[pptv]refactor

src/you_get/extractors/pptv.py

@@ -1,15 +1,218 @@
 #!/usr/bin/env python
 
-__all__ = ['pptv_download', 'pptv_download_by_id']
+#__all__ = ['pptv_download', 'pptv_download_by_id']
 
 from ..common import *
+from ..extractor import VideoExtractor
 
 import re
 import time
 import urllib
-from random import random
+import random
+import binascii
+from xml.dom.minidom import parseString
 
 
+def lshift(a, b):
+    return (a << b) & 0xffffffff
+def rshift(a, b):
+    if a >= 0:
+        return a >> b
+    return (0x100000000 + a) >> b
+
+def le32_pack(b_str):
+    result = 0
+    result |= b_str[0]
+    result |= (b_str[1] << 8)
+    result |= (b_str[2] << 16)
+    result |= (b_str[3] << 24)
+    return result
+
+def tea_core(data, key_seg):
+    delta = 2654435769
+
+    d0 = le32_pack(data[:4])
+    d1 = le32_pack(data[4:8])
+
+    sum_ = 0
+    for rnd in range(32):
+        sum_ = (sum_ + delta) & 0xffffffff
+        p1 = (lshift(d1, 4) + key_seg[0]) & 0xffffffff
+        p2 = (d1 + sum_) & 0xffffffff
+        p3 = (rshift(d1, 5) + key_seg[1]) & 0xffffffff
+
+        mid_p = p1 ^ p2 ^ p3
+        d0 = (d0 + mid_p) & 0xffffffff
+
+        p4 = (lshift(d0, 4) + key_seg[2]) & 0xffffffff
+        p5 = (d0 + sum_) & 0xffffffff
+        p6 = (rshift(d0, 5) + key_seg[3]) & 0xffffffff
+
+        mid_p = p4 ^ p5 ^ p6
+        d1 = (d1 + mid_p) & 0xffffffff
+
+    return bytes(unpack_le32(d0) + unpack_le32(d1))
+
+def ran_hex(size):
+    result = []
+    for i in range(size):
+        result.append(hex(int(15 * random.random()))[2:])
+    return ''.join(result)
+
+def zpad(b_str, size):
+    size_diff = size - len(b_str)
+    return b_str + bytes(size_diff)
+
+def gen_key(t):
+    key_seg = [1896220160,101056625, 100692230, 7407110]
+    t_s = hex(int(t))[2:].encode('utf8')
+    input_data = zpad(t_s, 16)
+    out = tea_core(input_data, key_seg)
+    return binascii.hexlify(out[:8]).decode('utf8') + ran_hex(16)
+
+def unpack_le32(i32):
+    result = []
+    result.append(i32 & 0xff)
+    i32 = rshift(i32, 8)
+    result.append(i32 & 0xff)
+    i32 = rshift(i32, 8)
+    result.append(i32 & 0xff)
+    i32 = rshift(i32, 8)
+    result.append(i32 & 0xff)
+    return result
+
+def get_elem(elem, tag):
+    return elem.getElementsByTagName(tag)
+
+def get_attr(elem, attr):
+    return elem.getAttribute(attr)
+
+def get_text(elem):
+    return elem.firstChild.nodeValue
+
+def shift_time(time_str):
+    ts = time_str[:-4]
+    return time.mktime(time.strptime(ts)) - 60
+
+def parse_pptv_xml(dom):
+    channel = get_elem(dom, 'channel')[0]
+    title = get_attr(channel, 'nm')
+    file_list = get_elem(channel, 'file')[0]
+    item_list = get_elem(file_list, 'item')
+    streams_cnt = len(item_list)
+    item_mlist = []
+    for item in item_list:
+        rid = get_attr(item, 'rid')
+        file_type = get_attr(item, 'ft')
+        size = get_attr(item, 'filesize')
+        width = get_attr(item, 'width')
+        height = get_attr(item, 'height')
+        bitrate = get_attr(item, 'bitrate')
+        res = '{}x{}@{}kbps'.format(width, height, bitrate)
+        item_meta = (file_type, rid, size, res)
+        item_mlist.append(item_meta)
+
+    dt_list = get_elem(dom, 'dt')
+    dragdata_list = get_elem(dom, 'dragdata')
+
+    stream_mlist = []
+    for dt in dt_list:
+        file_type = get_attr(dt, 'ft')
+        serv_time = get_text(get_elem(dt, 'st')[0])
+        expr_time = get_text(get_elem(dt, 'key')[0])
+        serv_addr = get_text(get_elem(dt, 'sh')[0])
+        stream_meta = (file_type, serv_addr, expr_time, serv_time)
+        stream_mlist.append(stream_meta)
+
+    segs_mlist = []
+    for dd in dragdata_list:
