git-analysis.py

import pandas as pd

# Load the two CSV files
commit_data = pd.read_csv('git_log_summary.csv')  # This contains author, email, date, and total lines changed
author_org_data = pd.read_csv('git_log_grouped_by_author.csv')  # This contains author, email, org, and other fields

# Merge the two dataframes on author and email
merged_data = pd.merge(commit_data, author_org_data[['Author', 'Email', 'Organization']], on=['Author', 'Email'], how='left')

# Fill missing organization names with "Community"
merged_data['Organization'] = merged_data['Organization'].fillna('Community')

# Ensure the 'Date' column has only valid dates
# Coerce invalid dates to NaT (Not a Time), which are ignored in the next step
merged_data['Date'] = pd.to_datetime(merged_data['Date'], errors='coerce')

# Drop any rows where the 'Date' is NaT (invalid dates)
merged_data = merged_data.dropna(subset=['Date'])

# Add a new column for year-month
merged_data['Year-Month'] = merged_data['Date'].dt.to_period('M')

# Group the data by 'Year-Month' and 'Organization' to calculate commit count and total lines changed
grouped_data = merged_data.groupby(['Year-Month', 'Organization']).agg(
    Commit_Count=('Date', 'size'),
    Total_Lines_Changed=('Total Lines Changed', 'sum')
).reset_index()

# Save the final result to a new CSV file
grouped_data.to_csv('git_log_grouped_by_month_and_org.csv', index=False)

print("CSV file 'git_log_grouped_by_month_and_org.csv' has been created.")

git-summary.py

import subprocess
import re
import csv
from collections import defaultdict

# Function to get git log data
def get_git_log():
    # Run git log command to get author, email, date, and insertions/deletions
    log_output = subprocess.run(
        ['git', 'log', '--shortstat', '--pretty=format:%an,%ae,%ad', '--date=short'],
        capture_output=True,
        text=True
    )
    return log_output.stdout.splitlines()

# Function to parse the git log and create CSV files
def parse_git_log_to_csv():
    # Get git log output
    lines = get_git_log()
    
    # Variables to store parsed data
    data = []
    commit_count_by_author = defaultdict(int)  # To count commits by author
    author_changes_summary = defaultdict(lambda: [0, 0])  # To store total changes by author
    author = ""
    email = ""
    date = ""

    # Regex to match lines containing "files changed" (i.e., lines with insertions and deletions)
    pattern = re.compile(r"(\d+) insertions\(\+\), (\d+) deletions\(-\)")

    # Process lines
    for line in lines:
        if re.match(r".+@.+\.\w{2,},\d{4}-\d{2}-\d{2}", line):  # Detect author, email, and date lines
            parts = line.split(',')
            author = parts[0].strip()
            email = parts[1].strip()
            date = parts[2].strip()
            commit_count_by_author[(author, email)] += 1  # Increment commit count for the author-email pair
        elif "files changed" in line:  # Detect lines with changes
            match = pattern.search(line)
            if match:
                insertions = int(match.group(1))
                deletions = int(match.group(2))
                total_changes = insertions + deletions
                # Append a row to the data list
                data.append([author, email, date, total_changes])
                # Sum the changes for each author-email pair
                author_changes_summary[(author, email)][0] += total_changes
                author_changes_summary[(author, email)][1] += 1  # Increment commit count

    # Write the data to a CSV file
    with open('git_log_summary.csv', 'w', newline='') as file:
        writer = csv.writer(file)
        writer.writerow(["Author", "Email", "Date", "Total Lines Changed"])
        writer.writerows(data)

    print("CSV file 'git_log_summary.csv' has been created.")

    # Prepare data for the grouped CSV, adding an empty 'organization' field
    grouped_data = [
        [author, email, commit_count, "", author_changes_summary[(author, email)][0]]
        for (author, email), commit_count in sorted(commit_count_by_author.items(), key=lambda x: x[1], reverse=True)
    ]

    # Write the grouped data to a CSV file
    with open('git_log_grouped_by_author.csv', 'w', newline='') as file:
        writer = csv.writer(file)
        writer.writerow(["Author", "Email", "Number of Commits", "Organization", "Total Lines Changed"])
        writer.writerows(grouped_data)

    print("CSV file 'git_log_grouped_by_author.csv' has been created.")

# Call the function to parse git log and create CSVs
parse_git_log_to_csv()