python 自动下载 voa MP3

hmzone

　　因为英语学习的需要，经常会去网上下载一些VOA的MP3，常去的一个网站是http://www.51voa.com/
　　要想下载该网站上的MP3，需要手动选择要下载的篇幅，打开之后再选择要下载的MP3文件。要下载单独一个MP3文件还好，但要是想把某一时间内的所有MP3文件都下载下来，就很繁琐，需要重复做那些无聊的操作。能否用python来做一个下载voa MP3的工具呢？
　　设计思路如下：
　　一、打开http://www.51voa.com/主页，分析html，解析出主页上 VOA美国之音听力最近更新 文件列表，生成的dictionary
　　二、对已生成的dictionary按照当前日期过滤，得到能下载的当天的VOA MP3
　　三、对过滤后的dictionary遍历，同时进行下载操作
　　其中所使用的技术：
　　一、解析html，可以使用standar library中的HTMLParser，或者SGMLParser，也可以选择3rd party的解析库，比如BeautifulSoup（对html和xml都能很好的支持），本文采用BeautifulSoup
　　二、下载MP3，采用urllib，为提高效率，使用多线程进行下载，url header中可以使用Range参数分片下载，这样一来就能多部分协同操作。
　　
　　具体代码如下：
　　一、多线程下载部分代码
　　



#!/usr/bin/env python
# -*- coding :utf-8 -*-
"""
It is a multi-thread downloading tool
"""
import sys
import os
import time
import urllib2
import urllib
from threading import Thread
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\n' %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, url, filename, ranges)
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'\rFilesize: %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!\n")
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'
if __name__ == '__main__':
url = r'http://www.51voa.com/path.asp?url=/201008/hennessy_africa_wildlife_18aug10-32b.mp3'
output = r"D:\Voa\Study:'Shoot to Kill' Policy in Africa's Parks Abuses Human Rights.mp3"
downLoadFile(url, output, blocks = 4)

　　
　　二、解析voa页面部分代码
　　



#!/usr/bin/env python
# -*- coding:utf-8 -*-
import urllib2
import chardet
import os
import time
import string
import re
from HTMLParser import HTMLParser
import sys
from BeautifulSoup import BeautifulSoup
import multiThreadDownloadTool

