import csv
import json
import re

def parse_description(description_text):
    """
    Parses the description text into a structured dictionary.
    """
    # Headers to look for in the description text.
    # The order is important for splitting.
    headers = [
        "Background",
        "Problem Description",
        "Description",
        "Impact / Why this problem needs to be solved",
        "Impact",
        "Expected Solution",
        "Expected Outcomes",
        "Relevant Stakeholders / Beneficiaries",
        "Supporting Data",
        "Objective",
        "Technical Scope",
        "Data Acquisition & Sampling",
        "Data Processing & Analysis",
        "Output & Storage",
        "Key Performance Parameters",
        "Eligibility",
        "Evaluation Criteria",
        "Deliverables",
        "Conclusion",
        "Innovative Features",
        "Key Features",
        "Additional Features",
        "Digital Tourist ID Generation Platform",
        "Mobile Application for Tourists",
        "AI-Based Anomaly Detection",
        "Tourism Department & Police Dashboard",
        "IoT Integration (Optional)",
        "Multilingual Support",
        "Data Privacy & Security"
    ]

    # Create a regex pattern to split the text by the headers.
    # The pattern looks for a header at the beginning of a line.
    pattern = r'^\s*(' + '|'.join(re.escape(h) for h in headers) + r')\s*$'
    
    # Split the text by the headers
    parts = re.split(pattern, description_text, flags=re.MULTILINE)
    
    details = {}
    # The first part is the introduction if it's not a header
    if parts[0].strip():
        details['introduction'] = parts[0].strip()

    # The rest of the parts are pairs of (header, content)
    it = iter(parts[1:])
    for header in it:
        content = next(it, "").strip()
        if header and content:
            # Normalize header to be used as a json key
            key = header.strip().lower().replace(' ', '_').replace('/', '_').replace('(', '').replace(')', '')
            details[key] = content

    # If details is empty, it means no headers were found.
    # In this case, the whole text is the description.
    if not details:
        return {'full_text': description_text.strip()}
        
    return details

def convert_csv_to_json(csv_file_path, json_file_path):
    """
    Converts a CSV file to a structured JSON file.
    """
    problems = []
    with open(csv_file_path, mode='r', encoding='utf-8') as csv_file:
        # Use DictReader to read CSV rows as dictionaries
        csv_reader = csv.DictReader(csv_file)
        
        for row in csv_reader:
            # Clean up keys from the CSV header
            cleaned_row = {key.strip().replace(' ', '_').replace('.', ''): value for key, value in row.items()}
            
            problem_data = {
                's_no': int(cleaned_row.get('SNo', 0)),
                'organization': cleaned_row.get('Organization', ''),
                'title': cleaned_row.get('Problem_Statement_Title', ''),
                'category': cleaned_row.get('Category', ''),
                'ps_number': cleaned_row.get('PS_Number', ''),
                'submitted_ideas_count': int(cleaned_row.get('Submitted_Ideas_Count', 0)),
                'theme': cleaned_row.get('Theme', ''),
                'details': parse_description(cleaned_row.get('Problem_Description', ''))
            }
            problems.append(problem_data)

    with open(json_file_path, mode='w', encoding='utf-8') as json_file:
        json.dump({"problems": problems}, json_file, indent=2)

if __name__ == '__main__':
    csv_input_file = 'SIH_Problem_Statements.csv'
    json_output_file = 'SIH_Problem_Statements_structured.json'
    
    convert_csv_to_json(csv_input_file, json_output_file)
    
    print(f"Successfully converted '{csv_input_file}' to '{json_output_file}'")