Merge remote-tracking branch 'sahutd/master'

[youtube-dl] / youtube_dl / extractor / common.py

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index f498bcf6f5af1feb05eb8df0184f77e6336ac878..f7478d4598e8f15ea802ec9451aff53ce452fe7a 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1,4 +1,6 @@
 import base64
+import hashlib
+import json
 import os
 import re
 import socket
@@ -61,18 +63,24 @@ class InfoExtractor(object):
                     * tbr        Average bitrate of audio and video in KBit/s
                     * abr        Average audio bitrate in KBit/s
                     * acodec     Name of the audio codec in use
+                    * asr        Audio sampling rate in Hertz
                     * vbr        Average video bitrate in KBit/s
                     * vcodec     Name of the video codec in use
+                    * container  Name of the container format
                     * filesize   The number of bytes, if known in advance
                     * player_url SWF Player URL (used for rtmpdump).
                     * protocol   The protocol that will be used for the actual
                                  download, lower-case.
-                                 "http", "https", "rtsp", "rtmp" or so.
+                                 "http", "https", "rtsp", "rtmp", "m3u8" or so.
                     * preference Order number of this format. If this field is
                                  present and not None, the formats get sorted
                                  by this field.
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
+                    * quality    Order number of the video quality of this
+                                 format, irrespective of the file format.
+                                 -1 for default (order by other properties),
+                                 -2 or smaller for less than default.
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)
@@ -214,6 +222,8 @@ class InfoExtractor(object):
                           webpage_bytes[:1024])
             if m:
                 encoding = m.group(1).decode('ascii')
+            elif webpage_bytes.startswith(b'\xff\xfe'):
+                encoding = 'utf-16'
             else:
                 encoding = 'utf-8'
         if self._downloader.params.get('dump_intermediate_pages', False):
@@ -229,6 +239,9 @@ class InfoExtractor(object):
                 url = url_or_request.get_full_url()
             except AttributeError:
                 url = url_or_request
+            if len(url) > 200:
+                h = u'___' + hashlib.md5(url.encode('utf-8')).hexdigest()
+                url = url[:200 - len(h)] + h
             raw_filename = ('%s_%s.dump' % (video_id, url))
             filename = sanitize_filename(raw_filename, restricted=True)
             self.to_screen(u'Saving request to ' + filename)
@@ -256,6 +269,15 @@ class InfoExtractor(object):
             xml_string = transform_source(xml_string)
         return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
 
+    def _download_json(self, url_or_request, video_id,
+                       note=u'Downloading JSON metadata',
+                       errnote=u'Unable to download JSON metadata'):
+        json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        try:
+            return json.loads(json_string)
+        except ValueError as ve:
+            raise ExtractorError('Failed to download JSON', cause=ve)
+
     def report_warning(self, msg, video_id=None):
         idstr = u'' if video_id is None else u'%s: ' % video_id
         self._downloader.report_warning(
@@ -444,6 +466,9 @@ class InfoExtractor(object):
         return RATING_TABLE.get(rating.lower(), None)
 
     def _sort_formats(self, formats):
+        if not formats:
+            raise ExtractorError(u'No video formats found')
+
         def _formats_key(f):
             # TODO remove the following workaround
             from ..utils import determine_ext
@@ -483,9 +508,11 @@ class InfoExtractor(object):
 
             return (
                 preference,
+                f.get('quality') if f.get('quality') is not None else -1,
                 f.get('height') if f.get('height') is not None else -1,
                 f.get('width') if f.get('width') is not None else -1,
                 ext_preference,
+                f.get('tbr') if f.get('tbr') is not None else -1,
                 f.get('vbr') if f.get('vbr') is not None else -1,
                 f.get('abr') if f.get('abr') is not None else -1,
                 audio_ext_preference,