HistoricalWeatherTW: A Python repository from AaronCHH

1 HistoricalWeatherTW 台灣歷史天氣爬蟲

This script is to crawl the information of 觀測資料查詢 website

2 Usage

Prepare station.csv.

Note

station.csv (you can find it from the https://e-service.cwb.gov.tw/wdps/obs/state.htm)

run the function of collect_weather_tw you will see the result!

def collect_weather_tw(station_csv_path: Path, output_path,
                       end_date: datetime.date, begin_date: datetime.date,
                       query_format,
                       convert2num):

You can refer to __init__.py for more help

Prepare config.yaml and use this path as input parameter

run __init__.py

2.1 QUICKLY START

from Carson.Tool.HistoricalWeatherTW import collect_weather_tw, QueryFormat
from pathlib import Path
import datetime
import os

if __name__ == '__main__':
    STATION_CSV = '../config/CSV/station_test.csv'
    OUTPUT_PATH = Path(f'../temp/year_result.csv')
    BEGIN_DATE = datetime.date(2019, 10, 1)
    END_DATE = datetime.date(2019, 10, 2)
    QUERY_FORMAT = QueryFormat.DAY
    CONVERT2NUM = True
    collect_weather_tw(Path(STATION_CSV), OUTPUT_PATH,
                       BEGIN_DATE, END_DATE,
                       QUERY_FORMAT,
                       CONVERT2NUM)
    os.startfile(OUTPUT_PATH)

3 Data

The output depends on QueryFormat!

Note

There have some non-numeric forms of the original data. Such as indefinite wind direction V, rain track T, etc. the variable of CONVERT2NUM will replace them with numbers.

4 Release note

4.1 v4.0

Encapsulated as API

4.2 v3.0

Features:

All outputs into a single file. (making it easier to use for SQL)

The output header field will automatically grab the content on the web page (not use hard coding)

You can choose the type of query (year, month, day) according to your needs.

Other:

Make the code easier to read.

4.3 V2.0

加入全台觀測站

4.4 V1.0

第一版

AaronCHH/HistoricalWeatherTW