fix(language_server): Replace invalid LSP_MAX_INT ranges with zero ranges

[Boshen]

Summary

Fixes "Range#create called with invalid arguments" error reported in oxc-project/coc-oxc#50

When diagnostics had no related span information, the language server was using LSP_MAX_INT (2147483647) as a fallback position, which is invalid for LSP ranges and caused editors to reject the diagnostic.

Changes

• Changed the fallback range from LSP_MAX_INT to (0, 0) when no span information is available
• Removed unused LSP_MAX_INT constant
• Updated test snapshot to reflect the new valid range behavior

Test Plan

• All existing tests pass
• Updated snapshot test confirms the fix works correctly
• The fix uses a zero range (0, 0) which is a standard practice for diagnostics without specific location information

Fixes oxc-project/coc-oxc#50

🤖 Generated with Claude Code

[Copilot]

Pull Request Overview

This PR fixes an LSP protocol violation where invalid range values were causing diagnostic errors in editors. The language server was using LSP_MAX_INT (2147483647) as fallback positions for diagnostics without span information, which exceeded LSP specification limits.

• Replace invalid LSP_MAX_INT fallback ranges with valid zero ranges (0, 0)
• Remove the unused LSP_MAX_INT constant from the codebase
• Update range comparison logic to use u32::MAX for internal comparisons while maintaining valid LSP output

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
main.rs	Removes the unused LSP_MAX_INT constant definition
error_with_position.rs	Updates fallback range logic to use zero ranges and fixes comparison initialization
debugger.ts.snap	Updates test snapshot to reflect the new valid zero range behavior

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[graphite-app]

How to use the Graphite Merge Queue

Add either label to this PR to merge it via the merge queue:

• 0-merge - adds this PR to the back of the merge queue
• hotfix - for urgent hot fixes, skip the queue and merge this PR next

You must have a Graphite account in order to use the merge queue. Sign up using this link.

An organization admin has enabled the Graphite Merge Queue in this repository.

Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.

[Boshen]

@Sysix not entirely sure if this is the right fix, or should we filter out these invalid spans

[graphite-app]

The use of u32::MAX (4294967295) here is inconsistent with the PR's approach of using (0, 0) for invalid ranges. While this works for comparison purposes, u32::MAX still exceeds the LSP specification's maximum value (2^31 - 1).

Consider an alternative approach for the initial comparison:

• Use an Option<Range> that starts as None
• Use a boolean flag to track if a valid range has been found
• Or initialize with the first range found, then compare subsequent ranges

This would maintain consistency with the PR's goal of eliminating invalid range values throughout the codebase.

Suggested change

	start: Position { line: u32::MAX, character: u32::MAX },
	end: Position { line: u32::MAX, character: u32::MAX },
	start: Position { line: 0, character: 0 },
	end: Position { line: 0, character: 0 },

Spotted by Diamond



Is this helpful? React 👍 or 👎 to let us know.

[Sysix]

Yes! IntelliJ expects to be in a valid LSP Range and will fail. We had it on u32::MAX and needed to change it for IntelliJ integration

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #13992 will not alter performance

_{Comparing fix/lsp-invalid-range (8e72e2a) with main (3ce0775)}

Summary

✅ 37 untouched

[Sysix]

I do not think the server is in fault of the nvim error.
I am fine to set everything to 0, but u32::MAX is not a valid integer in LSP.

Would close it in favor of #14010

[Sysix]

Yes! IntelliJ expects to be in a valid LSP Range and will fail. We had it on u32::MAX and needed to change it for IntelliJ integration

[Sysix]

next part is here: #14057