This PR fixes a panic that occurs when formatting files from VSCode if the file contains multi-byte characters (e.g.,→,🐈 or Japanese characters like日本語) and ends with 2+ newlines.

Problem

The panic occurred in thePositionToLineAndCharacter function ininternal/ls/converters.go when trying to slice text with bounds[start:position] whereposition could exceed the actual text length:

panic: runtime error: slice bounds out of range [:10] with length 9

Root Cause

When the formatter generates text edits for files with multi-byte characters and trailing newlines, it can create ranges that extend beyond the actual text length. ThePositionToLineAndCharacter function didn't handle this case properly, leading to a "slice bounds out of range" panic.

Solution

Added a bounds check to ensure thatposition never exceeds the actual text length before attempting to slice the text:

// Ensure position doesn't exceed text length to avoid slice bounds errorstext:=script.Text()textLen:=core.TextPos(len(text))position=min(position,textLen)

This ensures that even if the formatter or other components pass positions beyond the text length, the converter will handle them gracefully by treating them as positions at the end of the text.

Testing

• Added comprehensive tests that specifically reproduce the original issue
• Verified the fix handles positions at and beyond text length
• Added a specific test case for the exact scenario from the issue
• All existing tests continue to pass

The fix is minimal, safe, and maintains backward compatibility while resolving the panic.

Fixes#1389.

💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Clickhere to start the survey.