Nathan Ridge

Results 1499 comments of Nathan Ridge

Implementation of onTypeFormatting assumes that the position in the request is *after* the inserted character

> Does clangd assumes that the newline has already been written, meaning the newline has already been inserted into the document? Ah, yes, clangd does assume that a `didChange` for...

clang(invalid_consteval_call) false positive call

Please share a complete code example that demonstrates the issue.

clangd suggests to include <cstdlib> instead of <stdlib.h> in .c files

Likely duplicate of https://github.com/clangd/clangd/issues/376

clangd suggests to include <cstdlib> instead of <stdlib.h> in .c files

@kglundgren is this fixed for you in clangd 20 or later? (#376 was fixed in clangd 20)

Merging settings from the custom "settings.json" with those from "compile_flags.txt", "compile_commands.json", and ".clangd".

The relationship between these four sources of settings is as follows: * `compile_flags.txt` and `compile_commands.json` are two different supported formats for a "compilation database", which specifies what compile flags were...

vscode-clangd cannot navigate to __nanosleep

Failure to navigate is caused by compiler errors in the code (which is also shown in your screenshot), which in turn is usually caused by a configuration issue. In this...

vscode-clangd cannot navigate to __nanosleep

The first thing I look for in the logs in a situation like this is, what is the **first** error diagnostic that the server produces. That is usually what needs...

Ability to move a project's clangd background index without triggering reindexing

> 2. I tried using the [Remote Index](https://clangd.llvm.org/design/remote-index.html) option which works great if we don't make any changes in the "copied_project" but I am not sure what would happen when...

Ability to move a project's clangd background index without triggering reindexing

See also https://github.com/clangd/clangd/issues/587 which is very similar

Ability to move a project's clangd background index without triggering reindexing

> Has there been any progress on reusing background index data? Not that I'm aware of. > I'm imagining some sort of mapping option such that the initial background index...