from __future__ import annotations

import os


class ExplainModule:
    """Google Gemini text explainer."""

    def __init__(self):
        self.model = os.environ.get("GEMINI_EXPLAIN_MODEL", "gemini-3.1-flash-lite-preview")
        self.client = None
        try:
            from google import genai

            self.client = genai.Client(api_key=os.environ.get("GEMINI_API_KEY", ""))
        except Exception as exc:
            print(f"ExplainModule using Gemini fallback: {exc}")

    def explain(
        self,
        fakescore,
        s1,
        s2,
        s3,
        weights,
        attribution,
        segments,
        top_generator,
    ) -> str:
        verdict = "FAKE" if fakescore > 0.5 else "REAL"
        confidence = (
            "high"
            if abs(fakescore - 0.5) > 0.3
            else "moderate"
            if abs(fakescore - 0.5) > 0.15
            else "low"
        )
        seg_text = ""
        if segments:
            seg_text = "Flagged timestamps: " + ", ".join(
                [f"{segment['time']}s (score={segment['score']})" for segment in segments[:5]]
            )
        attr_text = ""
        if attribution:
            top3 = sorted(attribution.items(), key=lambda item: -item[1])[:3]
            attr_text = "Top generators: " + ", ".join(
                [f"{name}: {prob * 100:.1f}%" for name, prob in top3]
            )
        prompt = f"""You are a forensic AI analyst. Analyze these deepfake detection results. Be specific about evidence.

Results:
- Verdict: {verdict} (FakeScore: {fakescore:.3f}, confidence: {confidence})
- Lip-Sync (M1): {s1:.3f} (weight: {weights.get('lip_sync', 'N/A')})
- Fingerprint (M2): {s2:.3f} (weight: {weights.get('fingerprint', 'N/A')})
- Graph-GNN (M3): {s3:.3f} (weight: {weights.get('graph_gnn', 'N/A')})
{seg_text}
{attr_text}
- Most likely generator: {top_generator}

Write 3-5 sentences. Reference specific scores and timestamps."""

        try:
            if self.client is None:
                raise RuntimeError("Gemini client unavailable")
            from google.genai import types

            response = self.client.models.generate_content(
                model=self.model,
                contents=prompt,
                config=types.GenerateContentConfig(
                    system_instruction="You are a forensic deepfake analyst. Be precise.",
                    max_output_tokens=300,
                    temperature=0.3,
                ),
            )
            return (response.text or "").strip()
        except Exception:
            return self._fallback(verdict, fakescore, s1, s2, s3, top_generator, confidence)

    def _fallback(self, verdict, fakescore, s1, s2, s3, top_gen, conf):
        if verdict == "FAKE":
            return (
                f"Video classified as {verdict} with {conf} confidence "
                f"(FakeScore: {fakescore:.3f}). "
                f"Lip-sync scored {s1:.2f}, indicating "
                f"{'significant' if s1 > 0.7 else 'moderate' if s1 > 0.5 else 'minimal'} "
                f"audio-visual inconsistency. "
                f"Style fingerprinting scored {s2:.2f}, top attribution: {top_gen}. "
                f"Graph analysis scored {s3:.2f}."
            )
        return (
            f"Video classified as {verdict} with {conf} confidence "
            f"(FakeScore: {fakescore:.3f}). "
            f"All modules returned scores below detection threshold."
        )