E.g. in https://github.com/Rdatatable/data.table/pull/5875/files -- translator missed the detail that, when updating a translation marked fuzzy, we should also remove the , fuzzy notation and the #| comments with the guessed translation. It's also easy for a reviewer to miss this in a huge batch of translations.
This creates some toil when we need to (as developer) read through the old PR diff to see which translations were touched/not to tell whether they can safely have , fuzzy removed, or if they still need a review from translator eyes.
Not sure the right interface for this, but such tooling for developers to vet incoming translation PRs would be something to build on starting with this.
E.g. in https://github.com/Rdatatable/data.table/pull/5875/files -- translator missed the detail that, when updating a translation marked
fuzzy, we should also remove the, fuzzynotation and the#|comments with the guessed translation. It's also easy for a reviewer to miss this in a huge batch of translations.This creates some toil when we need to (as developer) read through the old PR diff to see which translations were touched/not to tell whether they can safely have
, fuzzyremoved, or if they still need a review from translator eyes.Not sure the right interface for this, but such tooling for developers to vet incoming translation PRs would be something to build on starting with this.