Refactored HackerOne report scraper with error handling and memory optimizations

[thiezn]

Hi batman & batman!

Thanks for sharing both the code and interesting talk. Nice example of leveraging AI for fuzzy data analytics.

I wanted to grab Hacktivity reports for myself for a long time but never got around to it. I saw the hackerone.py script which already figured out how the graphql + report endpoints work so thought I bite the bullet.

This pull request refactors the hackerone.py script to make it a bit more robust by:

• handling errors and rate limits
• leverages generators to avoid having to keep all the reports / urls in memory during retrieval
• incrementally adds reports to the output file so a crash of the script won't lose all of the results

Create a new python virtual environment and install the dependency (Friends don't let other friends use system-level Python dependencies)

python3 -m venv .venv
source .venv/bin/activate
python3 -m pip install httpx

You can run the script as follows (use -h to see optional cli arguments like the cookie and/or filename from the command line):

chmod +x hackerone.py
./hackerone.py

Kind regards,
Robin

[thiezn]

Just to give some idea, when I ran the script today it took me 137 minutes and gave me 12.668 reports.