[Python x Zapier] Get alert information and notify with Slack

What you want to do (BEFORE)

** Get the warning information of the Japan Meteorological Agency and notify by Slack if there is a warning in the specified area **

This article What you want to do is Don Pisha! I thought that I was proceeding according to the article, but every time the data of the Japan Meteorological Agency is updated, I am running the task of Zapier (a tool that can automate the work) My plan exceeds the maximum task ...

So I narrowed down the scope of what I wanted to do.

What you want to do (AFTER)

** Get the warning information of the Japan Meteorological Agency at "7:00 every morning" and notify you with Slack if there is a warning in the designated area ** ▼ It looks like this (it looks like I got a warning instead because there was no alarm when I moved it) I got the information only once every morning at 7 o'clock and set the number of task executions to once a day.

Zap overview

1. Every Day: Set to run the task here every morning at 7am
2. Run Python: Get / extract information from Japan Meteorological Agency (code below)
3. Only continue if ...: If there is an output in 2, proceed to the next (if an alarm is issued)
4. Send Channel Message in Slack: Notify on the channel specified by Slack

Python code

I found out how to scrape and wrote it because I thought I had to study because what I wanted to do changed a little. First Python! When I was researching, I found out that I would use something called Beautiful Soup, but Zapier Help ) Wrote something like this. Oh, oh, I can't use Beautiful Soup! There is a feeling of force, but I will extract only the desired part (alarm part) without using Beautiful Soup.

import re
import requests
import time

code_list = [['Sapporo', '0110000'], ['Eastern Sendai City', '0410001'], ['Saitama City', '1110000']]
output = {'text': ''}

for code in code_list:
    html = requests.get('https://www.jma.go.jp/jp/warn/f_' + code[1] + '.html').text
    data_list = re.findall('<span style="color:#FF2800">(.*?)</span>', html)

    if len(data_list) == 0:
        continue

    text = '【' + code[0] + '】'
    i = 1
    for data in data_list:
        if i == len(data_list):
            text += data
        else:
            data = re.sub('alarm', '', data)
            text  += data + '，'
        i += 1
    output['text'] += text + '\n'
    time.sleep(1)

Since the font color of the alarm part is # FF2800 on the page of each region, all the parts surrounded by<span style = "color: # FF2800"> ~ </ span>are extracted. Also, since it scrapes multiple pages, rest for 1 second with time.sleep (1).

reference

Obtain information from the Japan Meteorological Agency and notify Slack of weather warnings in the 23 wards of Tokyo I was very helpful including how to use Zapier.