+        file_type = get_attr(dd, 'ft')
+        seg_list = get_elem(dd, 'sgm')
+        segs = []
+        segs_size = []
+        for seg in seg_list:
+            rid = get_attr(seg, 'rid')
+            size = get_attr(seg, 'fs')
+            segs.append(rid)
+            segs_size.append(size)
+        segs_meta = (file_type, segs, segs_size)
+        segs_mlist.append(segs_meta)
+    return title, item_mlist, stream_mlist, segs_mlist
+
+#mergs 3 meta_data
+def merge_meta(item_mlist, stream_mlist, segs_mlist):
+    streams = {}
+    for i in range(len(segs_mlist)):
+        streams[str(i)] = {}
+
+    for item in item_mlist:
+        stream = streams[item[0]]
+        stream['rid'] = item[1]
+        stream['size'] = item[2]
+        stream['res'] = item[3]
+
+    for s in stream_mlist:
+        stream = streams[s[0]]
+        stream['serv_addr'] = s[1]
+        stream['expr_time'] = s[2]
+        stream['serv_time'] = s[3]
+
+    for seg in segs_mlist:
+        stream = streams[seg[0]]
+        stream['segs'] = seg[1]
+        stream['segs_size'] = seg[2]
+
+    return streams
+
+
+def make_url(stream):
+    host = stream['serv_addr']
+    rid = stream['rid']
+    key = gen_key(shift_time(stream['serv_time']))
+    key_expr = stream['expr_time']
+
+    src = []
+    for i, seg in enumerate(stream['segs']):
+        url = 'http://{}/{}/{}?key={}&k={}'.format(host, i, rid, key, key_expr)
+        url += '&fpp.ver=1.3.0.4&type=web.fpp'
+        src.append(url)
+    return src
+
+class PPTV(VideoExtractor):
+    name = 'PPTV'
+    stream_types = [
+            {'itag': '4'},
+            {'itag': '3'},
+            {'itag': '2'},
+            {'itag': '1'},
+            {'itag': '0'},
+    ]
+
+    def prepare(self, **kwargs):
+        if self.url and not self.vid:
+            if not re.match(r'http://v.pptv.com/show/(\w+)\.html', self.url):
+                raise('Unknown url pattern')
+            page_content = get_content(self.url)
+            self.vid = match1(page_content, r'webcfg\s*=\s*{"id":\s*(\d+)')
+
+        if not self.vid:
+            raise('Cannot find id')
+        api_url = 'http://web-play.pptv.com/webplay3-0-{}.xml'.format(self.vid)
+        api_url += '?type=web.fpp&version=4'
+        dom = parseString(get_content(api_url))
+        self.title, m_items, m_streams, m_segs = parse_pptv_xml(dom)
+        xml_streams = merge_meta(m_items, m_streams, m_segs)
+        for stream_id in xml_streams:
+            stream_data = xml_streams[stream_id]
+            src = make_url(stream_data)
+            self.streams[stream_id] = {
+                    'container': 'mp4',
+                    'video_profile': stream_data['res'],
+                    'size': int(stream_data['size']),
+                    'src': src
+            }
+
+'''
 def constructKey(arg):
 
     def str2hex(s):
@@ -143,12 +346,14 @@ def pptv_download_by_id(id, title = None, output_dir = '.', merge = True, info_o
             pptv_download_by_id(id, output_dir = output_dir, merge = merge, info_only = info_only)
 
 def pptv_download(url, output_dir = '.', merge = True, info_only = False, **kwargs):
-    assert re.match(r'http://v.pptv.com/show/(\w+)\.html$', url)
+    assert re.match(r'http://v.pptv.com/show/(\w+)\.html', url)
     html = get_html(url)
     id = r1(r'webcfg\s*=\s*{"id":\s*(\d+)', html)
     assert id
     pptv_download_by_id(id, output_dir = output_dir, merge = merge, info_only = info_only)
-
-site_info = "PPTV.com"
-download = pptv_download
+'''
+site = PPTV()
+#site_info = "PPTV.com"
+#download = pptv_download
+download = site.download_by_url
 download_playlist = playlist_not_supported('pptv')