def get_video_url(url, DEBUG=0):
scraper = cfscrape.create_scraper()
try:
response = scraper.get(
'http://www.fastvideo.me/embed-%s-607x360.html' %
re.search('me/([^/]+)/', url).group(1))
except:
return False, "Url not found"
if 'p,a,c,k,e,d' in response.content:
try:
packed = re.search(
r'<script[^>]+>eval\((function\(p,a,c,k,e,d\).*\)\))\).*?<\/script>',
response.content, re.S).group(1)
except:
packed = None
return False, "Packed regex error"
if packed:
import jsbeautifier.unpackers.packer as packer
jwconfig = packer.unpack(packed)
url = re.search('file:"([^"]+)"', jwconfig).group(1)
return True, url
else:
return False, "Packed not found"
def is_artefact_packed(filename, byte_content):
"""
Returns a tuple (is_packed, unpacked_byte_content) if the artefact is packed using an algorithm
similar to Dave Edward's p.a.c.k.e.r algorithm. This appears to generate syntax errors compounding feature analysis,
but we try anyway...
"""
# dont load the file again if we already have the byte-content...
if byte_content is None:
with open(filename, 'rb') as fp:
byte_content = fp.read()
try:
source = fast_decode(byte_content)
is_packed = packer.detect(source)
if is_packed:
ret = packer.unpack(source)
# TODO FIXME... yuk this is so silly! But we have to return bytes since extract-features.jar program expects that...
# fortunately most artefacts arent packed so this is not as expensive as it could be...
result = decode(encode(ret, 'utf-8', 'backslashreplace'),
'unicode-escape').encode()
assert isinstance(result, bytes)
assert len(result) > 0
return (True, result)
# else
return (False, byte_content)
except Exception as e:
# we fail an artefact that is purportedly packed, but which cannot be unpacked...
raise e
def _get_streams(self):
headers = {
'User-Agent': useragents.CHROME,
'Referer': self.url
}
res = http.get(self.url)
html_text = res.text
soup = BeautifulSoup(html_text, 'html.parser')
scripts = soup.find_all('script')
for script in scripts:
if script.string is not None:
if "document.getElementsByTagName('video')[0].volume = 1.0;" in script.string:
code = script.string
startidx = code.find('eval(function(p,a,c,k,e,')
endidx = code.find('hunterobfuscator =')
code = code[startidx:endidx]
# Here is the call to the first part of the deobfuscation i.e. getting packed code
#code = aadecode(code)
#code = code.aadecode()
if not code.replace(' ', '').startswith('eval(function(p,a,c,k,e,'):
code = 'fail'
break
else:
code = 'fail'
else:
code = 'fail'
#The second part of deobfuscation occurs here. Using module jsbeautifier.
if code != 'fail':
unpacked = packer.unpack(code)
video_location = unpacked[unpacked.rfind('http'):unpacked.rfind('m3u8')+4]
url = video_location
if url:
self.logger.debug('HLS URL: {0}'.format(url))
yield 'live', HLSStream(self.session, url, headers=headers)
def get_video_url(url):
list = []
from xbmctools import getnet, finddata
data = getnet(url)
regx = '''<script type="text/javascript"(.*)</script>'''
#print "jucie",_filter(data)
#print"data1", data
data2 = finddata(data.replace("\n", ""), 'eval(', '</script')
unpack = packer.unpack(data2)
print 'unpack', unpack
regx = '''http://(.*?)mkv'''
try:
link = 'http://' + re.findall(regx, unpack, re.M | re.I)[0] + 'mkv'
except:
regx = '''http://(.*?)mp4'''
link = 'http://' + re.findall(regx, unpack, re.M | re.I)[0] + 'mp4'
print 'link', link
return link
def extract_ports(js_url):
dict = {}
# Download the JS file.
html = download_webpage(js_url)
if html is None:
log.error('Failed to download webpage: %s', js_url)
return dict
try:
# For now, try and extract out the css/port pairs from the JS.
# This likely is a really back way of doing it, I'll revisit later.
unpack = packer.unpack(html)
unpack = unpack.replace("$(document).ready(function(){", "")
unpack = unpack.replace("});", "")
unpack = unpack.replace("\\", "")
unpack = unpack.replace("'", "")
unpack = unpack.replace(".", "")
# Pull out everything that is within a bracket.
parts = re.findall('\((.*?)\)', unpack)
# Now convert the list into a dictionary.
# Every other entry in the list is a pair css/port.
i = 0
while i < len(parts):
dict[parts[i]] = parts[i + 1]
i += 2
return dict
except Exception as e:
log.exception('Failed do extract ports from %s: %s.', js_url, e)
return dict
def play_video(selection):
r = requests.get(arconaitv_url+selection)
html_text = r.text
soup = BeautifulSoup(html_text, 'html.parser')
scripts = soup.find_all('script')
for script in scripts:
if script.string is not None:
if "document.getElementsByTagName('video')[0].volume = 1.0;" in script.string:
code = script.string
startidx = code.find('eval(function(p,a,c,k,e,')
endidx = code.find('hunterobfuscator =')
code = code[startidx:endidx]
# Here is the call to the first part of the deobfuscation i.e. getting packed code
#code = aadecode(code)
#code = code.aadecode()
if not code.replace(' ', '').startswith('eval(function(p,a,c,k,e,'):
code = 'fail'
break
else:
code = 'fail'
else:
code = 'fail'
#The second part of deobfuscation occurs here. Using module jsbeautifier.
if code != 'fail':
unpacked = packer.unpack(code)
video_location = unpacked[unpacked.rfind('http'):unpacked.rfind('m3u8')+4]
play_item = xbmcgui.ListItem(path=video_location+'|User-Agent=%s' % urllib2.quote(USER_AGENT, safe=''))
xbmcplugin.setResolvedUrl(addon_handle, True, listitem=play_item)
else:
xbmcgui.Dialog().ok('Sorry','Could not deobfuscate the code.')
def parse(self, response):
script = response.xpath('//script[1]/text()').extract()[0]
parse_str = script.strip().split('\n')[2]
# print(parse_str)
pages = packer.unpack(parse_str.strip())
# pages = packer.unpack(str(parse_str.strip()))
pages = pages.replace("var pages=pages=\\'[",
"").replace(";", "").replace("\\", "").replace(
'"', "").replace("[",
"").replace("]",
"").split(",")
# for page in pages:
item = LearningItem()
# item['image_urls'] = "https://images.dmzj.com/" + page
item['image_urls'] = map(lambda x: "https://images.dmzj.com/" + x,
pages)
item['img_dirname2'] = response.xpath(
"//span[@class='redhotl']/text()").extract()[0]
item['img_dirname1'] = response.xpath(
"//a[@class='redhotl']/text()").extract()[0]
# item['image_urls'].append("https://images.dmzj.com/" + pages[0])
# for img_link in pages:
# item["image_urls"] = "https://images.dmzj.com/" + img_link
yield item
def test_unpack(self):
"""Test unpack() function."""
check = lambda inp, out: self.assertEqual(unpack(inp), out)
check("eval(function(p,a,c,k,e,r){e=String;if(!''.replace(/^/,String)"
"){while(c--)r[c]=k[c]||c;k=[function(e){return r[e]}];e="
"function(){return'\\\\w+'};c=1};while(c--)if(k[c])p=p.replace("
"new RegExp('\\\\b'+e(c)+'\\\\b','g'),k[c]);return p}('0 2=1',"
"62,3,'var||a'.split('|'),0,{}))", 'var a=1')
def test_unpack(self):
"""Test unpack() function."""
