Find The Most Pulled Containers From Docker Hub

popular_containers.py

import csv
import re
import requests
import time
import os
from datetime import timedelta
from more_itertools import unique_everseen
from operator import itemgetter

start = time.time()

containers = []
index = []


def _get_containers():
    global containers
    page = 1
    while page <= 10:
        params = (
            ('page_size', '100'),
            ('page', (page)),
            ('q', '+'),
            ('source', 'community'),
            ('type', 'imgage,bundle'),
            ('sort', 'popularity'),
        )
        response = requests.get(
            'https://store.docker.com/api/content/v1/products/search',
            params=params)
        data = response.json()
        try:
            for summaries in data['summaries']:
                try:
                    name = summaries['name']
                    popularity = summaries['popularity']
                    popularity = format(popularity, ",")
                    containers.append([name, popularity, page])
                except KeyError:
                    continue
        except KeyError:
            continue
        page = page + 1
        print(page)
        time.sleep(1)
    containers = list(unique_everseen(containers))
    containers = sorted(containers, key=itemgetter(1))


def _add_to_index():
    _get_containers()
    global containers
    global index
    containerstemp = []
    print("Building Index")
    for elem in containers:
        container, popularity, page, = elem
        url = 'https://hub.docker.com/v2/repositories/%s' % container
        response = requests.get((url))
        data = response.json()
        pulls = (data['pull_count'])
        pulls = format(pulls, ',')
        last_updated = (data['last_updated'])
        try:
            last_updated = re.sub(r'\s*T.*', '', last_updated)
        except TypeError:
            last_updated = ""
        containerstemp.append([container, pulls, last_updated])
    containers = containerstemp


def _write_csv():
    global containers
    _add_to_index()
    with open('popular.csv', 'w', newline='') as csvfile:
        writer = csv.DictWriter(csvfile,
                                fieldnames=["Container",
                                            "Pulls",
                                            "Last Updated",
                                            ])
        writer.writeheader()
        writer = csv.writer(csvfile)
        writer.writerows(containers)

def main():
    _write_csv()
    _cleanup()
    elapsed = (time.time() - start)
    print("\n\nDone In:")
    print(str(timedelta(seconds=elapsed)))


main()