2016-02-06 07:27:04 -06:00
|
|
|
from __future__ import unicode_literals
|
2014-08-27 18:04:43 -05:00
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
import base64
|
2014-09-28 01:53:52 -05:00
|
|
|
import datetime
|
2014-01-17 07:47:46 -06:00
|
|
|
import hashlib
|
2014-01-07 02:35:34 -06:00
|
|
|
import json
|
2014-07-11 03:57:08 -05:00
|
|
|
import netrc
|
2013-06-23 12:57:38 -05:00
|
|
|
import os
|
|
|
|
import re
|
|
|
|
import socket
|
|
|
|
import sys
|
2014-07-11 03:57:08 -05:00
|
|
|
import time
|
2016-02-02 11:07:07 -06:00
|
|
|
import math
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2014-11-02 04:23:40 -06:00
|
|
|
from ..compat import (
|
2014-11-29 17:03:59 -06:00
|
|
|
compat_cookiejar,
|
2015-07-29 17:20:37 -05:00
|
|
|
compat_cookies,
|
2015-08-15 10:55:07 -05:00
|
|
|
compat_getpass,
|
2013-06-23 12:57:38 -05:00
|
|
|
compat_http_client,
|
|
|
|
compat_urllib_error,
|
2015-08-01 14:13:21 -05:00
|
|
|
compat_urllib_parse,
|
2014-09-24 07:16:56 -05:00
|
|
|
compat_urlparse,
|
2013-06-23 12:57:38 -05:00
|
|
|
compat_str,
|
2015-10-25 14:04:55 -05:00
|
|
|
compat_etree_fromstring,
|
2014-11-02 04:23:40 -06:00
|
|
|
)
|
|
|
|
from ..utils import (
|
2015-06-28 11:56:45 -05:00
|
|
|
NO_DEFAULT,
|
2015-01-07 00:20:20 -06:00
|
|
|
age_restricted,
|
2015-04-17 07:55:24 -05:00
|
|
|
bug_reports_message,
|
2013-06-23 12:57:38 -05:00
|
|
|
clean_html,
|
|
|
|
compiled_regex_type,
|
2015-07-15 14:15:15 -05:00
|
|
|
determine_ext,
|
2015-12-19 19:00:39 -06:00
|
|
|
error_to_compat_str,
|
2013-06-23 12:57:38 -05:00
|
|
|
ExtractorError,
|
2015-07-15 14:14:08 -05:00
|
|
|
fix_xml_ampersands,
|
2014-09-28 03:34:55 -05:00
|
|
|
float_or_none,
|
2014-07-28 08:25:56 -05:00
|
|
|
int_or_none,
|
2016-01-15 12:36:02 -06:00
|
|
|
parse_iso8601,
|
2013-10-23 07:38:03 -05:00
|
|
|
RegexNotFoundError,
|
2013-10-28 04:44:02 -05:00
|
|
|
sanitize_filename,
|
2015-11-21 10:18:17 -06:00
|
|
|
sanitized_Request,
|
2013-07-17 03:38:23 -05:00
|
|
|
unescapeHTML,
|
2015-10-01 11:18:59 -05:00
|
|
|
unified_strdate,
|
2015-08-01 14:13:21 -05:00
|
|
|
url_basename,
|
2015-08-09 08:07:18 -05:00
|
|
|
xpath_text,
|
|
|
|
xpath_with_ns,
|
2015-12-05 14:14:43 -06:00
|
|
|
determine_protocol,
|
2016-02-02 11:07:07 -06:00
|
|
|
parse_duration,
|
2016-02-20 15:02:03 -06:00
|
|
|
mimetype2ext,
|
2013-06-23 12:57:38 -05:00
|
|
|
)
|
2015-06-28 11:56:45 -05:00
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
class InfoExtractor(object):
|
|
|
|
"""Information Extractor class.
|
|
|
|
|
|
|
|
Information extractors are the classes that, given a URL, extract
|
|
|
|
information about the video (or videos) the URL refers to. This
|
|
|
|
information includes the real video URL, the video title, author and
|
|
|
|
others. The information is stored in a dictionary which is then
|
2014-12-21 09:58:29 -06:00
|
|
|
passed to the YoutubeDL. The YoutubeDL processes this
|
2013-06-23 12:57:38 -05:00
|
|
|
information possibly downloading the video to the file system, among
|
|
|
|
other possible outcomes.
|
|
|
|
|
2015-04-29 10:03:10 -05:00
|
|
|
The type field determines the type of the result.
|
2014-11-20 09:47:59 -06:00
|
|
|
By far the most common value (and the default if _type is missing) is
|
|
|
|
"video", which indicates a single video.
|
|
|
|
|
|
|
|
For a video, the dictionaries must include the following fields:
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
id: Video identifier.
|
|
|
|
title: Video title, unescaped.
|
2013-12-16 07:13:40 -06:00
|
|
|
|
2013-12-24 04:56:02 -06:00
|
|
|
Additionally, it must contain either a formats entry or a url one:
|
2013-12-16 07:13:40 -06:00
|
|
|
|
2013-12-24 04:56:02 -06:00
|
|
|
formats: A list of dictionaries for each format available, ordered
|
|
|
|
from worst to best quality.
|
|
|
|
|
|
|
|
Potential fields:
|
2013-12-16 07:13:40 -06:00
|
|
|
* url Mandatory. The URL of the video file
|
2015-07-23 12:37:45 -05:00
|
|
|
* ext Will be calculated from URL if missing
|
2013-12-16 07:13:40 -06:00
|
|
|
* 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
|
2013-12-26 14:19:00 -06:00
|
|
|
("mp4_h264_opus" or "19").
|
|
|
|
Technically optional, but strongly recommended.
|
2013-12-16 07:13:40 -06:00
|
|
|
* 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
|
2013-12-24 04:56:02 -06:00
|
|
|
* resolution Textual description of width and height
|
2013-12-25 08:18:40 -06:00
|
|
|
* tbr Average bitrate of audio and video in KBit/s
|
2013-12-16 07:13:40 -06:00
|
|
|
* abr Average audio bitrate in KBit/s
|
|
|
|
* acodec Name of the audio codec in use
|
2014-01-18 22:47:20 -06:00
|
|
|
* asr Audio sampling rate in Hertz
|
2013-12-16 07:13:40 -06:00
|
|
|
* vbr Average video bitrate in KBit/s
|
2014-10-30 03:34:13 -05:00
|
|
|
* fps Frame rate
|
2013-12-16 07:13:40 -06:00
|
|
|
* vcodec Name of the video codec in use
|
2014-01-23 16:54:06 -06:00
|
|
|
* container Name of the container format
|
2013-12-16 07:13:40 -06:00
|
|
|
* filesize The number of bytes, if known in advance
|
2014-07-21 05:02:44 -05:00
|
|
|
* filesize_approx An estimate for the number of bytes
|
2013-12-16 07:13:40 -06:00
|
|
|
* player_url SWF Player URL (used for rtmpdump).
|
2013-12-24 16:32:04 -06:00
|
|
|
* protocol The protocol that will be used for the actual
|
|
|
|
download, lower-case.
|
2015-01-30 08:53:16 -06:00
|
|
|
"http", "https", "rtsp", "rtmp", "rtmpe",
|
|
|
|
"m3u8", or "m3u8_native".
|
2013-12-24 04:56:02 -06:00
|
|
|
* preference Order number of this format. If this field is
|
2014-01-01 13:23:47 -06:00
|
|
|
present and not None, the formats get sorted
|
2014-03-23 11:41:43 -05:00
|
|
|
by this field, regardless of all other values.
|
2013-12-24 04:56:02 -06:00
|
|
|
-1 for default (order by other properties),
|
|
|
|
-2 or smaller for less than default.
|
2015-01-03 11:33:38 -06:00
|
|
|
< -1000 to hide the format (if there is
|
|
|
|
another one which is strictly better)
|
2016-01-01 06:28:45 -06:00
|
|
|
* language Language code, e.g. "de" or "en-US".
|
|
|
|
* language_preference Is this in the language mentioned in
|
|
|
|
the URL?
|
2014-11-20 05:06:33 -06:00
|
|
|
10 if it's what the URL is about,
|
|
|
|
-1 for default (don't know),
|
|
|
|
-10 otherwise, other values reserved for now.
|
2014-01-06 10:15:27 -06:00
|
|
|
* 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.
|
2014-10-24 17:10:11 -05:00
|
|
|
* source_preference Order number for this video source
|
|
|
|
(quality takes higher priority)
|
|
|
|
-1 for default (order by other properties),
|
|
|
|
-2 or smaller for less than default.
|
2014-08-23 18:31:35 -05:00
|
|
|
* http_headers A dictionary of additional HTTP headers
|
|
|
|
to add to the request.
|
2015-01-09 22:45:51 -06:00
|
|
|
* stretched_ratio If given and not 1, indicates that the
|
2015-01-25 11:09:48 -06:00
|
|
|
video's pixels are not square.
|
|
|
|
width : height ratio as float.
|
|
|
|
* no_resume The server does not support resuming the
|
|
|
|
(HTTP or RTMP) download. Boolean.
|
|
|
|
|
2013-12-15 21:09:30 -06:00
|
|
|
url: Final video URL.
|
2013-06-23 12:57:38 -05:00
|
|
|
ext: Video filename extension.
|
2013-12-16 07:13:40 -06:00
|
|
|
format: The video format, defaults to ext (used for --get-format)
|
|
|
|
player_url: SWF Player URL (used for rtmpdump).
|
2013-10-04 04:09:43 -05:00
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
The following fields are optional:
|
|
|
|
|
2014-12-11 20:34:28 -06:00
|
|
|
alt_title: A secondary title of the video.
