feat: add support for Co-authored-by commit trailers with username correlation

.env-example

@@ -4,6 +4,9 @@ END_DATE = ""
 ORGANIZATION = "organization"
 REPOSITORY = "organization/repository"
 START_DATE = ""
+SPONSOR_INFO = "False"
+LINK_TO_PROFILE = "True"
+ACKNOWLEDGE_COAUTHORS = "True"
 
 # GITHUB APP
 GH_APP_ID = ""

README.md

@@ -84,20 +84,23 @@ This action can be configured to authenticate with GitHub App Installation or Pe
 
 #### Other Configuration Options
 
-| field               | required                                        | default           | description                                                                                                                                                                                                              |
-| ------------------- | ----------------------------------------------- | ----------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
-| `GH_ENTERPRISE_URL` | False                                           | ""                | The `GH_ENTERPRISE_URL` is used to connect to an enterprise server instance of GitHub. github.com users should not enter anything here.                                                                                  |
-| `ORGANIZATION`      | Required to have `ORGANIZATION` or `REPOSITORY` |                   | The name of the GitHub organization which you want the contributor information of all repos from. ie. github.com/github would be `github`                                                                                |
-| `REPOSITORY`        | Required to have `ORGANIZATION` or `REPOSITORY` |                   | The name of the repository and organization which you want the contributor information from. ie. `github/contributors` or a comma separated list of multiple repositories `github/contributor,super-linter/super-linter` |
-| `START_DATE`        | False                                           | Beginning of time | The date from which you want to start gathering contributor information. ie. Aug 1st, 2023 would be `2023-08-01`.                                                                                                        |
-| `END_DATE`          | False                                           | Current Date      | The date at which you want to stop gathering contributor information. Must be later than the `START_DATE`. ie. Aug 2nd, 2023 would be `2023-08-02`                                                                       |
-| `SPONSOR_INFO`      | False                                           | False             | If you want to include sponsor information in the output. This will include the sponsor count and the sponsor URL. This will impact action performance. ie. SPONSOR_INFO = "False" or SPONSOR_INFO = "True"              |
-| `LINK_TO_PROFILE`   | False                                           | True              | If you want to link usernames to their GitHub profiles in the output. ie. LINK_TO_PROFILE = "True" or LINK_TO_PROFILE = "False"                                                                                          |
+| field                   | required                                        | default           | description                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |
+| ----------------------- | ----------------------------------------------- | ----------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `GH_ENTERPRISE_URL`     | False                                           | ""                | The `GH_ENTERPRISE_URL` is used to connect to an enterprise server instance of GitHub. github.com users should not enter anything here.                                                                                                                                                                                                                                                                                                                                       |
+| `ORGANIZATION`          | Required to have `ORGANIZATION` or `REPOSITORY` |                   | The name of the GitHub organization which you want the contributor information of all repos from. ie. github.com/github would be `github`                                                                                                                                                                                                                                                                                                                                     |
+| `REPOSITORY`            | Required to have `ORGANIZATION` or `REPOSITORY` |                   | The name of the repository and organization which you want the contributor information from. ie. `github/contributors` or a comma separated list of multiple repositories `github/contributor,super-linter/super-linter`                                                                                                                                                                                                                                                      |
+| `START_DATE`            | False                                           | Beginning of time | The date from which you want to start gathering contributor information. ie. Aug 1st, 2023 would be `2023-08-01`.                                                                                                                                                                                                                                                                                                                                                             |
+| `END_DATE`              | False                                           | Current Date      | The date at which you want to stop gathering contributor information. Must be later than the `START_DATE`. ie. Aug 2nd, 2023 would be `2023-08-02`                                                                                                                                                                                                                                                                                                                            |
+| `SPONSOR_INFO`          | False                                           | False             | If you want to include sponsor information in the output. This will include the sponsor count and the sponsor URL. This will impact action performance. ie. SPONSOR_INFO = "False" or SPONSOR_INFO = "True"                                                                                                                                                                                                                                                                   |
+| `LINK_TO_PROFILE`       | False                                           | True              | If you want to link usernames to their GitHub profiles in the output. ie. LINK_TO_PROFILE = "True" or LINK_TO_PROFILE = "False"                                                                                                                                                                                                                                                                                                                                               |
+| `ACKNOWLEDGE_COAUTHORS` | False                                           | True              | If you want to include co-authors from commit messages as contributors. Co-authors are identified via the `Co-authored-by:` trailer in commit messages. The action will extract GitHub usernames from GitHub noreply emails (e.g., `username@users.noreply.github.com`) or use the full email address for other email domains. This will impact action performance as it requires scanning all commits. ie. ACKNOWLEDGE_COAUTHORS = "True" or ACKNOWLEDGE_COAUTHORS = "False" |
 
 **Note**: If `start_date` and `end_date` are specified then the action will determine if the contributor is new. A new contributor is one that has contributed in the date range specified but not before the start date.
 
