Property files for localization requires to be encoded in ISO-8859-1.
It results difficulties in reviewing localizing pull requires such as https://github.com/jenkinsci/maven-plugin/pull/86.
It would be nice if property files in source directories can be written in UTF-8 letters.
There are two approaches:
- Encode property files when compiling.
- maven-resources-plugin copies files in the resources directory into the target directory. It doesn't provide a feature to encode property files.
- Though native2ascii-maven-plugin encodes property files, it is provided only as beta for long time.
- This might mean encoding property files in compilation time isn't a standard approach.
- maven-antrun-plugin can run native2ascii (But it feels strange approach to me).
- This approach would cause problems when testing plugins with "mvn hpi:run".
- As property files aren't compiled and read from source directories directly at that time.
- Make libraries reading property files accept UTF-8 letters.
Actually, I'm not sure this is a good idea.
I created this ticket and will create pull requests to request comments about this idea.