Maintainance and proliferation of upstream grammars

[dschrempf]

Recently I became aware of some upstream grammars being poorly maintained. For example, https://github.com/tree-sitter/tree-sitter-haskell is lagging behind two forks I know of: https://github.com/tek/tree-sitter-haskell and https://github.com/tree-sitter-grammars/tree-sitter-haskell.

See also the discussion in tek/tree-sitter-haskell#11.

For reference, neovim has recently switched over to using https://github.com/tree-sitter-grammars/tree-sitter-haskell.

What is the stance of the Emacs users on the maintainance and proliferation of upstream grammars? Do you think we should follow neovims lead, and also switch to grammar providers that are more actively maintained?

I am asking, because in my particular case I have reported a bug to the Haskell grammar, which was only fixed in an upstream fork of the grammar, and not in https://github.com/tree-sitter/tree-sitter-haskell.

Thank you!

[jcs090218]

Many existing grammars have switched their upstream to tree-sitter-grammars for this reason, so yes—I believe this is a good direction. :)

[dschrempf]

Hi! You marked this as a good first issue. Could you give us some guidance as to how we would proceed in implementing this? The grammars seem to be git sub-repositories. Should we just point these sub-repos to the tree-sitter-grammars organization?

[jcs090218]

That sounds correct. You can find the full details here: https://github.com/emacs-tree-sitter/tree-sitter-langs?tab=readme-ov-file#building-grammars-from-source. You just need to remove the old source and add or redirect it to the new upstream. :)