Analyzing Git Commit History and Contributor Statistics with Python#

1
import git
2
import os
3

4
# Path to the Git repository
5
repo_path = '/path/to/your/repository'
6

7
# Ensure the path is a valid Git repository
8
if git.is_git_repository(repo_path) and os.path.exists(os.path.join(repo_path, '.git')):
9
    repo = git.Repo(repo_path)
10
    print(f"Successfully opened repository: {repo_path}")
11
else:
12
    print(f"Error: {repo_path} is not a valid Git repository.")
13
    repo = None

1
if repo:
2
    # Iterate through all commits in the default branch (usually 'HEAD')
3
    # You can specify a branch or commit hash to start from
4
    print("\nRecent commits:")
5
    for commit in repo.iter_commits('HEAD', max_count=10):
6
        print(f"SHA: {commit.hexsha[:7]}")
7
        print(f"Author: {commit.author.name}")
8
        print(f"Date: {commit.authored_date}")
9
        print(f"Message: {commit.summary}")
10
        print("-" * 20)

1
if repo:
2
    print("\nStats for a recent commit:")
3
    try:
4
        latest_commit = repo.head.commit
5
        print(f"Commit: {latest_commit.hexsha[:7]}")
6
        stats = latest_commit.stats.total
7
        print(f"Lines Added: {stats['insertions']}")
8
        print(f"Lines Deleted: {stats['deletions']}")
9
        print(f"Total Lines Changed: {stats['lines']}")
10
        # You can also access stats['files'] for per-file changes
11
    except Exception as e:
12
        print(f"Could not get stats for latest commit: {e}")

1
import git
2
import pandas as pd
3
from datetime import datetime
4
import os

1
repo_path = '/path/to/your/repository' # Replace with the actual path

1
try:
2
    if git.is_git_repository(repo_path) and os.path.exists(os.path.join(repo_path, '.git')):
3
        repo = git.Repo(repo_path)
4
        print(f"Loading repository from {repo_path}")
5
    else:
6
        print(f"Error: {repo_path} is not a valid Git repository or does not exist.")
7
        repo = None
8
except git.exc.InvalidGitRepositoryError:
9
    print(f"Error: {repo_path} is not a valid Git repository.")
10
    repo = None
11
except Exception as e:
12
    print(f"An unexpected error occurred: {e}")
13
    repo = None

1
commit_list = []
2

3
if repo:
4
    print("Extracting commit data...")
5
    try:
6
        # Iterate through all commits in the history
7
        # Consider adding filtering arguments like max_count, since, until if history is very large
8
        for commit in repo.iter_commits('HEAD'):
9
            try:
10
                # Attempt to get stats; this can sometimes fail for merge commits or empty commits
11
                stats = commit.stats.total
12
                insertions = stats.get('insertions', 0)
13
                deletions = stats.get('deletions', 0)
14
                total_lines = stats.get('lines', 0) # lines is insertions + deletions
15
            except Exception:
16
                # Handle cases where stats might not be available or raise errors
17
                insertions = 0
18
                deletions = 0
19
                total_lines = 0
20

21
            commit_data = {
22
                'hexsha': commit.hexsha,
23
                'author_name': commit.author.name,
24
                'author_email': commit.author.email,
25
                'committed_date': datetime.fromtimestamp(commit.committed_date),
26
                'message': commit.message.strip(),
27
                'insertions': insertions,
28
                'deletions': deletions,
29
                'total_lines_changed': total_lines
30
            }
31
            commit_list.append(commit_data)
32

33
        print(f"Extracted {len(commit_list)} commits.")
34

35
    except git.exc.GitCommandError as e:
36
        print(f"Git command failed during iteration: {e}")
37
    except Exception as e:
38
        print(f"An error occurred during commit extraction: {e}")

1
if commit_list:
2
    df = pd.DataFrame(commit_list)
3

4
    # Ensure committed_date is in datetime format for time-based analysis
5
    df['committed_date'] = pd.to_datetime(df['committed_date'])
6

7
    print("\nDataFrame created:")
8
    print(df.head())
9
    print(f"\nDataFrame shape: {df.shape}")
10
else:
11
    print("\nNo commit data to process.")
12
    df = pd.DataFrame() # Create an empty DataFrame if no data

1
import matplotlib.pyplot as plt
2
import seaborn as sns
3

4
# Configure matplotlib for better display
5
plt.style.use('seaborn-v0_8-whitegrid')
6
sns.set_theme()
7

8
if not df.empty:
9
    # Visualization 1: Top N contributors by commit count
10
    top_n = 10
11
    if len(commits_per_author) > 0:
12
        plt.figure(figsize=(12, 6))
13
        sns.barplot(x='Total Commits', y='Author', data=commits_per_author.nlargest(top_n, 'Total Commits'))
14
        plt.title(f'Top {top_n} Contributors by Commit Count')
15
        plt.xlabel('Number of Commits')
16
        plt.ylabel('Author')
17
        plt.tight_layout()
18
        plt.show()
19
    else:
20
         print("\nNot enough data to plot Commits per Author.")
21

22

23
    # Visualization 2: Commit frequency over time (Weekly)
24
    if len(commits_over_week) > 1: # Need at least two points for a meaningful line plot
25
        plt.figure(figsize=(15, 6))
26
        sns.lineplot(x='committed_date', y='Commit Count', data=commits_over_week)
27
        plt.title('Weekly Commit Frequency')
28
        plt.xlabel('Week')
29
        plt.ylabel('Number of Commits')
30
        plt.xticks(rotation=45)
31
        plt.tight_layout()
32
        plt.show()
33
    else:
34
        print("\nNot enough data to plot Weekly Commit Frequency.")
35

36
    # Visualization 3: Top N contributors by lines changed
37
    if len(lines_per_author) > 0:
38
        plt.figure(figsize=(12, 6))
39
        sns.barplot(x='Calculated Total Lines Changed', y='Author', data=lines_per_author.nlargest(top_n, 'Calculated Total Lines Changed'))
40
        plt.title(f'Top {top_n} Contributors by Lines Changed (Insertions + Deletions)')
41
        plt.xlabel('Total Lines Changed')
42
        plt.ylabel('Author')
43
        plt.tight_layout()
44
        plt.show()
45
    else:
46
        print("\nNot enough data to plot Lines Changed per Author.")
47

48
else:
49
    print("\nNo data available for visualization.")