为了一个issue，重写了下载和合并模块，结果这个issue的问题还没解决。。。。。
但是倒是支持更多下载器了：aria2c,wget,axel,curl 。
合并模块请大家赐教。
论怎么把代码从34行代码 变成31行 变成17行 变成12行：
'''
data_list = data.split('\r')
for lines in data_list:
lines = str(lines)
if '<url>' in lines:
if 'youku' in lines:
url = lines[17:-9]
elif 'sina' in lines:
url = lines[16:-9]
elif 'qq.com' in lines:
url = lines[17:-9]
elif 'letv.com' in lines:
url = lines[17:-9]
break
elif 'acgvideo' in lines:
url = lines[17:-9]
is_local = 1
rawurl.append(url)
if 'backup_url' in lines and is_local is 1:
break'''
'''
output = commands.getstatusoutput('ffmpeg --help')
if str(output[0]) != '32512':
is_ffmpeg = 1
output = commands.getstatusoutput('avconv --help')
if str(output[0]) != '32512':
is_avconv = 1
output = commands.getstatusoutput('aria2c --help')
if str(output[0]) != '32512':
is_aria2c = 1
output = commands.getstatusoutput('aria2c --help')
if str(output[0]) != '32512':
is_aria2c = 1
output = commands.getstatusoutput('wget --help')
if str(output[0]) != '32512':
is_wget = 1
output = commands.getstatusoutput('curl --help')
if str(output[0]) == '32512':
is_curl = 1
if is_ffmpeg == 1:
concat_software = 'ffmpeg'
elif is_avconv == 1:
concat_software = 'avconv'
else:
print('ERROR: Cannot find any concating software!\nPlease get a ffmpeg or avconv!')
exit()
if is_aria2c == 1:
download_software = 'aria2c'
elif is_wget == 1:
is_aria2c = 'wget'
elif is_curl == 1:
is_aria2c = 'curl'
else:
print('ERROR: Cannot find any downloading software!\nPlease get a aria2c, wget or, at very least, curl!')
exit()
'''
'''
concat_software = ''
download_software = ''
output = commands.getstatusoutput('ffmpeg --help')
if str(output[0]) == '32512':
output = commands.getstatusoutput('avconv --help')
if str(output[0]) == '32512':
print('ERROR: Cannot find any concating software!\nPlease get a ffmpeg or avconv!')
exit()
else:
concat_software = 'avconv'
else:
concat_software = 'ffmpeg'
output = commands.getstatusoutput('aria2c --help')
if str(output[0]) == '32512':
output = commands.getstatusoutput('axel --help')
if str(output[0]) == '32512':
output = commands.getstatusoutput('wget --help')
if str(output[0]) == '32512':
output = commands.getstatusoutput('curl --help')
if str(output[0]) == '32512':
print('ERROR: Cannot find any downloading software!\nPlease get a aria2c, wget or, at very least, curl!')
exit()
else:
download_software = 'curl'
else:
download_software = 'wget'
else:
download_software = 'axel'
else:
download_software = 'aria2c'
'''
'''
if download_software.strip().lower() not in download_software_list:
print('WARNING: Requested Software not supported!\nBiligrab only support these following software(s):\n' + str(download_software_list) + 'Trying to find avalable one...')
for software in download_software_list:
output = commands.getstatusoutput(software + ' --help')
if str(output[0]) != '32512': #If exist
download_software = software
break
if concat_software.strip().lower() not in concat_software_list:
print('WARNING: Requested Software not supported!\nBiligrab only support these following software(s):\n' + str(concat_software_list) + 'Trying to find avalable one...')
for software in concat_software_list:
output = commands.getstatusoutput(software + ' --help')
if str(output[0]) != '32512': #If exist
concat_software = software
break'''总之，上代码，老地方。
https://github.com/cnbeining/Biligrab
#!/usr/bin/env python
#coding:utf-8
# Author: Beining --<ACICFG>
# Purpose: Yet another danmaku and video file downloader of Bilibili.
# Created: 11/03/2013
'''
Biligrab 0.91
Beining@ACICFG
cnbeining[at]gmail.com
http://www.cnbeining.com
https://github.com/cnbeining/Biligrab
MIT licence
'''
import sys
import os
from StringIO import StringIO
import gzip
import urllib
import urllib2
import sys
import commands
import hashlib
import getopt
from xml.dom.minidom import parse, parseString
import xml.dom.minidom
reload(sys)
sys.setdefaultencoding('utf-8')
global vid, cid, partname, title, videourl, part_now, is_first_run, appkey, secretkey
appkey='85eb6835b0a1034e';
secretkey = '2ad42749773c441109bdc0191257a664'
#----------------------------------------------------------------------
def list_del_repeat(list):
"""delete repeating items in a list, and keep the order.
