app.py

import json
import re
import threading
import gradio as gr

# ==============================================================================
# 1. CONFIGURATION
# ==============================================================================

# --- General Configuration ---
CONTACT_NUMBER = "+91-8977513427"
COURSE_CATALOG = []
# UPDATED: The script now reads your nested file directly.
JSON_FILE_PATH = "final_merged_course_data.json"

# ==============================================================================
# 2. COURSE CATALOG AND SEARCH LOGIC
# ==============================================================================

ABBREVIATION_MAP = { "cs": "computer science", "cse": "computer science engineering", "llb": "bachelor laws", "mca": "master computer applications", "dvoc": "diploma vocation" }
QUERY_STOP_WORDS = {'courses', 'course', 'in', 'of', 'what', 'are', 'the', 'offered', 'show', 'me', 'tell', 'and', 'vs', 'versus', 'compare'}

def generate_keywords(text: str) -> set:
    if not isinstance(text, str):
        return set()
    text = re.sub(r'[^\w\s.-]', '', text).lower().replace('.', '')
    words = text.split()
    words = [word for word in words if word not in QUERY_STOP_WORDS]
    expanded_words = set(words)
    for word in words:
        if word in ABBREVIATION_MAP:
            expanded_words.update(ABBREVIATION_MAP[word].split())
    return expanded_words

# ==============================================================================
# NEW DATA LOADING LOGIC TO HANDLE NESTED JSON
# ==============================================================================

def parse_nested_courses(parent_name, data, course_list):
    """
    A recursive function to navigate the nested dictionary and extract course info.
    """
    if isinstance(data, dict):
        # BASE CASE: We've found a course entry if it has these keys.
        if 'duration' in data and 'admission_fee' in data:
            course_entry = {
                "name": parent_name.strip(),
                "duration": data.get("duration"),
                "admission_fee": data.get("admission_fee")
            }
            course_list.append(course_entry)
            return

        # RECURSIVE STEP: It's a category, so go deeper.
        for key, value in data.items():
            # Create a meaningful name by combining the parent and current keys.
            if parent_name:
                # Avoid redundant names like "B.Com (Commerce)" if key is already in parent.
                if key.lower() in parent_name.lower() or key.lower() == 'general':
                    new_name = parent_name
                else:
                    new_name = f"{parent_name} ({key})"
            else:
                # This is for the top-level keys
                new_name = key
            
            parse_nested_courses(new_name, value, course_list)


def initialize_course_catalog():
    """
    Initializes the course catalog by loading and flattening the nested JSON file.
    """
    global COURSE_CATALOG
    print(f"Attempting to initialize course catalog from nested file '{JSON_FILE_PATH}'...")
    
    try:
        with open(JSON_FILE_PATH, 'r', encoding='utf-8') as f:
            nested_json_data = json.load(f)

        # This list will hold the simple, flattened course data.
        flat_course_list = []
        # Start the recursive parsing process.
        parse_nested_courses("", nested_json_data, flat_course_list)

        if not flat_course_list:
            print(f"🔴 FATAL ERROR: Could not find any valid course entries in '{JSON_FILE_PATH}'.")
            return
            
        # Now, build the final COURSE_CATALOG with keywords, just like before.
        catalog = []
        for course_data in flat_course_list:
            if isinstance(course_data, dict) and 'name' in course_data:
                details = course_data
                keywords = generate_keywords(details['name'])
                catalog.append({"name": details['name'], "keywords": keywords, "details": details})
            else:
                print(f"⚠️ WARNING: Skipping invalid course entry after processing: {course_data}")

        COURSE_CATALOG = catalog
        print(f"✅ Successfully loaded and processed {len(COURSE_CATALOG)} courses from the nested file.")

    except FileNotFoundError:
        print(f"🔴 FATAL ERROR: The file '{JSON_FILE_PATH}' was not found.")
        raise
    except json.JSONDecodeError:
        print(f"🔴 FATAL ERROR: Could not decode JSON from '{JSON_FILE_PATH}'. Please check for syntax errors.")
        raise
    except Exception as e:
        print(f"🔴 FATAL ERROR during course initialization: {e}")
        raise e

# ==============================================================================
# (The rest of the code is unchanged)
# ==============================================================================

def find_best_course_match(query: str, force_single_result: bool = False) -> dict | None:
    if not COURSE_CATALOG: return None
    query_keywords = generate_keywords(query)
    if not query_keywords: return None

    scored_matches = []
    for course in COURSE_CATALOG:
        intersection = query_keywords.intersection(course['keywords'])
        score = len(intersection) * 10
        if query_keywords.issubset(course['keywords']):
            score += 20
        score -= len(course['keywords']) - len(intersection)
        if score > 0:
            scored_matches.append((score, course))

    if not scored_matches: return None

    scored_matches.sort(key=lambda x: x[0], reverse=True)
    best_score, best_match_course = scored_matches[0]
    result = {"best_match": best_match_course["details"]}

    is_ambiguous = (not force_single_result and len(scored_matches) > 1 and (best_score - scored_matches[1][0] < 5))
    if is_ambiguous:
        result["suggestions"] = [match[1]['name'] for match in scored_matches]
    return result

