Add PR help message functionality and update dependencies

[mrT23]

User description

• Implement PRHelpMessage class to provide AI-powered assistance for pull requests.
• Add methods for similarity search using local, S3, and Pinecone databases.
• Update requirements.txt to include new dependencies for langchain and chromadb.
• Modify configuration.toml to include force_local_db setting for PR help.
• Update aiohttp and openai package versions.

---

PR Type

Enhancement

---

Description

• Implemented AI-powered PR help functionality in the PRHelpMessage class
• Added methods for similarity search using local, S3, and Pinecone databases
• Improved error handling and logging throughout the PR help process
• Updated PR help message generation to support different Git providers
• Added 'force_local_db' configuration option in configuration.toml
• Updated aiohttp and openai package versions in requirements.txt
• Added new dependencies for langchain, chromadb, and related packages to support AI-powered functionality

---

Changes walkthrough 📝

	Relevant files
Enhancement	pr_help_message.py Implement AI-powered PR help functionality                              pr_agent/tools/pr_help_message.py Implemented PRHelpMessage class with AI-powered assistance for pull requests Added methods for similarity search using local, S3, and Pinecone databases Improved error handling and logging Updated PR help message generation for different Git providers +340/-96
Configuration changes	configuration.toml Add force_local_db configuration option                                    pr_agent/settings/configuration.toml Added 'force_local_db' option to pr_help section +1/-0
Dependencies	requirements.txt Update dependencies and add new packages                                  requirements.txt Updated aiohttp and openai package versions Added new dependencies for langchain, chromadb, and related packages +8/-2

---

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

[mrT23]

/help "how to change the code suggestion to be commitable ?"

[codiumai-pr-agent-pro]

Question:

how to change the code suggestion to be commitable ?

Answer:

To make code suggestions commitable, you can use the --pr_code_suggestions.commitable_code_suggestions=true parameter when invoking the improve tool. You can do this manually by commenting on a PR with:

/improve --pr_code_suggestions.commitable_code_suggestions=true

For automatic runs, you can set this in the configuration file under the bitbucket_app section:

[bitbucket_app]
pr_commands = [
  "/improve --pr_code_suggestions.commitable_code_suggestions=true --pr_code_suggestions.suggestions_score_threshold=7",
]

This will present the suggestions as commitable code comments instead of a single table comment.

Relevant Sources:

• https://pr-agent-docs.codium.ai/usage-guide/automations_and_usage#bitbucket-app
• https://pr-agent-docs.codium.ai/tools/improve#usage-tips

[Zrmikstri]

/help "how to use pr-agent with self hosted llm"

[codiumai-pr-agent-pro]

Question:

how to use pr-agent with self hosted llm

Answer:

To use PR-Agent with a self-hosted LLM, you need to self-host PR-Agent itself. Here's how to do it:

1. Acquire the necessary tokens:

   • An API key for your chosen language model (e.g., OpenAI key with GPT-4 access, or a key for another supported LLM)
   • A personal access token (classic) from your Git provider (GitHub, GitLab, or BitBucket) with repo scope

2. Choose a deployment method for self-hosting PR-Agent:

   • Locally
   • GitHub
   • GitLab
   • BitBucket
   • Azure DevOps

3. Follow the installation guide for your chosen deployment method.

When self-hosting PR-Agent, you use your own API keys, which means the interaction with the LLM provider is directly between you and them. PR-Agent doesn't send your code data to its servers, ensuring better privacy for your code.

For detailed installation instructions, refer to the PR-Agent documentation for your specific deployment method.

Relevant Sources:

• https://pr-agent-docs.codium.ai/installation/index#installation
• https://pr-agent-docs.codium.ai/overview/data_privacy#self-hosted-pr-agent
• https://pr-agent-docs.codium.ai/faq/index#faq

[mrT23]

/improve

[mrT23]

/review

[codiumai-pr-agent-pro]