http://www.cnblogs.com/infim/archive/2011/03/10/1979615.html"""
l2 = []
[l2.append(i) for i in list if not i in l2]
return(l2)
#----------------------------------------------------------------------
def calc_sign(string):
"""str/any->str
return MD5."""
return str(hashlib.md5(str(string).encode('utf-8')).hexdigest())
#----------------------------------------------------------------------
def read_cookie(cookiepath):
""""""
try:
cookies_file = open(cookiepath, 'r')
cookies = cookies_file.readlines()
cookies_file.close()
#print(cookies)
return cookies
except:
print('Cannot read cookie, may affect some videos...')
return ''
#----------------------------------------------------------------------
def find_cid_api(vid, p, cookies):
"""find cid and print video detail"""
global cid, partname, title, videourl
cid = 0
title = ''
partname = ''
if str(p) is '0' or str(p) is '1':
str2Hash = 'appkey=' + str(appkey) + '&id=' + str(vid) + '&type=xml' + str(secretkey)
biliurl = 'https://api.bilibili.com/view?appkey=' + str(appkey) + '&id=' + str(vid) + '&type=xml&sign=' + calc_sign(str2Hash)
print(biliurl)
else:
str2Hash = 'appkey=' + str(appkey) + '&id=' + str(vid) + '&page=' + str(p) + '&type=xml' + str(secretkey)
biliurl = 'https://api.bilibili.com/view?appkey=' + str(appkey) + '&id=' + str(vid) + '&page=' + str(p) + '&type=xml&sign=' + calc_sign(str2Hash)
#print(biliurl)
videourl = 'http://www.bilibili.com/video/av'+ str(vid)+'/index_'+ str(p)+'.html'
print('Fetching webpage...')
try:
request = urllib2.Request(biliurl, headers={ 'User-Agent' : 'Biligrab /0.9 (cnbeining@gmail.com)', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache' , 'Cookie': cookies})
response = urllib2.urlopen(request)
data = response.read()
dom = parseString(data)
for node in dom.getElementsByTagName('cid'):
if node.parentNode.tagName == "info":
cid = node.toxml()[5:-6]
print('cid is ' + cid)
break
for node in dom.getElementsByTagName('partname'):
if node.parentNode.tagName == "info":
partname = node.toxml()[10:-11].strip()
print('partname is ' + partname)
break
for node in dom.getElementsByTagName('title'):
if node.parentNode.tagName == "info":
title = node.toxml()[7:-8].strip()
print('Title is ' + title)
except: #If API failed
print('ERROR: Cannot connect to API server!')
#----------------------------------------------------------------------
def find_cid_flvcd(videourl):
""""""
global vid, cid, partname, title
print('Fetching webpage via Flvcd...')
request = urllib2.Request(videourl, headers={ 'User-Agent' : 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache' })
request.add_header('Accept-encoding', 'gzip')
response = urllib2.urlopen(request)
if response.info().get('Content-Encoding') == 'gzip':
buf = StringIO( response.read())
f = gzip.GzipFile(fileobj=buf)
data = f.read()
data_list = data.split('\n')
#Todo: read title
for lines in data_list:
if 'cid=' in lines:
cid = lines.split('&')
cid = cid[0].split('=')
cid = cid[-1]
print('cid is ' + str(cid))
break
#----------------------------------------------------------------------
def find_link_flvcd(videourl):
""""""
print('Finding link via Flvcd...')
