Techatrix

> does this fix #1071 since files are more eagerly analyzed? We still search for references in dependencies instead of dependents which this PR doesn't fix. But eagerly opening files...

There is no conflict so I don't see why that would be necessary.

That's good to know. When a client only supports full semantic tokens then we can continue to provide it so that we don't lose semantic tokens. The more interesting case...

> fyi - it seems that can't really be relied upon > > https://gist.github.com/dotcarmen/7f186613f1f839427a839c33306af9dd what happens when a module is imported by multiple roots? Resolve it to all of them....

The stack trace appears to point to the following line in the extensions code: https://github.com/ziglang/vscode-zig/blob/59b1a6c3f31e7a166534c121b2ba007f05276fa9/src/zigSetup.ts#L92 This section of the code takes a non `-dev` version of Zig like `0.14.0` and...

It was a deliberate to use `mutable` over `readonly`. The reasoning can be found in https://github.com/zigtools/zls/pull/2251. If you still think that using `readonly` is the better choice, you are welcome...

To avoid any confusion, I have to clarify that my usage of the term "standard" and "non-standard" token types/modifers can be misleading. I should rather call them "non-predefined" or "custom"...

> I've reverted that change because the rest is otherwise valid, plus someone building ZLS from scratch will probably be using a newer Zig version anyway. Once the binaries start...

Is this issue still happening? If so, please provide more information on how the editor has been configured (plugins, config files, etc.). I had previously tested this with relatively bare...

So I guess there is nothing to do here. Closing.