[utils] Remove Content-encoding from headers after decompression

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index d00b14b86d8ea8d43aef6bd7cf5a575cda5f94af..0ed6c45c8090e7d328ec4e09d12875adc768c0f9 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -373,6 +373,13 @@ def sanitize_path(s):
     return os.path.join(*sanitized_path)
 
 
     return os.path.join(*sanitized_path)
 
 

+# Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of
+# unwanted failures due to missing protocol
+def sanitized_Request(url, *args, **kwargs):
+    return compat_urllib_request.Request(
+        'http:%s' % url if url.startswith('//') else url, *args, **kwargs)
+
+

 def orderedSet(iterable):
     """ Remove all duplicates from the input iterable """
     res = []
 def orderedSet(iterable):
     """ Remove all duplicates from the input iterable """
     res = []


@@ -656,6 +663,16 @@ def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
     return hc
 
 
     return hc
 
 

+def handle_youtubedl_headers(headers):
+    filtered_headers = headers
+
+    if 'Youtubedl-no-compression' in filtered_headers:
+        filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding')
+        del filtered_headers['Youtubedl-no-compression']
+
+    return filtered_headers
+
+

 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
     """Handler for HTTP requests and responses.
 
 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
     """Handler for HTTP requests and responses.
 


@@ -663,7 +680,7 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
     the standard headers to every HTTP request and handles gzipped and
     deflated responses from web servers. If compression is to be avoided in
     a particular request, the original request in the program code only has
     the standard headers to every HTTP request and handles gzipped and
     deflated responses from web servers. If compression is to be avoided in
     a particular request, the original request in the program code only has

-    to include the HTTP header "Youtubedl-No-Compression", which will be
+    to include the HTTP header "Youtubedl-no-compression", which will be

     removed before making the real request.
 
     Part of this code was copied from:
     removed before making the real request.
 
     Part of this code was copied from:


@@ -724,10 +741,8 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
             # The dict keys are capitalized because of this bug by urllib
             if h.capitalize() not in req.headers:
                 req.add_header(h, v)
             # The dict keys are capitalized because of this bug by urllib
             if h.capitalize() not in req.headers:
                 req.add_header(h, v)

-        if 'Youtubedl-no-compression' in req.headers:
-            if 'Accept-encoding' in req.headers:
-                del req.headers['Accept-encoding']
-            del req.headers['Youtubedl-no-compression']
+
+        req.headers = handle_youtubedl_headers(req.headers)

 
         if sys.version_info < (2, 7) and '#' in req.get_full_url():
             # Python 2.6 is brain-dead when it comes to fragments
 
         if sys.version_info < (2, 7) and '#' in req.get_full_url():
             # Python 2.6 is brain-dead when it comes to fragments


@@ -758,11 +773,13 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
                     raise original_ioerror
             resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg
                     raise original_ioerror
             resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg

+            del resp.headers['Content-encoding']

         # deflate
         if resp.headers.get('Content-encoding', '') == 'deflate':
             gz = io.BytesIO(self.deflate(resp.read()))
             resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg
         # deflate
         if resp.headers.get('Content-encoding', '') == 'deflate':
             gz = io.BytesIO(self.deflate(resp.read()))
             resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg

+            del resp.headers['Content-encoding']

         # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
         # https://github.com/rg3/youtube-dl/issues/6457).
         if 300 <= resp.code < 400:
         # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
         # https://github.com/rg3/youtube-dl/issues/6457).
         if 300 <= resp.code < 400:


@@ -925,6 +942,21 @@ def determine_ext(url, default_ext='unknown_video'):
     guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return guess
     guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return guess

