groq_api.py

import os
import requests

GROQ_API_KEY = os.getenv("GROQ_API_KEY") or "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"

def summarize_match(job_description, cv_names, cv_snippets):
    if not GROQ_API_KEY:
        return "❌ GROQ_API_KEY not set."

    try:
        # Limit content length per CV to avoid token overflow
        cv_snippets = [text.strip()[:1500] or "[No content]" for text in cv_snippets[:3]]
        cv_names = [name[:60] for name in cv_names[:3]]

        # Create structured prompt
        prompt = f"""
You are an AI recruitment assistant helping to match candidates to job descriptions.

### Job Description:
{job_description}

### Candidate CVs:
1. {cv_names[0]}:
{cv_snippets[0]}

2. {cv_names[1]}:
{cv_snippets[1]}

3. {cv_names[2]}:
{cv_snippets[2]}

Analyze how well each candidate matches the job requirements, especially in terms of:
- PHP programming
- Software or web development
- Relevant technical experience

Clearly identify which candidates are suitable and why.
""".strip()

        # Debug info (optional)
        print("📦 Prompt length:", len(prompt))
        if len(prompt) > 8000:
            return "❌ Prompt too long. Please shorten the CVs or JD."

        # Groq API call
        response = requests.post(
            url="https://api.groq.com/openai/v1/chat/completions",
            headers={
                "Authorization": f"Bearer {GROQ_API_KEY}",
                "Content-Type": "application/json"
            },
            json={
                "model": "mixtral-8x7b-32768",  # Or change to a different supported Groq model
                "messages": [{"role": "user", "content": prompt}],
                "temperature": 0.4
            },
            timeout=30
        )

        response.raise_for_status()
        return response.json()["choices"][0]["message"]["content"]

    except Exception as e:
        return f"❌ Groq API error: {e}"