requests下载文件并重新上传

Posted 2021-02-11 li1992

import re
import requests
from io import BytesIO
from django.core.files.uploadedfile import InMemoryUploadedFile

u = "http[s]?://(?:[a-zA-Z]|[0-9]|[[email protected]&+]|[!*(),]|(?:%[0-9a-fA-F][0-9a-fA-F]))+"


def match_by_re(str):
    pattern = re.compile(u, re.S)
    res = pattern.findall(string=str)
    if res: return res[0]


def get_suffix(url):
    res = url.split(‘.‘)
    return res[-1]


def get_name_content_type(url):
    suffix = get_suffix(url)
    image_suffix = [‘BMP‘, ‘JPG‘, ‘JPEG‘, ‘PNG‘, ‘GIF‘]
    text_suffix = [‘JSON‘]
    audio_suffix = [‘ACT‘, ‘REC‘, ‘AAC‘, ‘SC4‘, ‘DVF‘, ‘MSC‘, ‘WMA‘, ‘MP3‘, ‘WAV‘]
    name = ‘temp.{}‘.format(suffix)
    if suffix.upper() in image_suffix:
        return name, ‘image/jpeg‘
    elif suffix.upper() in text_suffix:
        return name, ‘application/json‘
    elif suffix.upper() in audio_suffix:
        return name, ‘audio/mp3‘


class DownloadTool():

    def __init__(self):
        self.los = {}

    def get_stream_data(self, url):
        stream_data = self.los.get(url)
        return stream_data

    def stream_download(self, url):
        url = match_by_re(url)
        stream = self.get_stream_data(url)
        if stream:
            return stream
        r = requests.get(url, stream=True)
        self.los[url] = r.content
        return r.content

    def get_file_obj(self, url):
        stream = self.stream_download(url)
        fd = BytesIO(stream)
        name, content_type = get_name_content_type(url)
        file_obj = InMemoryUploadedFile(fd, ‘url‘, name, content_type, len(stream), charset=‘utf-8‘)
        return file_obj

file_obj时Django的上传文件对象，可以作为model表中FileField字段的值