Hi all,
I have been working on a spellchecker for Sanskrit (an Indian language) using Hunspell. I want to improve upon it to support contextual spelling correction which is not possible in Hunspell.
I have already tried developing a few rules following the instructions in the Development Overview and LT seems to good for my requirements.
But before diving in full, I have few questions:
Has there been any previous attempt/work in LT for Sanskrit?
Are you accepting contributions for new languages?
Can my dictionary created for Hunspell be used here? Or creating it using Morphologick will be more efficient?
How much of LT is open source? I don’t want my contribution to be vendor locked-in.
How are my contributions licenced? I am doing it as part of my PhD.
We haven’t added a new language for quite some time, at least not in the sense that the new language will actually appear in the add-ons and on languagetool.org. We might add it to the source code, though, so it can be used via API. The reason for being hesitant about adding new languages is that it causes a lot of work for us, and the support for many languages is already unmaintained in LT.
If you want your changes to be merged into LT some day, they’ll need to be licensed under LGPL, and you’d need to follow these steps. I suggest you talk to your university to make sure you’re allowed to do so.