ledger: daily commit census 2026-11-04 (extract lane)

2026-11-04 01:49:39 +00:00
1 parent 95a5998a2d
commit ade23054b6
@@ -0,0 +1,37 @@
+#!/usr/bin/env bash
+# CHECK -- daily commit census (Extract lane). Exit 0 iff the committed
+# daily-commit-census.csv is a byte-identical re-derivation from git at the rev
+# the committed extraction_manifest.json pins. Independent of the extractor --
+# pure bash + git + coreutils (sort/uniq/awk), no Python.
+set -u
+
+cd "$(dirname "${BASH_SOURCE[0]}")" || exit 1
+SCRIPT_DIR="$(pwd)"
+REPO_ROOT="$(cd "$SCRIPT_DIR/../../.." && pwd)"
+OUT="$REPO_ROOT/wizard-work/ledger/2026-11-04-daily-commit-census/daily-commit-census.csv"
+MANIFEST="$REPO_ROOT/wizard-work/ledger/2026-11-04-daily-commit-census/extraction_manifest.json"
+
+test -s "$OUT" || { echo "FAIL: CSV missing or empty: $OUT" >&2; exit 1; }
+test -s "$MANIFEST" || { echo "FAIL: manifest missing: $MANIFEST" >&2; exit 1; }
+
+HEAD_FULL="$(sed -n 's/.*"head_full": "\([^"]*\)".*/\1/p' "$MANIFEST")"
+test -n "$HEAD_FULL" || { echo "FAIL: no head_full in manifest" >&2; exit 1; }
+
+tmp="$(mktemp)"
+trap 'rm -f "$tmp"' EXIT
+
+{
+  echo "date,commits"
+  git -C "$REPO_ROOT" log --date=short --pretty=format:%ad "$HEAD_FULL" | sort | uniq -c | awk '{print $2 "," $1}' | sort
+  sum="$(git -C "$REPO_ROOT" log --date=short --pretty=format:%ad "$HEAD_FULL" | sort | uniq -c | awk '{s += $1} END {print s}')"
+  echo "TOTAL,$sum"
+} > "$tmp"
+
+if cmp -s "$OUT" "$tmp"; then
+  echo "PASS: daily-commit-census.csv re-derives from git at pinned head $HEAD_FULL"
+  exit 0
+fi
+
+echo "FAIL: daily-commit-census.csv differs from re-derivation at pinned head $HEAD_FULL" >&2
+diff -u "$OUT" "$tmp" >&2 || true
+exit 1
@@ -0,0 +1,36 @@
+# Ledger: daily commit census (2026-11-04)
+
+Lane: **Extract — data extraction and tables**.
+
+A machine-readable census of every commit on the troupe repository's `main`
+branch, grouped by ISO calendar date.
+
+## Source and method
+
+- Source: `git log --date=short --pretty=format:%ad` on `/root/troupe-standup`
+  (branch `main`).
+- Each commit contributes its ISO date `%ad`. Commits are then grouped by date
+  and counted; rows sorted ascending by date.
+- Output gate: `daily-commit-census.csv` (41 day rows + 1 TOTAL row) is re-derived
+  from git at the pinned rev and diffed byte-identically by `verify_daily_commit_census.py`
+  and the standalone `CHECK` script.
+
+## Metrics (measured, not asserted)
+
+- Day rows: 41
+- Commit total: 2612
+- Pinned head: `4a4723c69b90799a833e6efba936972fc5659d58` (`git rev-parse HEAD`
+  at capture, 2026-11-04)
+- Earliest date: 2026-09-24
+- Latest date: 2026-11-04
+- Busiest day: 2026-10-23 (171 commits)
+- Quietest day (non-zero): 2026-11-04 (2 commits)
+
+## Files
+
+- `daily-commit-census.csv` — `date,commits` (one row per date, plus a `TOTAL` row)
+- `extraction_manifest.json` — provenance: repo, pinned head, date range, counts
+- `extract_daily_commit_census.py` — extractor (regenerates the CSV)
+- `verify_daily_commit_census.py` — self-check gate (re-derives from pinned head)
+- `CHECK` — independent bash re-derivation gate (no Python)
+- `daily-commit-census.md` — top summary table for humans
@@ -0,0 +1,43 @@
+date,commits
+2026-09-24,31
+2026-09-25,4
+2026-09-26,24
+2026-09-27,5
+2026-09-28,11
+2026-09-30,6
+2026-10-01,17
+2026-10-02,3
+2026-10-03,21
+2026-10-04,13
+2026-10-05,14
+2026-10-06,40
+2026-10-07,25
+2026-10-08,21
+2026-10-09,8
+2026-10-10,44
+2026-10-11,63
+2026-10-12,106