|
2014-03-03 05:06:28 -06:00
|
|
|
display_id An alternative identifier for the video, not necessarily
|
|
|
|
unique, but available before title. Typically, id is
|
|
|
|
something like "4234987", title "Dancing naked mole rats",
|
|
|
|
and display_id "dancing-naked-mole-rats"
|
2014-06-07 08:33:45 -05:00
|
|
|
thumbnails: A list of dictionaries, with the following entries:
|
2015-01-24 19:38:47 -06:00
|
|
|
* "id" (optional, string) - Thumbnail format ID
|
2014-06-07 08:33:45 -05:00
|
|
|
* "url"
|
2015-01-24 19:38:47 -06:00
|
|
|
* "preference" (optional, int) - quality of the image
|
2014-06-07 08:33:45 -05:00
|
|
|
* "width" (optional, int)
|
|
|
|
* "height" (optional, int)
|
|
|
|
* "resolution" (optional, string "{width}x{height"},
|
|
|
|
deprecated)
|
2013-06-23 12:57:38 -05:00
|
|
|
thumbnail: Full URL to a video thumbnail image.
|
2014-12-11 20:34:28 -06:00
|
|
|
description: Full video description.
|
2013-06-23 12:57:38 -05:00
|
|
|
uploader: Full name of the video uploader.
|
2016-03-02 11:06:39 -06:00
|
|
|
license: License name the video is licensed under.
|
2015-02-03 03:58:28 -06:00
|
|
|
creator: The main artist who created the video.
|
2015-09-26 10:07:54 -05:00
|
|
|
release_date: The date (YYYYMMDD) when the video was released.
|
2014-03-13 12:21:55 -05:00
|
|
|
timestamp: UNIX timestamp of the moment the video became available.
|
2013-06-23 12:57:38 -05:00
|
|
|
upload_date: Video upload date (YYYYMMDD).
|
2014-03-13 12:21:55 -05:00
|
|
|
If not explicitly set, calculated from timestamp.
|
2013-06-23 12:57:38 -05:00
|
|
|
uploader_id: Nickname or id of the video uploader.
|
2016-03-02 11:31:24 -06:00
|
|
|
uploader_url: Full URL to a personal webpage of the video uploader.
|
2014-08-26 18:44:47 -05:00
|
|
|
location: Physical location where the video was filmed.
|
2015-02-15 11:03:41 -06:00
|
|
|
subtitles: The available subtitles as a dictionary in the format
|
|
|
|
{language: subformats}. "subformats" is a list sorted from
|
|
|
|
lower to higher preference, each element is a dictionary
|
|
|
|
with the "ext" entry and one of:
|
|
|
|
* "data": The subtitles file contents
|
2015-07-23 12:37:45 -05:00
|
|
|
* "url": A URL pointing to the subtitles file
|
2015-10-04 09:33:42 -05:00
|
|
|
"ext" will be calculated from URL if missing
|
2015-02-16 14:44:17 -06:00
|
|
|
automatic_captions: Like 'subtitles', used by the YoutubeIE for
|
|
|
|
automatically generated captions
|
2015-12-03 08:55:02 -06:00
|
|
|
duration: Length of the video in seconds, as an integer or float.
|
2013-06-29 09:32:28 -05:00
|
|
|
view_count: How many users have watched the video on the platform.
|
2013-12-05 11:29:07 -06:00
|
|
|
like_count: Number of positive ratings of the video
|
|
|
|
dislike_count: Number of negative ratings of the video
|
2015-10-17 22:34:54 -05:00
|
|
|
repost_count: Number of reposts of the video
|
2015-02-11 11:39:31 -06:00
|
|
|
average_rating: Average rating give by users, the scale used depends on the webpage
|
2013-12-05 11:29:07 -06:00
|
|
|
comment_count: Number of comments on the video
|
2015-01-09 16:59:18 -06:00
|
|
|
comments: A list of comments, each with one or more of the following
|
|
|
|
properties (all but one of text or html optional):
|
|
|
|
* "author" - human-readable name of the comment author
|
|
|
|
* "author_id" - user ID of the comment author
|
|
|
|
* "id" - Comment ID
|
|
|
|
* "html" - Comment as HTML
|
|
|
|
* "text" - Plain text of the comment
|
|
|
|
* "timestamp" - UNIX timestamp of comment
|
|
|
|
* "parent" - ID of the comment this one is replying to.
|
|
|
|
Set to "root" to indicate that this is a
|
|
|
|
comment to the original video.
|
2013-10-05 23:06:30 -05:00
|
|
|
age_limit: Age restriction for the video, as an integer (years)
|
2015-07-23 12:37:45 -05:00
|
|
|
webpage_url: The URL to the video webpage, if given to youtube-dl it
|
2013-11-03 05:11:13 -06:00
|
|
|
should allow to get the same result again. (It will be set
|
|
|
|
by YoutubeDL if it's missing)
|
2014-05-15 05:41:42 -05:00
|
|
|
categories: A list of categories that the video falls in, for example
|
|
|
|
["Sports", "Berlin"]
|
2015-07-28 16:43:03 -05:00
|
|
|
tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"]
|
2014-09-19 02:57:53 -05:00
|
|
|
is_live: True, False, or None (=unknown). Whether this video is a
|
|
|
|
live stream that goes on instead of a fixed-length video.
|
2015-07-20 14:10:28 -05:00
|
|
|
start_time: Time in seconds where the reproduction should start, as
|
2015-07-23 12:37:45 -05:00
|
|
|
specified in the URL.
|
2015-07-23 06:20:21 -05:00
|
|
|
end_time: Time in seconds where the reproduction should end, as
|
2015-07-23 12:37:45 -05:00
|
|
|
specified in the URL.
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2015-12-30 15:10:44 -06:00
|
|
|
The following fields should only be used when the video belongs to some logical
|
|
|
|
chapter or section:
|
|
|
|
|
|
|
|
chapter: Name or title of the chapter the video belongs to.
|
2016-01-01 08:26:56 -06:00
|
|
|
chapter_number: Number of the chapter the video belongs to, as an integer.
|
|
|
|
chapter_id: Id of the chapter the video belongs to, as a unicode string.
|
2015-12-30 15:10:44 -06:00
|
|
|
|
|
|
|
The following fields should only be used when the video is an episode of some
|
|
|
|
series or programme:
|
|
|
|
|
|
|
|
series: Title of the series or programme the video episode belongs to.
|
|
|
|
season: Title of the season the video episode belongs to.
|
2016-01-01 08:26:56 -06:00
|
|
|
season_number: Number of the season the video episode belongs to, as an integer.
|
|
|
|
season_id: Id of the season the video episode belongs to, as a unicode string.
|
2015-12-30 15:10:44 -06:00
|
|
|
episode: Title of the video episode. Unlike mandatory video title field,
|
|
|
|
this field should denote the exact title of the video episode
|
|
|
|
without any kind of decoration.
|
2016-01-01 08:26:56 -06:00
|
|
|
episode_number: Number of the video episode within a season, as an integer.
|
|
|
|
episode_id: Id of the video episode, as a unicode string.
|
2015-12-30 15:10:44 -06:00
|
|
|
|
2013-10-04 03:40:42 -05:00
|
|
|
Unless mentioned otherwise, the fields should be Unicode strings.
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2014-10-03 13:17:10 -05:00
|
|
|
Unless mentioned otherwise, None is equivalent to absence of information.
|
|
|
|
|
2014-11-20 09:47:59 -06:00
|
|
|
|
|
|
|
_type "playlist" indicates multiple videos.
|
2014-12-06 07:02:19 -06:00
|
|
|
There must be a key "entries", which is a list, an iterable, or a PagedList
|
|
|
|
object, each element of which is a valid dictionary by this specification.
|
2014-11-20 09:47:59 -06:00
|
|
|
|
2015-08-09 08:09:50 -05:00
|
|
|
Additionally, playlists can have "title", "description" and "id" attributes
|
|
|
|
with the same semantics as videos (see above).
|
2014-11-20 09:47:59 -06:00
|
|
|
|
|
|
|
|
|
|
|
_type "multi_video" indicates that there are multiple videos that
|
|
|
|
form a single show, for examples multiple acts of an opera or TV episode.
|
|
|
|
It must have an entries key like a playlist and contain all the keys
|
|
|
|
required for a video at the same time.
|
|
|
|
|
|
|
|
|
|
|
|
_type "url" indicates that the video must be extracted from another
|
|
|
|
location, possibly by a different extractor. Its only required key is:
|
|
|
|
"url" - the next URL to extract.
|
2014-12-09 03:58:06 -06:00
|
|
|
The key "ie_key" can be set to the class name (minus the trailing "IE",
|
|
|
|
e.g. "Youtube") if the extractor class is known in advance.
|
|
|
|
Additionally, the dictionary may have any properties of the resolved entity
|
|
|
|
known in advance, for example "title" if the title of the referred video is
|
2014-11-20 09:47:59 -06:00
|
|
|
known ahead of time.
|
|
|
|
|
|
|
|
|
|
|
|
_type "url_transparent" entities have the same specification as "url", but
|
|
|
|
indicate that the given additional information is more precise than the one
|
|
|
|
associated with the resolved URL.
|
|
|
|
This is useful when a site employs a video service that hosts the video and
|
|
|
|
its technical metadata, but that video service does not embed a useful
|
|
|
|
title, description etc.
|
|
|
|
|
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
Subclasses of this one should re-define the _real_initialize() and
|
|
|
|
_real_extract() methods and define a _VALID_URL regexp.
|
|
|
|
Probably, they should also be added to the list of extractors.
|
|
|
|
|
|
|
|
Finally, the _WORKING attribute should be set to False for broken IEs
|
|
|
|
in order to warn the users and skip the tests.
|
|
|
|
"""
|
|
|
|
|
|
|
|
_ready = False
|
|
|
|
_downloader = None
|
|
|
|
_WORKING = True
|
|
|
|
|
|
|
|
def __init__(self, downloader=None):
|
|
|
|
"""Constructor. Receives an optional downloader."""
|
|
|
|
self._ready = False
|
|
|
|
self.set_downloader(downloader)
|
|
|
|
|
|
|
|
@classmethod
|
|
|
|
def suitable(cls, url):
|
|
|
|
"""Receives a URL and returns True if suitable for this IE."""
