Refactoring AI Handlers and Adding Support for Multiple AI Models

[mrT23]

User description

Type

Enhancement

---

Description

This PR introduces significant changes to the AI handling system of the PR Agent. The main changes include:

• Refactoring of the AI handler classes, introducing a new abstract base class BaseAiHandler to define a common interface for all AI handlers.
• Addition of new AI handlers: OpenAIHandler, LangChainOpenAIHandler, and LiteLLMAIHandler. These handlers are designed to interact with different AI models.
• Modification of the PRAgent and related classes to support the new AI handlers. The AI handler to be used can now be injected into these classes upon instantiation.
• Addition of the langchain package to the requirements (commented out as it's optional).
• Minor changes in logging and error handling.

---

PR changes walkthrough

	Relevant files
Enhancement	13 files base_ai_handler.py                                                                                    pr_agent/algo/ai_handlers/base_ai_handler.py Introduced a new abstract base class `BaseAiHandler` that defines a common interface for all AI handlers. This class includes an abstract `chat_completion` method that all subclasses must implement. +28/-0 openai_ai_handler.py                                                                                pr_agent/algo/ai_handlers/openai_ai_handler.py Added a new `OpenAIHandler` class that extends `BaseAiHandler`. This class is designed to interact with the OpenAI API for chat completions. +67/-0 langchain_ai_handler.py                                                                          pr_agent/algo/ai_handlers/langchain_ai_handler.py Added a new `LangChainOpenAIHandler` class that extends `BaseAiHandler`. This class is designed to interact with the LangChain AI model. +49/-0 litellm_ai_handler.py                                                                              pr_agent/algo/ai_handlers/litellm_ai_handler.py Renamed `AiHandler` to `LiteLLMAIHandler` and made it extend `BaseAiHandler`. This class is designed to interact with the LiteLLM AI model. +3/-2 pr_agent.py                                                                                                  pr_agent/agent/pr_agent.py Modified the `PRAgent` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +10/-5 pr_reviewer.py                                                                                            pr_agent/tools/pr_reviewer.py Modified the `PRReviewer` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +7/-3 pr_description.py                                                                                      pr_agent/tools/pr_description.py Modified the `PRDescription` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +5/-3 pr_generate_labels.py                                                                              pr_agent/tools/pr_generate_labels.py Modified the `PRGenerateLabels` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +5/-3 pr_code_suggestions.py                                                                            pr_agent/tools/pr_code_suggestions.py Modified the `PRCodeSuggestions` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +5/-3 pr_update_changelog.py                                                                            pr_agent/tools/pr_update_changelog.py Modified the `PRUpdateChangelog` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +4/-3 pr_add_docs.py                                                                                            pr_agent/tools/pr_add_docs.py Modified the `PRAddDocs` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +5/-3 pr_information_from_user.py                                                                  pr_agent/tools/pr_information_from_user.py Modified the `PRInformationFromUser` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +5/-3 pr_questions.py                                                                                          pr_agent/tools/pr_questions.py Modified the `PRQuestions` class to support the new AI handlers. The AI handler to be used can now be injected into this class upon instantiation. +4/-3
Dependencies	1 files requirements.txt                                                                                        requirements.txt Added the `langchain` package to the requirements (commented out as it's optional). +1/-0

---

Type

enhancement

---

Description

• Introduced a new AI handling system with an abstract base class BaseAiHandler and specific implementations for LiteLLM (LiteLLMAIHandler), OpenAI (OpenAIHandler), and LangChain (LangChainOpenAIHandler).
• Refactored PRAgent and related tools (PRAddDocs, PRCodeSuggestions, etc.) to use dependency injection for AI handlers, allowing for flexible AI model usage.
• Added error handling and retry logic in AI handlers for robust operation.
• Made LangChain dependency optional and updated requirements.txt accordingly.

---

Changes walkthrough

	Relevant files
Enhancement	pr_agent.py Refactor PRAgent to Use New AI Handler System                                    pr_agent/agent/pr_agent.py Introduced BaseAiHandler and LiteLLMAIHandler for AI handling. Added ai_handler parameter to PRAgent constructor for dependency injection. Modified handle_request method to use ai_handler for AI operations. +9/-5      base_ai_handler.py Introduce BaseAiHandler Abstract Class                                                  pr_agent/algo/ai_handlers/base_ai_handler.py Created an abstract base class BaseAiHandler defining the interface for AI handlers. Defined abstract methods for AI operations. +28/-0    langchain_ai_handler.py Add LangChain AI Handler Implementation                                                pr_agent/algo/ai_handlers/langchain_ai_handler.py Implemented LangChainOpenAIHandler extending BaseAiHandler. Added error handling and retry logic for AI operations. +49/-0    litellm_ai_handler.py Update LiteLLM AI Handler to New Interface                                          pr_agent/algo/ai_handlers/litellm_ai_handler.py Renamed AiHandler to LiteLLMAIHandler and made it extend BaseAiHandler. Updated chat_completion method to align with the new interface. +3/-2      openai_ai_handler.py Implement OpenAI Handler                                                                              pr_agent/algo/ai_handlers/openai_ai_handler.py Introduced OpenAIHandler class extending BaseAiHandler. Implemented AI operations with OpenAI API, including error handling and retries. +67/-0    pr_add_docs.py Refactor PRAddDocs to Use Injected AI Handler                                    pr_agent/tools/pr_add_docs.py Added ai_handler parameter to constructor for dependency injection. Updated AI operations to use the injected AI handler. +5/-3
Miscellaneous	utils.py Minor Formatting Fix in Utils                                                                    pr_agent/algo/utils.py Minor formatting change in clip_tokens function. +1/-1
Configuration changes	requirements.txt Update Requirements with Optional LangChain Dependency                  requirements.txt Added commented-out langchain package to requirements. +1/-0

