Author reconstruction for 'GitHub' user

[Leo-Send]

This pr adresses issue #52

Fixes this issue by adding the correct author to field 'event_info_2' in the issue data. This allows for reconstruction of the commit's author during author postprocessing, if the user that envoked the event is the 'GitHub' user.

Since parts of the changes take place in functions relying on codeface, testing all the changes is currently not possible.

[bockthom]

Thanks for this PR @Leo-Send.
I don't have run your changes yet, but here are already two comments:

Could you please update your copyright header in the files you have changed?

[Leo-Send]

This PR should also be ready for review. I was unable to test my changes in python 2, but since we are porting to python 3 in the near future I hope this is not an issue. I do also have the tested python 3 code locally, which could make @maxloeffler work a bit easier.

[Copilot]

Pull Request Overview

This PR addresses issue #52 by implementing author reconstruction for events triggered by the 'GitHub' user. The main purpose is to capture and preserve commit author information in event data so that the actual commit author can be reconstructed during post-processing, even when the event appears to be triggered by the generic 'GitHub' user.

• Adds commit author information to event_info_2 field for commit_added events
• Implements connected events filtering and matching logic to handle related issue connections
• Updates author post-processing to use the preserved author information as a fallback when commit data is unavailable

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
issue_processing/issue_processing.py	Adds author preservation logic, connected events handling, and updates user processing to include commit author information
author_postprocessing/author_postprocessing.py	Implements fallback author reconstruction using preserved author names when commit hash lookup fails

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[bockthom]

Could you please address the following issues, as well as the issues previously pointed out by Copilot, as well as the other unresolved one regarding necessary comments from a previous review of mine?

Thank you!

[bockthom]

And the copyright header in issue_processing/jira_issue_processing.py needs to be updated as well.

[Copilot]

Pull Request Overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 6 comments.

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Pull Request Overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 7 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[bockthom]

One more issue:

The anonymization script cannot handle our manually introduced Copilot user as it is not part of the authors.list file:

File "anonymization/anonymization.py", line 236, in run_anonymization
    new_author = author_to_anonymized_author[(issue_event[9], issue_event[10])]
KeyError: ('Copilot', 'copilot@example.com')

Solution: Add Copilot to the authors.list if it is not present there (during author_postprocessing).

However, we might also think about whether we ignore it in anonymization. Please check whether we ignore the GitHub user in anonymization or not.

[bockthom]

I found some time to run the issue analysis on our favorite test project.
This way, I encountered a number of issues that still need to be fixed. I guess the majority of them are easy to fix - only one or two might need a little more thinking on how to fix it. Please see my detailed comments below.

[bockthom]

This PR looks good to me now, so far. I ran it on one project for testing reasons and did not encounter any unexpected behavior any more. 🥳

Let's wait for the last tiny little piece (the reason for locking), and then this PR can be merged.
Meanwhile, as we are close to merging already, I will ask @copilot to review this PR for us.

[Copilot]

Pull request overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 11 comments.

Comments suppressed due to low confidence (1)

issue_processing/issue_processing.py:603

• reformat_events now mutates issue_data in-place and returns None, but its docstring still says it returns updated issue data. Please update the docstring (or return the list) to keep the API self-documenting.

def reformat_events(issue_data, filtered_connected_events, external_connected_events):
    """
    Re-format event information dependent on the event type.

    :param issue_data: the data of all issues that shall be re-formatted
    :return: the issue data with updated event information
    """

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 5 comments.

Comments suppressed due to low confidence (1)

bot_processing/bot_processing.py:137

• predicted_bots is documented/used as “usernames of the bots predicted to be a bot”, but the current code collects usernames from all rows in bot_data (which includes humans too per load_bot_data). This can prevent known bots/agents from being added because they will appear as “predicted” even if predicted as Human. Filter predicted_bots by the prediction column (e.g., only rows with label Bot/Agent) before doing membership checks.

    # Get the GitHub usernames of the bots predicted to be a bot
    predicted_bots = [bot[0] if len(bot) > 0 else "" for bot in bot_data]

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[bockthom]

I didn't spot any errors or inconsistencies in my test run, so let's assume everything works as intended.
Thanks for your efforts @Leo-Send.