Update language_detector.py

language_detector.py

@@ -1,4 +1,4 @@
-# language_detector.py
+# language_detector.py - FINAL EXPERT VERSION
 from langdetect import detect, DetectorFactory
 import re
 
@@ -7,7 +7,6 @@ DetectorFactory.seed = 0
 def detect_input_language(text):
     """User input ki language detect kare"""
     try:
-        # Clean text
         clean_text = re.sub(r'[^\w\s]', '', text)
         if not clean_text.strip():
             return 'mixed'
@@ -23,45 +22,125 @@ def detect_input_language(text):
         return 'mixed'
 
 def get_system_prompt(lang_mode, username):
-    """Language ke hisaab se system prompt generate kare"""
+    """AI ko expert coding ke liye train kare"""
     
-    hindi_prompt = f"""
-    तुम AumCore AI हो - Senior AI Architect।
-    यूजर: {username}।
+    # COMMON RULES FOR ALL LANGUAGES
+    common_rules = f"""
+    YOU ARE AUMCORE AI - SENIOR AI ARCHITECT & CODING EXPERT.
+    USER: {username}.
     
-    सख्त नियम:
-    1. भाषा: 100% हिंदी (देवनागरी लिपि)
-    2. कोड: अगर {username} कोड मांगे, तो 350+ लाइन का production-ready Python script दो
-    3. स्टाइल: Professional, powerful, direct (मैक्स 4 लाइन)
-    4. व्यक्तित्व: Master AI की तरह बात करो, basic chatbot नहीं
-    """
+    ABSOLUTE RULES:
+    1. CODING: When user asks for code, provide FULL PRODUCTION-READY CODE (300+ lines).
+    2. CODE FORMAT: Output ONLY RAW CODE, NO markdown blocks (```python```), NO explanations.
+    3. CODE QUALITY: Include error handling, logging, documentation, modular functions.
+    4. RESPONSE STYLE: Concise, powerful, direct (Max 4 lines for non-code responses).
     
-    english_prompt = f"""
-    You are AumCore AI - Senior AI Architect.
-    User: {username}.
+    CODING EXAMPLES:
+    - User: "google drive mount code"
+    - You: "from google.colab import drive\ndrive.mount('/content/gdrive')"
     
-    Strict Rules:
-    1. Language: 100% English
-    2. Code: If {username} asks for code, provide 350+ lines of production-ready Python script
-    3. Style: Professional, powerful, direct (Max 4 lines)
-    4. Persona: Talk like a Master AI, not a basic chatbot
+    - User: "web app code"
+    - You: 300+ lines of Flask/FastAPI code
     """
     
-    mixed_prompt = f"""
-    You are AumCore AI - Senior AI Architect.
-    User: {username}.
+    # LANGUAGE SPECIFIC INSTRUCTIONS
+    language_instructions = {
+        'hindi': """
+        भाषा: 100% हिंदी (कोड के अलावा)
+        उदाहरण: "नमस्ते, कोड बताओ" → आप सिर्फ कोड दो, हिंदी explanation नहीं
+        """,
+        
+        'english': """
+        Language: 100% English (except code)
+        Example: "hello, give me code" → You provide only code, no English explanation
+        """,
+        
+        'mixed': """
+        Language: 60% English + 40% Hindi blended
+        Example: "hi bhai, code de do" → You provide only code, no mixed explanation
+        """
+    }
     
-    Strict Rules:
-    1. Language: 60% English + 40% Hindi (blended in single response)
-    2. Code: If {username} asks for code, provide 350+ lines of production-ready Python script
-    3. Style: Professional, powerful, direct (Max 4 lines)
-    4. Persona: Talk like a Master AI, not a basic chatbot
-    """
+    # FINAL PROMPT
+    final_prompt = common_rules + language_instructions.get(lang_mode, language_instructions['mixed'])
     
-    prompts = {
-        'hindi': hindi_prompt,
-        'english': english_prompt,
-        'mixed': mixed_prompt
+    return final_prompt.strip()
+
+# TEST FUNCTION FOR CODE GENERATION
+def generate_expert_code(task_description):
+    """Expert code generation logic (for future enhancement)"""
+    code_templates = {
+        'web': """
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+import uvicorn
+import logging
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+app = FastAPI(title="Professional Web Service")
+
+class Item(BaseModel):
+    name: str
+    price: float
+    quantity: int
+
+@app.get("/")
+async def root():
+    return {"message": "Welcome to AumCore AI Web Service"}
+
+@app.post("/items/")
+async def create_item(item: Item):
+    try:
+        logger.info(f"Creating item: {item.name}")
+        # Business logic here
+        return {"status": "success", "item": item.dict()}
+    except Exception as e:
+        logger.error(f"Error: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+if __name__ == "__main__":
+    uvicorn.run(app, host="0.0.0.0", port=8000)
+""",
+        'data': """
+import pandas as pd
+import numpy as np
+from sklearn.model_selection import train_test_split
+from sklearn.ensemble import RandomForestClassifier
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+class DataAnalyzer:
+    def __init__(self, filepath):
+        self.df = pd.read_csv(filepath)
+        self.results = {}
+    
+    def analyze(self):
+        # Comprehensive data analysis
+        self.results['shape'] = self.df.shape
+        self.results['columns'] = list(self.df.columns)
+        self.results['missing'] = self.df.isnull().sum()
+        return self.results
+    
+    def visualize(self):
+        # Professional visualizations
+        fig, axes = plt.subplots(2, 2, figsize=(15, 10))
+        # Plotting logic...
+        plt.tight_layout()
+        return fig
+
+# Usage example
+if __name__ == "__main__":
+    analyzer = DataAnalyzer("data.csv")
+    print(analyzer.analyze())
+"""
     }
     
-    return prompts.get(lang_mode, mixed_prompt)
+    # Simple keyword matching (can be enhanced with AI)
+    if 'web' in task_description.lower() or 'app' in task_description.lower():
+        return code_templates['web']
+    elif 'data' in task_description.lower() or 'analy' in task_description.lower():
+        return code_templates['data']
+    else:
+        return code_templates['web']  # Default