X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=inline;f=youtube_dl%2Fextractor%2Fcommon.py;h=fb2d50a098992f8088c41259b73d653c6f6f173d;hb=801dbbdffd9df57e4051e8fd01efbb6104f99c26;hp=8009c2d85708638509ce1539198ecbd6db1dbafa;hpb=a3f62b8255120acda6d429d2a3ce53b13e871c5d;p=youtube-dl.git
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 8009c2d85..fb2d50a09 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -14,6 +14,8 @@ from ..utils import (
clean_html,
compiled_regex_type,
ExtractorError,
+ RegexNotFoundError,
+ sanitize_filename,
unescapeHTML,
)
@@ -35,6 +37,8 @@ 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)
@@ -47,12 +51,31 @@ class InfoExtractor(object):
uploader_id: Nickname or id of the video uploader.
location: Physical location of the video.
player_url: SWF Player URL (used for rtmpdump).
- subtitles: The subtitle file contents.
+ subtitles: The subtitle file contents as a dictionary in the format
+ {language: subtitles}.
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.
+ age_limit: Age restriction for the video, as an integer (years)
+ 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 the format_id, width, height.
+ and format_note fields if missing.
+ * format_id A short description of the format
+ ("mp4_h264_opus" or "19")
+ * format_note Additional info about the format
+ ("3D" or "DASH video")
+ * width Width of the video, if known
+ * height Height of the video, if known
+ webpage_url: The url to the video webpage, if given to youtube-dl it
+ should allow to get the same result again. (It will be set
+ by YoutubeDL if it's missing)
+
+ 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.
@@ -113,6 +136,11 @@ class InfoExtractor(object):
"""Real extraction process. Redefine in subclasses."""
pass
+ @classmethod
+ def ie_key(cls):
+ """A string for getting the InfoExtractor with get_info_extractor"""
+ return cls.__name__[:-2]
+
@property
def IE_NAME(self):
return type(self).__name__[:-2]
@@ -128,7 +156,7 @@ class InfoExtractor(object):
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
if errnote is None:
errnote = u'Unable to download webpage'
- raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
+ raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err)
def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
""" Returns a tuple (page content as string, URL handle) """
@@ -139,12 +167,17 @@ 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']+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()
@@ -153,6 +186,17 @@ class InfoExtractor(object):
self.to_screen(u'Dumping request to ' + url)
dump = base64.b64encode(webpage_bytes).decode('ascii')
self._downloader.to_screen(dump)
+ if self._downloader.params.get('write_pages', False):
+ try:
+ url = url_or_request.get_full_url()
+ except AttributeError:
+ url = url_or_request
+ raw_filename = ('%s_%s.dump' % (video_id, url))
+ filename = sanitize_filename(raw_filename, restricted=True)
+ self.to_screen(u'Saving request to ' + filename)
+ with open(filename, 'wb') as outf:
+ outf.write(webpage_bytes)
+
content = webpage_bytes.decode(encoding, 'replace')
return (content, urlh)
@@ -203,7 +247,7 @@ class InfoExtractor(object):
Perform a regex search on the given string, using a single or a list of
patterns returning the first matching group.
In case of failure return a default value or raise a WARNING or a
- ExtractorError, depending on fatal, specifying the field name.
+ RegexNotFoundError, depending on fatal, specifying the field name.
"""
if isinstance(pattern, (str, compat_str, compiled_regex_type)):
mobj = re.search(pattern, string, flags)
@@ -223,7 +267,7 @@ class InfoExtractor(object):
elif default is not None:
return default
elif fatal:
- raise ExtractorError(u'Unable to extract %s' % _name)
+ raise RegexNotFoundError(u'Unable to extract %s' % _name)
else:
self._downloader.report_warning(u'unable to extract %s; '
u'please report this issue on http://yt-dl.org/bug' % _name)
@@ -278,7 +322,9 @@ class InfoExtractor(object):
if name is None:
name = 'OpenGraph %s' % prop
escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
- return unescapeHTML(escaped)
+ if not escaped is None:
+ return unescapeHTML(escaped)
+ return None
def _og_search_thumbnail(self, html, **kargs):
return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
@@ -289,10 +335,19 @@ class InfoExtractor(object):
def _og_search_title(self, html, **kargs):
return self._og_search_property('title', html, **kargs)
- def _og_search_video_url(self, html, name='video url', **kargs):
- return self._html_search_regex([self._og_regex('video:secure_url'),
- self._og_regex('video')],
- html, name, **kargs)
+ def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
+ regexes = [self._og_regex('video')]
+ if secure: regexes.insert(0, self._og_regex('video:secure_url'))
+ return self._html_search_regex(regexes, html, name, **kargs)
+
+ def _rta_search(self, html):
+ # See http://www.rtalabel.org/index.php?content=howtofaq#single
+ if re.search(r'(?ix)