Ask HN: Extract text and translate on every PR
One of the most tedious parts of frontend dev for our team is making sure all text is extracted to query strings for translation. We use i18n, and it’s great, but a lot falls through the cracks.
Have you set something up for your org on every pull request that pulls out text?
The last time I worked on a localized project, we had a technical fake language present in non-release envs. It was generated with every build (it was just a simple .po file, so easy to handle), and it was the same as English, but had an "x" at the start and end of each text, so you knew that it was "translated" (or at least it was translatable, and not hardcoded). QA included checking this fake language - if they found any text without X's around it, it was a clear sign of a i18n bug.
That's cool! We used machine translate to hindi for this which made it super obvious what was english / not as it was different character set (but made it hard to navigate)
Looking for something to help with DX though here, not QA