request = urllib2.Request('http://www.flvcd.com/parse.php?'+urllib.urlencode([('kw', videourl)]), headers={ 'User-Agent' : 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache' })
request.add_header('Accept-encoding', 'gzip')
response = urllib2.urlopen(request)
data = response.read()
data_list = data.split('\n')
for items in data_list:
if 'name' in items and 'inf' in items and 'input' in items:
c = items
rawurlflvcd = c[59:-5]
rawurlflvcd = rawurlflvcd.split('|')
return rawurlflvcd
#----------------------------------------------------------------------
def check_dependencies(download_software, concat_software):
"""None->str,str
Will give softwares for concat or download."""
concat_software_list = ['ffmpeg', 'avconv']
download_software_list = ['aria2c', 'axel', 'wget', 'curl']
name_list = [[concat_software, concat_software_list], [download_software, download_software_list]]
for name in name_list:
if name[0].strip().lower() not in name[1] : # Unsupported software
if len(name[0].strip()) != 0: #Set a Unsupported software, not blank
print('WARNING: Requested Software not supported!\n Biligrab only support these following software(s):\n ' + str(name[1]) + '\n Trying to find available one...')
for software in name[1]:
output = commands.getstatusoutput(software + ' --help')
if str(output[0]) != '32512': #If exist
name[0] = software
break
if name[0] == '':
print('ERROR: Cannot find software in ' + str(name[1]) + ' !')
exit()
return name_list[0][0], name_list[1][0]
#----------------------------------------------------------------------
def download_video(part_number, download_software, video_link):
""""""
if download_software == 'aria2c':
os.system('aria2c -c -s16 -x16 -k1M --out '+part_number+'.flv "'+video_link+'"')
elif download_software == 'wget':
os.system('wget -c -O '+part_number+'.flv "'+video_link+'"')
elif download_software == 'curl':
os.system('curl -L -C -o '+part_number+'.flv "'+video_link+'"')
elif download_software == 'axel':
os.system('curl -n 20 -o '+part_number+'.flv "'+video_link+'"')
#----------------------------------------------------------------------
def concat_videos(concat_software, vid_num, filename):
"""str,str->None"""
if concat_software == 'ffmpeg':
f = open('ff.txt', 'w')
ff = ''
os.getcwd()
for i in range(vid_num):
ff = ff + 'file \'' + str(os.getcwd()) + '/'+ str(i) + '.flv\'\n'
ff = ff.encode("utf8")
f.write(ff)
f.close()
print('Concating videos...')
os.system('ffmpeg -f concat -i ff.txt -c copy "'+filename+'".mp4')
if os.path.isfile(str(filename+'.mp4')):
os.system('rm -r ff.txt')
for i in range(vid_num):
os.system('rm -r '+str(i)+'.flv')
print('Done, enjoy yourself!')
else:
print('ERROR: Cannot concatenative files, trying to make flv...')
os.system('ffmpeg -f concat -i ff.txt -c copy "'+filename+'".flv')
if os.path.isfile(str(filename+'.flv')):
print('FLV file made. Not possible to mux to MP4, highly likely due to audio format.')
os.system('rm -r ff.txt')
for i in range(vid_num):
os.system('rm -r '+str(i)+'.flv')
else:
print('ERROR: Cannot concatenative files!')
elif concat_software == 'avconv':
pass
#----------------------------------------------------------------------
def main(vid, p, oversea, cookies, download_software, concat_software):
global cid, partname, title, videourl, is_first_run
videourl = 'http://www.bilibili.com/video/av'+ str(vid)+'/index_'+ str(p)+'.html'
# Check both software
concat_software, download_software = check_dependencies(download_software, concat_software)
print(concat_software, download_software)
#Start to find cid, api-flvcd
find_cid_api(vid, p, cookies)
global cid
if cid is 0:
print('Cannot find cid, trying to do it brutely...')
find_cid_flvcd(videourl)
if cid is 0:
is_black3 = str(raw_input('Strange, still cannot find cid... Type y for trying the unpredictable way, or input the cid by yourself, press ENTER to quit.'))
if 'y' in str(is_black3):
vid = vid - 1
p = 1
find_cid_api(vid-1, p)
cid = cid + 1
elif str(is_black3) is '':
print('Cannot get cid anyway! Quit.')
exit()
else:
cid = str(is_black3)
#start to make folders...
if title is not '':
folder = title
else:
folder = cid
if len(partname) is not 0:
filename = partname
elif title is not '':
filename = title
else:
filename = cid
# In case make too much folders
folder_to_make = os.getcwd() + '/' + folder
if is_first_run == 0:
if not os.path.exists(folder_to_make):
os.makedirs(folder_to_make)
is_first_run = 1
os.chdir(folder_to_make)
# Download Danmaku
print('Fetching XML...')
os.system('curl -o "'+filename+'.xml" --compressed http://comment.bilibili.com/'+cid+'.xml')
os.system('gzip -d '+cid+'.xml.gz')
print('The XML file, ' + filename + '.xml should be ready...enjoy!')
print('Finding video location...')