VOA_URL = r'http://www.51voa.com'
DOWNLOAD_DIR = r'D:/Voa'
"""
File downloading from the web.
"""
def getURLContent(url):
    """
    get url content of the url, begin with html and ignor the doctype declarations
    """
    file = urllib2.urlopen(url)
    #print file.info()
    data = file.read()
    file.close()
    #return data.decode('utf-8')
    index = data.find('html')
    data = data[index - 1 :]
    return data
def getVOAURLs(content):
    """
    find the voa script urls in the content
    """
    urls = {}
    soup = BeautifulSoup(content)
    divs = soup.findAll('div', {'id':'rightContainer'})
    #print divs
    neededDiv = None
    if len(divs) >= 1:
        neededDiv = divs[0]
    if neededDiv != None:
        #pass the div
        #print neededDiv
        neededSpan = neededDiv.find('span', {'id' : 'list'})
        #print neededSpan
        lis = neededSpan.findAll('li')
        #print lis
        for li in lis:
            needAs = li.findAll('a')
            #got it
            #print needAs[1]
            #print needAs[1]['href']
            #print needAs[-1].string
            urls[needAs[-1].string] = VOA_URL + needAs[-1]['href']
    print "getVOAURLs() urls count is " , len(urls)
    return urls
def filterbyDate(urls ,date):
    """
    filter the urls by date
    """
    neededURLs = {}
    currentDate = time.localtime(time.time());
    #currentDateStr = time.strftime('%Y-%m-%d', currentDate)
    #currentDateStr = currentDate.tm_year + '-' + currentDate.tm_mon + ' ' + currentDate.tm_mday
    currentDateStr = "%s-%s-%s"  %(currentDate.tm_year, currentDate.tm_mon, currentDate.tm_mday)
    if(date != None):
        currentDateStr = date
    for url in urls.keys():
        name = url.lstrip().rstrip()
        length = len(name)
        publishDate = name[- len(currentDateStr) - 1 : -1]
        #print publishDate
        if publishDate == currentDateStr:
            neededURLs[name] = urls[url]
            print 'find ', name
    print 'After filter, the count is ' , len(neededURLs)
    return neededURLs
def findMP3FileInURL(url):
    """
    find MP3 files in a url
    """
    print 'parse the content of ', url
    urls = []
    #define a MP3 re string
    p = re.compile(r'/path.asp\?url=[-\w/]*\.mp3')
    #p = re.compile(r'/[-\w/]*\.mp3')
    content = getURLContent(url)
    matchLinks = p.findall(content)
    #print matchLinks
    for link in matchLinks:
        tmp = VOA_URL + link
        if tmp in urls: # check if exist already
            pass
        else:
            urls.append(tmp)
    print 'Current count of mp3 files is ', len(urls)
    return urls
def getHTMLFile(url, file_name):
    ifile = urllib2.urlopen(url)
    content = ifile.read()
    local_file = open(file_name, 'w')
    local_file.write(content)
    local_file.close()
def downloadFile(url, fileName2Store):
    """
    download file from url, and store it to local system using fileName2Store parameter
    """
    try:
        full_path = os.path.join(DOWNLOAD_DIR, fileName2Store)
        print 'begin to download url to ', full_path
        if os.path.isfile(full_path):
            #already exist
            print 'the file ', full_path, 'is alreasy exist, so just skip it!'
        else:
            print '\tDownloading the mp3 file...',  
            data=urllib2.urlopen(url).read()  
            print 'Done'  
            print '\tWriting data info file...',  
            f=file(full_path, 'wb')  
            f.write(data)  
            print 'Done'  
            f.close()
    except Exception, ex:
        print 'some exceptions occur when downloading ', ex
if __name__ == "__main__":
    try:
        #getHTMLFile(VOA_URL, r'.\Voa.html')
        context = getURLContent(VOA_URL)
        #file_read = open(r'.\Voa.html', 'r')
        #context = file_read.read()
        #print context
        #print '\n' * 5
#print chardet.detect(context)
print 'Begin to get download information, it may cost some minuts, please wait...'
files2download = getVOAURLs(context)
neededDownload = filterbyDate(files2download, None)
neededDownloadMp3s = {}
for name in neededDownload.keys():
fullURL = neededDownload[name]
formatedName = name[: -11].lstrip().rstrip()
#formatedName = formatedName.replace(' ', '-')
#print formatedName, ' ' * 5, fullURL
#print fullURL
mp3Names = findMP3FileInURL(fullURL)
if len(mp3Names) == 1:
#there is only on mp3 file in this file ,so we will use the formatedname
neededDownloadMp3s[formatedName] = mp3Names[0]
else:
for name in mp3Names:
print name
index_begin = name.rfind('/')
index_end = name.rfind('.')
tmpName = name[index_begin + 1 : index_end]
neededDownloadMp3s[tmpName] = name
print 'Now , the mp3 files  are :'
print neededDownloadMp3s
#findMP3FileInURL(r'http://www.51voa.com/VOA_Special_English/Phoning-Fertilizer-Philippine-Rice-Farmers--38545.html')
#findMP3FileInURL(r'http://www.51voa.com/Voa_English_Learning/Learn_A_Word_38412.html')
#down load file
for filename in neededDownloadMp3s.keys():
try:
full_path = os.path.join(DOWNLOAD_DIR, filename)
full_path = full_path + r'.mp3'
if full_path == r'D:\Voa\hennessy_africa_wildlife_18aug10-32b.mp3':
multiThreadDownloadTool.downLoadFile(neededDownloadMp3s[filename], full_path)
except Exception, ex:
print 'Some exceptions occur, when downloading file from %s, exception messages are %s' %(neededDownloadMp3s[filename], ex)
#downloadFile(r'http://www.51voa.com/path.asp?url=/201008/mercer_australia_election_16aug10-32b.mp3', 'test.mp3')
except Exception, ex:
print 'Exception caught, tracebacks are :',sys.exc_info(), ex
print 'download all completed!'
raw_input("Press any key to continue...")

　　
　　需要注意的地方：
　　在使用BeautifulSoup进行html解析的时候发现，BeautifulSoup对于

　　的支持不是很好，经常解析不出来，所以为了方便，在解析的时候先将源文件解析，只将之间的数据交与BeautifulSoup解析。具体为什么BeautifulSoup解析

　　DOCTYPE出错，我还没查出问题所在，希望有知道的朋友告知一声。