---

✨ Usage guide:

---

Overview:
The describe tool scans the PR code changes, and generates a description for the PR - title, type, summary, walkthrough and labels. The tool can be triggered automatically every time a new PR is opened, or can be invoked manually by commenting on a PR.

When commenting, to edit configurations related to the describe tool (pr_description section), use the following template:

/describe --pr_description.some_config1=... --pr_description.some_config2=...

With a configuration file, use the following template:

[pr_description]
some_config1=...
some_config2=...

Enabling\disabling automation When you first install the app, the default mode for the describe tool is: pr_commands = ["/describe --pr_description.add_original_user_description=true" "--pr_description.keep_original_user_title=true", ...] meaning the describe tool will run automatically on every PR, will keep the original title, and will add the original user description above the generated description. Markers are an alternative way to control the generated description, to give maximal control to the user. If you set: pr_commands = ["/describe --pr_description.use_description_markers=true", ...] the tool will replace every marker of the form pr_agent:marker_name in the PR description with the relevant content, where marker_name is one of the following: type: the PR type. summary: the PR summary. walkthrough: the PR walkthrough. Note that when markers are enabled, if the original PR description does not contain any markers, the tool will not alter the description at all.

Custom labels The default labels of the describe tool are quite generic: [Bug fix, Tests, Enhancement, Documentation, Other]. If you specify custom labels in the repo's labels page or via configuration file, you can get tailored labels for your use cases. Examples for custom labels: Main topic:performance - pr_agent:The main topic of this PR is performance New endpoint - pr_agent:A new endpoint was added in this PR SQL query - pr_agent:A new SQL query was added in this PR Dockerfile changes - pr_agent:The PR contains changes in the Dockerfile ... The list above is eclectic, and aims to give an idea of different possibilities. Define custom labels that are relevant for your repo and use cases. Note that Labels are not mutually exclusive, so you can add multiple label categories. Make sure to provide proper title, and a detailed and well-phrased description for each label, so the tool will know when to suggest it.

Inline File Walkthrough 💎 For enhanced user experience, the describe tool can add file summaries directly to the "Files changed" tab in the PR page. This will enable you to quickly understand the changes in each file, while reviewing the code changes (diffs). To enable inline file summary, set pr_description.inline_file_summary in the configuration file, possible values are: 'table': File changes walkthrough table will be displayed on the top of the "Files changed" tab, in addition to the "Conversation" tab. true: A collapsable file comment with changes title and a changes summary for each file in the PR. false (default): File changes walkthrough will be added only to the "Conversation" tab.

Utilizing extra instructions The describe tool can be configured with extra instructions, to guide the model to a feedback tailored to the needs of your project. Be specific, clear, and concise in the instructions. With extra instructions, you are the prompter. Notice that the general structure of the description is fixed, and cannot be changed. Extra instructions can change the content or style of each sub-section of the PR description. Examples for extra instructions: [pr_description] extra_instructions=""" - The PR title should be in the format: '<PR type>: <title>' - The title should be short and concise (up to 10 words) - ... """ Use triple quotes to write multi-line instructions. Use bullet points to make the instructions more readable.

More PR-Agent commands To invoke the PR-Agent, add a comment using one of the following commands: /review: Request a review of your Pull Request. /describe: Update the PR title and description based on the contents of the PR. /improve [--extended]: Suggest code improvements. Extended mode provides a higher quality feedback. /ask <QUESTION>: Ask a question about the PR. /update_changelog: Update the changelog based on the PR's contents. /add_docs 💎: Generate docstring for new components introduced in the PR. /generate_labels 💎: Generate labels for the PR based on the PR's contents. /analyze 💎: Automatically analyzes the PR, and presents changes walkthrough for each component. See the tools guide for more details. To list the possible configuration parameters, add a /config comment.

See the describe usage page for a comprehensive guide on using this tool.

[hussam789]

/describe

[codiumai-pr-agent-pro]

PR Description updated to latest commit (02871b1)

• Copy walkthrough table to "Files Changed" Tab