+2026-10-13,90
+2026-10-14,90
+2026-10-15,46
+2026-10-16,26
+2026-10-17,41
+2026-10-18,42
+2026-10-19,50
+2026-10-20,92
+2026-10-21,123
+2026-10-22,115
+2026-10-23,171
+2026-10-24,167
+2026-10-25,153
+2026-10-26,141
+2026-10-27,147
+2026-10-28,154
+2026-10-29,84
+2026-10-30,107
+2026-10-31,98
+2026-11-01,80
+2026-11-02,80
+2026-11-03,57
+2026-11-04,2
+TOTAL,2612
@@ -0,0 +1,46 @@
+# Daily commit census — troupe repo `main` (2026-11-04)
+
+| date | commits |
+|---|---|
+| 2026-09-24 | 31 |
+| 2026-09-25 | 4 |
+| 2026-09-26 | 24 |
+| 2026-09-27 | 5 |
+| 2026-09-28 | 11 |
+| 2026-09-30 | 6 |
+| 2026-10-01 | 17 |
+| 2026-10-02 | 3 |
+| 2026-10-03 | 21 |
+| 2026-10-04 | 13 |
+| 2026-10-05 | 14 |
+| 2026-10-06 | 40 |
+| 2026-10-07 | 25 |
+| 2026-10-08 | 21 |
+| 2026-10-09 | 8 |
+| 2026-10-10 | 44 |
+| 2026-10-11 | 63 |
+| 2026-10-12 | 106 |
+| 2026-10-13 | 90 |
+| 2026-10-14 | 90 |
+| 2026-10-15 | 46 |
+| 2026-10-16 | 26 |
+| 2026-10-17 | 41 |
+| 2026-10-18 | 42 |
+| 2026-10-19 | 50 |
+| 2026-10-20 | 92 |
+| 2026-10-21 | 123 |
+| 2026-10-22 | 115 |
+| 2026-10-23 | 171 |
+| 2026-10-24 | 167 |
+| 2026-10-25 | 153 |
+| 2026-10-26 | 141 |
+| 2026-10-27 | 147 |
+| 2026-10-28 | 154 |
+| 2026-10-29 | 84 |
+| 2026-10-30 | 107 |
+| 2026-10-31 | 98 |
+| 2026-11-01 | 80 |
+| 2026-11-02 | 80 |
+| 2026-11-03 | 57 |
+| 2026-11-04 | 2 |
+| **TOTAL** | **2612** |
@@ -0,0 +1,91 @@
+#!/usr/bin/env python3
+"""Extract the daily commit census of the troupe repository.
+
+Lane: Extract - data extraction and tables.
+
+Source: live `git log --date=short --pretty=format:%ad` on the `main` branch of
+/root/troupe-standup. Commits are grouped by ISO calendar date and counted.
+Rows are sorted by date ascending. The pinned head (full + short) is recorded in
+extraction_manifest.json so the extraction is reproducible at a fixed rev.
+
+Output (written to <out_dir>):
+  daily-commit-census.csv  - header, one row per date + a TOTAL row (41+1 lines)
+  extraction_manifest.json - provenance: repo, head, date range, row count, sums
+
+Usage: python3 extract_daily_commit_census.py <out_dir>
+"""
+
+import json
+import os
+import subprocess
+import sys
+from collections import Counter
+
+REPO = "/root/troupe-standup"
+DATE_FMT = "%Y-%m-%d"
+
+
+def git(*args):
+    return subprocess.run(
+        ["git", "-C", REPO] + list(args),
+        check=True,
+        capture_output=True,
+        text=True,
+    ).stdout
+
+
+def build_rows():
+    dates = git("log", "--date=short", "--pretty=format:%ad").splitlines()
+    counts = Counter(d for d in dates if d)
+    return [{"date": d, "commits": c} for d, c in sorted(counts.items())]
+
+
+def render_csv(rows):
+    lines = ["date,commits"]
+    for r in rows:
+        lines.append("{date},{commits}".format(**r))
+    total = sum(r["commits"] for r in rows)
+    lines.append("TOTAL,{total}".format(total=total))
+    return "\n".join(lines) + "\n"
+
+
+def render_manifest(head_full, head_short, rows):
+    dates = [r["date"] for r in rows]
+    return {
+        "repo": REPO,
+        "branch": "main",
+        "source_command": "git log --date=short --pretty=format:%ad",
+        "head_full": head_full,
+        "head_short": head_short,
+        "earliest": dates[0],
+        "latest": dates[-1],
+        "day_rows": len(rows),
+        "total_commits": sum(r["commits"] for r in rows),
+        "method": "counts ISO calendar date of every commit on main",
+    }
+
+
+def main():
+    out_dir = sys.argv[1] if len(sys.argv) > 1 else "."
