parse_sources.py

import re
import json

def parse_source_schema_v1(text_data):
    """
    Parses 'Source Information' blocks according to Schema Specification v1.0.
    Handles Audit Fields (Date-Entry-Created, etc.) and lists (Tags).
    """
    # Capture blocks starting with "# Source Information"
    entry_pattern = re.compile(r'# Source Information\n(.*?)(?=\n# Source Information|\Z)', re.DOTALL)
    
    # Capture "**Key:** Value"
    field_pattern = re.compile(r'\*\*(.*?):\*\*\s*(.*)')
    
    entries = entry_pattern.findall(text_data)
    parsed_data = []

    for entry in entries:
        record = {}
        lines = entry.strip().split('\n')
        
        for line in lines:
            match = field_pattern.match(line.strip())
            if match:
                key = match.group(1).strip()
                value = match.group(2).strip()
                
                # --- Type Conversion & Cleaning ---
                
                # Handle Nulls
                if value.lower() in ["not visible", "not displayed", "n/a", ""]:
                    value = None
                
                # Handle Integers
                elif key == "Stars":
                    try:
                        value = int(value)
                    except ValueError:
                        value = None # Keep null if strictly not a number
                        
                # Handle Lists (Tags)
                elif key == "Tags":
                    if value:
                        # Split by comma, strip whitespace, remove empty strings
                        value = [t.strip() for t in value.split(',') if t.strip()]
                    else:
                        value = []
                
                # Store the field
                record[key] = value
        
        if record:
            parsed_data.append(record)
            
    return parsed_data

# --- Test with the new Schema ---

if __name__ == "__main__":
    # Reading the file content (Simulated here with a string)
    with open('skills.md', 'r') as f:
        content = f.read()

    data = parse_source_schema_v1(content)
    
    # print output
    print(json.dumps(data, indent=2))