find the intersections of committers to various git projects

find.py

# python3 find.py linux llvm-project gcc binutils-gdb
import argparse
import pygit2

def parse_args():
    parser = argparse.ArgumentParser(description="Find kernel toolchain devs")
    parser.add_argument("kernel_dir", help="A git checkout of the Linux Kernel")
    parser.add_argument("toolchain_dir", help="A git checkout of a toolchain",
            nargs="+")
    return parser.parse_args()

def find_author_set(git_dir):
    print("Finding author set for ", git_dir)
    repo = pygit2.Repository(git_dir)
    last = repo[repo.head.target]
    devs = set()
    for commit in repo.walk(last.id):
        author = "%s <%s>" % (commit.author.name, commit.author.email)
        devs.add(author)
        # break
    return devs

def find(kernel_dir, toolchain_dirs):
    kernel_devs = find_author_set(kernel_dir)
    toolchains_devs = [find_author_set(toolchain_dir) for toolchain_dir in toolchain_dirs]
    whole_set = set()
    for toolchain_devs in toolchains_devs:
        intersection = kernel_devs & toolchain_devs
        whole_set &= intersection
        print(len(intersection), intersection)
    print(len(whole_set), whole_set)

if __name__ == "__main__":
    args = parse_args()
    print(args)
    find(args.kernel_dir, args.toolchain_dir)