pygount

Warnings

• breaking Pygount versions 3.x have removed support for older Python versions. Specifically, Python 3.7 and 3.8 are no longer supported. Ensure your environment uses Python 3.10 or newer.
  Fix: Upgrade your Python interpreter to version 3.10 or higher.
• gotcha The programmatic API (e.g., `SourceAnalysis`, `ProjectSummary`) is explicitly stated as 'work in progress and subject to change'. Breaking changes to the API might occur in future versions.
  Fix: Refer to the official documentation and 'Changes' log for each new release before upgrading when using the API programmatically. Pin `pygount` to a specific version if API stability is critical.
• gotcha Pygount automatically excludes certain files and folders from analysis by default. This includes files starting with a dot (`.`), files ending with a tilde (`~`), and folders named `.svn` or starting with a dot (`.`).
  Fix: Use the `--folders-to-skip` and `--names-to-skip` command-line options to override or extend the default exclusion patterns if you need to include such files/folders in your analysis.
• gotcha While robust, pygount can be notably slower than other command-line SLOC counting tools due to its detailed `pygments`-based parsing.
  Fix: Consider `pygount` for accuracy and language coverage rather than raw speed. For very large codebases where performance is critical, profile `pygount`'s execution or consider alternative tools if its speed is insufficient.

Install

• pip install pygount Install stable release

Imports

• SourceAnalysis

  from pygount import SourceAnalysis

• ProjectSummary

  from pygount import ProjectSummary

  While `pygount.source_analysis` is still available, the object-oriented API (`SourceAnalysis` and `ProjectSummary`) is the recommended approach for programmatic use, though still considered 'work in progress' and subject to change.

Quickstart

This quickstart demonstrates how to use `pygount`'s API to programmatically analyze source code files within a directory and summarize the results. It creates dummy files, analyzes them using `SourceAnalysis` and aggregates the results with `ProjectSummary`, then prints the language-specific counts.

import os
from pathlib import Path
from pygount import SourceAnalysis, ProjectSummary

# Create dummy files for demonstration
dummy_dir = Path('./my_project')
dummy_dir.mkdir(exist_ok=True)
(dummy_dir / 'main.py').write_text(
    """# This is a Python file
import sys

def hello(): # Code line
    print('Hello, world!') # Code line
    # Another comment
"""
)
(dummy_dir / 'README.md').write_text(
    """# My Project

This is a test project.
"""
)
(dummy_dir / '.hidden_file.txt').write_text("should not be counted")

project_summary = ProjectSummary()

# Analyze files in the dummy project directory
for source_path in dummy_dir.rglob('*'):
    if source_path.is_file():
        try:
            source_analysis = SourceAnalysis.from_file(source_path, 'my_project')
            project_summary.add(source_analysis)
        except UnicodeDecodeError as e:
            print(f"Warning: Could not decode {source_path}: {e}")

print("\n--- Analysis Results ---")
for language_summary in project_summary.language_to_language_summary_map.values():
    print(language_summary)

# Clean up dummy files
import shutil
shutil.rmtree(dummy_dir)