ledger: check-in census extraction (date x count table) + independent CHECK

2026-10-27 03:17:32 +00:00
1 parent 26a313c3bf
commit b0a2886707
@@ -0,0 +1,61 @@
+# Check-in census
+
+Extract lane table: per-calendar-day count of `checkins/*.md` files (README.md excluded).
+
+| date | checkin_files |
+|---|---|
+| 2026-09-24 | 2 |
+| 2026-09-25 | 1 |
+| 2026-09-26 | 2 |
+| 2026-09-27 | 1 |
+| 2026-09-28 | 2 |
+| 2026-09-30 | 1 |
+| 2026-10-01 | 6 |
+| 2026-10-02 | 1 |
+| 2026-10-03 | 3 |
+| 2026-10-04 | 2 |
+| 2026-10-05 | 3 |
+| 2026-10-06 | 9 |
+| 2026-10-07 | 6 |
+| 2026-10-08 | 2 |
+| 2026-10-09 | 2 |
+| 2026-10-10 | 3 |
+| 2026-10-11 | 3 |
+| 2026-10-12 | 3 |
+| 2026-10-13 | 4 |
+| 2026-10-14 | 3 |
+| 2026-10-15 | 4 |
+| 2026-10-16 | 3 |
+| 2026-10-17 | 3 |
+| 2026-10-18 | 3 |
+| 2026-10-19 | 3 |
+| 2026-10-20 | 3 |
+| 2026-10-21 | 3 |
+| 2026-10-22 | 2 |
+| 2026-10-23 | 3 |
+| 2026-10-24 | 2 |
+| 2026-10-25 | 3 |
+| 2026-10-26 | 2 |
+
+**Total:** 93 files across 32 dates (README.md excluded).
+
+## Per-wizard
+
+| wizard | files |
+|---|---|
+| zephyrus | 50 |
+| mercury | 17 |
+| volt | 9 |
+| merlin | 8 |
+| cassian | 3 |
+| zephyrus-wizards | 3 |
+| cassian-q5-census | 1 |
+| cassian-q5-recount | 1 |
+| gemini | 1 |
+
+## Provenance
+
+- Source: `checkins/*.md` in the troupe repo.
+- Dates parsed: 32; files excluded (no leading date): 1 (README.md).
+- `wizard` = filename with leading `YYYY-MM-DD-` and trailing `-{digits}` stripped.
+- Check: `./CHECK` recomputes this table from `checkins/` and diffs against the TSV.
@@ -0,0 +1,33 @@
+date,checkin_files
+2026-09-24,2
+2026-09-25,1
+2026-09-26,2
+2026-09-27,1
+2026-09-28,2
+2026-09-30,1
+2026-10-01,6
+2026-10-02,1
+2026-10-03,3
+2026-10-04,2
+2026-10-05,3
+2026-10-06,9
+2026-10-07,6
+2026-10-08,2
+2026-10-09,2
+2026-10-10,3
+2026-10-11,3
+2026-10-12,3
+2026-10-13,4
+2026-10-14,3
+2026-10-15,4
+2026-10-16,3
+2026-10-17,3
+2026-10-18,3
+2026-10-19,3
+2026-10-20,3
+2026-10-21,3
+2026-10-22,2
+2026-10-23,3
+2026-10-24,2
+2026-10-25,3
+2026-10-26,2
@@ -0,0 +1,34 @@
+#!/usr/bin/env bash
+# CHECK — verifier for the check-in census artifact (Extract lane: data extraction).
+#
+# Exit 0 iff: the committed TSV (wizard-work/ledger/2026-10-27-checkin-census.tsv)
+# is byte-identical to an independent recompute of the census straight from the
+# `checkins/*.md` source directory.
+#
+# The recompute uses only ls/grep/awk/sort/uniq (no Python, no shared extractor
+# logic), so a hand-edited or stale TSV cannot pass by accident.
+#
+# Repo root is derived from this file's directory, so the CHECK runs from any
+# working directory and does not depend on a specific clone path.
+
+set -u
+
+cd "$(dirname "${BASH_SOURCE[0]}")" || exit 1
+REPO_ROOT="$(cd ../.. && pwd)"
+SRC="$REPO_ROOT/checkins"
+OUT="$REPO_ROOT/wizard-work/ledger/2026-10-27-checkin-census.tsv"
+
+test -s "$OUT" || { echo "FAIL: TSV missing or empty: $OUT" >&2; exit 1; }
+
+# Independent recompute: header plus date prefix of each check-in file's basename.
