scraping_ym.py

import requests
import csv
from datetime import datetime, timedelta


def save_to_csv(data_list, output_file):
    """
    Saves a list of data dictionaries to a CSV file.
    Args:
        data_list (list): List of data rows.
        output_file (str): Path to the output CSV file.
    """
    if not data_list:
        print("No data to save.")
        return

    header = ["record_time", "obs_post_id", "obs_post_name", "obs_lat", "obs_lon", "wind_dir", "wind_speed"]

    with open(output_file, mode='w', newline='', encoding='utf-8') as file:
        writer = csv.writer(file)
        writer.writerow(header)
        writer.writerows(data_list)

    print(f"Data successfully saved to {output_file}")


if __name__ == '__main__':
    # API endpoint and required parameters
    url = 'http://www.khoa.go.kr/api/oceangrid/tideObsWind/search.do'
    authKey = 'zudwtvvTQ3EMLXJjltrCyQ=='  # API authentication key
    obsCode = 'DT_0005'  # Observation Code

    # Date range for data retrieval
    start_date = datetime.strptime('2023-01-01 00:00', '%Y-%m-%d %H:%M')
    end_date = datetime.strptime('2023-01-03 00:00', '%Y-%m-%d %H:%M')

    delta = timedelta(hours=1)  # Time step for hourly iteration
    all_data = []

    while start_date <= end_date:
        # Format the date parameter as YYYYMMDD (API accepts date without time)
        date_str = start_date.strftime('%Y%m%d')
        params = {
            'ServiceKey': authKey,
            'ObsCode': obsCode,
            'Date': date_str,
            'ResultType': 'json'  # Response format
        }

        try:
            response = requests.get(url, params=params)
            response.raise_for_status()  # Raise an HTTPError for bad responses (4xx and 5xx)
            data = response.json()

            if 'result' in data and 'data' in data['result']:
                for item in data['result']['data']:
                    row = [
                        item.get("record_time", "-99"),
                        item.get("obs_post_id", "-99"),
                        item.get("obs_post_name", "-99"),
                        item.get("obs_lat", "-99"),
                        item.get("obs_lon", "-99"),
                        item.get("wind_dir", "-99"),
                        item.get("wind_speed", "-99")
                    ]
                    all_data.append(row)

            print(f"Processed date: {start_date.strftime('%Y-%m-%d %H:%M')}.")

        except Exception as e:
            print(f"Error processing date {start_date.strftime('%Y-%m-%d %H:%M')}: {e}")

        # Increment the date by 1 hour
        start_date += delta

    # Save the collected data to a CSV file
    output_csv = 'khoa_wind_data.csv'
    save_to_csv(all_data, output_csv)