Add docs for including machine-generated Arabic translations in e2e tests #3877
Labels
♿️ aspect: a11y
Concerns related to the project's accessibility
🤖 aspect: dx
Concerns developers' experience with the codebase
🧰 goal: internal improvement
Improvement that benefits maintainers, not users
🟨 priority: medium
Not blocking but should be addressed soon
🧱 stack: documentation
Related to Sphinx documentation
Problem
We currently have an undocumented practice: Whenever we add new text to Openverse requiring translation, which is included in Visual regression tests whose snapshots are updated, we hardcode manual Arabic translations to the https://github.com/WordPress/openverse/blob/main/frontend/test/locales/ar.json file. This allows us to approximate how RTL and Arabic support looks in the actual site.
You can see this process in action on this PR.
Description
Add docs explaining how this works, and instructing folks to go to translate.google.com to get a machine translation, to https://docs.openverse.org/frontend/guides/translate.html.
The text was updated successfully, but these errors were encountered: