My personal dragonfly grammars
I need my grammars to detect which language the current profile uses, and translate the grammars accordingly. So I decided to make that happen.
I dislike the way natlink, dragonfly and aenea operate. Don't get me wrong, they are amazing tools. But they don't make for good or elegant code. I attempted to right some of the things that bother me most in this module (mostly regarding statefullness, mucking about in system, poorly documented config files and code-reuse), but it is still a highly statefull machine.
I attempted to use built-in Python tools as much as possible. I use gettext for translations, setuptools for installation and fresh instances and grammar registration instead of module reloading. It is compatible with existing tools, but it works slightly different.
Still, it is not nearly as pythonic as I would like and I might one day get my grammars to the point where contributing to the existing tools and improving them will be easy. Right now though, all my energy is going into the grammars themselves.
- install natlink, dragonfly & aenea
- compile translations with
python ./setup.py compile_catalog - install this with pip
- move _translated_grammars.py to MacroSystem directory
Make sure your grammars have load and unload functions, and edit init.py to have them called. Also make sure your grammar is reloaded there at the appropiate time.
Add your language in init.py and have it activated in load_grammars. Then make translations with setup.py.
Note: watch out with init_catalog, it will overwrite existing .po files
python ./setup.py extract_messages
python ./setup.py (init|update)_catalog --locale=<locale>
# edit relevant .po file in dragonfly_grammars/translations/language
python ./setup.py compile_catalog