import base64
import hashlib
import json
+import netrc
import os
import re
import socket
import sys
-import netrc
+import time
import xml.etree.ElementTree
from ..utils import (
clean_html,
compiled_regex_type,
ExtractorError,
+ int_or_none,
RegexNotFoundError,
sanitize_filename,
unescapeHTML,
* vcodec Name of the video codec in use
* container Name of the container format
* filesize The number of bytes, if known in advance
+ * filesize_approx An estimate for the number of bytes
* 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", "m3u8" or so.
* preference Order number of this format. If this field is
present and not None, the formats get sorted
- by this field.
+ by this field, regardless of all other values.
-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.
+ * http_referer HTTP Referer header value to set.
+ * http_method HTTP method to use for the download.
+ * http_headers A dictionary of additional HTTP headers
+ to add to the request.
+ * http_post_data Additional data to send with a POST
+ request.
url: Final video URL.
ext: Video filename extension.
format: The video format, defaults to ext (used for --get-format)
unique, but available before title. Typically, id is
something like "4234987", title "Dancing naked mole rats",
and display_id "dancing-naked-mole-rats"
- thumbnails: A list of dictionaries (with the entries "resolution" and
- "url") for the varying thumbnails
+ thumbnails: A list of dictionaries, with the following entries:
+ * "url"
+ * "width" (optional, int)
+ * "height" (optional, int)
+ * "resolution" (optional, string "{width}x{height"},
+ deprecated)
thumbnail: Full URL to a video thumbnail image.
description: One-line video description.
uploader: Full name of the video uploader.
+ timestamp: UNIX timestamp of the moment the video became available.
upload_date: Video upload date (YYYYMMDD).
+ If not explicitly set, calculated from timestamp.
uploader_id: Nickname or id of the video uploader.
location: Physical location of the video.
subtitles: The subtitle file contents as a dictionary in the format
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)
+ categories: A list of categories that the video falls in, for example
+ ["Sports", "Berlin"]
Unless mentioned otherwise, the fields should be Unicode strings.
_real_extract() methods and define a _VALID_URL regexp.
Probably, they should also be added to the list of extractors.
- _real_extract() must return a *list* of information dictionaries as
- described above.
-
Finally, the _WORKING attribute should be set to False for broken IEs
in order to warn the users and skip the tests.
"""
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))
+ basen = '%s_%s' % (video_id, url)
+ if len(basen) > 240:
+ h = u'___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
+ basen = basen[:240 - len(h)] + h
+ raw_filename = basen + '.dump'
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')
+ try:
+ content = webpage_bytes.decode(encoding, 'replace')
+ except LookupError:
+ content = webpage_bytes.decode('utf-8', 'replace')
+
+ if (u'<title>Access to this site is blocked</title>' in content and
+ u'Websense' in content[:512]):
+ msg = u'Access to this webpage has been blocked by Websense filtering software in your network.'
+ blocked_iframe = self._html_search_regex(
+ r'<iframe src="([^"]+)"', content,
+ u'Websense information URL', default=None)
+ if blocked_iframe:
+ msg += u' Visit %s for more details' % blocked_iframe
+ raise ExtractorError(msg, expected=True)
+
return (content, urlh)
def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
def _download_xml(self, url_or_request, video_id,
note=u'Downloading XML', errnote=u'Unable to download XML',
- transform_source=None):
+ transform_source=None, fatal=True):
"""Return the xml as an xml.etree.ElementTree.Element"""
- xml_string = self._download_webpage(url_or_request, video_id, note, errnote)
+ xml_string = self._download_webpage(
+ url_or_request, video_id, note, errnote, fatal=fatal)
+ if xml_string is False:
+ return xml_string
if transform_source:
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',
- transform_source=None):
- json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+ transform_source=None,
+ fatal=True):
+ json_string = self._download_webpage(
+ url_or_request, video_id, note, errnote, fatal=fatal)
+ if (not fatal) and json_string is False:
+ return None
if transform_source:
json_string = transform_source(json_string)
try:
else:
for p in pattern:
mobj = re.search(p, string, flags)
- if mobj: break
+ if mobj:
+ break
if os.name != 'nt' and sys.stderr.isatty():
_name = u'\033[0;34m%s\033[0m' % name
return (username, password)
+ def _get_tfa_info(self):
+ """
+ Get the two-factor authentication info
+ TODO - asking the user will be required for sms/phone verify
+ currently just uses the command line option
+ If there's no info available, return None
+ """
+ if self._downloader is None:
+ return None
+ downloader_params = self._downloader.params
+
+ if downloader_params.get('twofactor', None) is not None:
+ return downloader_params['twofactor']
+
+ return None
+
# Helper functions for extracting OpenGraph info
@staticmethod
def _og_regexes(prop):
return self._og_search_property('title', html, **kargs)
def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
- regexes = self._og_regexes('video')
- if secure: regexes = self._og_regexes('video:secure_url') + regexes
+ regexes = self._og_regexes('video') + self._og_regexes('video:url')
+ if secure:
+ regexes = self._og_regexes('video:secure_url') + regexes
return self._html_search_regex(regexes, html, name, **kargs)
- def _html_search_meta(self, name, html, display_name=None, fatal=False):
+ def _og_search_url(self, html, **kargs):
+ return self._og_search_property('url', html, **kargs)
+
+ def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
if display_name is None:
display_name = name
return self._html_search_regex(
r'''(?ix)<meta
- (?=[^>]+(?:itemprop|name|property)=["\']%s["\'])
+ (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?)
[^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
- html, display_name, fatal=fatal)
+ html, display_name, fatal=fatal, **kwargs)
def _dc_search_uploader(self, html):
return self._html_search_meta('dc.creator', html, 'uploader')
f.get('abr') if f.get('abr') is not None else -1,
audio_ext_preference,
f.get('filesize') if f.get('filesize') is not None else -1,
+ f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
f.get('format_id'),
)
formats.sort(key=_formats_key)
+ def http_scheme(self):
+ """ Either "https:" or "https:", depending on the user's preferences """
+ return (
+ 'http:'
+ if self._downloader.params.get('prefer_insecure', False)
+ else 'https:')
+
+ def _proto_relative_url(self, url, scheme=None):
+ if url is None:
+ return url
+ if url.startswith('//'):
+ if scheme is None:
+ scheme = self.http_scheme()
+ return scheme + url
+ else:
+ return url
+
+ def _sleep(self, timeout, video_id, msg_template=None):
+ if msg_template is None:
+ msg_template = u'%(video_id)s: Waiting for %(timeout)s seconds'
+ msg = msg_template % {'video_id': video_id, 'timeout': timeout}
+ self.to_screen(msg)
+ time.sleep(timeout)
+
+ def _extract_f4m_formats(self, manifest_url, video_id):
+ manifest = self._download_xml(
+ manifest_url, video_id, 'Downloading f4m manifest',
+ 'Unable to download f4m manifest')
+
+ formats = []
+ media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
+ for i, media_el in enumerate(media_nodes):
+ tbr = int_or_none(media_el.attrib.get('bitrate'))
+ format_id = 'f4m-%d' % (i if tbr is None else tbr)
+ formats.append({
+ 'format_id': format_id,
+ 'url': manifest_url,
+ 'ext': 'flv',
+ 'tbr': tbr,
+ 'width': int_or_none(media_el.attrib.get('width')),
+ 'height': int_or_none(media_el.attrib.get('height')),
+ })
+ self._sort_formats(formats)
+
+ return formats
+
class SearchInfoExtractor(InfoExtractor):
"""
projects / youtube-dl / blobdiff