"""Compute all scores and write the comparison site."""
from __future__ import annotations

import datetime as dt
import html
import json
import shutil
from pathlib import Path

import numpy as np

from .loading import Source, load_challenges, load_submissions, public_fields
from .metrics import compute_metric, metric_names
from .reference import collect_order, select_reference
from .wav import read_wav, resolve_output


def score_challenge(challenge: dict, submissions: list[dict], out_audio: Path | None = None) -> dict:
    """Load every output for one challenge, compute pairwise errors, choose the reference, and return a results dict."""
    spec = challenge["output"]
    names = metric_names(challenge)
    primary = challenge["metrics"]["primary"]

    signals: dict[str, np.ndarray] = {}
    rows: dict[str, dict] = {}
    for sub in submissions:
        src: Source = sub["_source"]
        for sol in sub["solutions"]:
            key = f"{sub['_key']}/{sol['solution_id']}"
            p = resolve_output(sol["output"], sub["_path"].parent)
            signals[key] = read_wav(p, spec)
            audio_rel = None
            if out_audio is not None:
                dest = out_audio / sub["_key"] / f"{sol['solution_id']}.wav"
                dest.parent.mkdir(parents=True, exist_ok=True)
                shutil.copyfile(p, dest)
                audio_rel = f"audio/{sub['_key']}/{sol['solution_id']}.wav"
            fields_url = None
            if "fields" in sol:
                fpath = sol["fields"]["path"]
                fields_url = fpath if fpath.startswith("http") else src.blob_url(sub["_path"].parent / fpath)
            rows[key] = {
                "key": key,
                "source": src.label,
                "source_url": src.url,
                "submission_id": sub["submission_id"],
                "submission_url": src.blob_url(sub["_path"]),
                "solution_id": sol["solution_id"],
                "submitter": sub["submitter"]["name"],
                "date": sub.get("date"),
                "method": sol.get("method"),
                "parameters": sol.get("parameters"),
                "timing_seconds": sol.get("timing_seconds"),
                "hardware": sol.get("hardware"),
                "code_url": sol.get("code_url") or sub.get("code_url"),
                "viz_url": sol.get("viz_url") or sub.get("viz_url"),
                "fields_url": fields_url,
                "more_accurate_than": [f"{sub['_key']}/{w}" for w in sol.get("more_accurate_than", [])],
                "notes": sol.get("notes"),
                "audio": audio_rel,
                "peak": float(np.max(np.abs(signals[key]))),
            }

    keys = list(signals)
    pairwise = {name: {a: {} for a in keys} for name in names}
    for name in names:
        params = challenge["metrics"][name]
        for a in keys:
            for r in keys:
                pairwise[name][a][r] = 0.0 if a == r else compute_metric(name, signals[a], signals[r], params)

    info = collect_order(submissions)
    explicit = challenge.get("reference_solution")
    if explicit is not None and explicit not in keys:
        raise KeyError(f"reference_solution {explicit!r} is not among the solutions")
    selection = select_reference(info, pairwise[primary], explicit)
    ref = selection["reference"]
    for key, row in rows.items():
        row["errors"] = {name: pairwise[name][key][ref] for name in names} if ref else {}
        row["is_reference"] = key == ref

    violations = []
    if ref:
        for worse, better, _sid in info.pairs:
            for name in names:
                if not pairwise[name][better][ref] < pairwise[name][worse][ref]:
                    violations.append({"metric": name, "worse": worse, "better": better})

    ordered = sorted(rows.values(), key=lambda r: (r["errors"].get(primary, float("inf")), r["key"]))
    return {
        "challenge_id": challenge["challenge_id"],
        "title": challenge.get("title"),
        "generated": dt.datetime.now(dt.timezone.utc).isoformat(timespec="seconds"),
        "metrics": names,
        "primary_metric": primary,
        "reference": ref,
        "reference_selection": selection,
        "order_violations_vs_reference": violations,
        "solutions": ordered,
        "pairwise": pairwise,
        "submissions": [
            {**{k: v for k, v in public_fields(sub).items() if k != "solutions"}, "key": sub["_key"], "source": sub["_source"].label, "source_url": sub["_source"].url}
            for sub in submissions
        ],
    }


