Python 爬取所有51VOA网站的Learn a words文本及mp3音频

Python 爬取所有51VOA网站的Learn a words文本及mp3音频

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#Python 爬取所有51VOA网站的Learn a words文本及mp3音频
import os
import sys
import time
import urllib as req
from threading import Thread
import urllib2
import urllib
from threading import Thread
import xml
import re
class MyWorkThread(Thread, urllib.FancyURLopener):
    """
    Multi-thread downloading class.
    run() is a vitual method of Thread
    """
    def __init__(self, threadname, url, filename, ranges = 0):
        Thread.__init__(self, name = threadname)
        urllib.FancyURLopener.__init__(self)
        self.name = threadname
        self.url = url
        self.filename = filename
        self.ranges = ranges
        self.downloaded = 0
    def run(self):
        """
        virtual function in Thread
        """
        try:
            self.downloaded = os.path.getsize(self.filename)
        except OSError:
            self.downloaded = 0
        #rebuild start point
        self.startpoint = self.ranges[0] + self.downloaded
         
        #if this part is completed
        if self.startpoint >= self.ranges[1]:
            print 'Part %s has been downloaded over.' % self.filename
            return
        self.oneTimeSize = 8 * 1024 #8K bytes / time
        print 'task %s will download from %d to %d' %(self.name, self.startpoint, self.ranges[1])
        self.addheader('Range', 'bytes=%d-%d' %(self.startpoint, self.ranges[1]))
        self.urlhandle = self.open(self.url)
        data = self.urlhandle.read(self.oneTimeSize)
        while data:
            filehandle = open(self.filename, 'ab+')
            filehandle.write(data)
            filehandle.close()
            self.downloaded += len(data)
            data = self.urlhandle.read(self.oneTimeSize)
def GetUrlFileSize(url):
    urlHandler = urllib.urlopen(url)
    headers = urlHandler.info().headers
    length = 0
    for header in headers:
        if header.find('Length') != -1:
            length = header.split(':')[-1].strip()
            length = int(length)
    return length
def SpliteBlocks(totalsize, blocknumber):
    blocksize = totalsize / blocknumber
    ranges = []
    for i in range(0, blocknumber -1):
        ranges.append((i * blocksize, i * blocksize + blocksize -1))
    ranges.append((blocksize * (blocknumber -1), totalsize -1))
    return ranges
def isLive(tasks):
    for task in tasks:
        if task.isAlive():
            return True
    return False
def downLoadFile(url, output, blocks = 6):
    sys.stdout.write('Begin to download from %s
' %url )
    sys.stdout.flush()
    size = GetUrlFileSize(url)
    ranges = SpliteBlocks(size, blocks)
     
    threadname = ["thread_%d" %i for i in range(0, blocks)]
    filename = ["tmpfile_%d" %i for i in range(0, blocks)]
    tasks = []
    for i in range(0, blocks):
        task = MyWorkThread(threadname[i], url, filename[i], ranges[i])
        task.setDaemon(True)
        task.start()
        tasks.append(task)
    time.sleep(2)
    while isLive(tasks):
        downloaded = sum([task.downloaded for task in tasks])
        process = downloaded / float(size) * 100
        show = u'
Filesize: %d Downloaded:%d Completed: %.2f%%' %(size, downloaded, process)
        sys.stdout.write(show)
        sys.stdout.flush
        time.sleep(1)
         
    output = formatFileName(output)
    filehandle = open(output, 'wb+')
    for i in filename:
        f = open(i, 'rb')
        filehandle.write(f.read())
        f.close()
        os.remove(i)
    filehandle.close()
    sys.stdout.write("Completed!
")
    sys.stdout.flush()
def formatFileName(filename):
    if isinstance(filename, str):
        header, tail = os.path.split(filename)
        if tail != '':
            tuple = ('\','/',':','*', '?', '"', '<', '>', '|')
            for char in tuple:
                if tail.find(char) != -1:
                    tail = tail.replace(char, ' ')
        filename = os.path.join(header, tail)
        #print filename
        return filename
    else:
        return 'None'

def remove_tags(raw_html):
  cleanr =re.compile('<.*?>')
  cleantext = re.sub(cleanr,'', raw_html)
  return cleantext

def saveword(url,name):
    res=req.urlopen(url)
    data=res.readlines()
    res.close()
    startag=r'id="mp3"'
    endtag=r'</div>'
    k=80
    data2=''
    data3=''
    data4=''
    while k<len(data)-10:
        if(data[k].find(startag)!=-1):
            data2=data[k]
        if(data[k].find('<div id="content">')!=-1):
            data3=data[k]
            if(data[k+1].find('<p>')!=-1):
                data4=data[k+1]
#                if(data4.rfind('...')!=-1):
#                    endid = data4.find('...')+3
#                else:
#                    endid = data4.find('</p>')
#                data4 = data4[3:endid]
                data4=remove_tags(data4)
        k=k+1   
#    print data2
##    data=str(data)
##    data2=data[(data.find(startag)+14):data.lower().find(endtag)+3]
##    data3=data[105]
#    print data3
    mp3url=data2[data2.find('http'):data2.find(''' title="''')-1]
    if(data3.find(endtag)!=-1):
        sent = data3[data3.find('今天我们要学'):data3.find(endtag)]
    else:
        sent = data3[data3.find('今天我们要学'):].strip('
').strip('
')+data4.strip('
')
#        sent = sent.replace('
','. ')   
#    print mp3url,sent
    f=open('LearningWord.txt','a+')
    sent=remove_tags(sent)    
    f.write(name+'
'+sent.strip('
')+'
')
    f.close()
#    print str(name)+'.mp3'
    if(data2.find(startag)!=-1):
        downLoadFile(mp3url,str(formatFileName(name.replace(':', ' ')))+'.mp3', blocks = 4)


