Finding commits by SHA-1 hash on Github repos forked from a given root repo.

findhash.py

""" Finding commits by SHA-1 hash on Github. This is a simple
brute-force search for a specific usecase: you assume that the commit is
within all repos forked from a certain root repo within the past n days.

This may result in many calls to the GitHub API, which in turn may result
in GitHub's rate limiter kicking in and forcing you to take a break. You
have been warned.

Built with Python 3.4 and github3.py 0.9.4.
"""
from datetime import *
from github3 import login

# your personal GitHub access token (needs no special scopes, just reading
# public stuff here), see https://github.com/settings/tokens
access_token = "INSERT_TOKEN_HERE"

sha_hash = "b90b16efe674be681c20f9bca76f82680299db66"
root_repo_owner = "rdpeng"
root_repo_name = "ProgrammingAssignment2"
age_days = 10

created_after = datetime.now(timezone.utc) - timedelta(days=age_days)

gh = login(token = access_token)
root_repo = gh.repository(root_repo_owner, root_repo_name)

print("searching repos and commits younger than", created_after)

count = 0
for fork in root_repo.iter_forks():
    count = count + 1
    if fork.created_at > created_after:
         for commit in fork.iter_commits(since=created_after):
             if commit.sha == sha_hash:
                 print("found " + commit.sha + " in " + fork.html_url)
                 break
    else:
         break

print(count, "repos searched")