Asymmetric context

[mrT23]

PR Type

Enhancement, Tests

---

Description

• Implemented separate parameters for extending patches before and after changes (patch_extra_lines_before and patch_extra_lines_after)
• Updated extend_patch function and related PR processing logic to support the new patch extension parameters
• Added tests for the new patch extension functionality
• Updated configuration and documentation to reflect the new patch extension parameters
• Increased max_context_tokens for PR code suggestions from 10000 to 16000
• Added support for the 'claude-3-5-sonnet' model
• Enabled extra lines in code suggestions by default

---

Changes walkthrough 📝

	Relevant files
Configuration changes	__init__.py Add new Claude model                                                                          pr_agent/algo/init.py Added 'claude-3-5-sonnet' model with a token limit of 100000 +1/-0      configuration.toml Update configuration for patch extension                                  pr_agent/settings/configuration.toml Changed patch_extra_lines to separate patch_extra_lines_before and patch_extra_lines_after Increased max_context_tokens for PR code suggestions from 10000 to 16000 +3/-2
Enhancement	git_patch_processing.py Refactor patch extension logic                                                      pr_agent/algo/git_patch_processing.py Modified extend_patch function to accept separate parameters for lines before and after Updated logic to extend patches based on new parameters +8/-19    pr_processing.py Update PR diff generation with new patch extension              pr_agent/algo/pr_processing.py Updated get_pr_diff and pr_generate_extended_diff functions to use separate before and after line parameters Modified patch extension calls to use new parameters +15/-17  pr_code_suggestions.py Enable extra lines in code suggestions                                      pr_agent/tools/pr_code_suggestions.py Changed disable_extra_lines parameter to False in _prepare_prediction method +1/-1
Tests	test_extend_patch.py Update and add tests for patch extension                                  tests/unittest/test_extend_patch.py Updated existing tests to use new patch extension parameters Added new test class PRProcessingTest with tests for extended patches +65/-45
Documentation	additional_configurations.md Update documentation for patch extension config                    docs/docs/usage-guide/additional_configurations.md Updated configuration example to use separate patch_extra_lines_before and patch_extra_lines_after parameters +2/-1

---

💡 PR-Agent usage:
Comment /help on the PR to get a list of all available PR-Agent tools and their descriptions

[codiumai-pr-agent-pro]

PR Reviewer Guide 🔍

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪

🏅 Score: 85

🧪 PR contains tests

🔒 No security concerns identified

🔀 Multiple PR themes Sub-PR theme: Implement separate parameters for extending patches before and after changes Relevant files: pr_agent/algo/git_patch_processing.py pr_agent/algo/pr_processing.py tests/unittest/test_extend_patch.py Sub-PR theme: Update configuration and increase token limits for PR code suggestions Relevant files: pr_agent/algo/init.py pr_agent/tools/pr_code_suggestions.py pr_agent/settings/configuration.toml

⚡ Key issues to review Code Refactoring The extend_patch function has been significantly modified to support separate parameters for extending patches before and after changes. This change might affect other parts of the codebase that rely on this function. Configuration Update The PR processing logic has been updated to use separate parameters for extending patches before and after changes. Ensure that this change is consistently applied throughout the codebase.

[mrT23]

/improve

[mrT23]

PR Code Suggestions ✨

Category	Suggestion	Score
Enhancement	Add error handling and logging to the model setting function The set_claude_model() function modifies global settings. Consider adding a warning log or raising an exception if the model is not available or if there are any issues setting it. pr_agent/git_providers/utils.py [56-63] def set_claude_model(): """ set the claude-sonnet-3.5 model easily (even by users), just by stating: --config.model='claude-3-5-sonnet' """ model_claude = "bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0" - get_settings().set('config.model', model_claude) - get_settings().set('config.model_turbo', model_claude) - get_settings().set('config.fallback_models', [model_claude]) + try: + get_settings().set('config.model', model_claude) + get_settings().set('config.model_turbo', model_claude) + get_settings().set('config.fallback_models', [model_claude]) + get_logger().info(f"Successfully set Claude model to {model_claude}") + except Exception as e: + get_logger().warning(f"Failed to set Claude model: {e}") Suggestion importance[1-10]: 7 Why: This suggestion adds important error handling and logging, improving the robustness of the function.	7
Performance	Evaluate the need for extra lines in code suggestions to optimize performance The disable_extra_lines parameter is set to False, which means extra lines will be included. This might increase the token count and potentially affect performance. Consider evaluating if this is the intended behavior for code suggestions. pr_agent/tools/pr_code_suggestions.py [285-289] self.patches_diff = get_pr_diff(self.git_provider, self.token_handler, model, add_line_numbers_to_hunks=True, - disable_extra_lines=False) + disable_extra_lines=True) # Consider disabling extra lines for code suggestions Suggestion importance[1-10]: 5 Why: The suggestion raises a valid point about performance, but the optimal setting depends on the specific use case.	5
Best practice	Use a more descriptive variable name for the caught exception Consider using a more descriptive variable name instead of e in the exception handling. This improves code readability and makes it easier to understand the nature of the exception being caught. pr_agent/algo/git_patch_processing.py [63-65] -except Exception as e: +except Exception as error: if get_settings().config.verbosity_level >= 2: - get_logger().error(f"Failed to extend patch: {e}") + get_logger().error(f"Failed to extend patch: {error}") Suggestion importance[1-10]: 3 Why: The suggestion is correct but minor, improving code readability slightly.	3