 **Performance Note:** Using start and end dates will reduce speed of the action by approximately 63X. ie without dates if the action takes 1.7 seconds, it will take 1 minute and 47 seconds.
 
+**Co-authors Note:** When `ACKNOWLEDGE_COAUTHORS` is enabled, the action will scan commit messages for `Co-authored-by:` trailers and include those users as contributors. For GitHub noreply email addresses (e.g., `username@users.noreply.github.com`), the username will be extracted. For other email addresses (e.g., `john@example.com`), the full email address will be used as the contributor identifier. See [GitHub's documentation on creating commits with multiple authors](https://docs.github.com/en/pull-requests/committing-changes-to-your-project/creating-and-editing-commits/creating-a-commit-with-multiple-authors).
+
 ### Example workflows
 
 **Be sure to change at least these values: `<YOUR_ORGANIZATION_GOES_HERE>`, `<YOUR_GITHUB_HANDLE_HERE>`**

contributors.py

@@ -1,6 +1,7 @@
 # pylint: disable=broad-exception-caught
 """This file contains the main() and other functions needed to get contributor information from the organization or repository"""
 
+import re
 from typing import List
 
 import auth
@@ -27,6 +28,7 @@ def main():
         end_date,
         sponsor_info,
         link_to_profile,
+        acknowledge_coauthors,
     ) = env.get_env_vars()
 
     # Auth to GitHub.com
@@ -46,7 +48,13 @@ def main():
 
     # Get the contributors
     contributors = get_all_contributors(
-        organization, repository_list, start_date, end_date, github_connection, ghe
+        organization,
+        repository_list,
+        start_date,
+        end_date,
+        github_connection,
+        ghe,
+        acknowledge_coauthors,
     )
 
     # Check for new contributor if user provided start_date and end_date
@@ -60,6 +68,7 @@ def main():
             end_date=start_date,
             github_connection=github_connection,
             ghe=ghe,
+            acknowledge_coauthors=acknowledge_coauthors,
         )
         for contributor in contributors:
             contributor.new_contributor = contributor_stats.is_new_contributor(
@@ -103,6 +112,7 @@ def get_all_contributors(
     end_date: str,
     github_connection: object,
     ghe: str,
+    acknowledge_coauthors: bool,
 ):
     """
     Get all contributors from the organization or repository
@@ -113,6 +123,8 @@ def get_all_contributors(
         start_date (str): The start date of the date range for the contributor list.
         end_date (str): The end date of the date range for the contributor list.
         github_connection (object): The authenticated GitHub connection object from PyGithub
+        ghe (str): The GitHub Enterprise URL to use for authentication
+        acknowledge_coauthors (bool): Whether to acknowledge co-authors from commit messages
 
     Returns:
         all_contributors (list): A list of ContributorStats objects
@@ -130,7 +142,14 @@ def get_all_contributors(
     all_contributors = []
     if repos:
         for repo in repos:
-            repo_contributors = get_contributors(repo, start_date, end_date, ghe)
+            repo_contributors = get_contributors(
+                repo,
+                start_date,
+                end_date,
+                ghe,
+                acknowledge_coauthors,
+                github_connection,
+            )
             if repo_contributors:
                 all_contributors.append(repo_contributors)
 
@@ -140,20 +159,91 @@ def get_all_contributors(
     return all_contributors
 
 
-def get_contributors(repo: object, start_date: str, end_date: str, ghe: str):
+def get_coauthors_from_message(
+    commit_message: str, github_connection: object = None
+) -> List[str]:
+    """
+    Extract co-author identifiers from a commit message.
+
+    Co-authored-by trailers follow the format:
+    Co-authored-by: Name <email>
+
+    For GitHub noreply emails (username@users.noreply.github.com), extracts the username.
+    For @github.com emails, extracts the username (part before @).
+    For other emails, uses GitHub Search Users API to find the username, or falls back to email.
+
+    Args:
+        commit_message (str): The commit message to parse
+        github_connection (object): The authenticated GitHub connection object from PyGithub
+
+    Returns:
+        List[str]: List of co-author identifiers (GitHub usernames or email addresses)
+    """
+    # Match Co-authored-by trailers - case insensitive
+    # Format: Co-authored-by: Name <email>
+    pattern = r"Co-authored-by:\s*[^<]*<([^>]+)>"
+    matches = re.findall(pattern, commit_message, re.IGNORECASE)
+
+    identifiers = []
+    for email in matches:
+        # Check if it's a GitHub noreply email format: username@users.noreply.github.com
+        noreply_pattern = r"^(\d+\+)?([^@]+)@users\.noreply\.github\.com$"
+        noreply_match = re.match(noreply_pattern, email)
+        if noreply_match:
+            # For GitHub noreply emails, extract just the username
+            identifiers.append(noreply_match.group(2))
+        elif email.endswith("@github.com"):
+            # For @github.com emails, extract the username (part before @)
+            username = email.split("@")[0]
+            identifiers.append(username)
+        else:
+            # For other emails, try to find GitHub username using Search Users API
+            if github_connection:
+                try:
+                    # Search for users by email
+                    search_result = github_connection.search_users(f"email:{email}")
+                    if search_result.totalCount > 0:
+                        # Use the first matching user's login
+                        identifiers.append(search_result[0].login)
+                    else:
+                        # If no user found, fall back to email address
+                        identifiers.append(email)
+                except Exception:
+                    # If API call fails, fall back to email address
+                    identifiers.append(email)
+            else:
+                # If no GitHub connection available, use the full email address
+                identifiers.append(email)
+    return identifiers
+
+
+def get_contributors(
+    repo: object,
+    start_date: str,
+    end_date: str,
+    ghe: str,
+    acknowledge_coauthors: bool,
+    github_connection: object,
+):
     """
     Get contributors from a single repository and filter by start end dates if present.
 
