Python实现视频爬取-Tutorial Python-php.cn

Python实现视频爬取

little bottle

Lepaskan： 2019-04-10 09:56:30

ke hadapan

3539 orang telah melayarinya

Python可以用来做什么？公司里主要是爬取数据，并把爬回来的数据进行分析和挖掘，然而我们自己可以用它来爬取一些资源去使用，比如，想看的剧。本文中，小编将分享爬取视频的代码，大家存起来试试吧！

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[&#39;content-length&#39;])
        print &#39;下载开始&#39;
        with open(path, "wb") as f:
            for chunk in r.iter_content(chunk_size=chunk_size):
                f.write(chunk)
if __name__ == &#39;__main__&#39;:
    url = &#39;就在原帖...&#39;
    path = &#39;想存哪都行&#39;
    download_file(url, path)

Salin selepas log masuk

遭遇当头一棒：

AttributeError: __exit__

Salin selepas log masuk

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[&#39;content-length&#39;])
        print &#39;下载开始&#39;
        with open(path, "wb") as f:
            for chunk in r.iter_content(chunk_size=chunk_size):
                f.write(chunk)

Salin selepas log masuk

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[&#39;content-length&#39;])
        print &#39;下载开始&#39;
        with open(path, "wb") as f:
            for chunk in r.iter_content(chunk_size=chunk_size):
                f.write(chunk)
                f.flush()
                os.fsync(f.fileno())

Salin selepas log masuk

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[&#39;content-length&#39;])
        print &#39;下载开始&#39;
        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 &#39;已下载{0:%}&#39;.format(loaded)
                n += 1

Salin selepas log masuk

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%

Salin selepas log masuk

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[&#39;content-length&#39;])
        print &#39;下载开始&#39;
        with open(path, "wb") as f:
            p = ProgressData(size = content_size, unit=&#39;Kb&#39;, 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=&#39;&#39;, ):
        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&#39;%s下载完成\r\n&#39;%self.file_name
        else:
            print u&#39;{0}下载进度{1:.2f}{2}/{3:.2f}{4} 下载速度{5:.2%} {6:.2f}{7}/s&#39;.\
                  format(self.file_name, loaded, self.unit,\
                  self.size, self.unit, progress, speed, self.unit)
            print &#39;%50s&#39;%(&#39;/&#39;*int((1-progress)*50))

Salin selepas log masuk

运行：

下载开始
下载进度10.24Kb/120174.05Kb 0.01% 下载速度4.75Kb/s 
///////////////////////////////////////////////// 
下载进度20.48Kb/120174.05Kb 0.02% 下载速度32.93Kb/s 
/////////////////////////////////////////////////

Salin selepas log masuk

看上去舒服多了。

下面要做的就是多线程同时下载了，主线程生产url放入队列，下载线程获取url：

# -*- coding: utf-8 -*-
import requests
from contextlib import closing
import time
import Queue
import hashlib
import threading
import os
def download_file(url, path):
    with closing(requests.get(url, stream=True)) as r:
        chunk_size = 1024*10
        content_size = int(r.headers[&#39;content-length&#39;])
        if os.path.exists(path) and os.path.getsize(path)>=content_size:
            print &#39;已下载&#39;
            return
        print &#39;下载开始&#39;
        with open(path, "wb") as f:
            p = ProgressData(size = content_size, unit=&#39;Kb&#39;, block=chunk_size, file_name=path)
            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=&#39;&#39;, ):
        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&#39;%s下载完成\r\n&#39;%self.file_name
        else:
            print u&#39;{0}下载进度{1:.2f}{2}/{3:.2f}{4} {5:.2%} 下载速度{6:.2f}{7}/s&#39;.\
                  format(self.file_name, loaded, self.unit,\
                  self.size, self.unit, progress, speed, self.unit)
            print &#39;%50s&#39;%(&#39;/&#39;*int((1-progress)*50))

queue = Queue.Queue()
def run():
    while True:
        url = queue.get(timeout=100)
        if url is None:
            print u&#39;全下完啦&#39;
            break
        h = hashlib.md5()
        h.update(url)
        name = h.hexdigest()
        path = &#39;e:/download/&#39; + name + &#39;.mp4&#39;
        download_file(url, path)
def get_url():
    queue.put(None)

if __name__ == &#39;__main__&#39;:
    get_url()
    for i in xrange(4):
        t = threading.Thread(target=run)
        t.daemon = True
        t.start()

Salin selepas log masuk

加了重复下载的判断，至于怎么源源不断的生产url，诸位摸索吧，保重身体！

【推荐课程：Python视频教程】

Atas ialah kandungan terperinci Python实现视频爬取. Untuk maklumat lanjut, sila ikut artikel berkaitan lain di laman web China PHP!