Random Wikipedia Articles using Python pool.appy_async

random_article_async.py

"""
Mit __iter__ kann man komplexe Iterationen abstrahieren
- Iteratoren sind Zustandsautomaten.
- Dadurch lässt sich das iterieren vom prozessieren trennen.
- So kann man den gleichen Iterator seriell oder parallel umsetzen
"""
import json
import time
import requests
import multiprocessing


def title(json_string):
    return json.loads(json_string).get('title')


class RandomArticles:
    def __init__(self, n, url):
        self._url = url
        self._n = n

    def _download(self, *args):
        return requests.get(self._url).content

    def __iter__(self):
        # Version 1
        for _ in range(self._n):
            yield self._download()

        # Version 2
        with multiprocessing.Pool() as pool:
            future_results = [
                pool.apply_async(self._download)
                for _ in range(self._n)
            ]
            for result in future_results:
                yield result.get()



class Stopwatch:
    def __enter__(self):
        self._start = time.time()

    def __exit__(self, *args):
        end = time.time()
        elapsed = end - self._start
        print(f'{elapsed:.2f}s')


if __name__ == '__main__':

    url = "https://en.wikipedia.org/api/rest_v1/page/random/summary"

    with Stopwatch():
        for article in RandomArticles(10, url):
            print(title(article))