Open BadWolf42 opened 2 months ago
Hello,
thanks for the feedback.
I think the reason is you didn't put "fr_FR" in the list of target languages; I understand that having the flag "generate_source_language_translations" to true brings confusion but the list of target languages take priority over this flag.
Was it for a test or did you had an actual need for this? because I think I should do a V2 and completely remove this feature and never generate the file matching source_language
Hi,
I think you're right, I didn't put "fr_FR" in the target languages.
Actually, I used in the past BabelEdit to translate jMQTT. This software handles the keys in json files just as keys, and matches it accros all json files, so the "real" texts are in values in each language file. I would have liked to keep the fr_FR json files to be able to continue to use BabelEdit on some occasions, but it is not very important, your workflow is easier and does a faster job.
P.S.: Do you have or planned to build a tool to "flip" the translation text between 2 languages? I would like to replace all French texts in the source code by their English counterpart, from the translation file, in order to have an English only source code and French only in fr_FR.json.
Hi,
Sorry, I didn't received the notification on answer.
I didn't plan that and I think it is out of scope / responsibility of this workflow On top, this won't be a recurrent task to do so we you probably don't need a workflow. I would suggest you build a dedicated "one-shot" script for this need.
Hello Mips,
Thanks for this github-action, it saves a lot of time and is very simple to deploy :)
I just tested on plugin-brother, with this workflow and got this result.
In workflow I have set
generate_source_language_translations: true
, but nocore/i18n/fr_FR.json
has been generated. I know for a face thatcore/i18n/fr_FR.json
is not good so it should be present in the branch/PR.Did I do something wrong?
Thanks again, Bad