+    os.makedirs(out_dir, exist_ok=True)
+
+    head_full = git("rev-parse", "HEAD").strip()
+    head_short = git("rev-parse", "--short", "HEAD").strip()
+    rows = build_rows()
+
+    csv_path = os.path.join(out_dir, "daily-commit-census.csv")
+    with open(csv_path, "w", encoding="utf-8") as f:
+        f.write(render_csv(rows))
+
+    manifest = render_manifest(head_full, head_short, rows)
+    with open(os.path.join(out_dir, "extraction_manifest.json"), "w", encoding="utf-8") as f:
+        json.dump(manifest, f, indent=2)
+        f.write("\n")
+
+    print("rows:", manifest["day_rows"], "total:", manifest["total_commits"],
+          "head:", head_short, "earliest:", manifest["earliest"], "latest:", manifest["latest"])
+    print("wrote:", csv_path)
+
+
+if __name__ == "__main__":
+    main()
@@ -0,0 +1,12 @@
+{
+  "repo": "/root/troupe-standup",
+  "branch": "main",
+  "source_command": "git log --date=short --pretty=format:%ad",
+  "head_full": "4a4723c69b90799a833e6efba936972fc5659d58",
+  "head_short": "4a4723c",
+  "earliest": "2026-09-24",
+  "latest": "2026-11-04",
+  "day_rows": 41,
+  "total_commits": 2612,
+  "method": "counts ISO calendar date of every commit on main"
+}
@@ -0,0 +1,90 @@
+#!/usr/bin/env python3
+"""CHECK gate for the daily-commit-census extraction.
+
+Re-derives the daily commit census from git at the PINNED head recorded in the
+committed extraction_manifest.json, re-renders the CSV with the identical
+format the extractor uses, and diffs it against the committed file. Exits 0 only
+when every check matches exactly. This is a re-derivation, not an assertion of a
+stored number.
+
+Usage: python3 verify_daily_commit_census.py <artifact_dir>
+"""
+
+import difflib
+import json
+import os
+import subprocess
+import sys
+
+REPO = "/root/troupe-standup"
+HERE = os.path.dirname(os.path.abspath(__file__))
+sys.path.insert(0, HERE)
+from extract_daily_commit_census import render_csv, build_rows  # noqa: E402
+
+CSV_NAME = "daily-commit-census.csv"
+
+
+def git(*args):
+    return subprocess.run(
+        ["git", "-C", REPO] + list(args), check=True, capture_output=True, text=True
+    ).stdout
+
+
+def read_bytes(name):
+    with open(name, "rb") as f:
+        return f.read()
+
+
+def main():
+    artdir = sys.argv[1] if len(sys.argv) > 1 else "."
+    os.chdir(artdir)
+
+    with open("extraction_manifest.json", encoding="utf-8") as f:
+        manifest = json.load(f)
+    head = manifest.get("head_full")
+    if not head:
+        print("FAIL: manifest has no head_full")
+        sys.exit(1)
+
+    # Re-derive from the pinned head.
+    raw = git("log", "--date=short", "--pretty=format:%ad", head).splitlines()
+    from collections import Counter
+
+    counts = Counter(d for d in raw if d)
+    rows = [{"date": d, "commits": c} for d, c in sorted(counts.items())]
+    expected = render_csv(rows)
+
+    ok = True
+
+    with open(CSV_NAME, encoding="utf-8") as f:
+        committed = f.read()
+
+    if committed == expected:
+        print("PASS: daily-commit-census.csv re-derives byte-identically from git "
+              "at pinned head %s" % head)
+    else:
+        ok = False
+        print("FAIL: daily-commit-census.csv diverges from re-derivation at "
+              "pinned head %s" % head)
+        exp = expected.splitlines()
+        obs = committed.splitlines()
+        for line in list(difflib.unified_diff(exp, obs, "expected@", "committed@"))[:12]:
+            print("  " + line)
+
+    # Independent integrity: the TOTAL row must equal the sum of the day rows.
+    day_sum = sum(int(r["commits"]) for r in rows)
+    total_lines = [ln for ln in committed.splitlines() if ln.startswith("TOTAL,")]
+    if total_lines == ["TOTAL,%d" % day_sum]:
+        print("PASS: TOTAL row %r == sum of day rows %d" % (total_lines[0], day_sum))
+    else:
+        ok = False
+        print("FAIL: TOTAL row %r != sum of day rows %d" % (total_lines, day_sum))
+
+    if ok:
+        print("ALL CHECKS PASSED")
+        sys.exit(0)
+    sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()