check = lambda inp, out: self.assertEqual(unpack(inp), out)
check(
"eval(function(p,a,c,k,e,r){e=String;if(!''.replace(/^/,String)"
"){while(c--)r[c]=k[c]||c;k=[function(e){return r[e]}];e="
"function(){return'\\\\w+'};c=1};while(c--)if(k[c])p=p.replace("
"new RegExp('\\\\b'+e(c)+'\\\\b','g'),k[c]);return p}('0 2=1',"
"62,3,'var||a'.split('|'),0,{}))", 'var a=1')
def fetch_images_url(chapter_url=''):
request_url = HOME_URL_PREFIX + chapter_url
req = requests.get(request_url, headers=HEADERS)
tmp_url = req.text.split('\n')[11] # html中第11行保存了图片的URL后缀
tmp_url = unpack(tmp_url)
pic_url_str = re.sub('var pages\s?=\s?pages\s?=\\\\\'\[\"', '', tmp_url)
pic_url_str = re.sub('\"\]\\\\\';\s?', '', pic_url_str)
pic_url_arr = [] # 保存最新一章的图片URL
for i in pic_url_str.replace('\\\/', '/').split('","'):
# url = PIC_URL_PREFIX + urllib.quote(i.decode('unicode-escape').encode('utf8'))
url = PIC_URL_PREFIX + i
pic_url_arr.append(url)
return pic_url_arr
def fetch_images_url(chapter_url=''):
request_url = HOME_URL_PREFIX + chapter_url
req = requests.get(request_url, headers=HEADERS)
tmp_url = req.text.split('\n')[11] # html中第11行保存了图片的URL后缀
tmp_url = unpack(tmp_url)
pic_url_str = re.sub('var pages\s?=\s?pages\s?=\\\\\'\[\"', '', tmp_url)
pic_url_str = re.sub('\"\]\\\\\';\s?', '', pic_url_str)
pic_url_arr = [] # 保存最新一章的图片URL
for i in pic_url_str.replace('\\\/', '/').split('","'):
# url = PIC_URL_PREFIX + urllib.quote(i.decode('unicode-escape').encode('utf8'))
url = PIC_URL_PREFIX + i
pic_url_arr.append(url)
return pic_url_arr
def solve_js(html):
js = '(function' + re.findall('function(.*?)</script>', html)[0]
encrpyted_js = js.split(',')[-3][1:-15]
decrypted_js = lzstring.LZString().decompressFromBase64(encrpyted_js)
original_js = js.split(',')
original_js[-3] = "'" + decrypted_js + "'.split('|')"
packed_js = 'eval(' + ','.join(original_js) + ')'
# print('packed_js', packed_js)
unpack = packer.unpack(packed_js)
print(unpack)
# js_result = jsbeautifier.beautify(unpack)
# print('js_result', js_result)
imgData = re.findall("SMH\.imgData\((.*?)\)\.preInit\(\)\;", unpack)[0]
res = json.loads(imgData)
print(res['bname'])
return res
def obtem_vidig2(url):
import jsbeautifier.unpackers.packer as packer
soup = BeautifulSoup(abrir_url(url))
codigo_fonte = packer.unpack(soup('script')[6].text)
#ow = open(r'G:\\html_list\\script4.html','w')
#ow.write(str(soup('script')[6]))
#ow.close()#"(http://.*?/v.flv)"
try:
url_video = re.findall(r"'(http://.*?/v.mp4)\\'",codigo_fonte)[0]
return [url_video+'|User-Agent=Mozilla/5.0 (Windows NT 6.3; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.104 Safari/537.36',"-"]
except:
pass
try:
url_video = re.findall(r'"(http://.*?/v.flv)"',codigo_fonte)[0]
return [url_video+'|User-Agent=Mozilla/5.0 (Windows NT 6.3; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.104 Safari/537.36',"-"]
except:
return ["-","-"]
def get_video_url(url):
list=[]
from xbmctools import getnet,finddata
data=getnet(url)
regx='''<script type="text/javascript"(.*)</script>'''
#print "jucie",_filter(data)
#print"data1", data
data2 =finddata(data.replace("\n",""),'eval(','</script')
try:
unpack = packer.unpack(data2)
except:
unpack=data
regx='''file:"(.*?)m3u8"'''
try:
m3u8=re.findall(regx,unpack, re.M|re.I)[0]+'m3u8'
