seb314

based on https://www.typescriptlang.org/docs/handbook/modules/theory.html, I stumbled on https://arethetypeswrong.github.io/?p=antlr4%404.13.1-patch-1 which seems to be a promising tool to analyze this issue. I've a patch that satisfies the attw checker in all except the...

> tbh I suspect that it might be impossible to satisfy all scenarios given the mess created over time by ts vs js and esm vs common js. Basically, we...

@jonaskello I also believe that it would in principle be preferable to have a single source of truth in .ts files. But 1) I believe the resulting js would not...

Thanks for the making this happen! I played around with the test cases a bit. Here is a test case that does not currently work as I'd expect: ``` #...

The suggested tokens for the test case look good now! Regarding the single parser context for all suggested tokens, I'm a bit puzzled why that works. Is this really always...

I tried to come up with an example where more than one parser context is consistent with the same input & cursor position: https://github.com/seb314/antlr4/commit/118e126db71e4b44e5c85a0c9465858b6df7266f Thanks for the JS version! I'll...

> The TS build is broken, but has been for a couple of weeks... I'm looking into it, I don't think it relates to this change As long as the...

I've now tried the js version with our regular test cases a bit. The suggested tokens look mostly good already (modulo a few unwanted suggestions that look like they are...

@ericvergnaud thanks for the clarification! I'll try the approach with separate parsing and then report back (might take a couple of days until I have time to test this) (Knowing...

My understanding of why we have a custom implementation is: 1. for certain grammar rules, our desired autocomplete suggestions are not individual tokens, but a sequence of tokens that corresponds...