Clarify that url and ext are optional when formats is given (#980)

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 77a13aea533d17aa57f17e01929ca3a276787844..69cdcdc1b5b75d1cac5733b34565f087c9dcddec 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -35,6 +35,8 @@ class InfoExtractor(object):
     title:          Video title, unescaped.
     ext:            Video filename extension.
 
+    Instead of url and ext, formats can also specified.
+
     The following fields are optional:
 
     format:         The video format, defaults to ext (used for --get-format)
@@ -52,8 +54,19 @@ class InfoExtractor(object):
     view_count:     How many users have watched the video on the platform.
     urlhandle:      [internal] The urlHandle to be used to download the file,
                     like returned by urllib.request.urlopen
-
-    The fields should all be Unicode strings.
+    formats:        A list of dictionaries for each format available, it must
+                    be ordered from worst to best quality. Potential fields:
+                    * url       Mandatory. The URL of the video file
+                    * ext       Will be calculated from url if missing
+                    * format    A human-readable description of the format
+                                ("mp4 container with h264/opus").
+                                Calculated from width and height if missing.
+                    * format_id A short description of the format
+                                ("mp4_h264_opus" or "19")
+                    * width     Width of the video, if known
+                    * height    Height of the video, if known
+
+    Unless mentioned otherwise, the fields should be Unicode strings.
 
     Subclasses of this one should re-define the _real_initialize() and
     _real_extract() methods and define a _VALID_URL regexp.
@@ -145,12 +158,17 @@ class InfoExtractor(object):
 
         urlh = self._request_webpage(url_or_request, video_id, note, errnote)
         content_type = urlh.headers.get('Content-Type', '')
+        webpage_bytes = urlh.read()
         m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
         if m:
             encoding = m.group(1)
         else:
-            encoding = 'utf-8'
-        webpage_bytes = urlh.read()
+            m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
+                          webpage_bytes[:1024])
+            if m:
+                encoding = m.group(1).decode('ascii')
+            else:
+                encoding = 'utf-8'
         if self._downloader.params.get('dump_intermediate_pages', False):
             try:
                 url = url_or_request.get_full_url()