Nathan Ridge
Nathan Ridge
> I just noticed: The upcoming spec version has a new client capability that's very close to the mentioned option: Nice find, I think this makes the spec's intention clear....
> @HighCommander4 Is this being worked on by anybody? Not that I'm aware of. > 1. As the missing tokens are lexical and well-defined by itself or its immediate surroundings,...
> I believe providing lexical highlighting tokens should be a fairly simple matter of iterating through the tokens in the TokenBuffer, selecting the subset for which we want to provide...
> ### Lexical Tokens > > There are roughly the following lexical tokens in the language > > * comments > > * keywords > > * number literal >...
Confirmed. This is similar to #432 and #810, and should be fixed by getting code completion to use `HeuristicResolver` as dicussed [here](https://github.com/clangd/clangd/issues/810#issuecomment-877379680).
Related to (possibly duplicate of ) #965
> How big of a deal would it be to change one of the datatypes in the index? Can clangd easily detect this and just rebuild the index or would...
Could you post a log with `--log=verbose` added to `clangd.arguments` please?
I don't think that worked, there are supposed to be additional log statements prefixed with `V` when running with `--log=verbose`. May just need to restart clangd for the change to...
Thanks. I have verified using the log that I'm following the same reproduction steps that you are (including sending the same `textDocument/completion` request that seems to be crashing for you)......