"""
Run v3 detector (framing_v2 signals + frame_library_v3 rules) on the
expanded corpus.

Writes labels/labels_detector_v3.json with per-doc v3 labels.
"""

from __future__ import annotations

import json
import sys
from pathlib import Path

REPO_ROOT = Path(__file__).resolve().parents[2]
STUDY_DIR = REPO_ROOT / "fvs_eval" / "validation_study"
CORPUS_DIR = STUDY_DIR / "corpus"
LABELS_DIR = STUDY_DIR / "labels"

sys.path.insert(0, str(REPO_ROOT))
sys.path.insert(0, str(STUDY_DIR))

from framing import detect_voice, temporal_orientation, detect_epistemic_basis  # noqa: E402
from framing_v2 import (  # noqa: E402
    detect_coverage_v2,
    detect_named_author_citation,
    detect_growth_vocabulary,
    detect_efficiency_vocabulary,
)
from frame_library_v3 import suggest_frames_v3, V3_STATUS  # noqa: E402


FRAME_IDS_ALL = [f"FVS-{i:03d}" for i in range(1, 21)]


def main():
    with open(CORPUS_DIR / "manifest.json") as f:
        manifest = json.load(f)

    out = {
        "version": 3,
        "labeler": "detector_v3",
        "labeler_note": (
            "framing_v2 signals (S-1 expanded uncertainty regex, S-2 "
            "named-author-citation, S-3 growth vocabulary, S-4 "
            "efficiency vocabulary) + frame_library_v3 rules "
            "(v2 rule revisions + signal-rewires per RULE_AUDIT + "
            "DESIGN_v2 §4)."
        ),
        "v3_status": V3_STATUS,
        "frame_universe": FRAME_IDS_ALL,
        "labels": {},
        "signals": {},
    }

    for doc in manifest["documents"]:
        doc_id = doc["doc_id"]
        path = CORPUS_DIR / f"{doc_id}.txt"
        if not path.exists():
            continue
        text = path.read_text(encoding="utf-8")

        cov_v2 = detect_coverage_v2(text)
        voice = detect_voice(text)
        temp = temporal_orientation(text)
        epist = detect_epistemic_basis(text)
        na = detect_named_author_citation(text)
        gv = detect_growth_vocabulary(text)
        ev = detect_efficiency_vocabulary(text)

        suggs = suggest_frames_v3(cov_v2, voice, temp, epist, na, gv, ev)
        suggested_ids = {s["fvs_id"] for s in suggs}

        out["labels"][doc_id] = {
            fid: {"exhibits": fid in suggested_ids} for fid in FRAME_IDS_ALL
        }

        out["signals"][doc_id] = {
            "coverage_v2": {k: (len(v) if isinstance(v, list) else v)
                            for k, v in cov_v2.items() if k != "categories"},
            "coverage_v2_density": {
                c: cov_v2["categories"][c]["density_per_1kw"]
                for c in cov_v2.get("categories", {})
            },
            "voice": voice.get("voice"),
            "named_author_density": na["density_per_1kw"],
            "named_author_matches": na["matches"],
            "growth_density": gv["density_per_1kw"],
            "efficiency_density": ev["density_per_1kw"],
            "temporal": {
                "past": temp.get("past_pct"),
                "present": temp.get("present_pct"),
                "future": temp.get("future_pct"),
            },
            "epistemic_sourced_pct": epist.get("sourced_pct"),
        }
        n = sum(1 for v in out["labels"][doc_id].values() if v["exhibits"])
        sigs_brief = f"[na={na['density_per_1kw']}, gv={gv['density_per_1kw']}, ev={ev['density_per_1kw']}, unc={cov_v2['categories'].get('uncertainty',{}).get('density_per_1kw',0)}]"
        print(f"  {doc_id}: {n} frames {sorted([fid for fid, e in out['labels'][doc_id].items() if e['exhibits']])} {sigs_brief}")

    with open(LABELS_DIR / "labels_detector_v3.json", "w") as f:
        json.dump(out, f, indent=2)
    print(f"\nWritten: {LABELS_DIR / 'labels_detector_v3.json'}")


if __name__ == "__main__":
    main()