app.py

import os
import re
from flask import Flask, request, jsonify
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import torch

app = Flask(__name__)

# --- MODEL LOADING ---
MODEL_NAME = "SEBIS/code_trans_t5_base_commit_generation"
print(f"--- AI Commit Generator Server ---")
print(f"Downloading/Loading Model: {MODEL_NAME}")

device = "cpu"

try:
    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, skip_special_tokens=True)
    model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME).to(device)
    print("✅ Model loaded successfully!")
except Exception as e:
    print(f"❌ Error loading model: {e}")
    exit(1)

def get_smart_fallback(diff_text, filename):
    """
    Analyzes the CODE itself to generate a specific message 
    when the AI model fails or hallucinates.
    """
    # 1. Detect Logging / Debugging
    if "console.log" in diff_text or "outputChannel.append" in diff_text or "print(" in diff_text:
        return f"Add debug logging to {filename}"

    # 2. Detect Error Handling
    if "try" in diff_text and "catch" in diff_text:
        return f"Improve error handling in {filename}"

    # 3. Detect Timing / Async Logic
    if "setTimeout" in diff_text or "debounce" in diff_text or "await" in diff_text:
        return f"Update async/debounce logic in {filename}"

    # 4. Detect Import Changes
    if "import " in diff_text or "require(" in diff_text:
        return f"Update imports in {filename}"

    # 5. Last Resort: Find the function name
    patterns = [r'function\s+([a-zA-Z0-9_]+)', r'const\s+([a-zA-Z0-9_]+)\s*=']
    for pattern in patterns:
        match = re.search(pattern, diff_text)
        if match:
            return f"Refactor '{match.group(1)}' logic"
            
    return f"Update logic in {filename}"

def preprocess_diff(diff_text):
    if not diff_text: return ""
    lines = diff_text.split('\n')
    cleaned_lines = []
    for line in lines:
        if (line.startswith('+') or line.startswith('-')):
            if line.startswith('+++') or line.startswith('---'): continue
            # Don't strip imports here, we might need them for context
            if len(line.strip()) < 4: continue 
            cleaned_lines.append(line.strip())
    return "\n".join(cleaned_lines)

def sanitize_summary(summary, diff_text, filename):
    summary_clean = summary.strip()
    
    # 1. Catch Hallucinated Jira Tickets
    ticket_pattern = re.compile(r'\b[A-Z]{3,}\s?-\s?\d+\b')
    match = ticket_pattern.search(summary_clean)
    
    if match:
        ticket = match.group()
        if ticket not in diff_text:
            print(f"⚠️ Hallucination Killed: '{ticket}' -> Using Smart Fallback")
            return get_smart_fallback(diff_text, filename)

    # 2. Catch Linguistic Nonsense
    forbidden_words = ["transitive verb", "intransitive", "adjective"]
    for word in forbidden_words:
        if word in summary_clean.lower():
            return get_smart_fallback(diff_text, filename)

    return summary_clean

def generate_summary(diff_text, filename):
    cleaned_diff = preprocess_diff(diff_text)
    
    if not cleaned_diff or len(cleaned_diff) < 20:
        return get_smart_fallback(diff_text, filename)

    input_ids = tokenizer.encode(cleaned_diff, return_tensors="pt", max_length=512, truncation=True).to(device)

    outputs = model.generate(
        input_ids, 
        max_length=60,         
        min_length=5, 
        num_beams=5, 
        early_stopping=True
    )
    
    raw_summary = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return sanitize_summary(raw_summary, diff_text, filename)

@app.route('/generate', methods=['POST'])
def generate_commit():
    data = request.json
    files = data.get('files', [])
    
    if not files: return jsonify({"commit_message": ""})

    final_message_parts = []
    
    for file_obj in files:
        name = file_obj.get('name', 'file')
        diff = file_obj.get('diff', '')
        
        if len(diff) > 12000:
            final_message_parts.append(f"{name}\nUpdate large file (chunked)")
            continue

        try:
            summary = generate_summary(diff, name)
            final_message_parts.append(f"{name}\n{summary}")
        except Exception as e:
            print(f"Error processing {name}: {e}")
            final_message_parts.append(f"{name}\nUpdate file")

    return jsonify({"commit_message": "\n\n".join(final_message_parts)})

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=7860)