git-commit-map

#!/usr/bin/env python3
"""Show commits in a range: file -> function -> lines.

Usage:
  git-commit-map                 # merge-base with main/master
  git-commit-map <branch>        # merge-base with <branch>
  git-commit-map <rev>..<rev>    # explicit range
  git-commit-map --from-init     # from first commit
"""
import re, subprocess, sys
from collections import defaultdict


def run(cmd, check=True):
    return subprocess.run(cmd, capture_output=True, text=True, check=check)


def resolve_range(arg):
    if arg == '--from-init':
        root = run(['git', 'rev-list', '--max-parents=0', 'HEAD']).stdout.strip()
        return f'{root}..HEAD'
    if '..' in arg or arg.startswith('^'):
        return arg
    r = run(['git', 'merge-base', 'HEAD', arg], check=False)
    return f'{r.stdout.strip()}..HEAD' if r.returncode == 0 else arg


def default_range():
    for b in ['main', 'master', 'origin/main', 'origin/master']:
        r = run(['git', 'merge-base', 'HEAD', b], check=False)
        if r.returncode == 0:
            return f'{r.stdout.strip()}..HEAD'
    return 'HEAD~10..HEAD'


def parse_diff(diff_text):
    """Return {file: {func: [(start, end)]}}."""
    files = {}
    cur = None
    for line in diff_text.splitlines():
        if line.startswith('+++ b/'):
            cur = line[6:]
            files.setdefault(cur, defaultdict(list))
        elif line.startswith('+++ /dev/null'):
            cur = None
        elif line.startswith('@@') and cur is not None:
            m = re.match(r'^@@ -\d+(?:,\d+)? \+(\d+)(?:,(\d+))? @@(.*)', line)
            if m:
                start = int(m.group(1))
                count = int(m.group(2)) if m.group(2) is not None else 1
                end = start + max(count - 1, 0)
                func = m.group(3).strip()
                files[cur][func].append((start, end))
    return files


def fmt(start, end):
    if end == 0:
        return f'after {start}'
    return f'{start}' if start == end else f'{start}-{end}'


def main():
    range_spec = resolve_range(sys.argv[1]) if len(sys.argv) > 1 else default_range()
    shas = run(['git', 'log', '--format=%H', range_spec]).stdout.strip().splitlines()
    if not shas:
        print('No commits found.')
        return

    for sha in reversed(shas):
        subject = run(['git', 'show', '--no-patch', '--format=%s', sha]).stdout.strip()
        diff = parse_diff(run(['git', 'show', '--unified=0', '--format=', sha]).stdout)

        print(f'{sha[:12]}  {subject}')
        for filepath, funcs in diff.items():
            print(f'  {filepath}')
            for func, ranges in funcs.items():
                locs = ', '.join(fmt(s, e) for s, e in ranges)
                if func:
                    print(f'    {func}')
                    print(f'      {locs}')
                else:
                    print(f'    {locs}')
        print()


if __name__ == '__main__':
    main()