#try api
sign_this = calc_sign('appkey=' + appkey + '&cid=' + cid + secretkey)
if oversea == '1':
try:
request = urllib2.Request('http://interface.bilibili.com/v_cdn_play?appkey=' + appkey + '&cid=' + cid + '&sign=' + sign_this, headers={ 'User-Agent' : 'Biligrab /0.9 (cnbeining@gmail.com)', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache' })
except:
print('ERROR: Cannot connect to CDN API server!')
elif oversea is '2':
#Force get oriurl
try:
request = urllib2.Request('http://interface.bilibili.com/player?appkey=' + appkey + '&cid=' + cid + '&sign=' + sign_this, headers={ 'User-Agent' : 'Biligrab /0.9 (cnbeining@gmail.com)', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache' })
except:
print('ERROR: Cannot connect to original source API server!')
else:
try:
request = urllib2.Request('http://interface.bilibili.com/playurl?appkey=' + appkey + '&cid=' + cid + '&sign=' + sign_this, headers={ 'User-Agent' : 'Biligrab /0.9 (cnbeining@gmail.com)', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache' })
except:
print('ERROR: Cannot connect to normal API server!')
response = urllib2.urlopen(request)
data = response.read()
rawurl = []
originalurl = ''
if oversea is '2':
data = data.split('\n')
for l in data:
if 'oriurl' in l:
originalurl = str(l[8:-9])
print('Original URL is ' + originalurl)
break
if originalurl is not '':
rawurl = find_link_flvcd(originalurl)
else:
print('Cannot get original URL! Using falloff plan...')
pass
else:
dom = parseString(data)
for node in dom.getElementsByTagName('url'):
if node.parentNode.tagName == "durl":
rawurl.append(node.toxml()[14:-9])
#print(str(node.toxml()[14:-9]))
pass
if len(rawurl) == 0: #hope this never happen
rawurl = find_link_flvcd(videourl)
#flvcd
vid_num = len(rawurl)
if vid_num is 0: # shit hit the fan
rawurl = list(str(raw_input('Cannot get download URL! If you know the url, please enter it now; URL1|URL2...'))).split('|')
vid_num = len(rawurl)
if vid_num is 0: # shit really hit the fan
print('Cannot get video URL anyway!')
exit()
print(str(vid_num) + ' videos in part ' + str(part_now) + ' to download, fetch yourself a cup of coffee...')
for i in range(vid_num):
video_link = rawurl[i]
part_number = str(i)
print('Downloading ' + str(i+1) + ' of ' + str(vid_num) + ' videos in part ' + str(part_now) + '...')
#Call a function to support multiple download softwares
download_video(part_number, download_software, video_link)
concat_videos(concat_software, vid_num, filename)
print('INFO: DONE!')
exit()
#----------------------------------------------------------------------
def get_full_p(p_raw):
"""str->list"""
p_list = []
p_raw = p_raw.split(',')
for item in p_raw:
if '~' in item:
#print(item)
lower = 0
higher = 0
item = item.split('~')
part_now = '0'
try:
lower = int(item[0])
except:
print('Cannot read lower!')
try:
higher = int(item[1])
except:
print('Cannot read higher!')
if lower == 0 or higher == 0:
if lower == 0 and higher != 0:
lower = higher
elif lower != 0 and higher == 0:
higher = lower
else:
print('Cannot find any higher or lower, ignoring...')
#break
mid = 0
if higher < lower:
mid = higher
higher = lower
lower = mid
p_list.append(lower)
while lower < higher:
lower = lower + 1
p_list.append(lower)
#break
else:
try:
p_list.append(int(item))
except:
print('Cannot read "'+str(item)+'", abondon it.')
