Consider using pywikiapi for the low-level api calls?

[nyurik]

Hi, I think my pywikiapi library would be a perfect fit for you -- it will take care of the low-level communication with the mediawiki, while you can focus on the lexeme-related logic? Just an idea. I have been using this lib in many of my projects, and it was originally written to be used by pywikibot library as a low-level layer (but hasn't been integrated yet).

P.S. I'm writing a big bot to import all of ru-wiktionary into lexemes at the moment -- https://github.com/nyurik/lexicator (a bit work in progress, the main.py might not work properly). It uses that lib too.

[Nudin]

I had considered building that into version 0.2 but postponed it to a later version – probably 0.3 – since I wanted to get version 0.2 out as soon as possible, since it contains already a big amount of changes and improvements.

[Nudin]

PRs are of course welcome! ;)

[nyurik]

hehe :)

The lexicator is going full steam ahead, already supporting a lot of Wiktionary parsing use cases for at least two languages. I think it could function as a wiktionary parsing platform for later importing things into wikidata.