Skip to content

Examples

Siphon CLI offers powerful options for managing and extracting Git repository contents efficiently. Below are detailed examples to showcase its versatility in various scenarios.

Example 1: Include Specific File Types

Extract all Python files from a Git repository and save the output in python_files.txt.

si -i "*.py" -o python_files.txt

Use Cases:

  • Ideal when you want to extract only code files for analysis.
  • Combine file types or directories to target specific file patterns.

Additional Example:

Include Python and Markdown files:

si -i "*.py" "*.md" -o code_and_docs.txt

Example 2: Exclude Directories or Files

Exclude specific directories or file types from the output. In this example, all files are included except those in the tests/ directory.

si -e "tests/*" -o no_tests.txt

Use Cases:

  • Skipping tests or documentation files in favor of core code.
  • Excluding large, unnecessary directories to reduce output size.

Additional Example:

Exclude temporary files and logs:

si -e "*.log" "*.tmp" -o clean_output.txt

Example 3: Interactive File Selection

Interactive mode allows you to choose files interactively during extraction. Siphon will prompt you for each file in the repository.

si --interactive -o selected_files.txt

Use Cases:

  • Great for when you only need to extract a few important files manually.
  • Useful when you're unsure of exact filenames and want to decide as you go.

Example 4: Creating a Compressed Archive

Package your repository contents into a .tar archive. This example compresses the output into a tarball file.

si -f tar -o repo_archive.tar

Use Cases:

  • Perfect for packaging a repository to send to another developer or store for later use.
  • Compressing large codebases or datasets into a single archive for easy storage.

Example 5: Estimating Token Counts for Large Language Models (LLMs)

Use the --tokenizer option to estimate token counts for specific LLMs, which is useful for determining how much text can be processed by models like GPT-3 or Claude.

si --tokenizer gpt3 -o context_with_tokens.txt

Use Cases:

  • Useful for preparing repository content to fit within token limits for LLM-based tools.
  • Estimate the number of tokens in the extracted content for accurate LLM queries.

Additional Example:

Estimate tokens for Claude:

si --tokenizer claude -o claude_context.txt

Example 6: Copy Output to Clipboard

Quickly copy the output to the clipboard instead of saving it to a file, which is especially useful in workflows where the output needs to be pasted directly into another tool.

si --clipboard

Use Cases:

  • Streamline workflows by directly copying extracted content to share in chats, emails, or documentation.
  • Quick access to repository contents without creating a file.

Example 7: Print to Standard Output

Print the extracted content to standard output, which is useful for quickly viewing or piping the results into other command-line tools.

si --stdout

Use Cases:

  • Ideal for quick inspection or further processing with tools like grep or sed.
  • Viewing output without writing to a file.

Additional Example:

Pipe the output into grep to filter for specific content:

si --stdout | grep "def"

This will extract the content and show only lines containing the def keyword (e.g., Python function definitions).

Example 8: Combining Multiple Options

You can combine multiple options for more complex workflows. In this example, Siphon extracts Python files, excludes test files, and copies the output to the clipboard.

si -i "*.py" -e "tests/*" --clipboard

Use Cases:

  • Combine various filters and output options for maximum flexibility.
  • Extract only relevant parts of a large repository and make the content available in your clipboard instantly.

Example 9: Caching and Future Features

Siphon’s future features will include caching for faster repeated extractions. You can enable caching now with the --cache option, though it's a placeholder for upcoming functionality.

si --cache -o cached_output.txt

Use Cases:

  • Speed up frequent extractions in large repositories.
  • Prepare for future updates to Siphon that will support more robust caching mechanisms.