PR Code Suggestions ✨

Category	Suggestion	Score
Error handling	Use more specific exception types for better error handling Consider using a more specific exception type instead of a bare except clause. This will make the error handling more precise and easier to debug. pr_agent/tools/pr_help_message.py [58-60] -except Exception as e: +except (jinja2.TemplateError, ValueError) as e: get_logger().error(f"Error while preparing prediction: {e}") return "" +except Exception as e: + get_logger().error(f"Unexpected error while preparing prediction: {e}") + return "" Apply this suggestion Suggestion importance[1-10]: 7 Why: This suggestion improves error handling by providing more specific exception types. It allows for more precise error identification and handling, which can aid in debugging and maintenance.	7
Organization best practice	Avoid using mutable default arguments in function definitions Instead of using a mutable default argument (args=None) in the init method, use None as the default and assign an empty list in the method body if args is None. This aligns with the company's best practice of avoiding mutable default arguments. pr_agent/tools/pr_help_message.py [35] def __init__(self, pr_url: str, args=None, ai_handler: partial[BaseAiHandler,] = LiteLLMAIHandler): + self.args = [] if args is None else args Apply this suggestion Suggestion importance[1-10]: 6 Why: This is a good practice to avoid potential bugs related to mutable default arguments. While the current code might not cause immediate issues, implementing this change can prevent future problems.	6
	Import the entire jinja2 module instead of individual components Instead of using individual imports from jinja2, consider importing the entire module. This aligns with the company's best practice for imports. pr_agent/tools/pr_help_message.py [8] -from jinja2 import Environment, StrictUndefined +import jinja2 Apply this suggestion Suggestion importance[1-10]: 3 Why: While this change aligns with some coding practices, it's a minor improvement that doesn't significantly impact functionality or readability. The current import style is also widely used and acceptable.	3
Best practice	Use a context manager for the S3 client to ensure proper resource management Consider using a context manager (with statement) for the S3 client to ensure proper resource management and cleanup. pr_agent/tools/pr_help_message.py [80-85] -s3 = boto3.client('s3') +with boto3.client('s3') as s3: + # Download the file from S3 to the temporary directory + bucket = 'pr-agent' + file_name = 'chroma_db.zip' + s3.download_file(bucket, file_name, local_file_path) -# Download the file from S3 to the temporary directory -bucket = 'pr-agent' -file_name = 'chroma_db.zip' -s3.download_file(bucket, file_name, local_file_path) - Apply this suggestion Suggestion importance[1-10]: 5 Why: Using a context manager for the S3 client is a good practice for resource management. However, in this case, the improvement is minor as the S3 client doesn't hold significant resources that require explicit cleanup.	5

[codiumai-pr-agent-pro]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 4 🔵🔵🔵🔵⚪

🏅 Score: 85

🧪 No relevant tests

🔒 Security concerns Sensitive information exposure: The code downloads files from an S3 bucket (pr-agent/chroma_db.zip) without any apparent authentication or authorization checks. This could potentially expose sensitive information if the S3 bucket is not properly secured or if the file contains any sensitive data. It's recommended to implement proper access controls and ensure that only authorized users can access this data.

🔀 Multiple PR themes Sub-PR theme: Implement AI-powered PR help functionality Relevant files: pr_agent/tools/pr_help_message.py Sub-PR theme: Update dependencies and configuration Relevant files: pr_agent/settings/configuration.toml requirements.txt

⚡ Recommended focus areas for review Error Handling The error handling in the similarity search methods (get_sim_results_from_s3_db, get_sim_results_from_local_db, get_sim_results_from_pinecone_db) could be improved. Currently, exceptions are caught and logged, but the methods continue execution, potentially with empty results. Consider adding more robust error handling or fallback mechanisms. Code Duplication There is significant code duplication in the run method when generating the PR comment for different Git providers. Consider refactoring this into separate methods or using a template to reduce duplication and improve maintainability. Configuration Management The code relies heavily on the global configuration (get_settings()). Consider passing necessary configuration as parameters to methods or using dependency injection to improve testability and reduce global state dependencies.

[mrT23]

/analyze

[codiumai-pr-agent-pro]

PR Analysis 🔬

• This screen contains a list of code components that were changed in this PR.
• You can initiate specific actions for each component, by checking the relevant boxes.
• After you check a box, the action will be performed automatically by PR-Agent.
• Results will appear as a comment on the PR, typically after 30-60 seconds.

file

Changed components

pr_help_message.py

Test Docs Improve Similar   extract_header (function)   +12/-0   Test Docs Improve Similar   __init__ (method of PRHelpMessage)   +13/-2   Test Docs Improve Similar   _prepare_prediction (method of PRHelpMessage)   +12/-0   Test Docs Improve Similar   parse_args (method of PRHelpMessage)   +6/-0   Test Docs Improve Similar   get_sim_results_from_s3_db (method of PRHelpMessage)   +29/-0   Test Docs Improve Similar   get_sim_results_from_local_db (method of PRHelpMessage)   +22/-0   Test Docs Improve Similar   get_sim_results_from_pinecone_db (method of PRHelpMessage)   +17/-0   Test Docs Improve Similar   run (method of PRHelpMessage)   +152/-87   Test Docs Improve Similar   prepare_relevant_snippets (method of PRHelpMessage)   +28/-0   Test Docs Improve Similar   generate_bbdc_table (function)   +18/-0

---

💡 Usage guide:

---

Using static code analysis capabilities, the analyze tool scans the PR code changes and find the code components (methods, functions, classes) that changed in the PR.

The tool can be triggered automatically every time a new PR is opened, or can be invoked manually by commenting on any PR:

/analyze

Language that are currently supported: Python, Java, C++, JavaScript, TypeScript, C#.
See more information about the tool in the docs.