mapping_to_yaml.py

import csv
import yaml
import os
from copy import deepcopy

with open("example.yml", "r") as file:
    yaml_template = yaml.safe_load(file)


def create_yaml_from_template(data, template, output_dir):
    # Use deepcopy to ensure the template is not modified
    new_yaml = deepcopy(template)
    print(data["Concept Name"])
    # Directly update the necessary fields
    new_yaml["name"] = data["Concept Name"]
    new_yaml["description"] = data["Description"]
    new_yaml["unit"] = data["Standard Unit"]
    new_yaml["upper_limit"] = data["Absolute Upper"]
    new_yaml["lower_limit"] = data["Absolute Lower"]
    new_yaml["identifiers"]["snomed"] = data["SCT CONEPT ID"]
    new_yaml["identifiers"]["loinc"] = data["LOINC Code"]
    new_yaml["mapper"][0]["source"] = "mimiciv"
    new_yaml["mapper"][0]["unit"] = data["MIMIC Unit"]
    new_yaml["mapper"][0]["params"]["schema"] = data["MIMICIV schema"]
    new_yaml["mapper"][0]["params"]["table"] = data["MIMICIV table"]
    new_yaml["mapper"][0]["params"]["constraints"]["itemid"] = data["MIMICIV IDs"]

    # Write the updated YAML to a new file
    output_file = os.path.join(output_dir, f"{data['Concept Name']}.yml")
    with open(output_file, "w") as outfile:
        yaml.dump(new_yaml, outfile, allow_unicode=True, default_flow_style=False)


# Read the CSV file and process flagged rows
csv_file = "main_mapping.csv"
output_directory = "concepts"

# Create the output directory if it doesn't exist
os.makedirs(output_directory, exist_ok=True)

with open(csv_file, "r") as csvfile:
    reader = csv.DictReader(csvfile)
    for row in reader:
        if row.get("ready for conversion") == "TRUE":
            create_yaml_from_template(row, yaml_template, output_directory)

print("Conversion completed. YAML files are created in the 'converted_yamls' directory.")