suggestion.py

import sys
import os
import json
import urllib.request
import urllib.error

def get_biotools_data(tool_id):
    """Fetches raw tool metadata from the bio.tools API."""
    url = f"https://bio.tools/api/tool/{tool_id}?format=json"
    print(f"Fetching metadata for '{tool_id}' from bio.tools...")
    try:
        with urllib.request.urlopen(url) as response:
            if response.status == 200:
                return json.loads(response.read().decode())
    except urllib.error.HTTPError as e:
        print(f"Error fetching data from bio.tools: {e}", file=sys.stderr)
        sys.exit(1)
    except urllib.error.URLError as e:
        print(f"Network error: {e}", file=sys.stderr)
        sys.exit(1)

def load_valid_operations(obo_path):
    """Reads the clean OBO file and collects all present operation codes."""
    valid_ops = set()
    if not os.path.exists(obo_path):
        print(f"Warning: '{obo_path}' not found. All codes will be marked as 'unknown code'.")
        return valid_ops

    with open(obo_path, "r", encoding="utf-8") as f:
        for line in f:
            if line.strip().startswith("id:"):
                # Extracts "operation_XXXX" from lines like "id: operation_XXXX" or URIs
                raw_id = line.split("id:")[1].strip()
                clean_id = raw_id.split("/")[-1]
                valid_ops.add(clean_id)
    return valid_ops

def parse_operations(data, tool_id, valid_ops):
    """Parses JSON payload, validates IDs against the OBO file, and formats lines."""
    formatted_operations = []
    functions = data.get("function", [])
    
    for func in functions:
        for op in func.get("operation", []):
            term_name = op.get("term")
            uri = op.get("uri", "")
            
            if term_name and uri:
                # Extract the operation ID from the URI
                operation_id = uri.split("/")[-1]
                
                # Convert operation_XXXX to DIO:000XXXX to match the new OBO format
                if "operation_" in operation_id:
                    num = operation_id.split("operation_")[1]
                    dio_id = f"DIO:{num.zfill(7)}"
                else:
                    dio_id = operation_id

                # Check if the operation code exists in our clean OBO set
                if dio_id in valid_ops:
                    formatted_line = f'"{term_name}": {dio_id}\t{tool_id}'
                else:
                    # Append 'unknown code' after the program name if missing
                    formatted_line = f'"{term_name}": {dio_id}\t{tool_id} unknown code'
                
                if formatted_line not in formatted_operations:
                    formatted_operations.append(formatted_line)
                
    return sorted(formatted_operations)

def main():
    if len(sys.argv) != 2:
        print("Usage: python3 script.py <name_of_program>", file=sys.stderr)
        sys.exit(1)
        
    tool_id = sys.argv[1]
    obo_file_path = "/data/dio.obo" # Points to the finalized DIO OBO file
    
    # 1. Load existing valid operations from the clean OBO file
    valid_ops = load_valid_operations(obo_file_path)
    
    # 2. Fetch & Parse data from bio.tools
    raw_data = get_biotools_data(tool_id)
    operations = parse_operations(raw_data, tool_id, valid_ops)
    
    # 3. Save formatted classifications to the designated file (append mode)
    output_filename = "/data/software_classification_suggestion"
    
    print(f"Adding classification suggestions to '{output_filename}'...")
    with open(output_filename, "a", encoding="utf-8") as out:
        for op in operations:
            out.write(op + "\n")
        
    print("Execution complete. Data saved successfully.")

if __name__ == "__main__":
    main()