# ==============================================================================
# 4. CORE CHATBOT LOGIC
# ==============================================================================
GREETING_WORDS = {"hello", "hi", "hey", "hlo", "yo"}

def get_user_intent(user_message: str) -> str:
    message = user_message.lower()
    if any(word in message for word in ["compare", "vs", "versus"]): return "compare_courses"
    words = set(message.split())
    if len(words) <= 2 and words.intersection(GREETING_WORDS): return "greeting"
    if any(word in message for word in ["fee", "fees", "cost", "price", "much"]): return "find_fee"
    if any(word in message for word in ["duration", "long", "length", "years"]): return "find_duration"
    return "general_inquiry"

def process_user_query(user_message: str, history: list) -> tuple:
    print(f"\nReceived new query: '{user_message}'")
    if not COURSE_CATALOG:
        bot_message = "🔴 Error: The course catalog isn't loaded. Please check the server logs for errors."
        history.append((user_message, bot_message))
        return "", history

    intent = get_user_intent(user_message)
    print(f"Detected intent: '{intent}'")
    bot_message = ""

    try:
        if intent == "greeting":
            bot_message = "Hello! I am your AI Course Assistant. How can I help?"
        elif intent == "compare_courses":
            parts = re.split(r'\s+(?:and|vs|versus)\s+', user_message, flags=re.IGNORECASE)
            if len(parts) < 2:
                bot_message = "Please tell me the two courses you want to compare. E.g., 'compare btech cse and llb'."
            else:
                details_list = []
                for part in parts:
                    if part.strip():
                        result = find_best_course_match(part.strip(), force_single_result=True)
                        if result and result.get("best_match"):
                            details_list.append(result["best_match"])
                if len(details_list) < 2:
                    bot_message = "I had trouble finding distinct details for both courses. Please try again."
                else:
                    header = "| Feature | " + " | ".join([d.get('name', 'N/A') for d in details_list]) + " |"
                    separator = "|---|" + "---|" * len(details_list)
                    fee_row = "| Admission Fee | " + " | ".join([str(d.get('admission_fee', 'N/A')) for d in details_list]) + " |"
                    duration_row = "| Duration | " + " | ".join([d.get('duration', 'N/A') for d in details_list]) + " |"
                    bot_message = f"{header}\n{separator}\n{fee_row}\n{duration_row}"
        else:
            result = find_best_course_match(user_message)
            if not result:
                bot_message = "I couldn't find any courses matching your query. Please try a different name."
            elif result.get("suggestions"):
                suggestions_text = "\n".join([f"• {name}" for name in result["suggestions"][:5]])
                bot_message = f"I found a few related courses. Did you mean one of these?\n\n{suggestions_text}\n\nPlease ask again with the full name for details."
            elif result.get("best_match"):
                details = result["best_match"]
                course_name = details.get("name", "N/A")
                response_parts = [f"Showing details for **{course_name}**:"]
                if intent == "find_fee" and details.get("admission_fee"):
                    response_parts.append(f"- **Admission Fee:** {details['admission_fee']}")
                elif intent == "find_duration" and details.get("duration"):
                    response_parts.append(f"- **Duration:** {details['duration']}")
                else:
                    if details.get("admission_fee"): response_parts.append(f"- **Admission Fee:** {details['admission_fee']}")
                    if details.get("duration"): response_parts.append(f"- **Duration:** {details['duration']}")
                if len(response_parts) == 1:
                    response_parts.append("- No specific fee or duration info found.")
                response_parts.append(f"\nFor more help, contact admissions at **{CONTACT_NUMBER}**.")
                bot_message = "\n".join(response_parts)

        history.append((user_message, bot_message))
        print(f"Generated response: '{bot_message[:100]}...'")
        return "", history

    except Exception as e:
        print(f"🔴 CRITICAL ERROR in process_user_query: {e}")
        error_message = "I'm sorry, I encountered a critical error on my server. Please try again later."
        history.append((user_message, error_message))
        return "", history

# ==============================================================================
# 5. GRADIO INTERFACE AND APP LAUNCH
# ==============================================================================

initialize_course_catalog()

with gr.Blocks(theme="soft", title="AATM Course Assistant") as iface:
    gr.Markdown("## 🤖 AI Course Assistant\nYour guide to course fees, duration, and comparisons.")
    chatbot = gr.Chatbot(height=600, bubble_full_width=False, label="AATM Chat", elem_id="chatbot")
    with gr.Row():
        msg = gr.Textbox(
            show_label=False,
            placeholder="E.g., 'fee for mca', or 'compare btech cse and llb'...",
            container=False,
            scale=7,
            elem_id="chatbot_input"
        )
        clear = gr.ClearButton([msg, chatbot], scale=1)

    msg.submit(process_user_query, [msg, chatbot], [msg, chatbot], api_name="process_user_query")

if __name__ == "__main__":
    iface.launch()