Merge branch 'fix_acfun' of https://github.com/rosynirvana/you-get into rosynirvana-fix_acfun

src/you_get/common.py

@@ -138,6 +138,29 @@ if sys.stdout.isatty():
 else:
     default_encoding = locale.getpreferredencoding().lower()
 
+def rc4(key, data):
+#all encryption algo should work on bytes
+    assert type(key)==type(data) and type(key) == type(b'')
+    state = list(range(256))
+    j = 0
+    for i in range(256):
+        j += state[i] + key[i % len(key)]
+        j &= 0xff
+        state[i], state[j] = state[j], state[i]
+
+    i = 0
+    j = 0
+    out_list = []
+    for char in data:
+        i += 1
+        i &= 0xff
+        j += state[i]
+        j &= 0xff
+        state[i], state[j] = state[j], state[i]
+        prn = state[(state[i] + state[j]) & 0xff]
+        out_list.append(char ^ prn)
+
+    return bytes(out_list)
 def maybe_print(*s):
     try: print(*s)
     except: pass

src/you_get/extractors/acfun.py

@@ -10,11 +10,32 @@ from .sina import sina_download_by_vid
 from .tudou import tudou_download_by_iid
 from .youku import youku_download_by_vid, youku_open_download_by_vid
 
-import json, re
+import json
+import re
+import base64
 
 def get_srt_json(id):
     url = 'http://danmu.aixifan.com/V2/%s' % id
-    return get_html(url)
+    return get_content(url)
+
+def youku_acfun_proxy(vid, sign):
+    url = 'http://aplay-vod.cn-beijing.aliyuncs.com/acfun/web?vid={}&ct=85&ev=3&sign={}'.format(vid, sign)
+    json_data = json.loads(get_content(url))['data']
+    enc_text = base64.b64decode(json_data)
+    dec_text = rc4(b'8bdc7e1a', enc_text).decode('utf8')
+    youku_json = json.loads(dec_text)
+
+    yk_streams = {}
+    for stream in youku_json['stream']:
+        tp = stream['stream_type']
+        yk_streams[tp] = [], stream['total_size']
+        if stream.get('segs'):
+            for seg in stream['segs']:
+                yk_streams[tp][0].append(seg['url'])
+        else:
+            yk_streams[tp] = stream['m3u8'], stream['total_size']
+
+    return yk_streams
 
 def acfun_download_by_vid(vid, title, output_dir='.', merge=True, info_only=False, **kwargs):
     """str, str, str, bool, bool ->None
@@ -26,7 +47,7 @@ def acfun_download_by_vid(vid, title, output_dir='.', merge=True, info_only=Fals
     """
 
     #first call the main parasing API
-    info = json.loads(get_html('http://www.acfun.tv/video/getVideo.aspx?id=' + vid))
+    info = json.loads(get_content('http://www.acfun.tv/video/getVideo.aspx?id=' + vid))
 
     sourceType = info['sourceType']
 
@@ -47,12 +68,23 @@ def acfun_download_by_vid(vid, title, output_dir='.', merge=True, info_only=Fals
         letvcloud_download_by_vu(sourceId, '2d8c027396', title, output_dir=output_dir, merge=merge, info_only=info_only)
     elif sourceType == 'zhuzhan':
         #As in Jul.28.2016, Acfun is using embsig to anti hotlink so we need to pass this
-        embsig =  info['encode']
+#In Mar. 2017 there is a dedicated ``acfun_proxy'' in youku cloud player
-        a = 'http://api.aixifan.com/plays/%s' % vid
+#old code removed
-        s = json.loads(get_content(a, headers={'deviceType': '2'}))
+        yk_streams = youku_acfun_proxy(info['sourceId'], info['encode'])
-        if s['data']['source'] == "zhuzhan-youku":
+        seq = ['mp4hd3', 'mp4hd2', 'mp4hd', 'flvhd']
-            sourceId = s['data']['sourceId']
+        for t in seq:
-            youku_open_download_by_vid(client_id='908a519d032263f8', vid=sourceId, title=title, output_dir=output_dir,merge=merge, info_only=info_only, embsig = embsig, **kwargs)
+            if yk_streams.get(t):
+                preferred = yk_streams[t]
+                break
+#total_size in the json could be incorrect(F.I. 0)
+        size = 0
+        for url in preferred[0]:
+            _, _, seg_size = url_info(url)
+            size += seg_size
+#fallback to flvhd is not quite possible
+        print_info(site_info, title, 'mp4', size)
+        if not info_only:
+            download_urls(preferred[0], title, 'mp4', size, output_dir=output_dir, merge=merge)
     else:
         raise NotImplementedError(sourceType)
 
@@ -71,7 +103,7 @@ def acfun_download_by_vid(vid, title, output_dir='.', merge=True, info_only=Fals
 
 def acfun_download(url, output_dir='.', merge=True, info_only=False, **kwargs):
     assert re.match(r'http://[^\.]+.acfun.[^\.]+/\D/\D\D(\d+)', url)
-    html = get_html(url)
+    html = get_content(url)
 
     title = r1(r'data-title="([^"]+)"', html)
     title = unescape_html(title)