[PYTHON] The Advent calendar has been posted! Notify Slack

It is the 9th day of aratana Advent Calendar 2019.

The company Slack has a channel for advent calendar, and when it is posted on the aratana advent calendar, I will notify Slack. (One hour interval I made it when I was in Advent a year ago, but I'll rewrite it and paste the code.

Main operation

1. Notify the "Title, Date, URL" of the posted article
2. Check for the entire period and notify the posted article (because it may be posted at a later date)
3. Notify each post

code

from abc import ABCMeta, abstractmethod
from pathlib import Path
from typing import List, Set

import requests
from bs4 import BeautifulSoup


class AdventCalendarEntry:
    def __init__(self, title: str, url: str, date_str: str) -> None:
        self.title = title
        self.url = url
        self.date_str = date_str

    def empty(self) -> bool:
        if self.title:
            return False

        return True


class AdventCalendarEntryList:
    ITEM_SELECTOR = ".adventCalendarItem"
    DATE_SELECTOR = ".adventCalendarItem_date"
    ENTRY_SELECTOR = ".adventCalendarItem_entry"

    def __init__(self, entry_list: List[AdventCalendarEntry]) -> None:
        self._entry_list = entry_list

    def __iter__(self):
        return iter(self._entry_list)

    @classmethod
    def from_html(cls, html_text: str) -> "AdventCalendarEntryList":
        bs = BeautifulSoup(html_text, "lxml")
        item_list_bs = bs.select(cls.ITEM_SELECTOR)

        entry_list = [cls._make_item_from_item_bs(item_bs) for item_bs in item_list_bs]

        return cls(entry_list)

    @staticmethod
    def _make_item_from_item_bs(item_bs):
        entry_date = item_bs.select_one(AdventCalendarEntryList.DATE_SELECTOR).text

        entry_bs = item_bs.select_one(AdventCalendarEntryList.ENTRY_SELECTOR)
        if not entry_bs:
            #If not posted
            return AdventCalendarEntry(title="", url="", date_str=entry_date)

        _anchor = entry_bs.select_one("a")
        entry_title = _anchor.text
        entry_url = _anchor.get("href")

        return AdventCalendarEntry(
            title=entry_title, url=entry_url, date_str=entry_date
        )


class AdventCalendarDateCache:
    def __init__(self, cache_path: str) -> None:
        self.cache_path = Path(cache_path)
        self.cached_date_set = self.load()

    def load(self) -> Set[str]:
        if not Path(self.cache_path).exists():
            return set()

        with open(self.cache_path, "r") as f:
            checked_list = [line.strip() for line in f.readlines()]

        return set(checked_list)

    def save(self) -> None:
        with open(self.cache_path, "w") as f:
            lines = "\n".join(list(self.cached_date_set))
            f.writelines(lines)

    def add(self, date_str: str) -> None:
        self.cached_date_set.add(date_str)

    def __contains__(self, date_str: str) -> bool:
        return date_str in self.cached_date_set


class Notification(metaclass=ABCMeta):
    @abstractmethod
    def run(self, title: str, entry: AdventCalendarEntry) -> None:
        pass


class PrintNotification(Notification):
    def run(self, title: str, entry: AdventCalendarEntry) -> None:
        message = (
            f"{title}\n"
            f"＜{entry.date_str}＞\n"
            f"TITLE: {entry.title}\n"
            f"URL  : {entry.url}"
        )

        print(message)


class SlackNotification(Notification):
    def __init__(self, webhook_url: str) -> None:
        self.webhook_url = webhook_url

    def run(self, title: str, entry: AdventCalendarEntry) -> None:
        message = (
            f"<!here>\n"
            f"*{title}*\n"
            f"> ＜{entry.date_str}＞\n"
            f"> TITLE: {entry.title}\n"
            f"> URL  : {entry.url}"
        )

        payload = {"text": message, "as_user": True}

        r = requests.post(self.webhook_url, json=payload)
        r.raise_for_status()


class AdventCalendarNotify:
    def __init__(
        self, message_title: str, cache_path: str, notify: Notification
    ) -> None:
        self.message_title = message_title
        self.cache_path = cache_path
        self.notify = notify

        self.date_cache = AdventCalendarDateCache(cache_path)

    def run(self, url: str) -> None:
        for entry in self.load_entry_list(url):
            if not self.new_entry(entry):
                continue

            self.notify.run(self.message_title, entry)
            self.date_cache.add(entry.date_str)

        self.date_cache.save()

    def load_entry_list(self, url: str) -> List[AdventCalendarEntry]:
        r = requests.get(url)
        r.raise_for_status()

        return list(AdventCalendarEntryList.from_html(r.text))

    def new_entry(self, entry: AdventCalendarEntry) -> bool:
        if entry.empty():
            return False

        #Notified is skipped
        if entry.date_str in self.date_cache:
            return False

        return True

Use

Enter the required values as shown below, execute it, and you will be notified of Slack!

url = "https://qiita.com/advent-calendar/2019/aratana"
cache_path = "./list.json"
title = "aratana Advent Calendar Notification of new posts"

slack_notification = SlackNotification(
    [INCOMING_WEBHOOK_Enter the URL!]
)

notify = AdventCalendarNotify(title, cache_path, slack_notification)
notify.run(url)

If you change slack_notification to PrintNotification (), you can check the operation without Slack notification!