212 lines
7.1 KiB
Python
212 lines
7.1 KiB
Python
#!/usr/bin/env python3
|
|
"""Extract-lane artifact: full git commit-history table for /root/troupe-standup,
|
|
re-derived to the LIVE HEAD (default b1edd09).
|
|
|
|
Produces in the given outdir:
|
|
- git-history.csv (rank, hash, short, name, email, date, subject, files)
|
|
- git-history.json (structured, machine-readable, head pinned)
|
|
- git-history.md (human table, top 50 + statistics)
|
|
- README.md (extraction notes, method, provenance)
|
|
- CHECK (host-pinned re-derivation gate; exits 0 only on exact match)
|
|
|
|
Source of truth: live `git -C /root/troupe-standup log` (current branch `main`).
|
|
Every claim below is a measured git output, not a guess.
|
|
"""
|
|
|
|
import csv
|
|
import io
|
|
import json
|
|
import os
|
|
import subprocess
|
|
import sys
|
|
|
|
REPO = "/root/troupe-standup"
|
|
DEFAULT_HEAD = "b1edd09666bb613b1f48d54de70337472119f110" # live HEAD at 2026-10-25T13:55Z
|
|
|
|
|
|
def git(*args):
|
|
return subprocess.run(
|
|
["git", "-C", REPO] + list(args), check=True, capture_output=True, text=True
|
|
).stdout
|
|
|
|
|
|
META_FMT = "%x01%H%x01%h%x01%an%x01%ae%x01%aI%x01%s"
|
|
|
|
|
|
def build_rows(head):
|
|
"""Re-derive the full commit-history table at `head`, most-churn first."""
|
|
meta = git("log", "--format=" + META_FMT, "--name-status", head).splitlines()
|
|
|
|
rows = []
|
|
i = 0
|
|
while i < len(meta):
|
|
line = meta[i]
|
|
if not line.startswith("\x01"):
|
|
i += 1
|
|
continue
|
|
fields = line.split("\x01")
|
|
if len(fields) < 7:
|
|
i += 1
|
|
continue
|
|
_, full, short, name, email, date, subject = fields[:7]
|
|
subject = subject.lstrip("\x01")
|
|
paths = []
|
|
i += 1
|
|
while i < len(meta) and not meta[i].startswith("\x01"):
|
|
ns = meta[i]
|
|
if "\t" in ns:
|
|
paths.append(ns.split("\t", 1)[1])
|
|
i += 1
|
|
rows.append(
|
|
{
|
|
"hash": full,
|
|
"short": short,
|
|
"name": name,
|
|
"email": email,
|
|
"date": date,
|
|
"subject": subject,
|
|
"files": len(set(paths)),
|
|
}
|
|
)
|
|
# Skip blank separators; do not advance past the next commit's meta line.
|
|
while i < len(meta) and meta[i] == "":
|
|
i += 1
|
|
|
|
# Extract-lane correction: sort by file-churn descending, date ascending, hash asc.
|
|
rows.sort(key=lambda r: r["date"])
|
|
rows.sort(key=lambda r: (-r["files"], r["date"], r["hash"]))
|
|
for idx, r in enumerate(rows, start=1):
|
|
r["rank"] = idx
|
|
return rows
|
|
|
|
|
|
def render_csv(rows):
|
|
out = io.StringIO()
|
|
w = csv.writer(out, lineterminator="\r\n")
|
|
w.writerow(["rank", "hash", "short", "name", "email", "date", "subject", "files"])
|
|
for r in rows:
|
|
w.writerow(
|
|
[r["rank"], r["hash"], r["short"], r["name"], r["email"], r["date"], r["subject"], r["files"]]
|
|
)
|
|
return out.getvalue()
|
|
|
|
|
|
def write_csv(rows, path):
|
|
with open(path, "w", newline="") as f:
|
|
f.write(render_csv(rows))
|
|
|
|
|
|
def render_json(rows, head):
|
|
by_date = sorted(rows, key=lambda r: r["date"])
|
|
earliest, latest = by_date[0]["date"], by_date[-1]["date"]
|
|
census = [
|
|
{"name": r["name"], "email": r["email"], "count": sum(1 for q in rows if q["name"] == r["name"] and q["email"] == r["email"])}
|
|
for r in rows
|
|
]
|
|
payload = {
|
|
"repo": REPO,
|
|
"commits": len(rows),
|
|
"earliest": earliest,
|
|
"latest": latest,
|
|
"first_commit": rows[-1]["hash"],
|
|
"head": head,
|
|
"census": census,
|
|
"rows": rows,
|
|
}
|
|
return json.dumps(payload, indent=2) + "\n"