|
2013-08-20 21:06:46 -05:00
|
|
|
|
|
|
|
# This does not use has/getattr intentionally - we want to know whether
|
|
|
|
# we have cached the regexp for *this* class, whereas getattr would also
|
|
|
|
# match the superclass
|
|
|
|
if '_VALID_URL_RE' not in cls.__dict__:
|
|
|
|
cls._VALID_URL_RE = re.compile(cls._VALID_URL)
|
|
|
|
return cls._VALID_URL_RE.match(url) is not None
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2014-09-28 02:31:58 -05:00
|
|
|
@classmethod
|
|
|
|
def _match_id(cls, url):
|
|
|
|
if '_VALID_URL_RE' not in cls.__dict__:
|
|
|
|
cls._VALID_URL_RE = re.compile(cls._VALID_URL)
|
|
|
|
m = cls._VALID_URL_RE.match(url)
|
|
|
|
assert m
|
|
|
|
return m.group('id')
|
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
@classmethod
|
|
|
|
def working(cls):
|
|
|
|
"""Getter method for _WORKING."""
|
|
|
|
return cls._WORKING
|
|
|
|
|
|
|
|
def initialize(self):
|
|
|
|
"""Initializes an instance (authentication, etc)."""
|
|
|
|
if not self._ready:
|
|
|
|
self._real_initialize()
|
|
|
|
self._ready = True
|
|
|
|
|
|
|
|
def extract(self, url):
|
|
|
|
"""Extracts URL information and returns it in list of dicts."""
|
2015-02-09 18:13:57 -06:00
|
|
|
try:
|
|
|
|
self.initialize()
|
|
|
|
return self._real_extract(url)
|
|
|
|
except ExtractorError:
|
|
|
|
raise
|
|
|
|
except compat_http_client.IncompleteRead as e:
|
2016-01-10 09:17:47 -06:00
|
|
|
raise ExtractorError('A network error has occurred.', cause=e, expected=True)
|
2015-02-10 08:55:51 -06:00
|
|
|
except (KeyError, StopIteration) as e:
|
2016-01-10 09:17:47 -06:00
|
|
|
raise ExtractorError('An extractor error has occurred.', cause=e)
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
def set_downloader(self, downloader):
|
|
|
|
"""Sets the downloader for this IE."""
|
|
|
|
self._downloader = downloader
|
|
|
|
|
|
|
|
def _real_initialize(self):
|
|
|
|
"""Real initialization process. Redefine in subclasses."""
|
|
|
|
pass
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
"""Real extraction process. Redefine in subclasses."""
|
|
|
|
pass
|
|
|
|
|
2013-07-08 08:14:27 -05:00
|
|
|
@classmethod
|
|
|
|
def ie_key(cls):
|
|
|
|
"""A string for getting the InfoExtractor with get_info_extractor"""
|
2015-10-31 12:12:57 -05:00
|
|
|
return compat_str(cls.__name__[:-2])
|
2013-07-08 08:14:27 -05:00
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
@property
|
|
|
|
def IE_NAME(self):
|
2015-10-31 12:12:57 -05:00
|
|
|
return compat_str(type(self).__name__[:-2])
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2013-12-08 18:49:01 -06:00
|
|
|
def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
|
2013-06-23 12:57:38 -05:00
|
|
|
""" Returns the response handle """
|
|
|
|
if note is None:
|
|
|
|
self.report_download_webpage(video_id)
|
|
|
|
elif note is not False:
|
2013-12-08 18:49:01 -06:00
|
|
|
if video_id is None:
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('%s' % (note,))
|
2013-12-08 18:49:01 -06:00
|
|
|
else:
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('%s: %s' % (video_id, note))
|
2013-06-23 12:57:38 -05:00
|
|
|
try:
|
2013-11-22 12:57:52 -06:00
|
|
|
return self._downloader.urlopen(url_or_request)
|
2013-06-23 12:57:38 -05:00
|
|
|
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
|
2013-12-20 10:05:28 -06:00
|
|
|
if errnote is False:
|
|
|
|
return False
|
2013-06-23 12:57:38 -05:00
|
|
|
if errnote is None:
|
2014-08-27 18:04:43 -05:00
|
|
|
errnote = 'Unable to download webpage'
|
2015-12-19 17:27:38 -06:00
|
|
|
|
2015-12-19 19:00:39 -06:00
|
|
|
errmsg = '%s: %s' % (errnote, error_to_compat_str(err))
|
2013-12-08 18:49:01 -06:00
|
|
|
if fatal:
|
|
|
|
raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
|
|
|
|
else:
|
|
|
|
self._downloader.report_warning(errmsg)
|
|
|
|
return False
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2015-03-20 23:21:27 -05:00
|
|
|
def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None):
|
2013-06-23 12:57:38 -05:00
|
|
|
""" Returns a tuple (page content as string, URL handle) """
|
2013-07-13 15:52:12 -05:00
|
|
|
# Strip hashes from the URL (#1038)
|
|
|
|
if isinstance(url_or_request, (compat_str, str)):
|
|
|
|
url_or_request = url_or_request.partition('#')[0]
|
|
|
|
|
2013-12-08 18:49:01 -06:00
|
|
|
urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal)
|
|
|
|
if urlh is False:
|
|
|
|
assert not fatal
|
|
|
|
return False
|
2015-03-20 23:21:27 -05:00
|
|
|
content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding)
|
2014-10-26 11:05:44 -05:00
|
|
|
return (content, urlh)
|
|
|
|
|
2015-03-20 23:21:27 -05:00
|
|
|
@staticmethod
|
|
|
|
def _guess_encoding_from_content(content_type, webpage_bytes):
|
2013-06-23 12:57:38 -05:00
|
|
|
m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
|
|
|
|
if m:
|
|
|
|
encoding = m.group(1)
|
|
|
|
else:
|
2013-08-29 04:35:15 -05:00
|
|
|
m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
|
2013-08-28 06:59:08 -05:00
|
|
|
webpage_bytes[:1024])
|
|
|
|
if m:
|
|
|
|
encoding = m.group(1).decode('ascii')
|
2014-01-20 18:39:39 -06:00
|
|
|
elif webpage_bytes.startswith(b'\xff\xfe'):
|
|
|
|
encoding = 'utf-16'
|
2013-08-28 06:59:08 -05:00
|
|
|
else:
|
|
|
|
encoding = 'utf-8'
|
2015-03-20 23:21:27 -05:00
|
|
|
|
|
|
|
return encoding
|
|
|
|
|
|
|
|
def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None):
|
|
|
|
content_type = urlh.headers.get('Content-Type', '')
|
|
|
|
webpage_bytes = urlh.read()
|
|
|
|
if prefix is not None:
|
|
|
|
webpage_bytes = prefix + webpage_bytes
|
|
|
|
if not encoding:
|
|
|
|
encoding = self._guess_encoding_from_content(content_type, webpage_bytes)
|
2013-06-23 12:57:38 -05:00
|
|
|
if self._downloader.params.get('dump_intermediate_pages', False):
|
|
|
|
try:
|
|
|
|
url = url_or_request.get_full_url()
|
|
|
|
except AttributeError:
|
|
|
|
url = url_or_request
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('Dumping request to ' + url)
|
2013-06-23 12:57:38 -05:00
|
|
|
dump = base64.b64encode(webpage_bytes).decode('ascii')
|
|
|
|
self._downloader.to_screen(dump)
|
2013-10-28 04:44:02 -05:00
|
|
|
if self._downloader.params.get('write_pages', False):
|
|
|
|
try:
|
|
|
|
url = url_or_request.get_full_url()
|
|
|
|
except AttributeError:
|
|
|
|
url = url_or_request
|
2014-05-15 05:39:33 -05:00
|
|
|
basen = '%s_%s' % (video_id, url)
|
2014-05-12 14:56:10 -05:00
|
|
|
if len(basen) > 240:
|
2014-08-27 18:04:43 -05:00
|
|
|
h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
|
2014-05-12 14:56:10 -05:00
|
|
|
basen = basen[:240 - len(h)] + h
|
|
|
|
raw_filename = basen + '.dump'
|
2013-10-28 04:44:02 -05:00
|
|
|
filename = sanitize_filename(raw_filename, restricted=True)
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('Saving request to ' + filename)
|
2014-10-14 09:43:48 -05:00
|
|
|
# Working around MAX_PATH limitation on Windows (see
|
|
|
|
# http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx)
|
|
|
|
if os.name == 'nt':
|
|
|
|
absfilepath = os.path.abspath(filename)
|
|
|
|
if len(absfilepath) > 259:
|
|
|
|
filename = '\\\\?\\' + absfilepath
|
2013-10-28 04:44:02 -05:00
|
|
|
with open(filename, 'wb') as outf:
|
|
|
|
outf.write(webpage_bytes)
|
|
|
|
|
2014-04-07 16:09:53 -05:00
|
|
|
try:
|
|
|
|
content = webpage_bytes.decode(encoding, 'replace')
|
|
|
|
except LookupError:
|
|
|
|
content = webpage_bytes.decode('utf-8', 'replace')
|
2014-04-02 23:07:35 -05:00
|
|
|
|
2014-08-27 18:04:43 -05:00
|
|
|
if ('<title>Access to this site is blocked</title>' in content and
|
|
|
|
'Websense' in content[:512]):
|
|
|
|
msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
|
2014-04-02 23:07:35 -05:00
|
|
|
blocked_iframe = self._html_search_regex(
|
|
|
|
r'<iframe src="([^"]+)"', content,
|
2014-08-27 18:04:43 -05:00
|
|
|
'Websense information URL', default=None)
|
2014-04-02 23:07:35 -05:00
|
|
|
if blocked_iframe:
|
2014-08-27 18:04:43 -05:00
|
|
|
msg += ' Visit %s for more details' % blocked_iframe
|
2014-04-02 23:07:35 -05:00
|
|
|
raise ExtractorError(msg, expected=True)
|
2015-02-21 07:51:07 -06:00
|
|
|
if '<title>The URL you requested has been blocked</title>' in content[:512]:
|
|
|
|
msg = (
|
|
|
|
'Access to this webpage has been blocked by Indian censorship. '
|
|
|
|
'Use a VPN or proxy server (with --proxy) to route around it.')
