src/services/repository_service.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

"""
Repository Service

This module provides functionality for cloning and managing Git repositories.
"""

import os
import shutil
import tempfile
import logging
import re
from git import Repo
from git.exc import GitCommandError

logger = logging.getLogger(__name__)

class RepositoryService:
    """
    Service for cloning and managing Git repositories.
    """
    
    def __init__(self, base_temp_dir=None):
        """
        Initialize the RepositoryService.
        
        Args:
            base_temp_dir (str, optional): Base directory for temporary repositories.
                If None, system temp directory will be used.
        """
        self.base_temp_dir = base_temp_dir or tempfile.gettempdir()
        self.repos = {}
        logger.info(f"Initialized RepositoryService with base temp dir: {self.base_temp_dir}")
    
    def validate_github_url(self, url):
        """
        Validate if the provided URL is a valid GitHub repository URL.
        
        Args:
            url (str): The GitHub repository URL to validate.
        
        Returns:
            bool: True if the URL is valid, False otherwise.
        """
        # GitHub URL patterns
        patterns = [
            r'^https?://github\.com/[\w.-]+/[\w.-]+(\.git)?$',  # https://github.com/user/repo[.git]
            r'^git@github\.com:[\w.-]+/[\w.-]+(\.git)?$',        # git@github.com:user/repo[.git]
        ]
        
        for pattern in patterns:
            if re.match(pattern, url):
                return True
        
        return False
    
    def normalize_github_url(self, url):
        """
        Normalize a GitHub URL to a consistent format.
        
        Args:
            url (str): The GitHub repository URL to normalize.
        
        Returns:
            str: The normalized URL.
        """
        # Convert SSH URL to HTTPS URL
        if url.startswith('git@github.com:'):
            user_repo = url[len('git@github.com:'):]
            if user_repo.endswith('.git'):
                user_repo = user_repo[:-4]
            return f"https://github.com/{user_repo}"
        
        # Ensure HTTPS URL ends without .git
        if url.startswith('http'):
            if url.endswith('.git'):
                return url[:-4]
        
        return url
    
    def extract_repo_name(self, url):
        """
        Extract repository name from a GitHub URL.
        
        Args:
            url (str): The GitHub repository URL.
        
        Returns:
            str: The repository name.
        """
        normalized_url = self.normalize_github_url(url)
        return normalized_url.split('/')[-1]
    
    def clone_repository(self, url, branch=None):
        """
        Clone a Git repository from the provided URL.
        
        Args:
            url (str): The repository URL to clone.
            branch (str, optional): The branch to checkout. If None, the default branch is used.
        
        Returns:
            str: The path to the cloned repository.
        
        Raises:
            ValueError: If the URL is not a valid GitHub repository URL.
            GitCommandError: If there's an error during the Git operation.
        """
        if not self.validate_github_url(url):
            raise ValueError(f"Invalid GitHub repository URL: {url}")
        
        repo_name = self.extract_repo_name(url)
        repo_dir = os.path.join(self.base_temp_dir, f"codereview_{repo_name}_{os.urandom(4).hex()}")
        
        logger.info(f"Cloning repository {url} to {repo_dir}")
        
        try:
            # Clone the repository
            if branch:
                repo = Repo.clone_from(url, repo_dir, branch=branch)
                logger.info(f"Cloned repository {url} (branch: {branch}) to {repo_dir}")
            else:
                repo = Repo.clone_from(url, repo_dir)
                logger.info(f"Cloned repository {url} (default branch) to {repo_dir}")
            
            # Store the repository instance
            self.repos[repo_dir] = repo
            
            return repo_dir
        
        except GitCommandError as e:
            logger.error(f"Error cloning repository {url}: {e}")
            # Clean up the directory if it was created
            if os.path.exists(repo_dir):
                shutil.rmtree(repo_dir, ignore_errors=True)
            raise
    
    def get_repository_info(self, repo_path):
        """
        Get information about a repository.
        
        Args:
            repo_path (str): The path to the repository.
        
        Returns:
            dict: A dictionary containing repository information.
        """
        if repo_path not in self.repos:
            try:
                self.repos[repo_path] = Repo(repo_path)
            except Exception as e:
                logger.error(f"Error opening repository at {repo_path}: {e}")
                return {}
        
        repo = self.repos[repo_path]
        
        try:
            # Get the active branch
            try:
                active_branch = repo.active_branch.name
            except TypeError:
                # Detached HEAD state
                active_branch = 'HEAD detached'
            
            # Get the latest commit
            latest_commit = repo.head.commit
            
            # Get remote URL
            try:
                remote_url = repo.remotes.origin.url
            except AttributeError:
                remote_url = 'No remote URL found'
            
            # Get repository size (approximate)
            repo_size = sum(os.path.getsize(os.path.join(dirpath, filename))
                           for dirpath, _, filenames in os.walk(repo_path)
                           for filename in filenames)
            
            # Count files
            file_count = sum(len(files) for _, _, files in os.walk(repo_path))
            
            return {
                'path': repo_path,
                'active_branch': active_branch,
                'latest_commit': {
                    'hash': latest_commit.hexsha,
                    'author': f"{latest_commit.author.name} <{latest_commit.author.email}>",
                    'date': latest_commit.committed_datetime.isoformat(),
                    'message': latest_commit.message.strip(),
                },
                'remote_url': remote_url,
                'size_bytes': repo_size,
                'file_count': file_count,
            }
        
        except Exception as e:
            logger.error(f"Error getting repository info for {repo_path}: {e}")
            return {
                'path': repo_path,
                'error': str(e),
            }
    
    def cleanup_repository(self, repo_path):
        """
        Clean up a cloned repository.
        
        Args:
            repo_path (str): The path to the repository to clean up.
        
        Returns:
            bool: True if the cleanup was successful, False otherwise.
        """
        logger.info(f"Cleaning up repository at {repo_path}")
        
        # Remove the repository from the tracked repos
        if repo_path in self.repos:
            del self.repos[repo_path]
        
        # Remove the directory
        try:
            if os.path.exists(repo_path):
                shutil.rmtree(repo_path, ignore_errors=True)
            return True
        except Exception as e:
            logger.error(f"Error cleaning up repository at {repo_path}: {e}")
            return False
    
    def cleanup_all_repositories(self):
        """
        Clean up all cloned repositories.
        
        Returns:
            bool: True if all cleanups were successful, False otherwise.
        """
        logger.info("Cleaning up all repositories")
        
        success = True
        for repo_path in list(self.repos.keys()):
            if not self.cleanup_repository(repo_path):
                success = False
        
        return success