list.append(('m3u8',m3u8))
except:
pass
regx=''',{file:"(.*?)mp4"'''
try:
mp4=re.findall(regx,unpack, re.M|re.I)[0]+'mp4'
list.append(('mp4',mp4))
except:
pass
print "list",list
return list
Pre_Stream_URL = re.search('file:"(.*)"', unpack).group(1)
print 'Pre_Stream_URL1',Pre_Stream_URL
Pre_Stream_URL = re.search('file:"(.*)",label', unpack).group(1)
print 'Pre_Stream_URL2',Pre_Stream_URL
sys.exit(0)
return Pre_Stream_URL
def fetch(self, url):
soup = self.requester.get(url)
test = re.search(r"eval(?:.+fullscreen.+\))", soup.text)
if test is None:
return
unpack = packer.unpack(test.group())
test = re.search("http:\/\/[^\s\"]+.\.(?:mp4|mpg|avi|flv)", unpack)
if test is None:
return
info = self.requester.send(test.group())
info = re.search(r"(['\"])(?P<url>https:.+)\1", info.text)
if info is None:
return
return info.group("url")
def parse_page_ext2(self, response):
"""
拿到跟图片地址有关的js
"""
manga_no = response.meta['manga_no']
chapter_no = response.meta['chapter_no']
manga_name = response.meta['manga_name']
chapter_name = response.meta['chapter_name']
manga_save_folder = response.meta['manga_save_folder']
data = response.meta['data']
current_page = response.meta['current_page']
page_count = response.meta['page_count']
try_repr = response.meta.setdefault('try_repr', False)
proxy = response.meta['proxy']
# 返回一段js代码
# jsbeautifier 1.7.5解码
# python3 兼容
response_body_str = str(response.body, response.encoding)
packed_js = response_body_str
try:
if try_repr == True:
packed_js = repr(packed_js)
unpacked_js = packer.unpack(packed_js).replace("\\'", "'")
# 解码后找出图片路径
cid = re.findall('var cid=(.*?);', unpacked_js)[0]
key = re.findall("var key='(.*?)';", unpacked_js)[0]
pix = re.findall('var pix="(.*?)";', unpacked_js)[0]
pvalue = re.findall('var pvalue=\[(.*?)\];', unpacked_js)[0]
pvalue = pvalue.replace('"', '')
if (',' in pvalue):
pvalue = pvalue.split(',')
else:
pvalue = [
pvalue,
]
image_url = pix + pvalue[0] + '?cid=' + cid + '&key=' + key + '&uk='
item = Dm5Item()
item['imgurl'] = image_url
item['imgname'] = image_url.split('/')[-1].split('_')[0].zfill(
3) + '.jpg'
# 不使用pipeline下载需要特殊处理下载地址
item['imgfolder'] = settings.get(
'IMAGES_STORE') + '/' + manga_save_folder + '/' + chapter_no
# 准备下载图片
image_save_path = os.path.join(item['imgfolder'], item['imgname'])
# 跳过已经下载的图片
if not os.path.exists(image_save_path):
if not os.path.exists(item['imgfolder']):
os.makedirs(item['imgfolder'])
print('当前页面', manga_name, chapter_no, chapter_name,
current_page, '/', page_count)
host = image_url.split('//')[1].split('/')[0]
headers = {
'Host':
host,
'User-Agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:63.0) Gecko/20100101 Firefox/63.0',
'Accept':
'*/*',
'Connection':
'close',
'Accept-Language':
'zh-CN,zh;q=0.8,en;q=0.5,en-US;q=0.3',
'Accept-Encoding':
'gzip, deflate',
'Referer':
'http://www.dm5.com/m' + str(data['cid']) + '-p' +
str(current_page) + '/',
'DNT':
'1',
}
# print 1111, image_url
# 使用urllib2下载图片
# req = urllib2.Request(image_url, headers=headers)
# res = urllib2.urlopen(req)
# # print res
# with open(image_save_path, 'wb') as f:
# f.write(res.read())
# 使用requests下载图片
# res = requests.get(image_url, headers=headers, proxies=proxyDict)
# with open(image_save_path, 'wb') as f:
# f.write(res.content)
# 使用自定义的imagepipeline下载图片
item['imgheaders'] = headers
