lang1] [lang2]

The OCR feature is definitely neat, and I can see how it would be useful if one is trying to catch up and link things. In our case, we are maintaining this as we go because we recognize that the screenshots are vital to help translators understand context (especially in languages where computer terminology is not well defined).

[github-actions[bot]]

This issue has been put aside. It is currently unclear if it will ever be implemented as it seems to cover too narrow of a use case or doesn't seem to fit into Weblate.

Please try to clarify the use case or consider proposing something more generic to make it useful to more users.

[tomkolp]

In our case, there are several strings/dozen strings in one screenshot. We don't use the screenshot feature entirely because adding a million screenshots manually would take several years. Currently, it takes several minutes to add one screenshot. It would be ideal if weblate would automatically match the screenshot for a larger number of strings. I don't know if it's technically possible at all.