+expected=$(echo "date,checkin_files"; ls "$SRC"/*.md 2>/dev/null \
+  | grep -v 'README\.md$' \
+  | awk -F/ '{n=split($0, a, "/"); name=a[n]; sub(/\.md$/, "", name); print substr(name, 1, 10)}' \
+  | sort \
+  | uniq -c \
+  | awk '{print $2 "," $1}')
+
+diff -u "$OUT" - <<<"$expected" >/dev/null \
+  || { echo "FAIL: census mismatch against independent recompute" >&2; exit 1; }
+
+echo "PASS: census rows=$(tail -n +2 "$OUT" | grep -c .); recomputed from $SRC"
@@ -0,0 +1,92 @@
+#!/usr/bin/env python3
+"""extract_checkin_census_2026-10-27.py -- Extract the check-in census table.
+
+Extract lane (data extraction + tables). Reads checkins/*.md in the troupe
+repo, groups files by their leading YYYY-MM-DD date, emits a date x count
+table plus a per-wizard tally.
+
+Outputs -- written next to this script, under wizard-work/ledger/:
+  * 2026-10-27-checkin-census.tsv   date,checkin_files (one row per date)
+  * 2026-10-27-checkin-census.md  human summary + provenance
+Reads use paths relative to this file's directory (repo-root agnostic).
+"""
+import collections
+import os
+import re
+
+HERE = os.path.dirname(os.path.abspath(__file__))
+REPO = os.path.dirname(os.path.dirname(HERE))   # troupe repo root
+SRC = os.path.join(REPO, "checkins")
+OUT_TSV = os.path.join(HERE, "2026-10-27-checkin-census.tsv")
+OUT_MD = os.path.join(HERE, "2026-10-27-checkin-census.md")
+
+DATE_RE = re.compile(r"^(20\d{2}-\d{2}-\d{2})-")
+SUFFIX_RE = re.compile(r"-\d+$")                  # trailing -NNNN -> strip
+README_RE = re.compile(r"README\.md$")
+
+
+def main():
+    counts = collections.Counter()                # date -> files
+    per_wizard = collections.Counter()            # wizard -> files
+    readme = []
+    with os.scandir(SRC) as it:
+        for entry in sorted(it, key=lambda e: e.name):
+            if not entry.name.endswith(".md"):
+                continue
+            m = DATE_RE.match(entry.name)
+            if not m:
+                readme.append(entry.name)
+                continue
+            date = m.group(1)
+            wizard = SUFFIX_RE.sub("", entry.name[len(m.group(1)) + 1:-3])
+            counts[date] += 1
+            per_wizard[wizard] += 1
+
+    rows = sorted(counts.items())
+
+    with open(OUT_TSV, "w", encoding="utf-8") as f:
+        f.write("date,checkin_files\n")
+        for date, n in rows:
+            f.write(f"{date},{n}\n")
+
+    total = sum(counts.values())
+    lines = [
+        "# Check-in census",
+        "",
+        "Extract lane table: per-calendar-day count of `checkins/*.md` files "
+        "(README.md excluded).",
+        "",
+        "| date | checkin_files |",
+        "|---|---|",
+    ]
+    lines += [f"| {date} | {n} |" for date, n in rows]
+    lines += [
+        "",
+        f"**Total:** {total} files across {len(rows)} dates (README.md excluded).",
+        "",
+        "## Per-wizard",
+        "",
+        "| wizard | files |",
+        "|---|---|",
+    ]
+    lines += [f"| {w} | {n} |" for w, n in sorted(per_wizard.items(), key=lambda kv: (-kv[1], kv[0]))]
+    lines += [
+        "",
+        "## Provenance",
+        "",
+        "- Source: `checkins/*.md` in the troupe repo.",
+        f"- Dates parsed: {len(rows)}; files excluded (no leading date): "
+        f"{len(readme)} ({', '.join(readme) if readme else 'none'}).",
+        "- `wizard` = filename with leading `YYYY-MM-DD-` and trailing `-{digits}` stripped.",
+        "- Check: `./CHECK` recomputes this table from `checkins/` and diffs against the TSV.",
+    ]
+    with open(OUT_MD, "w", encoding="utf-8") as f:
+        f.write("\n".join(lines) + "\n")
+
+    print(f"dates={len(rows)} files={total} excluded={len(readme)}")
+    print(f"wrote {OUT_TSV}")
+    print(f"wrote {OUT_MD}")
+
+
+if __name__ == "__main__":
+    main()