|
|
|
|
block_msg = self._html_search_regex(
|
|
|
|
r'</h1><p>(.*?)</p>',
|
|
|
|
content, 'block message', default=None)
|
|
|
|
if block_msg:
|
|
|
|
msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
|
|
|
|
raise ExtractorError(msg, expected=True)
|
2014-04-02 23:07:35 -05:00
|
|
|
|
2014-10-26 11:05:44 -05:00
|
|
|
return content
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2015-03-20 23:21:27 -05:00
|
|
|
def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None):
|
2013-06-23 12:57:38 -05:00
|
|
|
""" Returns the data of the page as a string """
|
2014-12-04 07:11:27 -06:00
|
|
|
success = False
|
|
|
|
try_count = 0
|
|
|
|
while success is False:
|
|
|
|
try:
|
2015-03-20 23:21:27 -05:00
|
|
|
res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding)
|
2014-12-04 07:11:27 -06:00
|
|
|
success = True
|
|
|
|
except compat_http_client.IncompleteRead as e:
|
|
|
|
try_count += 1
|
|
|
|
if try_count >= tries:
|
|
|
|
raise e
|
|
|
|
self._sleep(timeout, video_id)
|
2013-12-08 18:49:01 -06:00
|
|
|
if res is False:
|
|
|
|
return res
|
|
|
|
else:
|
|
|
|
content, _ = res
|
|
|
|
return content
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2013-11-27 22:47:50 -06:00
|
|
|
def _download_xml(self, url_or_request, video_id,
|
2014-08-27 18:04:43 -05:00
|
|
|
note='Downloading XML', errnote='Unable to download XML',
|
2015-03-20 23:21:27 -05:00
|
|
|
transform_source=None, fatal=True, encoding=None):
|
2013-11-24 07:59:19 -06:00
|
|
|
"""Return the xml as an xml.etree.ElementTree.Element"""
|
2014-04-21 06:45:27 -05:00
|
|
|
xml_string = self._download_webpage(
|
2015-03-20 23:21:27 -05:00
|
|
|
url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding)
|
2014-04-21 06:45:27 -05:00
|
|
|
if xml_string is False:
|
|
|
|
return xml_string
|
2013-12-10 05:45:22 -06:00
|
|
|
if transform_source:
|
|
|
|
xml_string = transform_source(xml_string)
|
2015-10-25 14:04:55 -05:00
|
|
|
return compat_etree_fromstring(xml_string.encode('utf-8'))
|
2013-11-24 07:59:19 -06:00
|
|
|
|
2014-01-07 02:35:34 -06:00
|
|
|
def _download_json(self, url_or_request, video_id,
|
2014-08-27 18:04:43 -05:00
|
|
|
note='Downloading JSON metadata',
|
|
|
|
errnote='Unable to download JSON metadata',
|
2014-07-22 18:27:25 -05:00
|
|
|
transform_source=None,
|
2015-03-20 23:21:27 -05:00
|
|
|
fatal=True, encoding=None):
|
2014-07-22 18:27:25 -05:00
|
|
|
json_string = self._download_webpage(
|
2015-03-20 23:21:27 -05:00
|
|
|
url_or_request, video_id, note, errnote, fatal=fatal,
|
|
|
|
encoding=encoding)
|
2014-07-22 18:27:25 -05:00
|
|
|
if (not fatal) and json_string is False:
|
|
|
|
return None
|
2014-12-05 05:07:06 -06:00
|
|
|
return self._parse_json(
|
|
|
|
json_string, video_id, transform_source=transform_source, fatal=fatal)
|
|
|
|
|
|
|
|
def _parse_json(self, json_string, video_id, transform_source=None, fatal=True):
|
2014-02-09 10:56:10 -06:00
|
|
|
if transform_source:
|
|
|
|
json_string = transform_source(json_string)
|
2014-01-07 02:35:34 -06:00
|
|
|
try:
|
|
|
|
return json.loads(json_string)
|
|
|
|
except ValueError as ve:
|
2014-09-30 04:12:59 -05:00
|
|
|
errmsg = '%s: Failed to parse JSON ' % video_id
|
|
|
|
if fatal:
|
|
|
|
raise ExtractorError(errmsg, cause=ve)
|
|
|
|
else:
|
|
|
|
self.report_warning(errmsg + str(ve))
|
2014-01-07 02:35:34 -06:00
|
|
|
|
2013-12-23 08:57:43 -06:00
|
|
|
def report_warning(self, msg, video_id=None):
|
2014-08-27 18:04:43 -05:00
|
|
|
idstr = '' if video_id is None else '%s: ' % video_id
|
2013-12-23 08:57:43 -06:00
|
|
|
self._downloader.report_warning(
|
2014-08-27 18:04:43 -05:00
|
|
|
'[%s] %s%s' % (self.IE_NAME, idstr, msg))
|
2013-12-23 08:57:43 -06:00
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
def to_screen(self, msg):
|
|
|
|
"""Print msg to screen, prefixing it with '[ie_name]'"""
|
2014-08-27 18:04:43 -05:00
|
|
|
self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg))
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
def report_extraction(self, id_or_name):
|
|
|
|
"""Report information extraction."""
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('%s: Extracting information' % id_or_name)
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
def report_download_webpage(self, video_id):
|
|
|
|
"""Report webpage download."""
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('%s: Downloading webpage' % video_id)
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
def report_age_confirmation(self):
|
|
|
|
"""Report attempt to confirm age."""
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('Confirming age')
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2013-07-07 16:24:34 -05:00
|
|
|
def report_login(self):
|
|
|
|
"""Report attempt to log in."""
|
2014-08-27 18:04:43 -05:00
|
|
|
self.to_screen('Logging in')
|
2013-07-07 16:24:34 -05:00
|
|
|
|
2015-08-26 10:24:47 -05:00
|
|
|
@staticmethod
|
|
|
|
def raise_login_required(msg='This video is only available for registered users'):
|
|
|
|
raise ExtractorError(
|
|
|
|
'%s. Use --username and --password or --netrc to provide account credentials.' % msg,
|
|
|
|
expected=True)
|
|
|
|
|
2015-09-22 10:50:20 -05:00
|
|
|
@staticmethod
|
|
|
|
def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'):
|
|
|
|
raise ExtractorError(
|
|
|
|
'%s. You might want to use --proxy to workaround.' % msg,
|
|
|
|
expected=True)
|
|
|
|
|
2014-11-23 13:41:03 -06:00
|
|
|
# Methods for following #608
|
2013-12-19 13:28:52 -06:00
|
|
|
@staticmethod
|
2015-04-12 12:11:47 -05:00
|
|
|
def url_result(url, ie=None, video_id=None, video_title=None):
|
2015-07-23 12:37:45 -05:00
|
|
|
"""Returns a URL that points to a page that should be processed"""
|
2014-11-23 13:41:03 -06:00
|
|
|
# TODO: ie should be the class used for getting the info
|
2013-06-23 12:57:38 -05:00
|
|
|
video_info = {'_type': 'url',
|
|
|
|
'url': url,
|
|
|
|
'ie_key': ie}
|
2013-11-22 15:46:46 -06:00
|
|
|
if video_id is not None:
|
|
|
|
video_info['id'] = video_id
|
2015-04-12 12:11:47 -05:00
|
|
|
if video_title is not None:
|
|
|
|
video_info['title'] = video_title
|
2013-06-23 12:57:38 -05:00
|
|
|
return video_info
|
2014-11-23 13:41:03 -06:00
|
|
|
|
2013-12-19 13:28:52 -06:00
|
|
|
@staticmethod
|
2014-12-06 13:46:30 -06:00
|
|
|
def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
|
2013-06-23 12:57:38 -05:00
|
|
|
"""Returns a playlist"""
|
|
|
|
video_info = {'_type': 'playlist',
|
|
|
|
'entries': entries}
|
|
|
|
if playlist_id:
|
|
|
|
video_info['id'] = playlist_id
|
|
|
|
if playlist_title:
|
|
|
|
video_info['title'] = playlist_title
|
2014-12-06 13:46:30 -06:00
|
|
|
if playlist_description:
|
|
|
|
video_info['description'] = playlist_description
|
2013-06-23 12:57:38 -05:00
|
|
|
return video_info
|
|
|
|
|
2015-06-28 11:56:45 -05:00
|
|
|
def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
|
2013-06-23 12:57:38 -05:00
|
|
|
"""
|
|
|
|
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
|
2013-10-23 07:38:03 -05:00
|
|
|
RegexNotFoundError, depending on fatal, specifying the field name.
|
2013-06-23 12:57:38 -05:00
|
|
|
"""
|
|
|
|
if isinstance(pattern, (str, compat_str, compiled_regex_type)):
|
|
|
|
mobj = re.search(pattern, string, flags)
|
|
|
|
else:
|
|
|
|
for p in pattern:
|
|
|
|
mobj = re.search(p, string, flags)
|
2014-07-25 03:43:03 -05:00
|
|
|
if mobj:
|
|
|
|
break
|
2013-06-23 12:57:38 -05:00
|
|
|
|
2015-02-09 21:22:10 -06:00
|
|
|
if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty():
|
2014-08-27 18:04:43 -05:00
|
|
|
_name = '\033[0;34m%s\033[0m' % name
|
2013-06-23 12:57:38 -05:00
|
|
|
else:
|
|
|
|
_name = name
|
|
|
|
|
|
|
|
if mobj:
|
2014-11-04 16:14:16 -06:00
|
|
|
if group is None:
|
|
|
|
# return the first matching group
|
|
|
|
return next(g for g in mobj.groups() if g is not None)
|
|
|
|
else:
|
|
|
|
return mobj.group(group)
|
2015-06-28 11:56:45 -05:00
|
|
|
elif default is not NO_DEFAULT:
|
2013-06-23 12:57:38 -05:00
|
|
|
return default
|
|
|
|
elif fatal:
|
2014-08-27 18:04:43 -05:00
|
|
|
raise RegexNotFoundError('Unable to extract %s' % _name)
|
2013-06-23 12:57:38 -05:00
|
|
|
else:
|
2015-04-17 07:55:24 -05:00
|
|
|
self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message())
|
2013-06-23 12:57:38 -05:00
|
|
|
return None
|
|
|
|
|
2015-06-28 11:56:45 -05:00
|
|
|
def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
|
2013-06-23 12:57:38 -05:00
|
|
|
"""
|
|
|
|
Like _search_regex, but strips HTML tags and unescapes entities.
