fix: improved path detection in prompts, now works with more adjacent punctuation types

[ErikBjare]

… punctuation types

@Miyou reported this when it didn't read a file that he had followed with a ,

---

Important

Improved path detection in prompts to handle more punctuation types and added tests for this functionality.

• Behavior:
  • Improved path detection in _find_potential_paths() in chat.py to handle paths with adjacent punctuation (e.g., !, ?, ,, .).
  • Excludes paths within code blocks from detection.
• Testing:
  • Added test_find_potential_paths() in test_chat.py to verify detection of various path formats and exclusion of paths in code blocks.
  • Added test_find_potential_paths_empty() to check behavior with empty content.
  • Added test_find_potential_paths_punctuation() to test path detection with trailing punctuation.

^{This description was created by for c353e4c. It will automatically update as commits are pushed.}

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to c353e4c in 1 minute and 6 seconds

More details

• Looked at 215 lines of code in 2 files
• Skipped 0 files when reviewing.
• Skipped posting 2 drafted comments based on config settings.

1. gptme/chat.py:316

• Draft comment:
  Consider using a single rstrip call to remove all punctuation characters at once for better performance and readability.

word = word.rstrip('?.!,')

• Reason this comment was not posted:
  Confidence changes required: 50%
  The _find_potential_paths function uses multiple rstrip calls to remove punctuation, which can be optimized by using a single rstrip call with all characters.

2. gptme/chat.py:325

• Draft comment:
  Consider using a single rstrip call to remove all punctuation characters at once for better performance and readability.

word = word.rstrip('?.!,')

• Reason this comment was not posted:
  Confidence changes required: 50%
  The _find_potential_paths function uses multiple rstrip calls to remove punctuation, which can be optimized by using a single rstrip call with all characters. This is also applicable in another part of the function.

Workflow ID: wflow_LQ7agiPK6YqBzSQI

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 96.66667% with 1 line in your changes missing coverage. Please review.

Project coverage is 72.88%. Comparing base (d598ff3) to head (c353e4c).
Report is 1 commits behind head on master.

✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
gptme/chat.py	96.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #333      +/-   ##
==========================================
+ Coverage   72.77%   72.88%   +0.10%     
==========================================
  Files          67       67              
  Lines        4951     4967      +16     
==========================================
+ Hits         3603     3620      +17     
+ Misses       1348     1347       -1     

Flag	Coverage Δ
anthropic/claude-3-haiku-20240307	70.92% <96.66%> (+0.19%)	⬆️
openai/gpt-4o-mini	69.51% <96.66%> (-0.07%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.