image_downloader_mongo.py

import asyncio
import io
import imghdr
from PIL import Image

from aiohttp import ClientSession
import motor.motor_asyncio


client = motor.motor_asyncio.AsyncIOMotorClient()
db = client.test_database


async def save_to_mongo(document):
    result = await db.test_collection.insert_one(document)
    return result


def get_image_sizes(image_bytes):
    img = io.BytesIO(image_bytes)

    assert imghdr.what(img), 'It isnt an image'

    image = Image.open(img)
    return image.width, image.height


async def download(url):
    async with ClientSession() as session:
        async with session.get(url) as resp:
            assert resp.status == 200
            result = await resp.read()

    return result


async def process_images(urls):
    assert len(urls) < 50

    async def process_image(url):
        image_data = await download(url)
        width, height = get_image_sizes(image_data)
        result = await save_to_mongo({
            'url': url,
            'width': width,
            'height': height
        })
        return result

    await asyncio.gather(
        *[process_image(url) for url in urls],
        return_exceptions=True
    )


if __name__ == '__main__':
    urls = [
        "http://www.e1.ru",  # isn't image
        "https://static.ngs.ru/news/99/preview/f026eb9b2ec0ec137f7578384744cd3f085b5da3_1079.jpg",
        "https://static.ngs.ru/news/99/preview/6fc818ccd6eef6daa10e06045347c5972176888dd_700.JPG",
        "http://www.blog.pythonlibrary.org/wp-content/uploads/2018/07/jupy_wp_widget.jpg",
    ]
    loop = asyncio.get_event_loop()
    loop.run_until_complete(process_images(urls*10))