|
|
|
|
"""
|
2014-11-04 16:14:16 -06:00
|
|
|
res = self._search_regex(pattern, string, name, default, fatal, flags, group)
|
2013-06-23 12:57:38 -05:00
|
|
|
if res:
|
|
|
|
return clean_html(res).strip()
|
|
|
|
else:
|
|
|
|
return res
|
|
|
|
|
2013-07-07 16:24:34 -05:00
|
|
|
def _get_login_info(self):
|
|
|
|
"""
|
2015-04-29 10:03:10 -05:00
|
|
|
Get the login info as (username, password)
|
2013-07-07 16:24:34 -05:00
|
|
|
It will look in the netrc file using the _NETRC_MACHINE value
|
|
|
|
If there's no info available, return (None, None)
|
|
|
|
"""
|
|
|
|
if self._downloader is None:
|
|
|
|
return (None, None)
|
|
|
|
|
|
|
|
username = None
|
|
|
|
password = None
|
|
|
|
downloader_params = self._downloader.params
|
|
|
|
|
|
|
|
# Attempt to use provided username and password or .netrc data
|
2016-02-14 02:25:04 -06:00
|
|
|
if downloader_params.get('username') is not None:
|
2013-07-07 16:24:34 -05:00
|
|
|
username = downloader_params['username']
|
|
|
|
password = downloader_params['password']
|
|
|
|
elif downloader_params.get('usenetrc', False):
|
|
|
|
try:
|
|
|
|
info = netrc.netrc().authenticators(self._NETRC_MACHINE)
|
|
|
|
if info is not None:
|
|
|
|
username = info[0]
|
|
|
|
password = info[2]
|
|
|
|
else:
|
|
|
|
raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
|
|
|
|
except (IOError, netrc.NetrcParseError) as err:
|
2015-12-19 19:00:39 -06:00
|
|
|
self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err))
|
2014-11-23 13:41:03 -06:00
|
|
|
|
2013-07-07 16:24:34 -05:00
|
|
|
return (username, password)
|
|
|
|
|
2015-08-15 10:55:07 -05:00
|
|
|
def _get_tfa_info(self, note='two-factor verification code'):
|
2014-08-16 16:28:41 -05:00
|
|
|
"""
|
|
|
|
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
|
|
|
|
|
2016-02-14 02:25:04 -06:00
|
|
|
if downloader_params.get('twofactor') is not None:
|
2014-08-16 16:28:41 -05:00
|
|
|
return downloader_params['twofactor']
|
|
|
|
|
2015-08-15 10:55:07 -05:00
|
|
|
return compat_getpass('Type %s and press [Return]: ' % note)
|
2014-08-16 16:28:41 -05:00
|
|
|
|
2013-07-12 12:00:19 -05:00
|
|
|
# Helper functions for extracting OpenGraph info
|
|
|
|
@staticmethod
|
2013-11-15 05:24:54 -06:00
|
|
|
def _og_regexes(prop):
|
2015-10-17 22:11:02 -05:00
|
|
|
content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
|
2015-10-14 09:49:39 -05:00
|
|
|
property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
|
|
|
|
% {'prop': re.escape(prop)})
|
2013-11-15 05:54:13 -06:00
|
|
|
template = r'<meta[^>]+?%s[^>]+?%s'
|
2013-11-15 05:24:54 -06:00
|
|
|
return [
|
2013-11-15 05:54:13 -06:00
|
|
|
template % (property_re, content_re),
|
|
|
|
template % (content_re, property_re),
|
2013-11-15 05:24:54 -06:00
|
|
|
]
|
2013-07-12 12:00:19 -05:00
|
|
|
|
2015-07-28 16:43:03 -05:00
|
|
|
@staticmethod
|
|
|
|
def _meta_regex(prop):
|
|
|
|
return r'''(?isx)<meta
|
2015-08-15 04:58:30 -05:00
|
|
|
(?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1)
|
2015-07-28 16:43:03 -05:00
|
|
|
[^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop)
|
|
|
|
|
2013-07-13 13:39:47 -05:00
|
|
|
def _og_search_property(self, prop, html, name=None, **kargs):
|
2013-07-12 12:00:19 -05:00
|
|
|
if name is None:
|
2013-07-13 13:39:47 -05:00
|
|
|
name = 'OpenGraph %s' % prop
|
2013-11-15 05:24:54 -06:00
|
|
|
escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
|
2013-11-12 03:36:23 -06:00
|
|
|
if escaped is None:
|
|
|
|
return None
|
|
|
|
return unescapeHTML(escaped)
|
2013-07-12 12:00:19 -05:00
|
|
|
|
|
|
|
def _og_search_thumbnail(self, html, **kargs):
|
2015-07-23 12:37:45 -05:00
|
|
|
return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs)
|
2013-07-12 12:00:19 -05:00
|
|
|
|
|
|
|
def _og_search_description(self, html, **kargs):
|
|
|
|
return self._og_search_property('description', html, fatal=False, **kargs)
|
|
|
|
|
|
|
|
def _og_search_title(self, html, **kargs):
|
|
|
|
return self._og_search_property('title', html, **kargs)
|
|
|
|
|
2013-10-28 01:34:29 -05:00
|
|
|
def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
|
2014-08-21 06:05:24 -05:00
|
|
|
regexes = self._og_regexes('video') + self._og_regexes('video:url')
|
|
|
|
if secure:
|
|
|
|
regexes = self._og_regexes('video:secure_url') + regexes
|
2013-10-28 01:34:29 -05:00
|
|
|
return self._html_search_regex(regexes, html, name, **kargs)
|
2013-07-12 12:00:19 -05:00
|
|
|
|
2014-06-26 09:34:36 -05:00
|
|
|
def _og_search_url(self, html, **kargs):
|
|
|
|
return self._og_search_property('url', html, **kargs)
|
|
|
|
|
2014-07-11 08:38:18 -05:00
|
|
|
def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
|
2013-11-19 23:13:19 -06:00
|
|
|
if display_name is None:
|
|
|
|
display_name = name
|
|
|
|
return self._html_search_regex(
|
2015-07-28 16:43:03 -05:00
|
|
|
self._meta_regex(name),
|
2014-11-04 16:14:16 -06:00
|
|
|
html, display_name, fatal=fatal, group='content', **kwargs)
|
2013-11-19 23:13:19 -06:00
|
|
|
|
|
|
|
def _dc_search_uploader(self, html):
|
|
|
|
return self._html_search_meta('dc.creator', html, 'uploader')
|
|
|
|
|
2013-10-05 23:06:30 -05:00
|
|
|
def _rta_search(self, html):
|
|
|
|
# See http://www.rtalabel.org/index.php?content=howtofaq#single
|
|
|
|
if re.search(r'(?ix)<meta\s+name="rating"\s+'
|
|
|
|
r' content="RTA-5042-1996-1400-1577-RTA"',
|
|
|
|
html):
|
|
|
|
return 18
|
|
|
|
return 0
|
|
|
|
|
2013-11-19 23:13:19 -06:00
|
|
|
def _media_rating_search(self, html):
|
|
|
|
# See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/
|
|
|
|
rating = self._html_search_meta('rating', html)
|
|
|
|
|
|
|
|
if not rating:
|
|
|
|
return None
|
|
|
|
|
|
|
|
RATING_TABLE = {
|
|
|
|
'safe for kids': 0,
|
|
|
|
'general': 8,
|
|
|
|
'14 years': 14,
|
|
|
|
'mature': 17,
|
|
|
|
'restricted': 19,
|
|
|
|
}
|
2016-02-14 02:25:04 -06:00
|
|
|
return RATING_TABLE.get(rating.lower())
|
2013-11-19 23:13:19 -06:00
|
|
|
|
2015-02-08 09:39:00 -06:00
|
|
|
def _family_friendly_search(self, html):
|
2015-02-14 15:20:24 -06:00
|
|
|
# See http://schema.org/VideoObject
|
2015-02-08 09:39:00 -06:00
|
|
|
family_friendly = self._html_search_meta('isFamilyFriendly', html)
|
|
|
|
|
|
|
|
if not family_friendly:
|
|
|
|
return None
|
|
|
|
|
|
|
|
RATING_TABLE = {
|
|
|
|
'1': 0,
|
|
|
|
'true': 0,
|
|
|
|
'0': 18,
|
|
|
|
'false': 18,
|
|
|
|
}
|
2016-02-14 02:25:04 -06:00
|
|
|
return RATING_TABLE.get(family_friendly.lower())
|
2015-02-08 09:39:00 -06:00
|
|
|
|
2014-01-29 11:03:32 -06:00
|
|
|
def _twitter_search_player(self, html):
|
|
|
|
return self._html_search_meta('twitter:player', html,
|
2014-11-23 14:39:15 -06:00
|
|
|
'twitter card player')
|
2014-01-29 11:03:32 -06:00
|
|
|
|
2016-01-16 06:45:36 -06:00
|
|
|