# ---------------------------------------------------------------- HTML

_CSS = """
:root { --fg: #1b1b1b; --bg: #ffffff; --muted: #666; --line: #ddd; --accent: #2a6db0; --ref: #e8f4e8; }
@media (prefers-color-scheme: dark) { :root { --fg: #e6e6e6; --bg: #151515; --muted: #999; --line: #333; --accent: #7ab3e8; --ref: #1f3320; } }
body { font-family: system-ui, sans-serif; color: var(--fg); background: var(--bg); margin: 0; padding: 16px; line-height: 1.45; }
main { max-width: 1300px; margin: 0 auto; }
a { color: var(--accent); }
table { border-collapse: collapse; width: 100%; font-size: 0.92em; }
th, td { border-bottom: 1px solid var(--line); padding: 6px 8px; text-align: left; vertical-align: top; }
th { position: sticky; top: 0; background: var(--bg); }
td.num { text-align: right; font-variant-numeric: tabular-nums; white-space: nowrap; }
tr.ref { background: var(--ref); }
.muted { color: var(--muted); }
.wrap { overflow-x: auto; }
details { margin: 1em 0; }
code { font-size: 0.9em; }
"""


def _fmt(x, digits=4):
    if x is None:
        return ""
    if isinstance(x, float):
        if x == 0:
            return "0"
        return f"{x:.{digits}g}"
    return html.escape(str(x))


def _link(url, text):
    if not url:
        return ""
    return f'<a href="{html.escape(url)}">{html.escape(text)}</a>'


def render_challenge_page(res: dict, challenge: dict, links: dict) -> str:
    names = res["metrics"]
    primary = res["primary_metric"]
    head = "".join(f"<th>{html.escape(n)}{' (primary)' if n == primary else ''}</th>" for n in names)
    body = []
    for r in res["solutions"]:
        errs = "".join(f'<td class="num">{_fmt(r["errors"].get(n))}</td>' for n in names)
        params = html.escape(json.dumps(r["parameters"])) if r["parameters"] else ""
        mat = ", ".join(html.escape(k.rsplit("/", 1)[1]) for k in r["more_accurate_than"])
        audio = f'<div class="cplayer" data-src="{html.escape(r["audio"])}"></div>' if r["audio"] else ""
        linkbits = " ".join(x for x in [
            _link(r["audio"], "wav"), _link(r["fields_url"], "fields"), _link(r["code_url"], "code"), _link(r["viz_url"], "viz"),
        ] if x)
        sub_cell = _link(r["submission_url"], r["submission_id"]) if r["submission_url"] else html.escape(r["submission_id"])
        src_cell = _link(r["source_url"], r["source"]) if r["source"] else ""
        body.append(
            f'<tr class="{"ref" if r["is_reference"] else ""}">'
            f'<td>{sub_cell}<br><span class="muted">{src_cell}{" · " if src_cell else ""}{html.escape(r["submitter"])}</span></td>'
            f'<td>{html.escape(r["solution_id"])}{" <b>(reference)</b>" if r["is_reference"] else ""}</td>'
            f'<td>{html.escape(r["method"] or "")}<br><span class="muted"><code>{params}</code></span></td>'
            f"{errs}"
            f'<td class="num">{_fmt(r["timing_seconds"], 3)}</td>'
            f'<td>{html.escape(r["hardware"] or "")}</td>'
            f'<td class="muted">{mat}</td>'
            f"<td>{audio}<br>{linkbits}</td>"
            "</tr>"
        )
    sel = res["reference_selection"]
    cand_rows = "".join(
        f"<tr><td>{html.escape(c['candidate'])}</td><td class='num'>{c['pairs_preserved']}/{c['pairs_total']}</td>"
        f"<td class='num'>{_fmt(c['convergence_gap'])}</td><td class='num'>{_fmt(c['mean_error_to_other_candidates'])}</td>"
        f"<td>{html.escape(', '.join(f'{w} vs {b}' for w, b in c['violated_pairs']))}</td></tr>"
        for c in sel.get("candidates", [])
    )
    viol = res["order_violations_vs_reference"]
    viol_html = (
        "<p>All asserted accuracy orderings are preserved by both metrics under the chosen reference.</p>"
        if not viol
        else "<ul>" + "".join(
            f"<li>{html.escape(v['metric'])}: {html.escape(v['better'])} asserted more accurate than {html.escape(v['worse'])}, but its error is not smaller</li>"
            for v in viol
        ) + "</ul>"
    )
    pair_tables = []
    keys = [r["key"] for r in res["solutions"]]
    for n in names:
        hdr = "".join(f"<th>{html.escape(k)}</th>" for k in keys)
        rws = "".join(
            f"<tr><th>{html.escape(a)}</th>" + "".join(f"<td class='num'>{_fmt(res['pairwise'][n][a][b], 3)}</td>" for b in keys) + "</tr>"
            for a in keys
        )
        pair_tables.append(f"<h4>{html.escape(n)}</h4><div class='wrap'><table><tr><th></th>{hdr}</tr>{rws}</table></div>")
    spec_link = links.get("spec")
    playback_gain = float((challenge.get("playback") or {}).get("gain_db", 20))
    return f"""<!doctype html>
<html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width, initial-scale=1">
<title>{html.escape(res["challenge_id"])} results</title><style>{_CSS}</style></head>
<body><main>
<p><a href="../">All challenges</a></p>
<h1>{html.escape(res["challenge_id"])}</h1>
<p>{html.escape(res.get("title") or "")}. {"<a href='" + html.escape(spec_link) + "'>Challenge specification</a>. " if spec_link else ""}<a href="results.json">results.json</a>. Generated {html.escape(res["generated"])}.</p>
<p>Errors are relative to the reference solution <b>{html.escape(res["reference"] or "none")}</b>, chosen by rule <i>{html.escape(sel["rule"])}</i>. Rows are sorted by the primary metric. Submissions are named as <i>repository/submission</i>.</p>
<div id="playback"></div>
<div class="wrap"><table>
<tr><th>Submission</th><th>Solution</th><th>Method</th>{head}<th>Time (s)</th><th>Hardware</th><th>Claimed more accurate than</th><th>Output</th></tr>
{"".join(body)}
</table></div>
<details><summary>Reference selection</summary>
<p>Candidates are the maximal elements of each submission's asserted order. The score is the number of asserted pairs from other submissions whose order is preserved by the primary error to this candidate. Ties are broken first by the convergence gap (the primary error between the candidate and the nearest solution of its own submission asserted to be less accurate; smaller means a tighter convergence claim), then by the mean primary error to the other candidates.</p>
<div class="wrap"><table><tr><th>Candidate</th><th>Pairs preserved</th><th>Convergence gap</th><th>Mean error to other candidates</th><th>Violated pairs</th></tr>{cand_rows}</table></div>
<h4>Order check against the chosen reference</h4>{viol_html}
</details>
<details><summary>Pairwise errors (row vs column as reference)</summary>{"".join(pair_tables)}</details>
</main>
<script src="../player.js"></script>
<script>ChallengePlayer.controls(document.getElementById("playback"), {{gainDb: {playback_gain}, limiter: true}}); ChallengePlayer.mount();</script>
</body></html>
"""