+    elif guess.rstrip('/') in (
+            'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
+            'flv', 'f4v', 'f4a', 'f4b',
+            'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
+            'mkv', 'mka', 'mk3d',
+            'avi', 'divx',
+            'mov',
+            'asf', 'wmv', 'wma',
+            '3gp', '3g2',
+            'mp3',
+            'flac',
+            'ape',
+            'wav',
+            'f4f', 'f4m', 'm3u8', 'smil'):
+        return guess.rstrip('/')

     else:
         return default_ext
 
     else:
         return default_ext
 


@@ -1376,6 +1408,15 @@ def remove_end(s, end):
     return s
 
 
     return s
 
 

+def remove_quotes(s):
+    if s is None or len(s) < 2:
+        return s
+    for quote in ('"', "'", ):
+        if s[0] == quote and s[-1] == quote:
+            return s[1:-1]
+    return s
+
+

 def url_basename(url):
     path = compat_urlparse.urlparse(url).path
     return path.strip('/').split('/')[-1]
 def url_basename(url):
     path = compat_urlparse.urlparse(url).path
     return path.strip('/').split('/')[-1]


@@ -1668,7 +1709,13 @@ def urlencode_postdata(*args, **kargs):
 
 
 def encode_dict(d, encoding='utf-8'):
 
 
 def encode_dict(d, encoding='utf-8'):

-    return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items())
+    def encode(v):
+        return v.encode(encoding) if isinstance(v, compat_basestring) else v
+    return dict((encode(k), encode(v)) for k, v in d.items())
+
+
+def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
+    return string if isinstance(string, compat_str) else compat_str(string, encoding, errors)

 
 
 US_RATINGS = {
 
 
 US_RATINGS = {


@@ -1765,6 +1812,15 @@ def args_to_str(args):
     return ' '.join(shlex_quote(a) for a in args)
 
 
     return ' '.join(shlex_quote(a) for a in args)
 
 

+def error_to_compat_str(err):
+    err_str = str(err)
+    # On python 2 error byte string must be decoded with proper
+    # encoding rather than ascii
+    if sys.version_info[0] < 3:
+        err_str = err_str.decode(preferredencoding())
+    return err_str
+
+

 def mimetype2ext(mt):
     _, _, res = mt.rpartition('/')
 
 def mimetype2ext(mt):
     _, _, res = mt.rpartition('/')
 


@@ -1935,15 +1991,15 @@ def match_filter_func(filter_str):
 
 def parse_dfxp_time_expr(time_expr):
     if not time_expr:
 
 def parse_dfxp_time_expr(time_expr):
     if not time_expr:

-        return 0.0
+        return

 
     mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
     if mobj:
         return float(mobj.group('time_offset'))
 
 
     mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
     if mobj:
         return float(mobj.group('time_offset'))
 

-    mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:\.\d+)?)$', time_expr)
+    mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)

     if mobj:
     if mobj:

-        return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3))
+        return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))

 
 
 def srt_subtitles_timecode(seconds):
 
 
 def srt_subtitles_timecode(seconds):


@@ -1979,10 +2035,15 @@ def dfxp2srt(dfxp_data):
         raise ValueError('Invalid dfxp/TTML subtitle')
 
     for para, index in zip(paras, itertools.count(1)):
         raise ValueError('Invalid dfxp/TTML subtitle')
 
     for para, index in zip(paras, itertools.count(1)):

-        begin_time = parse_dfxp_time_expr(para.attrib['begin'])
+        begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))

         end_time = parse_dfxp_time_expr(para.attrib.get('end'))
         end_time = parse_dfxp_time_expr(para.attrib.get('end'))

+        dur = parse_dfxp_time_expr(para.attrib.get('dur'))
+        if begin_time is None:
+            continue

         if not end_time:
         if not end_time:

-            end_time = begin_time + parse_dfxp_time_expr(para.attrib['dur'])
+            if not dur:
+                continue
+            end_time = begin_time + dur

         out.append('%d\n%s --> %s\n%s\n\n' % (
             index,
             srt_subtitles_timecode(begin_time),
         out.append('%d\n%s --> %s\n%s\n\n' % (
             index,
             srt_subtitles_timecode(begin_time),