Github Document Loader

​

Node Details

• Name: Github_DocumentLoaders
• Type: Document
• Version: 2.0
• Category: Document Loaders

​

Parameters

​

Credential (Optional)

• Label: Connect Credential
• Name: credential
• Type: credential
• Description: Only needed when accessing a private repo
• Credential Names: githubApi

​

Inputs

1. Repo Link (required)
   • Type: string
   • Description: The URL of the GitHub repository
   • Example: https://github.com/Ardor_Cerebrum/Ardor
2. Branch (required)
   • Type: string
   • Default: “main”
   • Description: The branch of the repository to load from
3. Recursive (optional)
   • Type: boolean
   • Description: Whether to recursively traverse the repository
4. Max Concurrency (optional)
   • Type: number
   • Description: Maximum number of concurrent operations
5. Ignore Paths (optional)
   • Type: string (JSON array)
   • Description: An array of paths to be ignored
   • Example: [“*.md”]
6. Max Retries (optional)
   • Type: number
   • Description: Maximum number of retries for a single call
   • Default: 2
7. Text Splitter (optional)
   • Type: TextSplitter
   • Description: A text splitter to apply to the loaded documents
8. Additional Metadata (optional)
   • Type: JSON
   • Description: Additional metadata to be added to the extracted documents
9. Omit Metadata Keys (optional)
   • Type: string
   • Description: Comma-separated list of metadata keys to omit from the documents

​

Functionality

1. Initializes a GithubRepoLoader with the provided repository link and options.
2. Loads documents from the specified GitHub repository.
3. Applies text splitting if a TextSplitter is provided.
4. Adds custom metadata to the documents if specified.
5. Omits specified metadata keys from the documents.

​

Output

​

Use Cases

• Loading source code for analysis or processing
• Fetching documentation or README files from GitHub projects
• Integrating GitHub-hosted content into document processing workflows
• Analyzing changes across different branches of a repository

​

Notes

• When accessing private repositories, make sure to provide the appropriate GitHub API credentials.
• The node supports various options for customizing the loading process, including recursive traversal, concurrency control, and retry logic.
• Use the text splitter option to break down large documents into smaller chunks if needed.
• The additional metadata and metadata key omission features allow for fine-grained control over the document metadata.