lead-scraper/app/api/enrich-leads/route.ts

import { NextRequest, NextResponse } from "next/server";
import { prisma } from "@/lib/db";

const ENRICHMENT_PROMPT = `Du analysierst eine Unternehmenswebsite für ein Solarunternehmen und extrahierst strukturierte Informationen.

Antworte ausschließlich mit JSON:
{
  "companyType": "stadtwerke" | "industrie" | "sonstiges",
  "topics": ["photovoltaik"|"speicher"|"emobilitaet"|"waerme"|"netz"|"wasserstoff"],
  "salesScore": 1-5,
  "salesReason": "Ein Satz warum dieser Lead gut/schlecht passt",
  "offerPackage": "solar-basis" | "solar-pro" | "solar-speicher" | "komplett"
}

Bewertungskriterien salesScore:
5 = Stadtwerk oder Industriebetrieb mit explizitem Energiethema, perfekter Fit
4 = Energierelevante Branche, wahrscheinlich guter Fit
3 = Mittelständischer Betrieb, möglicher Fit
2 = Kleiner Betrieb oder unklare Relevanz
1 = Offensichtlich nicht relevant (Privatpersonen, Behörden ohne Energie-Relevanz)

Angebotspaket-Logik:
- solar-basis: Kleine Betriebe <50 Mitarbeiter, residentiell
- solar-pro: Mittelstand, Gewerbe
- solar-speicher: Energieintensive Betriebe, Stadtwerke
- komplett: Stadtwerke, Großindustrie, wenn topics viele Themen enthält

Website-Text:
[TEXT]`;

async function fetchWebsiteText(domain: string): Promise<string> {
  const controller = new AbortController();
  const timeout = setTimeout(() => controller.abort(), 3000);
  try {
    const url = domain.startsWith("http") ? domain : `https://${domain}`;
    const res = await fetch(url, {
      signal: controller.signal,
      headers: { "User-Agent": "Mozilla/5.0 (compatible; LeadBot/1.0)" },
    });
    clearTimeout(timeout);
    const html = await res.text();
    // Strip tags, extract text
    const text = html
      .replace(/<script[\s\S]*?<\/script>/gi, "")
      .replace(/<style[\s\S]*?<\/style>/gi, "")
      .replace(/<[^>]+>/g, " ")
      .replace(/\s+/g, " ")
      .trim()
      .slice(0, 2000);
    return text;
  } catch {
    clearTimeout(timeout);
    return "";
  }
}

interface EnrichmentResult {
  companyType: string;
  topics: string[];
  salesScore: number;
  salesReason: string;
  offerPackage: string;
}

async function enrichWithAI(text: string, apiKey: string): Promise<EnrichmentResult | null> {
  if (!text.trim()) return null;

  const prompt = ENRICHMENT_PROMPT.replace("[TEXT]", text);

  const res = await fetch("https://openrouter.ai/api/v1/chat/completions", {
    method: "POST",
    headers: {
      "Authorization": `Bearer ${apiKey}`,
      "Content-Type": "application/json",
      "HTTP-Referer": "https://mein-solar.com",
      "X-Title": "MeinSolar Leads",
    },
    body: JSON.stringify({
      model: "openai/gpt-4o-mini",
      temperature: 0.2,
      max_tokens: 256,
      messages: [
        { role: "user", content: prompt },
      ],
    }),
  });

  if (!res.ok) return null;

  const data = await res.json() as { choices: Array<{ message: { content: string } }> };
  const raw = data.choices[0]?.message?.content?.trim() ?? "";

  try {
    const parsed = JSON.parse(raw) as EnrichmentResult;
    return parsed;
  } catch {
    const match = raw.match(/\{[\s\S]*\}/);
    if (!match) return null;
    try {
      return JSON.parse(match[0]) as EnrichmentResult;
    } catch {
      return null;
    }
  }
}

export async function POST(req: NextRequest) {
  try {
    const { jobId } = await req.json() as { jobId: string };

    if (!jobId) {
      return NextResponse.json({ error: "jobId fehlt" }, { status: 400 });
    }

    const apiKey = process.env.OPENROUTER_API_KEY;
    if (!apiKey) {
      return NextResponse.json({ error: "OpenRouter API Key nicht konfiguriert" }, { status: 500 });
    }

    // Load all LeadResults for this job
    const results = await prisma.leadResult.findMany({
      where: { jobId },
    });

    // Find corresponding leads with domains
    const domainsToEnrich = results
      .filter(r => r.domain)
      .map(r => r.domain as string);

    if (domainsToEnrich.length === 0) {
      return NextResponse.json({ enriched: 0, skipped: 0 });
    }

    // Find leads matching these domains
    const leads = await prisma.lead.findMany({
      where: {
        domain: { in: domainsToEnrich },
        salesScore: null, // only enrich if not already enriched
      },
    });

    let enriched = 0;
    let skipped = 0;

    for (const lead of leads) {
      if (!lead.domain) { skipped++; continue; }

      const text = await fetchWebsiteText(lead.domain);
      if (!text) { skipped++; continue; }

      const result = await enrichWithAI(text, apiKey);
      if (!result) { skipped++; continue; }

      await prisma.lead.update({
        where: { id: lead.id },
        data: {
          companyType: result.companyType || null,
          topics: result.topics ? JSON.stringify(result.topics) : null,
          salesScore: typeof result.salesScore === "number" ? result.salesScore : null,
          salesReason: result.salesReason || null,
          offerPackage: result.offerPackage || null,
        },
      });

      enriched++;
    }

    return NextResponse.json({ enriched, skipped, total: leads.length });
  } catch (err) {
    console.error("POST /api/enrich-leads error:", err);
    return NextResponse.json({ error: "Enrichment failed" }, { status: 500 });
  }
}