@lyxal I can't remember if you already gave me your verdict on this but what was your opinion on rewriting the parser (and also separating the tokenizer from it)?
I am drafting up a file structure so we can discuss the components and maybe soon start reimplementing a bit to get a workable version that can run like 1 + 2
although i would split different utility types like array, string, math, meta (e.g. all the vectorization behavior functions) although there might be too much overlap to properly do that
I mostly want to split the tokenizer out of the parser, so if we think rewriting it is a waste of time, just splitting the file is fine for me. And even then I don't really care too much about rewriting it. It's mostly that all the digraph and string parsing behavior feels a bit... cursed.
I think someone mentioned it would be good for Vylight if there were a separate tokenizer
I think it's pretty good, though I also have no experience complying with a code style standard; I usually pretty much just write stuff and then it works and I'm happy.
room topic changed to Vyxal: For discussion about Vyxal. Website: github.com/Vyxal/Vyxal | Org: github.com/Vyxal | Interpreter: lyxal.pythonanywhere.com | Type !!/info for more info [chat-bot] [code-golf] [epic-gamers] [never-gonna-give-you-up] [rickroll] [unicode] [vyxal]
Given that only two people have sent me their emails and we haven't really gotten any progress (people are probably busy especially since university has started for lyxal again - I'm not saying we need to try to force progress but I am just making this observation), I think we should abandon Jira for now and just try to get things going in the first place.
Since things are slow I think it's unlikely we'll get unorganized quickly enough for things to go bad again, and we can decide to return to it in the future if we decide we need it, but for now, we should just discuss things here and on GitHub and try to actually get some implementing going.
actually yeah i'll just stick to the existing format just for consistency and so we can reference the old files if we need to (though ideally we shouldn't)