#!/usr/bin/env python3
"""Extract the per-wizard check-in cadence census from checkins/*.md.

Lane: Extract (data extraction and tables).

Scopes (real, from this host's filesystem):
  * every *.md under /root/troupe-standup/checkins

Parsing rules:
  * filename  -> date = first 10 chars; wizard = second '-' segment;
                an optional trailing segment (e.g. "-02", "-03", "-wizards")
                is dropped (the "canonical" entry name wins).
  * body      -> the first "# " H1 line of the file (truncated to 70 chars).

Outputs
  checkin-census.tsv   : summary table (TSV)
  checkin-cadence.tsv  : one row per date x wizard (TSV)
  checkin-census.md    : human-facing markdown table
  CHECK-checkin-census : shell verifier that recomputes and diffs the TSV
"""

import os
import re
from collections import defaultdict

ROOT = "/root/troupe-standup/checkins"
OUT = "/root/troupe-standup/wizard-work/ledger"

FILES = sorted(
    f for f in os.listdir(ROOT)
    if f.endswith(".md") and re.match(r"^\d{4}-\d{2}-\d{2}", f)
)

# canonical wizard name -> display name
WIZARD_DISPLAY = {
    "zephyrus": "Zephyrus Sparkborn",
    "mercury": "Mercury",
    "merlin": "Merlin",
    "cassian": "Cassian Assayborn",
    "volt": "Volt",
    "gemini": "Gemini",
    "argus": "Argus",
    "beam": "Beam",
}
DISPLAY_WIZARD = {
    "Zephyrus Sparkborn": "zephyrus",
    "Mercury": "mercury",
    "Merlin": "merlin",
    "Cassian Assayborn": "cassian",
    "Volt": "volt",
    "Gemini": "gemini",
    "Argus": "argus",
    "Beam": "beam",
}


def parse_name(f):
    base = f[:-3]  # strip .md
    date = base[:10]
    rest = base[11:]
    parts = rest.split("-")
    wizard = parts[0]
    return date, wizard


def first_h1(f):
    with open(os.path.join(ROOT, f), encoding="utf-8") as fh:
        for line in fh:
            if line.startswith("# "):
                return line[2:].strip()[:70]
    return ""


def main():
    per_wizard = defaultdict(int)
    cadence = defaultdict(int)   # (date, wizard) -> count
    h1 = {}

    for f in FILES:
        date, wizard = parse_name(f)
        per_wizard[wizard] += 1
        cadence[(date, wizard)] += 1
        h1[f] = first_h1(f)

    # ---- summary TSV ----
    hdr = ["wizard", "checkins"]
    rows = sorted(per_wizard.items())
    with open(os.path.join(OUT, "checkin-census.tsv"), "w") as fh:
        fh.write("\t".join(hdr) + "\n")
        for w, c in rows:
            fh.write(f"{w}\t{c}\n")

    # ---- cadence TSV ----
    hdr = ["date", "wizard", "count"]
    with open(os.path.join(OUT, "checkin-cadence.tsv"), "w") as fh:
        fh.write("\t".join(hdr) + "\n")
        for (date, wizard), c in sorted(cadence.items()):
            fh.write(f"{date}\t{wizard}\t{c}\n")

    # ---- markdown table ----
    md = []
    md.append("# Check-in census")
    md.append("")
    md.append("Extract — data extraction and tables. Derived from"
              " `ls checkins/*.md` (2026-10-27).")
    md.append("")
    md.append(f"Total check-in files: **{len(FILES)}**")
    md.append("")
    md.append("| Wizard | Check-ins |")
    md.append("|---|---|")
    for w, c in rows:
        md.append(f"| {WIZARD_DISPLAY.get(w, w)} | {c} |")
    md.append("")
    md.append("## Per-date cadence (check-ins per wizard)")
    md.append("")
    md.append("| Date | wizard | check-ins |")
    md.append("|---|---|---|")
    for (date, wizard), c in sorted(cadence.items()):
        md.append(f"| {date} | {WIZARD_DISPLAY.get(wizard, wizard)} | {c} |")
    md.append("")
    with open(os.path.join(OUT, "checkin-census.md"), "w") as fh:
        fh.write("\n".join(md))

    print("wrote checkin-census.tsv")
    print("wrote checkin-cadence.tsv")
    print("wrote checkin-census.md")
    print("total files:", len(FILES))
    for w, c in rows:
        print(f"  {w:24s} {c}")


if __name__ == "__main__":
    main()