def _search_json_ld(self, html, video_id, **kwargs):
|
2016-01-15 12:36:02 -06:00
|
|
|
json_ld = self._search_regex(
|
|
|
|
r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
|
2016-01-16 06:45:36 -06:00
|
|
|
html, 'JSON-LD', group='json_ld', **kwargs)
|
2016-01-15 12:36:02 -06:00
|
|
|
if not json_ld:
|
|
|
|
return {}
|
2016-01-16 06:45:36 -06:00
|
|
|
return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True))
|
2016-01-15 12:36:02 -06:00
|
|
|
|
|
|
|
def _json_ld(self, json_ld, video_id, fatal=True):
|
|
|
|
if isinstance(json_ld, compat_str):
|
|
|
|
json_ld = self._parse_json(json_ld, video_id, fatal=fatal)
|
|
|
|
if not json_ld:
|
|
|
|
return {}
|
|
|
|
info = {}
|
|
|
|
if json_ld.get('@context') == 'http://schema.org':
|
|
|
|
item_type = json_ld.get('@type')
|
|
|
|
if item_type == 'TVEpisode':
|
|
|
|
info.update({
|
|
|
|
'episode': unescapeHTML(json_ld.get('name')),
|
|
|
|
'episode_number': int_or_none(json_ld.get('episodeNumber')),
|
|
|
|
'description': unescapeHTML(json_ld.get('description')),
|
|
|
|
})
|
|
|
|
part_of_season = json_ld.get('partOfSeason')
|
|
|
|
if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason':
|
|
|
|
info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
|
|
|
|
part_of_series = json_ld.get('partOfSeries')
|
|
|
|
if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries':
|
|
|
|
info['series'] = unescapeHTML(part_of_series.get('name'))
|
|
|
|
elif item_type == 'Article':
|
|
|
|
info.update({
|
|
|
|
'timestamp': parse_iso8601(json_ld.get('datePublished')),
|
|
|
|
'title': unescapeHTML(json_ld.get('headline')),
|
|
|
|
'description': unescapeHTML(json_ld.get('articleBody')),
|
|
|
|
})
|
|
|
|
return dict((k, v) for k, v in info.items() if v is not None)
|
|
|
|
|
2015-07-10 10:49:09 -05:00
|
|
|
@staticmethod
|
2015-07-14 11:36:30 -05:00
|
|
|
def _hidden_inputs(html):
|
2015-09-11 10:07:32 -05:00
|
|
|
html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)
|
2015-08-15 10:52:22 -05:00
|
|
|
hidden_inputs = {}
|
2015-09-11 09:43:05 -05:00
|
|
|
for input in re.findall(r'(?i)<input([^>]+)>', html):
|
2015-09-05 20:20:47 -05:00
|
|
|
if not re.search(r'type=(["\'])(?:hidden|submit)\1', input):
|
2015-08-15 10:52:22 -05:00
|
|
|
continue
|
|
|
|
name = re.search(r'name=(["\'])(?P<value>.+?)\1', input)
|
|
|
|
if not name:
|
|
|
|
continue
|
|
|
|
value = re.search(r'value=(["\'])(?P<value>.*?)\1', input)
|
|
|
|
if not value:
|
|
|
|
continue
|
|
|
|
hidden_inputs[name.group('value')] = value.group('value')
|
|
|
|
return hidden_inputs
|
2015-07-10 10:49:09 -05:00
|
|
|
|
2015-07-14 11:38:10 -05:00
|
|
|
def _form_hidden_inputs(self, form_id, html):
|
|
|
|
form = self._search_regex(
|
2015-09-11 09:43:05 -05:00
|
|
|
r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id,
|
2015-07-14 11:38:10 -05:00
|
|
|
html, '%s form' % form_id, group='form')
|
|
|
|
return self._hidden_inputs(form)
|
|
|
|
|
2015-04-20 10:13:31 -05:00
|
|
|
def _sort_formats(self, formats, field_preference=None):
|
2014-01-27 00:31:54 -06:00
|
|
|
if not formats:
|
2014-08-27 18:04:43 -05:00
|
|
|
raise ExtractorError('No video formats found')
|
2014-01-27 00:31:54 -06:00
|
|
|
|
2016-01-27 09:11:17 -06:00
|
|
|
for f in formats:
|
|
|
|
# Automatically determine tbr when missing based on abr and vbr (improves
|
|
|
|
# formats sorting in some cases)
|
2016-01-29 13:47:46 -06:00
|
|
|
if 'tbr' not in f and f.get('abr') is not None and f.get('vbr') is not None:
|
2016-01-27 09:11:17 -06:00
|
|
|
f['tbr'] = f['abr'] + f['vbr']
|
|
|
|
|
2013-12-24 05:25:22 -06:00
|
|
|
def _formats_key(f):
|
2013-12-24 05:40:23 -06:00
|
|
|
# TODO remove the following workaround
|
|
|
|
from ..utils import determine_ext
|
|
|
|
if not f.get('ext') and 'url' in f:
|
|
|
|
f['ext'] = determine_ext(f['url'])
|
|
|
|
|
2015-04-20 10:13:31 -05:00
|
|
|
if isinstance(field_preference, (list, tuple)):
|
|
|
|
return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference)
|
|
|
|
|
2013-12-24 05:25:22 -06:00
|
|
|
preference = f.get('preference')
|
|
|
|
if preference is None:
|
2015-12-05 14:14:43 -06:00
|
|
|
preference = 0
|
2013-12-24 05:25:22 -06:00
|
|
|
if f.get('ext') in ['f4f', 'f4m']: # Not yet supported
|
|
|
|
preference -= 0.5
|
|
|
|
|
2015-12-05 14:14:43 -06:00
|
|
|
proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1
|
|
|
|
|
2013-12-24 05:25:22 -06:00
|
|
|
if f.get('vcodec') == 'none': # audio only
|
|
|
|
if self._downloader.params.get('prefer_free_formats'):
|
2014-08-27 18:04:43 -05:00
|
|
|
ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus']
|
2013-12-24 05:25:22 -06:00
|
|
|
else:
|
2014-08-27 18:04:43 -05:00
|
|
|
ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a']
|
2013-12-24 05:25:22 -06:00
|
|
|
ext_preference = 0
|
|
|
|
try:
|
|
|
|
audio_ext_preference = ORDER.index(f['ext'])
|
|
|
|
except ValueError:
|
|
|
|
audio_ext_preference = -1
|
|
|
|
else:
|
|
|
|
if self._downloader.params.get('prefer_free_formats'):
|
2014-08-27 18:04:43 -05:00
|
|
|
ORDER = ['flv', 'mp4', 'webm']
|
2013-12-24 05:25:22 -06:00
|
|
|
else:
|
2014-08-27 18:04:43 -05:00
|
|
|
ORDER = ['webm', 'flv', 'mp4']
|
2013-12-24 05:25:22 -06:00
|
|
|
try:
|
|
|
|
ext_preference = ORDER.index(f['ext'])
|
|
|
|
except ValueError:
|
|
|
|
ext_preference = -1
|
|
|
|
audio_ext_preference = 0
|
|
|
|
|
|
|
|
return (
|
|
|
|
preference,
|
2014-11-20 05:06:33 -06:00
|
|
|
f.get('language_preference') if f.get('language_preference') is not None else -1,
|
2014-01-06 10:15:27 -06:00
|
|
|
f.get('quality') if f.get('quality') is not None else -1,
|
2014-01-07 03:25:34 -06:00
|
|
|
f.get('tbr') if f.get('tbr') is not None else -1,
|
2015-02-15 21:37:55 -06:00
|
|
|
f.get('filesize') if f.get('filesize') is not None else -1,
|
2013-12-24 05:25:22 -06:00
|
|
|
f.get('vbr') if f.get('vbr') is not None else -1,
|
2015-02-03 03:53:05 -06:00
|
|
|
f.get('height') if f.get('height') is not None else -1,
|
|
|
|
f.get('width') if f.get('width') is not None else -1,
|
2015-12-05 14:14:43 -06:00
|
|
|
proto_preference,
|
2015-02-06 08:16:43 -06:00
|
|
|
ext_preference,
|
2013-12-24 05:25:22 -06:00
|
|
|
f.get('abr') if f.get('abr') is not None else -1,
|
|
|
|
audio_ext_preference,
|
2014-10-30 03:40:52 -05:00
|
|
|
f.get('fps') if f.get('fps') is not None else -1,
|
2014-07-21 05:02:44 -05:00
|
|
|
f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
|
2014-10-24 17:10:11 -05:00
|
|
|
f.get('source_preference') if f.get('source_preference') is not None else -1,
|
2015-05-06 10:24:24 -05:00
|
|
|
f.get('format_id') if f.get('format_id') is not None else '',
|
2013-12-24 05:25:22 -06:00
|
|
|
)
|
|
|
|
formats.sort(key=_formats_key)
|
2013-11-19 23:13:19 -06:00
|
|
|
|
2015-01-25 12:32:31 -06:00
|
|
|
