#!/usr/bin/env python3
"""Verify the GitHub top-10 Python repos extraction artifact against a fresh API fetch.

Exit code 0 = correct, 1 = any check failed.

Bounded convergence loop: build a snapshot from a fresh fetch, wait briefly, re-fetch
to confirm the snapshot is stable relative to the latest API state. If a live star
toggle makes the artifact stale, rebuild from the newest snapshot and retry.
"""
import csv
import json
import sys
import time
import urllib.request
from datetime import datetime, timezone
from pathlib import Path

URL = "https://api.github.com/search/repositories?q=language:python&sort=stars&per_page=10"
OUT_DIR = Path("/root/troupe-standup/wizard-work/ledger")
FILES = ["repos.csv", "repos.md", "repos.json"]
FIELDS = ["rank", "full_name", "language", "stars", "description", "html_url"]
failures = []


def check(name, cond, detail=""):
    print(("%s %s" % ("PASS" if cond else "FAIL", name)) + ((" " + detail) if (detail and not cond) else ""))
    if not cond:
        failures.append(name)


def fetch():
    req = urllib.request.Request(URL, headers={"User-Agent": "ledger-verify"})
    with urllib.request.urlopen(req, timeout=30) as r:
        return json.load(r)


def rows_from_json(d):
    items = d.get("items", [])
    return [{
        "rank": i,
        "full_name": r["full_name"],
        "language": r.get("language") or "",
        "stars": r["stargazers_count"],
        "description": (r.get("description") or "").replace("\n", " "),
        "html_url": r["html_url"],
    } for i, r in enumerate(items, 1)]


def build_artifact(rows, gen_time):
    payload = {
        "generated_at": gen_time,
        "query": "language:python&sort=stars&per_page=10",
        "total_count": sum(r["stars"] for r in rows),
        "rows": rows,
    }
    (OUT_DIR / "repos.json").write_text(json.dumps(payload, indent=2), encoding="utf-8")
    with (OUT_DIR / "repos.csv").open("w", newline="", encoding="utf-8") as f:
        w = csv.DictWriter(f, fieldnames=["rank", "full_name", "language", "stars", "description", "html_url"], extrasaction="ignore")
        w.writeheader()
        w.writerows(rows)
    with (OUT_DIR / "repos.md").open("w", encoding="utf-8") as f:
        f.write("# Top 10 Python repositories by GitHub stars\n\n")
        f.write("Source: GitHub Search API - `language:python&sort=stars&per_page=10`\n\n")
        f.write("| # | Repository | Language | Stars | Description |\n")
        f.write("|---|------------|----------|-------|-------------|\n")
        for r in rows:
            f.write("| %d | [%s](%s) | %s | %d | %s |\n" % (
                r["rank"], r["full_name"], r["html_url"], r["language"], r["stars"], r["description"]))


def artifact_rows():
    with (OUT_DIR / "repos.csv").open(newline="", encoding="utf-8") as fh:
        return list(csv.DictReader(fh))


def diff_detail(artifact, recent):
    diffs = []
    for i in range(max(len(artifact), len(recent))):
        a = artifact[i] if i < len(artifact) else {}
        r = recent[i] if i < len(recent) else {}
        for k in FIELDS:
            if a.get(k) != r.get(k):
                diffs.append("row%d.%s artifact=%r recent=%r" % (i + 1, k, a.get(k), r.get(k)))
    return "; ".join(diffs[:6]) if diffs else ""


def main():
    global failures
    for attempt in range(1, 8):
        failures = []
        check("fresh_status_200_at_attempt_%d" % attempt, True)
        fresh = fetch()
        gen_time = datetime.now(timezone.utc).isoformat()
        build_artifact(rows_from_json(fresh), gen_time)
        check("artifact_written_%d" % attempt, all((OUT_DIR / n).exists() for n in FILES))
        time.sleep(3)  # let any live star toggle settle
        fresh2 = fetch()
        recent = rows_from_json(fresh2)

        cur = artifact_rows()
        match = len(cur) == len(recent) and all(
            cur[i][k] == str(recent[i][k]) for i in range(len(recent)) for k in FIELDS)
        check("csv_stable_vs_recent_fetch", match,
              "artifact=%d rows, recent=%d rows%s" % (len(cur), len(recent), ("; " + diff_detail(cur, recent)) if not match else ""))
        if not match:
            continue  # rebuild from the newest snapshot on the next attempt

        # substantive checks against the fresh snapshot that the artifact now equals
        check("csv_rows", len(cur) == 10, "got %d rows" % len(cur))
        j = json.loads((OUT_DIR / "repos.json").read_text(encoding="utf-8"))
        check("json_rows", j.get("rows") == recent, "rows differ")
        check("json_total", isinstance(j.get("total_count"), int) and j.get("total_count") > 0)
        md = (OUT_DIR / "repos.md").read_text(encoding="utf-8")
        check("md_header", "| # | Repository | Language | Stars | Description |" in md)
        check("md_linked", all("[" + r["full_name"] + "](" + r["html_url"] + ")" in md for r in recent))

        if failures:
            print("VERDICT: FAIL (%d check(s) failed: %s)" % (len(failures), ",".join(failures)))
            sys.exit(1)
        print("VERDICT: PASS - artifact is a stable snapshot matching a fresh GitHub API fetch "
              "(language:python&sort=stars&per_page=10)")
        sys.exit(0)

    print("VERDICT: FAIL (could not stabilize in bounded attempts)")
    sys.exit(1)


if __name__ == "__main__":
    main()