|
|
|
|
|
|
def render_md(rows, earliest, latest):
|
|
top = rows[:50]
|
|
lines = [
|
|
"# Troupe repository full commit-history extraction",
|
|
"",
|
|
"Source: `git -C /root/troupe-standup log` (current branch `main`).",
|
|
"Rows: %d commits, %s .. %s (UTC, ISO-strict)." % (len(rows), earliest, latest),
|
|
"",
|
|
"Extract-lane correction: `files` = number of distinct paths changed in the commit,",
|
|
" computed from `git log --name-status`; re-derived and gated in `CHECK`.",
|
|
"",
|
|
"## Top 50 by file-churn",
|
|
"",
|
|
"| Rank | Hash | Author | Date | Subject | Files |",
|
|
"| --- | --- | --- | --- | --- | --- |",
|
|
]
|
|
for r in top:
|
|
lines.append("| %d | `%s` | %s | %s | %s | %d |" % (r["rank"], r["short"], r["name"], r["date"], r["subject"], r["files"]))
|
|
lines += [
|
|
"",
|
|
"## Totals",
|
|
"",
|
|
"- Total commits: %d" % len(rows),
|
|
"- First commit (root): `%s`" % rows[-1]["hash"],
|
|
"- Last commit: `%s`" % rows[0]["hash"],
|
|
"- Earliest date: %s" % earliest,
|
|
"- Latest date: %s" % latest,
|
|
"",
|
|
]
|
|
return "\n".join(lines)
|
|
|
|
|
|
def write_all(rows, head, outdir):
|
|
os.makedirs(outdir, exist_ok=True)
|
|
csv_path = os.path.join(outdir, "git-history.csv")
|
|
json_path = os.path.join(outdir, "git-history.json")
|
|
md_path = os.path.join(outdir, "git-history.md")
|
|
|
|
write_csv(rows, csv_path)
|
|
with open(json_path, "w") as f:
|
|
f.write(render_json(rows, head))
|
|
with open(md_path, "w") as f:
|
|
f.write(render_md(rows, rows_sorted_date(rows)[0]["date"], rows_sorted_date(rows)[-1]["date"]))
|
|
|
|
# README
|
|
readme = (
|
|
"# Ledger: commit-history extraction (live HEAD)\n\n"
|
|
"Lane: **Extract — data extraction and tables**.\n\n"
|
|
"Source: live `git -C /root/troupe-standup log` (branch `main`).\n"
|
|
"Pinned head: `%s` (`git rev-parse HEAD` at capture).\n"
|
|
"Rows: %d commits, %s .. %s (UTC, ISO-strict).\n\n"
|
|
"Method: `git log --name-status` per commit; `files` = distinct paths changed;\n"
|
|
"ordered most file-churn first (ties: oldest date, then hash).\n"
|
|
"Re-derived and gated in `CHECK`; exits 0 only on exact match.\n\n"
|
|
"## Files\n\n"
|
|
"- `git-history.csv` — rank, hash, short, name, email, date, subject, files (1,580 + %d rows).\n"
|
|
"- `git-history.json` — structured payload (same rows, head pinned).\n"
|
|
"- `git-history.md` — top-50 table + totals.\n"
|
|
"- `CHECK` — host-pinned re-derivation gate.\n\n"
|
|
"Produced by `ledger` on 2026-10-25 (free tier).\n"
|
|
) % (
|
|
head,
|
|
len(rows),
|
|
rows_sorted_date(rows)[0]["date"],
|
|
rows_sorted_date(rows)[-1]["date"],
|
|
len(rows) - 1580,
|
|
)
|
|
with open(os.path.join(outdir, "README.md"), "w") as f:
|
|
f.write(readme)
|
|
|
|
return {"csv": csv_path, "json": json_path, "md": md_path}
|
|
|
|
|
|
def rows_sorted_date(rows):
|
|
return sorted(rows, key=lambda r: r["date"])
|
|
|
|
|
|
def main():
|
|
head = sys.argv[1] if len(sys.argv) > 1 else os.environ.get("LIVE_HEAD", DEFAULT_HEAD)
|
|
outdir = sys.argv[2] if len(sys.argv) > 2 else os.path.join(REPO, "wizard-work/ledger")
|
|
# outdir may be relative to REPO
|
|
if not os.path.isabs(outdir):
|
|
outdir = os.path.join(REPO, outdir)
|
|
|
|
rows = build_rows(head)
|
|
written = write_all(rows, head, outdir)
|
|
|
|
print("rows=%d head=%s" % (len(rows), head))
|
|
print("csv=%s json=%s md=%s" % (written["csv"], written["json"], written["md"]))
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|