def _check_formats(self, formats, video_id):
|
|
|
|
if formats:
|
|
|
|
formats[:] = filter(
|
|
|
|
lambda f: self._is_valid_url(
|
|
|
|
f['url'], video_id,
|
|
|
|
item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'),
|
|
|
|
formats)
|
|
|
|
|
2016-02-21 13:19:39 -06:00
|
|
|
@staticmethod
|
|
|
|
def _remove_duplicate_formats(formats):
|
|
|
|
format_urls = set()
|
|
|
|
unique_formats = []
|
|
|
|
for f in formats:
|
|
|
|
if f['url'] not in format_urls:
|
|
|
|
format_urls.add(f['url'])
|
|
|
|
unique_formats.append(f)
|
|
|
|
formats[:] = unique_formats
|
|
|
|
|
2015-01-25 12:32:31 -06:00
|
|
|
def _is_valid_url(self, url, video_id, item='video'):
|
2015-03-02 10:38:44 -06:00
|
|
|
url = self._proto_relative_url(url, scheme='http:')
|
|
|
|
# For now assume non HTTP(S) URLs always valid
|
|
|
|
if not (url.startswith('http://') or url.startswith('https://')):
|
|
|
|
return True
|
2015-01-25 12:32:31 -06:00
|
|
|
try:
|
2015-02-17 10:31:35 -06:00
|
|
|
self._request_webpage(url, video_id, 'Checking %s URL' % item)
|
2015-01-25 12:32:31 -06:00
|
|
|
return True
|
|
|
|
except ExtractorError as e:
|
2015-10-24 05:25:04 -05:00
|
|
|
if isinstance(e.cause, compat_urllib_error.URLError):
|
2015-05-16 15:59:35 -05:00
|
|
|
self.to_screen(
|
|
|
|
'%s: %s URL is invalid, skipping' % (video_id, item))
|
2015-01-25 12:32:31 -06:00
|
|
|
return False
|
|
|
|
raise
|
|
|
|
|
2014-05-04 20:12:41 -05:00
|
|
|
def http_scheme(self):
|
2014-10-24 08:34:19 -05:00
|
|
|
""" Either "http:" or "https:", depending on the user's preferences """
|
2014-05-04 20:12:41 -05:00
|
|
|
return (
|
|
|
|
'http:'
|
|
|
|
if self._downloader.params.get('prefer_insecure', False)
|
|
|
|
else 'https:')
|
|
|
|
|
2014-05-13 02:42:38 -05:00
|
|
|
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
|
|
|
|
|
2014-07-11 03:57:08 -05:00
|
|
|
def _sleep(self, timeout, video_id, msg_template=None):
|
|
|
|
if msg_template is None:
|
2014-08-27 18:04:43 -05:00
|
|
|
msg_template = '%(video_id)s: Waiting for %(timeout)s seconds'
|
2014-07-11 03:57:08 -05:00
|
|
|
msg = msg_template % {'video_id': video_id, 'timeout': timeout}
|
|
|
|
self.to_screen(msg)
|
|
|
|
time.sleep(timeout)
|
|
|
|
|
2015-07-16 23:02:49 -05:00
|
|
|
def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
|
2015-10-01 12:03:31 -05:00
|
|
|
transform_source=lambda s: fix_xml_ampersands(s).strip(),
|
|
|
|
fatal=True):
|
2014-07-28 08:42:19 -05:00
|
|
|
manifest = self._download_xml(
|
|
|
|
manifest_url, video_id, 'Downloading f4m manifest',
|
2015-07-15 14:14:08 -05:00
|
|
|
'Unable to download f4m manifest',
|
|
|
|
# Some manifests may be malformed, e.g. prosiebensat1 generated manifests
|
|
|
|
# (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244)
|
2015-10-01 12:03:31 -05:00
|
|
|
transform_source=transform_source,
|
|
|
|
fatal=fatal)
|
|
|
|
|
|
|
|
if manifest is False:
|
2015-12-27 08:33:39 -06:00
|
|
|
return []
|
2014-07-28 08:25:56 -05:00
|
|
|
|
|
|
|
formats = []
|
2014-10-23 23:17:39 -05:00
|
|
|
manifest_version = '1.0'
|
2014-08-25 06:03:08 -05:00
|
|
|
media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
|
2014-10-23 16:11:10 -05:00
|
|
|
if not media_nodes:
|
2014-10-23 23:17:39 -05:00
|
|
|
manifest_version = '2.0'
|
2014-10-23 16:11:10 -05:00
|
|
|
media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media')
|
2015-11-21 06:01:39 -06:00
|
|
|
base_url = xpath_text(
|
|
|
|
manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'],
|
|
|
|
'base URL', default=None)
|
|
|
|
if base_url:
|
|
|
|
base_url = base_url.strip()
|
2014-08-25 06:03:08 -05:00
|
|
|
for i, media_el in enumerate(media_nodes):
|
2014-10-23 23:17:39 -05:00
|
|
|
if manifest_version == '2.0':
|
2015-07-15 14:25:33 -05:00
|
|
|
media_url = media_el.attrib.get('href') or media_el.attrib.get('url')
|
|
|
|
if not media_url:
|
|
|
|
continue
|
2015-07-15 14:14:52 -05:00
|
|
|
manifest_url = (
|
|
|
|
media_url if media_url.startswith('http://') or media_url.startswith('https://')
|
2015-11-21 06:01:39 -06:00
|
|
|
else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url))
|
2015-07-15 14:15:15 -05:00
|
|
|
# If media_url is itself a f4m manifest do the recursive extraction
|
|
|
|
# since bitrates in parent manifest (this one) and media_url manifest
|
|
|
|
# may differ leading to inability to resolve the format by requested
|
|
|
|
# bitrate in f4m downloader
|
|
|
|
if determine_ext(manifest_url) == 'f4m':
|
2015-12-28 12:58:24 -06:00
|
|
|
formats.extend(self._extract_f4m_formats(
|
|
|
|
manifest_url, video_id, preference, f4m_id, fatal=fatal))
|
2015-07-15 14:15:15 -05:00
|
|
|
continue
|
2014-08-25 06:03:08 -05:00
|
|
|
tbr = int_or_none(media_el.attrib.get('bitrate'))
|
2014-07-28 08:25:56 -05:00
|
|
|
formats.append({
|
2015-04-04 12:05:25 -05:00
|
|
|
'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])),
|
2014-07-28 08:25:56 -05:00
|
|
|
'url': manifest_url,
|
|
|
|
'ext': 'flv',
|
2014-08-25 06:03:08 -05:00
|
|
|
'tbr': tbr,
|
2014-07-28 08:25:56 -05:00
|
|
|
'width': int_or_none(media_el.attrib.get('width')),
|
|
|
|
'height': int_or_none(media_el.attrib.get('height')),
|
2015-02-05 10:16:27 -06:00
|
|
|
'preference': preference,
|
2014-07-28 08:25:56 -05:00
|
|
|
})
|
|
|
|
self._sort_formats(formats)
|
|
|
|
|
|
|
|
return formats
|
|
|
|
|
2014-09-24 07:16:56 -05:00
|
|
|
def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
|
2015-02-05 10:16:27 -06:00
|
|
|
entry_protocol='m3u8', preference=None,
|
2015-07-05 19:39:38 -05:00
|
|
|
m3u8_id=None, note=None, errnote=None,
|
|
|
|
fatal=True):
|
2014-09-24 07:16:56 -05:00
|
|
|
|
2014-08-26 05:51:13 -05:00
|
|
|
formats = [{
|
2015-03-06 10:53:53 -06:00
|
|
|
'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
|
2014-08-26 05:51:13 -05:00
|
|
|
'url': m3u8_url,
|
|
|
|
'ext': ext,
|
|
|
|
'protocol': 'm3u8',
|
2015-02-22 15:30:10 -06:00
|
|
|
'preference': preference - 1 if preference else -1,
|
2014-08-26 05:51:13 -05:00
|
|
|
'resolution': 'multiple',
|
|
|
|
'format_note': 'Quality selection URL',
|
|
|
|
}]
|
|
|
|
|
2014-09-24 07:16:56 -05:00
|
|
|
format_url = lambda u: (
|
|
|
|
u
|
|
|
|
if re.match(r'^https?://', u)
|
|
|
|
else compat_urlparse.urljoin(m3u8_url, u))
|
|
|
|
|
2015-10-31 13:01:34 -05:00
|
|
|
res = self._download_webpage_handle(
|
2014-10-26 20:28:37 -05:00
|
|
|
m3u8_url, video_id,
|
2015-06-07 03:33:22 -05:00
|
|
|
note=note or 'Downloading m3u8 information',
|
2015-07-05 19:39:38 -05:00
|
|
|
errnote=errnote or 'Failed to download m3u8 information',
|
|
|
|
fatal=fatal)
|
2015-10-31 13:01:34 -05:00
|
|
|
if res is False:
|
2015-12-27 08:33:39 -06:00
|
|
|
return []
|
2015-10-31 13:01:34 -05:00
|
|
|
m3u8_doc, urlh = res
|
2015-09-08 13:35:41 -05:00
|
|
|
m3u8_url = urlh.geturl()