def render_index(results: list[dict], sources: list[Source], links: dict) -> str:
    items = "".join(
        f'<li><a href="{html.escape(r["challenge_id"])}/">{html.escape(r["challenge_id"])}</a>: {html.escape(r.get("title") or "")} '
        f'<span class="muted">({len(r["solutions"])} solutions, reference {html.escape(r["reference"] or "none")})</span></li>'
        for r in results
    )
    srcs = "".join(
        f"<li>{_link(s.url, s.repo or s.label or str(s.dir)) if s.url else html.escape(str(s.dir))}"
        f"{' <span class=muted>(' + html.escape(s.branch) + ')</span>' if s.branch else ''}"
        f"{': ' + html.escape(s.description) if s.description else ''}</li>"
        for s in sources
    )
    repo = links.get("challenge_repo")
    return f"""<!doctype html>
<html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width, initial-scale=1">
<title>Challenge results</title><style>{_CSS}</style></head>
<body><main>
<h1>Challenge results</h1>
<p>{"Specifications and registry: <a href='" + html.escape(repo) + "'>" + html.escape(repo) + "</a>." if repo else ""}</p>
<h2>Challenges</h2><ul>{items}</ul>
<h2>Registered solution repositories</h2><ul>{srcs}</ul>
</main></body></html>
"""


def build(challenges_dir: Path, sources: list[Source], out: Path, challenge_repo_url: str | None = None) -> list[dict]:
    challenges = load_challenges(challenges_dir)
    submissions = load_submissions(sources)
    out.mkdir(parents=True, exist_ok=True)
    results = []
    for cid, ch in challenges.items():
        subs = [s for s in submissions if s["challenge_id"] == cid]
        cdir = out / cid
        cdir.mkdir(parents=True, exist_ok=True)
        res = score_challenge(ch, subs, cdir / "audio") if subs else {
            "challenge_id": cid, "title": ch.get("title"), "generated": dt.datetime.now(dt.timezone.utc).isoformat(timespec="seconds"),
            "metrics": metric_names(ch), "primary_metric": ch["metrics"]["primary"], "reference": None,
            "reference_selection": {"rule": "none", "candidates": []}, "order_violations_vs_reference": [], "solutions": [], "pairwise": {}, "submissions": [],
        }
        with open(cdir / "results.json", "w") as f:
            json.dump(res, f, indent=1)
        links = {"spec": f"{challenge_repo_url}/blob/main/challenges/{cid}/README.md" if challenge_repo_url else None}
        (cdir / "index.html").write_text(render_challenge_page(res, ch, links))
        results.append(res)
    (out / "index.html").write_text(render_index(results, sources, {"challenge_repo": challenge_repo_url}))
    (out / "player.js").write_text(player_js())
    return results


def player_js() -> str:
    """The Web Audio player shipped with every site (also copied by solution repositories' site builders)."""
    from importlib import resources

    return resources.files("challenge_tools").joinpath("player.js").read_text()