#break
p_list = list_del_repeat(p_list)
return p_list
#----------------------------------------------------------------------
def usage():
""""""
print('''
Biligrab
https://github.com/cnbeining/Biligrab
http://www.cnbeining.com/
Beining@ACICFG
Usage:
python biligrab.py (-h) (-a) (-p) (-s) (-c) (-d) (-v)
-h: Default: None
Print this usage file.
-a: Default: None
The av number.
If not set, Biligrab will use the falloff interact mode.
-p: Default: 1
The part number.
Support "~", "," and mix use.
Examples:
Input Output
1 [1]
1,2 [1, 2]
1~3 [1, 2, 3]
1,2~3 [1, 2, 3]
-s: Default: 0
Source to download.
0: The original API source, can be Letv backup,
and can failed if the original video is not avalable(e.g., deleted)
1: The CDN API source, "oversea accelerate".
Can be MINICDN backup in Mainland China or oversea.
Good to bypass some bangumi's limit.
2: Force to use the original source.
Use Flvcd to parase the video, but would fail if
1) The original source DNE, e.g., some old videos
2) The original source is Letvcloud itself.
3) Other unknown reason(s) that stops Flvcd from parase the video.
For any video that failed to parse, Biligrab will try to use Flvcd.
(Mainly for oversea users regarding to copyright-restricted bangumies.)
-c: Default: ./bilicookies
The path of cookies.
Use cookies to visit member-only videos.
-d: Default: None
Set the desired download software.
Biligrab supports aria2c(16 threads), axel(20 threads), wget and curl by far.
If not set, Biligrab will detect an avalable one;
If none of those is avalable, Biligrab will quit.
For more software support, please open an issue at https://github.com/cnbeining/Biligrab/issues/
-v Default:None
Set the desired download software.
Biligrab supports ffmpeg by far.
If not set, Biligrab will detect an avalable one;
If none of those is avalable, Biligrab will quit.
For more software support, please open an issue at https://github.com/cnbeining/Biligrab/issues/
Make sure you include a *working* command line example of this software!''')
#----------------------------------------------------------------------
if __name__=='__main__':
is_first_run = 0
argv_list = []
argv_list = sys.argv[1:]
p_raw, vid, oversea, cookiepath, download_software, concat_software = '', '', '', '', '', ''
try:
opts, args = getopt.getopt(argv_list, "ha:p:s:c:d:v:", ['help', "av",'part', 'source', 'cookie', 'download', 'concat'])
except getopt.GetoptError:
usage()
exit()
for o, a in opts:
if o in ('-h', '--help'):
usage()
exit()
if o in ('-a', '--av'):
vid = a
try:
argv_list.remove('-a')
except:
break
if o in ('-p', '--part'):
p_raw = a
try:
argv_list.remove('-p')
except:
break
if o in ('-s', '--source'):
oversea = a
try:
argv_list.remove('-s')
except:
break
if o in ('-c', '--cookie'):
cookiepath = a
try:
argv_list.remove('-c')
except:
print('No cookie path set, use default: ./bilicookies')
cookiepath = './bilicookies'
break
if o in ('-d', '--download'):
download_software = a
try:
argv_list.remove('-d')
except:
break
if o in ('-v', '--concat'):
concat_software = a
try:
argv_list.remove('-v')
except:
break
if len(vid) == 0:
vid = str(raw_input('av'))
p_raw = str(raw_input('P'))
oversea = str(raw_input('Source?'))
cookiepath = './bilicookies'
'''
if len(vid) == 0:
vid = str(raw_input('Please input: av'))
if len(vid) == 0:
print('Cannot download nothing!')
exit()
'''
if len(cookiepath) == 0:
cookiepath = './bilicookies'
if len(p_raw) == 0:
print('No part number set, download part 1.')
p_raw = '1'
if len(oversea) == 0:
oversea = '0'
print('Oversea not set, use original API(methon 0).')
print('Your targe download is av' + vid + ', part ' + p_raw + ', from source ' + oversea)
p_list = get_full_p(p_raw)
cookies = read_cookie(cookiepath)
for p in p_list:
reload(sys)
sys.setdefaultencoding('utf-8')
part_now = str(p)
main(vid, p, oversea, cookies, download_software, concat_software)
exit()