窗外下着小雨，作为单身程序员的我逛着逛着发现一篇好东西，来自知乎 你都用 Python 来做什么？的第一个高亮答案。
到上面去看了看，地址都是明文的，得，赶紧开始吧。
下载流式文件，requests库中请求的stream设为True就可以啦，文档在此。
先找一个视频地址试验一下：
# -*- coding: utf-8 -*-
import requests
def download_file(url, path):
with requests.get(url, stream=True) as r:
chunk_size = 1024
content_size = int(r.headers['content-length'])
print '下载开始'
with open(path, "wb") as f:
for chunk in r.iter_content(chunk_size=chunk_size):
f.write(chunk)
if __name__ == '__main__':
url = '就在原帖...'
path = '想存哪都行'
download_file(url, path)
遭遇当头一棒：
AttributeError: __exit__
这文档也会骗人的么！
看样子是没有实现上下文需要的__exit__方法。既然只是为了保证要让r最后close以释放连接池，那就使用contextlib的closing特性好了：
# -*- coding: utf-8 -*-
import requests
from contextlib import closing
def download_file(url, path):
with closing(requests.get(url, stream=True)) as r:
chunk_size = 1024
content_size = int(r.headers['content-length'])
print '下载开始'
with open(path, "wb") as f:
for chunk in r.iter_content(chunk_size=chunk_size):
f.write(chunk)
程序正常运行了，不过我盯着这文件，怎么大小不见变啊，到底是完成了多少了呢？还是要让下好的内容及时存进硬盘，还能省点内存是不是：
# -*- coding: utf-8 -*-
import requests
from contextlib import closing
import os
def download_file(url, path):
with closing(requests.get(url, stream=True)) as r:
chunk_size = 1024
content_size = int(r.headers['content-length'])
print '下载开始'
with open(path, "wb") as f:
for chunk in r.iter_content(chunk_size=chunk_size):
f.write(chunk)
f.flush()
<b>本文来源gao@!dai!ma.com搞$$代^@码5网@</b>os.fsync(f.fileno())
文件以肉眼可见的速度在增大，真心疼我的硬盘，还是最后一次写入硬盘吧，程序中记个数就好了：
def download_file(url, path):
with closing(requests.get(url, stream=True)) as r:
chunk_size = 1024
content_size = int(r.headers['content-length'])
print '下载开始'
with open(path, "wb") as f:
n = 1
for chunk in r.iter_content(chunk_size=chunk_size):
loaded = n*1024.0/content_size
f.write(chunk)
print '已下载{0:%}'.format(loaded)
n += 1
结果就很直观了：
已下载2.579129% 已下载2.581255% 已下载2.583382% 已下载2.585508%
心怀远大理想的我怎么会只满足于这一个呢，写个类一起使用吧：
# -*- coding: utf-8 -*-
import requests
from contextlib import closing
import time
def download_file(url, path):
with closing(requests.get(url, stream=True)) as r:
chunk_size = 1024*10
content_size = int(r.headers['content-length'])
print '下载开始'
with open(path, "wb") as f:
p = ProgressData(size = content_size, unit='Kb', block=chunk_size)
for chunk in r.iter_content(chunk_size=chunk_size):
f.write(chunk)
p.output()
class ProgressData(object):
def __init__(self, block,size, unit, file_name='', ):
self.file_name = file_name
self.block = block/1000.0
self.size = size/1000.0
self.unit = unit
self.count = 0
self.start = time.time()
def output(self):
self.end = time.time()
self.count += 1
speed = self.block/(self.end-self.start) if (self.end-self.start)>0 else 0
self.start = time.time()
loaded = self.count*self.block
progress = round(loaded/self.size, 4)
if loaded >= self.size:
print u'%s下载完成\r\n'%self.file_name
else:
print u'{0}下载进度{1:.2f}{2}/{3:.2f}{4} 下载速度{5:.2%} {6:.2f}{7}/s'.\
format(self.file_name, loaded, self.unit,\
self.size, self.unit, progress, speed, self.unit)
print '%50s'%('/'*int((1-progress)*50))
搞代码网（gaodaima.com）提供的所有资源部分来自互联网，如果有侵犯您的版权或其他权益，请说明详细缘由并提供版权或权益证明然后发送到邮箱[email protected]，我们会在看到邮件的第一时间内为您处理，或直接联系QQ：872152909。本网站采用BY-NC-SA协议进行授权
转载请注明原文链接：Python爬取视频(其实是一篇福利)过程解析
转载请注明原文链接：Python爬取视频(其实是一篇福利)过程解析