item['imgproxy'] = proxy
yield item
# 准备访问下一页
current_page += 1
if int(current_page) > int(page_count):
return
data['page'] = current_page
pages_url = 'http://www.dm5.com/m' + str(
data['cid']) + '/chapterfun.ashx?' + urllib.parse.urlencode(
data)
yield scrapy.Request(pages_url,
cookies=cookies,
headers={
'Referer':
'http://www.dm5.com/m' +
str(data['cid']) + '/'
},
meta={
'proxy': proxy,
'manga_no': manga_no,
'chapter_no': chapter_no,
'manga_name': manga_name,
'chapter_name': chapter_name,
'manga_save_folder': manga_save_folder,
'data': data,
'current_page': current_page,
'page_count': page_count
},
callback=self.parse_page_ext2)
except UnpackingError as e:
print('解包js错误 访问地址', response.url)
print('当前页面', current_page)
response_body_str = str(response.body, response.encoding)
print('返回内容', response_body_str)
if try_repr == True:
# 准备访问下一页
current_page += 1
if int(current_page) > int(page_count):
return
data['page'] = current_page
pages_url = 'http://www.dm5.com/m' + str(
data['cid']
) + '/chapterfun.ashx?' + urllib.parse.urlencode(data)
yield scrapy.Request(pages_url,
cookies=cookies,
headers={
'Referer':
'http://www.dm5.com/m' +
str(data['cid']) + '/'
},
meta={
'proxy': proxy,
'manga_no': manga_no,
'chapter_no': chapter_no,
'manga_name': manga_name,
'chapter_name': chapter_name,
'manga_save_folder':
manga_save_folder,
'data': data,
'current_page': current_page,
'page_count': page_count
},
callback=self.parse_page_ext2)
else:
yield scrapy.Request(response.url,
cookies=cookies,
meta={
'proxy': proxy,
'manga_no': manga_no,
'chapter_no': chapter_no,
'manga_name': manga_name,
'chapter_name': chapter_name,
'manga_save_folder':
manga_save_folder,
'data': data,
'current_page': current_page,
'page_count': page_count,
'try_repr': True
},
callback=self.parse_page_ext2,
dont_filter=True)
def check(inp, out):
return detect(inp) and self.assertEqual(unpack(inp), out)
req = requests.Request(
'GET',
base_url +
'/Search?query=1%27+and+1%3D0+union+select+1%2C2%2CContent%2C4%2Cnull%2Cnull+from+Reviews+where+id%3D10+--+-',
headers=headers,
cookies=cookies)
r = requestDebug(s, req)
img = BeautifulSoup(r.text, 'lxml').find('img', {'src': '2'}).get('alt')
first_obfuscated_part = img.split('\n', 3)[2]
second_obfuscated_part = img.split('\n', 4)[3]
#
# Reverse algorithm in second_obfuscated_part to get the "good" input
#
second_part_unpacked = packer.unpack(img.split('\n', 4)[3]).split(';', 2)[0]
flag = eval(second_part_unpacked.split()[1].split('"')[1])
myDebug('1:\n%s\n' % flag)
input = flag[::-1]
myDebug('2:\n%s\n' % input)
intermediate = ""
for i in range(len(input)):
intermediate += chr(input[i] - 20) if i % 2 == 0 else chr(input[i])
myDebug('3:\n%s\n' % intermediate)
input = intermediate.replace('\\x', '')
myDebug('4:\n%s\n' % input)
intermediate = b''
for i in range(0, len(input), 2):
intermediate += codecs.decode(input[i:i + 2], 'hex')
input = codecs.decode(intermediate, 'base64').decode()
myDebug('5:\n%s\n' % input)
def check(inp, out):
return self.assertEqual(unpack(inp), out)
def check(inp, out):
return detect(inp) and self.assertEqual(unpack(inp), out)
def check(inp, out): return self.assertEqual(unpack(inp), out)
check("eval(function(p,a,c,k,e,r){e=String;if(!''.replace(/^/,String)"
