#!/usr/bin/env python3
"""Extract the reply index from the troupe-standup replies/ directory.

Lane: Extract -- data extraction and tables.
Writes a per-reply catalog (TSV) plus a per-recipient summary (Markdown table).

Accepted filenames:
  YYYY-MM-DD-to-<recipient>-<N>.md
  YYYY-MM-DD-to-<recipient>-interview.md   (interview, no number by construction)
  YYYY-MM-DD-to-<recipient>.md             (bare reply, no number)
"""
import glob
import os
import re
import sys
from collections import Counter
from datetime import date

REPO = "/root/troupe-standup"
REPLIES_DIR = os.path.join(REPO, "replies")
OUT_DIR = os.path.join(REPO, "wizard-work", "ledger")
CATALOG_TSV = os.path.join(OUT_DIR, "replies-catalog-2026-11-02.tsv")
SUMMARY_MD = os.path.join(OUT_DIR, "replies-catalog-2026-11-02.md")

FILENAME_RE = re.compile(
    r"^(?P<date>\d{4}-\d{2}-\d{2})-to-(?P<recipient>[A-Za-z0-9_-]+)"
    r"(?:-(?P<suffix>\d+|interview|preview))?\.md$"
)


def categorize(suffix):
    if suffix == "interview":
        return "interview"
    if suffix == "preview":
        return "preview"
    return "reply"


def main():
    md_files = sorted(glob.glob(os.path.join(REPLIES_DIR, "*.md")))
    assert md_files, "no .md files found"

    rows = []
    other_files = []
    for path in md_files:
        fn = os.path.basename(path)
        m = FILENAME_RE.match(fn)
        if m:
            suffix = m.group("suffix") or ""
            rows.append({
                "date": m.group("date"),
                "recipient": m.group("recipient"),
                "number": suffix if suffix.isdigit() else "",
                "category": categorize(suffix),
                "file": fn,
            })
        else:
            other_files.append(fn)

    os.makedirs(OUT_DIR, exist_ok=True)
    with open(CATALOG_TSV, "w", encoding="utf-8", newline="") as f:
        f.write("date\trecipient\tnumber\tcategory\tfile\n")
        for r in rows:
            f.write("%s\t%s\t%s\t%s\t%s\n" % (
                r["date"], r["recipient"], r["number"], r["category"], r["file"]))

    recipient_counts = Counter(r["recipient"] for r in rows)
    total = len(rows)
    by_date = Counter(r["date"] for r in rows)
    cat_counts = Counter(r["category"] for r in rows)

    lines = []
    lines.append("# Reply index — extracted %s" % date.today().isoformat())
    lines.append("")
    lines.append("Source: `%s/replies/*.md` (files matching `YYYY-MM-DD-to-<recipient>[-<N>|-interview|-preview].md`)."
                 % REPLIES_DIR)
    lines.append("")
    lines.append("## Per-recipient reply counts")
    lines.append("")
    lines.append("| recipient | replies |")
    lines.append("|---|---|")
    for recip in sorted(recipient_counts):
        lines.append("| %s | %d |" % (recip, recipient_counts[recip]))
    lines.append("| **total** | **%d** |" % total)
    lines.append("")
    lines.append("## Per-day reply counts")
    lines.append("")
    lines.append("| date | replies |")
    lines.append("|---|---|")
    for d in sorted(by_date):
        lines.append("| %s | %d |" % (d, by_date[d]))
    lines.append("| **total** | **%d** |" % total)
    lines.append("")
    lines.append("## Categories")
    lines.append("")
    lines.append("| category | count |")
    lines.append("|---|---|")
    for c in sorted(cat_counts):
        lines.append("| %s | %d |" % (c, cat_counts[c]))
    lines.append("| **total** | **%d** |" % total)
    lines.append("")

    with open(SUMMARY_MD, "w", encoding="utf-8") as f:
        f.write("\n".join(lines))

    print("rows written:", total)
    print("categories:", dict(cat_counts))
    print("recipients:", len(recipient_counts))
    print("other_files:", other_files)
    print("catalog:", CATALOG_TSV)
    print("summary:", SUMMARY_MD)


if __name__ == "__main__":
    sys.exit(main())