Open Akemi-Madoka opened 7 months ago
The screenshots in the glossary are more a side effect of using the same model as translations, rather than an intentional feature. Can you please elaborate a bit more about using screenshots in glossaries?
Using external server for screenshots is covered by https://github.com/WeblateOrg/weblate/issues/3366.
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.
@nijel
When we previously translated game projects on Crowdin, there were many personal names in the dialogue text, and it was difficult to obtain screenshots for these dialogues in bulk. Therefore, our solution was to directly insert the corresponding person's avatar into the glossary, which we were very satisfied with. However, due to cost concerns, we plan to migrate to Weblate, but this feature has become cumbersome there.
Describe the problem
The current Glossary feature allows the upload of contextual images. However, to view these images, one must click into the corresponding Glossary entry, which is cumbersome. This is particularly problematic for specialized terms such as personal names, equipment names, or other texts that require translators to understand the context, as they must click on the Glossary and wait for it to load to see what it is (via images uploaded by developers).
Describe the solution you would like
Therefore, I hope for two improvements:
The Glossary should support external URLs instead of requiring file uploads, allowing for simpler integration with our CDN and image servers. On the translation interface, Glossary terms that are highlighted should show a pop-up with images when hovered over (if available), and the Glossary sidebar should ideally display image thumbnails as well.
Describe alternatives you have considered
No response
Screenshots
No response
Additional context
No response