def savepage(url):
    res=req.urlopen(url)
    data=res.read()
    res.close()
    startag='''<ul><li>'''
    endtag='''</li></ul>'''
    data=str(data)
    data2=data[data.find(startag)+12:data.find(endtag)]
    linestart='href'
    meddle = '''" target'''
    lineend = '</a>'
    urls=[]
    words = []
    i=data2.find(linestart)
    while(i!=-1):
        k = data2.find(meddle)
        j = data2.find(lineend)
        url = 'http://www.51voa.com/'+data2[i+6:k]
        urls = urls+[url]
        word = data2[k+16:j]
        print i,k,j, word,url
        words = words + [word]
        data2=data2[j+3:]
        saveword(url,word)
        i=data2.find(linestart)
#        break


#下载所有单词
f=open('LearningWord.txt','w')
f.close()
i=53
while i<=54:
    url = 'http://www.51voa.com/Learn_A_Word_'+str(i)+'.html'
    savepage(url)
    i=i+1

#下载指定单词
#url = "http://www.51voa.com/Voa_English_Learning/Learn_A_Word_21951.html"
#name ='9:pop up'
#saveword(url,name)

下载单词文本示例：(全部单词文本下载地址：http://pan.baidu.com/s/1o8pmojS)

2650 endorse
今天我们要学的词是 endorse. Endorse 作为动词，有支持的意思。Senator Ted Cruz endorsed Donald Trump, but later said the decision was “agonizing.” 美国联邦参议员克鲁兹支持川普，但是后来又表示，他做出这一决定十分痛苦。The New York Times endorsed Hillary Clinton for president in a Saturday editorial, and dismissed Donald Trump as “the worst nominee put forward by a major party in modern American history.” 纽约时报在星期六的社论中支持希拉里.克林顿当总统，并批评说，川普是“美国现代史上主要政党推举出的最差劲的候选人”。好的，我们今天学习的词是 endorse, endorse, endorse...
2649 deportation
今天我们要学的词是 deportation. Deportation 名词，驱逐出境，递解出境。The Obama administration said it would fully resume deportations of undocumented Haitian immigrants. 奥巴马政府表示，将全面恢复对无证海地移民的遣返工作。China and Canada have reached a new border agreement that would speed up the deportation of Chinese nationals inadmissible in Canada. 中国和加拿大达成新的边境协议，加快遣返那些本不该被允许进入加拿大的中国公民。好的，我们今天学习的词是 deportation, deportation, deportation...
2648 voluntarily
今天我们要学的词是 voluntarily. Voluntarily 副词，自愿地。The International Organization for Migrants says that more people are voluntarily returning to their home countries. 国际移民组织说，越来越多的人开始自愿返回自己的祖国。A high-tech diagnostic company voluntarily withdrew its Zika virus blood test from FDA approval. 一家高科技诊断公司自愿撤回递交美国食品药物管理局的寨卡病毒血液检测批准申请。好的，我们今天学习的词是 voluntarily, voluntarily, voluntarily...
2647 guerrilla
今天我们要学的词是 guerrilla. Guerrilla 形容词，游击队的。The Columbian government signed a peace agreement on Monday with the Revolutionary Armed Forces of Columbia (FARC), a national guerrilla movement. 哥伦比亚政府星期一跟全国游击队运动“哥伦比亚革命武装力量”签署了和平协议。The agreement needs to be approved by an Oct. 2 referendum before roughly 7,000 guerrilla fighters start their transition to civilian life. 这项协议还需经过10月2号全民公决批准，大约七千名游击队员才会开始向平民生活过渡。好的，我们今天学习的词是 guerrilla, guerrilla, guerrilla...
2646 curfew
今天我们要学的词是 curfew. Curfew 名词，宵禁。The city of Charlotte in North Carolina has lifted its midnight curfew, but the state of emergency remains in effect. 北卡罗来纳州夏洛特市取消了午夜宵禁，但是紧急状态依旧生效。Authorities in an Austrian city imposed a curfew on young immigrants following a series of sexual attacks at a local beer and wine festival. 奥地利一个城市的有关当局对未成年移民实施宵禁，此前当地一个啤酒葡萄酒节期间发生了一系列性侵事件。 好的，我们今天学习的词是 curfew, curfew, curfew...
2645 estimate
今天我们要学的词是 estimate. Estimate 动词，估计。A recent study estimates that the Indonesian forest fires that created a smoky haze last year may have caused more than 100,000 premature deaths. 一项最新研究估计，去年印尼山火引发的雾霾可能造成了10万人过早死亡。A new survey estimates that Americans own 265 million guns, but half of these guns are in the hands of only 3% of Americans. 最新调查估计，美国人拥有枪支总数2.65亿支，但其中半数都集中在3%的人手中。好的，我们今天学习的词是 estimate, estimate, estimate...
2644 mercy killing
今天我们要学的词是 mercy killing. Mercy killing 名词，安乐死。A terminally ill 17-year-old has become the first minor to be euthanized in Belgium since the age restrictions on such mercy killings were lifted in 2014. 比利时一个17岁绝症男孩安乐死，他是比利时2014年取消对安乐死年龄限制以来第一个安乐死的未成年人。The United Arab Emirates passed a new law banning all mercy killings. 阿联酋通过新法律，禁止安乐死。好的，我们今天学习的词是 mercy killing, mercy killing, mercy killing...