Import improvements

[fingolfin]

I just discovered the fork of this repository by @reiniscirpons. Really cool! Opening this PR so I won't forget where to look for it in the future.

If you are interested in "completing" this, I'd be happy to migrate the repository to the gap-system GitHub organization and give you direct write access there, to make it more official and perhaps invite others to contribute. Thoughts?

[reiniscirpons]

Hi @fingolfin, thank you for the kind words! I actually meant to get in contact with you about exactly this, but marking etc. got in the way and I forgot.

I would like to finishing the grammar. The main hurdle left at the moment is handling leading period floats e.g. .1 and correctly distinguishing them from record selectors (it seems this is not really used much in the GAP corpus, but is the last big language feature missing). The suggestion to migrate it to the gap-system organization is good and I would very much appreciate write access. I am also no expert in writing tree-sitter grammars, so contributions and suggestions from others would be greatly appreciated. Overall I agree with your proposal.

Having the grammar in as part of the official gap-system repository also opens up new possibilities of integrating tree-sitter with GAP itself, e.g. improving syntax highlighting in the interactive GAP sessions. I would also like to mention the tree-sitter-gaptst grammar I've been working on for parsing .tst files, which uses tree-sitter-gap for language injections. Do you think it would make sense to migrate tree-sitter-gaptst to the gap-system organization as well, or should this project be separate at the moment?

[fingolfin]

Great to hear you are interested in finishing this :-)

Regarding float literals like .1 -- IMHO it would also be OK to ignore them for now -- as you say, they are rare to use (and I think it was a mistake to introduce them in the first place; I've been thinking about pushing towards deprecating and ultimately removing this, see gap-system/gap#3280). That said, for now it is there, and I wouldn't want to stop you from implementing it -- but if that's really the only thing stopping this from being "ready", we shouldn't hesitate to make a first "release" now (I guess that would include registering it with npm, so that npm install tree-sitter-gap tree-sitter would work, and submitting it for inclusion in the list on https://tree-sitter.github.io/tree-sitter/)

Having tree-sitter-gaptst on gap-system sounds sensible as well.

My plan would to migrate this repository here to gap-system, then give you write access, so that you can merge / push / integrate your work as you deem fit. I'll wait a bit with that in case you have a better suggestion.

[reiniscirpons]

I believe using the external scanner I should be able to implement leading period floats, its just that until now there were lower hanging fruit which I've prioritized. The suggestion about making a first release and submitting it to registries is good, I will look into that once the migration is done.

I agree with the proposed plan, no better suggestions here.

[fingolfin]

@reiniscirpons I've transferred this repository and gave you "maintainer" access (if that turns out to be too limited I can still upgrade it to admin, just let me know if you run into blocks).

I've also sent you an invite to the gap-system GitHub org. Once you accepted that, you should be able to transfer your other treesitter repository.

Let me know if you need assistance with any of that.

Finally, you may wish to consider joining the GAP Slack (a multi user chat system) to be able to chat with others about this or other GAP related stuff: https://gap-system.org/slack should allow you to join.