     Args:
         repo (object): The repository object from PyGithub
         start_date (str): The start date of the date range for the contributor list.
         end_date (str): The end date of the date range for the contributor list.
+        ghe (str): The GitHub Enterprise URL to use for authentication
+        acknowledge_coauthors (bool): Whether to acknowledge co-authors from commit messages
+        github_connection (object): The authenticated GitHub connection object from PyGithub
 
     Returns:
         contributors (list): A list of ContributorStats objects
     """
     all_repo_contributors = repo.contributors()
     contributors = []
+    # Track usernames already added as contributors
+    contributor_usernames = set()
+
     try:
         for user in all_repo_contributors:
             # Ignore contributors with [bot] in their name
@@ -187,6 +277,19 @@ def get_contributors(repo: object, start_date: str, end_date: str, ghe: str):
                 "",
             )
             contributors.append(contributor)
+            contributor_usernames.add(user.login)
+
+        # Get co-authors from commit messages if enabled
+        if acknowledge_coauthors:
+            coauthor_contributors = get_coauthor_contributors(
+                repo,
+                start_date,
+                end_date,
+                ghe,
+                github_connection,
+            )
+            contributors.extend(coauthor_contributors)
+
     except Exception as e:
         print(f"Error getting contributors for repository: {repo.full_name}")
         print(e)
@@ -195,5 +298,72 @@ def get_contributors(repo: object, start_date: str, end_date: str, ghe: str):
     return contributors
 
 
+def get_coauthor_contributors(
+    repo: object,
+    start_date: str,
+    end_date: str,
+    ghe: str,
+    github_connection: object,
+) -> List[contributor_stats.ContributorStats]:
+    """
+    Get contributors who were co-authors on commits in the repository.
+
+    Args:
+        repo (object): The repository object
+        start_date (str): The start date of the date range for the contributor list.
+        end_date (str): The end date of the date range for the contributor list.
+        ghe (str): The GitHub Enterprise URL
+        github_connection (object): The authenticated GitHub connection object from PyGithub
+
+    Returns:
+        List[ContributorStats]: A list of ContributorStats objects for co-authors
+    """
+    coauthor_counts: dict = {}  # username -> count
+    endpoint = ghe if ghe else "https://github.com"
+
+    try:
+        # Get all commits in the date range
+        if start_date and end_date:
+            commits = repo.commits(since=start_date, until=end_date)
+        else:
+            commits = repo.commits()
+
+        for commit in commits:
+            # Get commit message from the commit object
+            commit_message = commit.commit.message if commit.commit else ""
+            if not commit_message:
+                continue
+
+            # Extract co-authors from commit message
+            coauthors = get_coauthors_from_message(commit_message, github_connection)
+            for username in coauthors:
+                coauthor_counts[username] = coauthor_counts.get(username, 0) + 1
+
+    except Exception as e:
+        print(f"Error getting co-authors for repository: {repo.full_name}")
+        print(e)
+        return []
+
+    # Create ContributorStats objects for co-authors
+    coauthor_contributors = []
+    for username, count in coauthor_counts.items():
+        if start_date and end_date:
+            commit_url = f"{endpoint}/{repo.full_name}/commits?author={username}&since={start_date}&until={end_date}"
+        else:
+            commit_url = f"{endpoint}/{repo.full_name}/commits?author={username}"
+
+        contributor = contributor_stats.ContributorStats(
+            username,
+            False,
+            "",  # No avatar URL available for co-authors
+            count,
+            commit_url,
+            "",
+        )
+        coauthor_contributors.append(contributor)
+
+    return coauthor_contributors
+
+
 if __name__ == "__main__":
     main()