学步园

本来已经有人写了python脚本从ted上下载字幕了，但是他的网站被墙同时有些ted的地址他解析不了，

所以我将他的python 脚本下载了下来，修改了一下。

谢谢：
http://tedtalksubtitledownload.appspot.com/

source 如下：

#! /usr/bin/env python<br /> import simplejson<br /> import pdb<br /> import urllib<br /> import sys<br /> import re<br /> def getFormatedTime(intvalue):<br /> mils = intvalue%1000<br /> segs = (intvalue/1000)%60<br /> mins = (intvalue/60000)%60<br /> hors = (intvalue/3600000)<br /> return "%02d:%02d:%02d,%03d"%(hors,mins,segs,mils)<br /> def availableSubs(subs):<br /> a = subs.find("LanguageCode")<br /> if a == -1:<br /> return []<br /> subs = subs[a+len("LanguageCode"):]<br /> return [re.search("%22([^A-Z]+)%22", subs).group(1)] + availableSubs(subs)<br /> def getVideoParameters(urldirection):<br /> ht = urllib.urlopen(urldirection).read()<br /> var = re.search('flashVars = {/n([^}]+)}', ht)<br /> if var:<br /> var = var.group(1)<br /> else:<br /> return None<br /> var = [a.replace('/t', '') for a in var.split('/n')]<br /> # debug pdb.set_trace()<br /> for a in range(len(var)):<br /> if var[a]:<br /> var[a] = var[a][:var[a].rfind(',')]<br /> resultado = []<br /> for a in var:<br /> l = a.find(':')<br /> if l != -1:<br /> resultado.append((a[:l], a[l+1:]))<br /> return dict(resultado)<br /> def downloadSub(idtalk, lang, timeIntro):<br /> print("Downloading subtitles for language %s"%lang)<br /> c = simplejson.load(urllib.urlopen('http://www.ted.com/talks/subtitles/id/%d/lang/%s'%(idtalk, lang)))<br /> salida = file('subs_%s_%s.srt'%(idtalk,lang), 'w')<br /> conta = 1</p> <p> c = c['captions']<br /> for linea in c:<br /> salida.write("%d/n"%conta)<br /> conta += 1<br /> salida.write("%s --> %s/n"%(getFormatedTime(timeIntro+linea['startTime']), getFormatedTime(timeIntro+linea['startTime']+linea['duration'])))<br /> salida.write("%s/n/n"%(linea['content'].encode('utf-8')))<br /> salida.close()<br /> def main(tedurl):<br /> print("Loading information about TED talk number %s..."%tedurl)<br /> vidpar = getVideoParameters(tedurl)<br /> if not vidpar:<br /> print("There was a problem fetching information about that TED Talk")<br /> sys.exit(1)<br /> print("Download all subtitles (write 'all' when prompted) or only one (specify wich)?")<br /> a = raw_input()<br /> availables = availableSubs(vidpar['languages'])<br /> idtalk = vidpar['ti']<br /> idtalk = int(idtalk[1:3])<br /> if a == "all":<br /> for lang in availables:<br /> downloadSub(idtalk, lang, int(vidpar['introDuration']))<br /> else:<br /> while a not in availables:<br /> print("We're sorry, the only available languages are:")<br /> for a in availables:<br /> print("/t"+a)<br /> a = raw_input()<br /> downloadSub(idtalk, a, int(vidpar['introDuration']))<br /> if __name__ == "__main__":<br /> if len(sys.argv) < 2:<br /> print("Usage: %s tedurl"%sys.argv[0])<br /> else:<br /> main(sys.argv[1])<br />

要使用它的话，需要先下载simplejson包，地址是： http://pypi.python.org/pypi/simplejson/

在通过http代理上网的环境中也可以使用。

具体使用例子如下：

D:/Document and Setting/test/My Documents/Downloads/TEDTalkSubtitles>TEDTalkSub
itles.py
http://www.ted.com/talks/barry_schwartz_on_the_paradox_of_choice.html

Loading information about TED talk number http://www.ted.com/talks/barry_schwar
z_on_the_paradox_of_choice.html...
Download all subtitles (write 'all' when prompted) or only one (specify wich)?
chi_hans

Downloading subtitles for language chi_hans

D:/Document and Setting/test/My Documents/Downloads/TEDTalkSubtitles>dir
 ドライブ D のボリューム ラベルは programe です
 ボリューム シリアル番号は 447B-7E2B です

 D:/Document and Setting/test/My Documents/Downloads/TEDTalkSubtitles のディレク
トリ

2011/04/15  14:16    <DIR>          .
2011/04/15  14:16    <DIR>          ..
2011/04/15  14:34            31,879 subs_93_chi_hans.srt
2011/04/15  14:16            31,928 subs_93_eng.srt
2011/04/15  14:26             2,639 TEDTalkSubtitles.py
               3 個のファイル              66,446 バイト
               2 個のディレクトリ  13,469,048,832 バイトの空き領域

D:/Document and Setting/test/My Documents/Downloads/TEDTalkSubtitles>

refs：

http://pythonconquerstheuniverse.wordpress.com/category/the-python-debugger/

http://meyerweb.com/eric/tools/dencoder/