|
2016-02-26 19:01:11 -06:00
|
|
|
|
|
|
|
# We should try extracting formats only from master playlists [1], i.e.
|
|
|
|
# playlists that describe available qualities. On the other hand media
|
|
|
|
# playlists [2] should be returned as is since they contain just the media
|
|
|
|
# without qualities renditions.
|
|
|
|
# Fortunately, master playlist can be easily distinguished from media
|
|
|
|
# playlist based on particular tags availability. As of [1, 2] master
|
|
|
|
# playlist tags MUST NOT appear in a media playist and vice versa.
|
|
|
|
# As of [3] #EXT-X-TARGETDURATION tag is REQUIRED for every media playlist
|
|
|
|
# and MUST NOT appear in master playlist thus we can clearly detect media
|
|
|
|
# playlist with this criterion.
|
|
|
|
# 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.4
|
|
|
|
# 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3
|
|
|
|
# 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.1
|
|
|
|
if '#EXT-X-TARGETDURATION' in m3u8_doc: # media playlist, return as is
|
2016-01-26 10:44:44 -06:00
|
|
|
return [{
|
|
|
|
'url': m3u8_url,
|
|
|
|
'format_id': m3u8_id,
|
|
|
|
'ext': ext,
|
|
|
|
'protocol': entry_protocol,
|
|
|
|
'preference': preference,
|
|
|
|
}]
|
2014-08-26 05:51:13 -05:00
|
|
|
last_info = None
|
2015-02-17 14:46:20 -06:00
|
|
|
last_media = None
|
2014-08-26 05:51:13 -05:00
|
|
|
kv_rex = re.compile(
|
|
|
|
r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
|
|
|
|
for line in m3u8_doc.splitlines():
|
|
|
|
if line.startswith('#EXT-X-STREAM-INF:'):
|
|
|
|
last_info = {}
|
|
|
|
for m in kv_rex.finditer(line):
|
|
|
|
v = m.group('val')
|
|
|
|
if v.startswith('"'):
|
|
|
|
v = v[1:-1]
|
|
|
|
last_info[m.group('key')] = v
|
2015-02-17 11:55:53 -06:00
|
|
|
elif line.startswith('#EXT-X-MEDIA:'):
|
|
|
|
last_media = {}
|
|
|
|
for m in kv_rex.finditer(line):
|
|
|
|
v = m.group('val')
|
|
|
|
if v.startswith('"'):
|
|
|
|
v = v[1:-1]
|
|
|
|
last_media[m.group('key')] = v
|
2014-08-26 05:51:13 -05:00
|
|
|
elif line.startswith('#') or not line.strip():
|
|
|
|
continue
|
|
|
|
else:
|
2014-08-28 06:04:49 -05:00
|
|
|
if last_info is None:
|
2014-09-24 07:16:56 -05:00
|
|
|
formats.append({'url': format_url(line)})
|
2014-08-28 03:55:32 -05:00
|
|
|
continue
|
2014-08-26 05:51:13 -05:00
|
|
|
tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
|
2015-03-06 10:52:50 -06:00
|
|
|
format_id = []
|
|
|
|
if m3u8_id:
|
|
|
|
format_id.append(m3u8_id)
|
2015-05-07 05:06:22 -05:00
|
|
|
last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None
|
2015-03-06 10:52:50 -06:00
|
|
|
format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats)))
|
2014-08-26 05:51:13 -05:00
|
|
|
f = {
|
2015-03-06 10:52:50 -06:00
|
|
|
'format_id': '-'.join(format_id),
|
2014-09-24 07:16:56 -05:00
|
|
|
'url': format_url(line.strip()),
|
2014-08-26 05:51:13 -05:00
|
|
|
'tbr': tbr,
|
|
|
|
'ext': ext,
|
2014-09-24 07:16:56 -05:00
|
|
|
'protocol': entry_protocol,
|
|
|
|
'preference': preference,
|
2014-08-26 05:51:13 -05:00
|
|
|
}
|
|
|
|
resolution = last_info.get('RESOLUTION')
|
|
|
|
if resolution:
|
|
|
|
width_str, height_str = resolution.split('x')
|
|
|
|
f['width'] = int(width_str)
|
|
|
|
f['height'] = int(height_str)
|
2016-02-26 18:48:13 -06:00
|
|
|
codecs = last_info.get('CODECS')
|
|
|
|
if codecs:
|
|
|
|
vcodec, acodec = [None] * 2
|
|
|
|
va_codecs = codecs.split(',')
|
|
|
|
if len(va_codecs) == 1:
|
|
|
|
# Audio only entries usually come with single codec and
|
|
|
|
# no resolution. For more robustness we also check it to
|
|
|
|
# be mp4 audio.
|
|
|
|
if not resolution and va_codecs[0].startswith('mp4a'):
|
|
|
|
vcodec, acodec = 'none', va_codecs[0]
|
|
|
|
else:
|
|
|
|
vcodec = va_codecs[0]
|
|
|
|
else:
|
|
|
|
vcodec, acodec = va_codecs[:2]
|
|
|
|
f.update({
|
|
|
|
'acodec': acodec,
|
|
|
|
'vcodec': vcodec,
|
|
|
|
})
|
2015-02-17 11:55:53 -06:00
|
|
|
if last_media is not None:
|
|
|
|
f['m3u8_media'] = last_media
|
|
|
|
last_media = None
|
2014-08-26 05:51:13 -05:00
|
|
|
formats.append(f)
|
|
|
|
last_info = {}
|
|
|
|
self._sort_formats(formats)
|
|
|
|
return formats
|
|
|
|
|
2015-08-01 14:13:21 -05:00
|
|
|
@staticmethod
|
|
|
|
def _xpath_ns(path, namespace=None):
|
|
|
|
if not namespace:
|
|
|
|
return path
|
|
|
|
out = []
|
|
|
|
for c in path.split('/'):
|
|
|
|
if not c or c == '.':
|
|
|
|
out.append(c)
|
|
|
|
else:
|
|
|
|
out.append('{%s}%s' % (namespace, c))
|
|
|
|
return '/'.join(out)
|
|
|
|
|
|
|
|
def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None):
|
|
|
|
smil = self._download_smil(smil_url, video_id, fatal=fatal)
|
|
|
|
|
2015-02-02 16:38:35 -06:00
|
|
|
if smil is False:
|
|
|
|
assert not fatal
|
|
|
|
return []
|
2014-12-09 10:28:28 -06:00
|
|
|
|
2015-08-01 14:31:17 -05:00
|
|
|
namespace = self._parse_smil_namespace(smil)
|
2015-08-01 14:13:21 -05:00
|
|
|
|
|
|
|
return self._parse_smil_formats(
|
|
|
|
smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params)
|
|
|
|
|
|
|
|
def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None):
|
|
|
|
smil = self._download_smil(smil_url, video_id, fatal=fatal)
|
|
|
|
if smil is False:
|
|
|
|
return {}
|
|
|
|
return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params)
|
|
|
|
|
|
|
|
def _download_smil(self, smil_url, video_id, fatal=True):
|
|
|
|
return self._download_xml(
|
|
|
|
smil_url, video_id, 'Downloading SMIL file',
|
|
|
|
'Unable to download SMIL file', fatal=fatal)
|
|
|
|
|
|
|
|
def _parse_smil(self, smil, smil_url, video_id, f4m_params=None):
|
2015-08-01 14:31:17 -05:00
|
|
|
namespace = self._parse_smil_namespace(smil)
|
2015-08-01 14:13:21 -05:00
|
|
|
|
|
|
|
formats = self._parse_smil_formats(
|
|
|
|
smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params)
|
|
|
|
subtitles = self._parse_smil_subtitles(smil, namespace=namespace)
|
|
|
|
|
|
|
|
video_id = os.path.splitext(url_basename(smil_url))[0]
|
|
|
|
title = None
|
|
|
|
description = None
|
2015-10-01 11:18:59 -05:00
|
|
|
upload_date = None
|
2015-08-01 14:13:21 -05:00
|
|
|
for meta in smil.findall(self._xpath_ns('./head/meta', namespace)):
|
|
|
|
name = meta.attrib.get('name')
|
|
|
|
content = meta.attrib.get('content')
|
|
|
|
if not name or not content:
|
|
|
|
continue
|
|
|
|
if not title and name == 'title':
|
|
|
|
title = content
|
|
|
|
elif not description and name in ('description', 'abstract'):
|
|
|
|
description = content
|
2015-10-01 11:18:59 -05:00
|
|
|
elif not upload_date and name == 'date':
|
|
|
|
upload_date = unified_strdate(content)
|
2015-08-01 14:13:21 -05:00
|
|
|
|
2015-10-01 11:08:16 -05:00
|
|
|
thumbnails = [{
|
|
|
|
'id': image.get('type'),
|
|
|
|
'url': image.get('src'),
|
|
|
|
'width': int_or_none(image.get('width')),
|
|
|
|
'height': int_or_none(image.get('height')),
|
|
|
|
} for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')]
|
|
|
|
|
2015-08-01 14:13:21 -05:00
|
|
|
return {
|
|
|
|
'id': video_id,
|
|
|
|
'title': title or video_id,
|
|
|
|
'description': description,
|
2015-10-01 11:18:59 -05:00
|
|
|
'upload_date': upload_date,
|
2015-10-01 11:08:16 -05:00
|
|
|
'thumbnails': thumbnails,
|
2015-08-01 14:13:21 -05:00
|
|
|
'formats': formats,
|
|
|
|
'subtitles': subtitles,
|
|
|
|
}
|
|
|
|
|
2015-08-01 14:31:17 -05:00
|
|
|
def _parse_smil_namespace(self, smil):
|
|
|
|
return self._search_regex(
|
|
|
|
r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None)
|
|
|
|
|
2015-08-19 10:11:25 -05:00
|
|
|
def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
|
2015-08-01 14:13:21 -05:00
|
|
|
base = smil_url
|
|
|
|
for meta in smil.findall(self._xpath_ns('./head/meta', namespace)):
|
|
|
|
b = meta.get('base') or meta.get('httpBase')
|
|
|
|
if b:
|
|
|
|
base = b
|
|
|
|
break
|
2014-12-09 10:28:28 -06:00
|
|
|
|
|
|
|
formats = []
|
|
|
|
rtmp_count = 0
|
2015-08-01 14:13:21 -05:00
|
|
|
http_count = 0
|
2016-01-26 10:44:44 -06:00
|
|
|
m3u8_count = 0
|
2015-08-01 14:13:21 -05:00
|
|
|
|
2016-02-11 10:58:48 -06:00
|
|
|
srcs = []
|
2015-08-01 14:13:21 -05:00
|
|
|
videos = smil.findall(self._xpath_ns('.//video', namespace))
|
|
|
|
for video in videos:
|
|
|
|
src = video.get('src')
|
2016-02-11 10:58:48 -06:00
|
|
|
if not src or src in srcs:
|
2015-08-01 14:13:21 -05:00
|
|
|
continue
|
2016-02-11 10:58:48 -06:00
|
|
|
srcs.append(src)
|
2015-08-01 14:13:21 -05:00
|
|
|
|
2015-10-01 10:51:04 -05:00
|
|
|
bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
|
2015-08-01 14:13:21 -05:00
|
|
|
filesize = int_or_none(video.get('size') or video.get('fileSize'))
|
|
|
|
width = int_or_none(video.get('width'))
|
|
|
|
height = int_or_none(video.get('height'))
|
|
|
|
proto = video.get('proto')
|
|
|
|
ext = video.get('ext')
|
|
|
|
src_ext = determine_ext(src)
|
|
|
|
streamer = video.get('streamer') or base
|
|
|
|
|
|
|
|
if proto == 'rtmp' or streamer.startswith('rtmp'):
|
|
|
|
rtmp_count += 1
|
|
|
|
formats.append({
|
|
|
|
'url': streamer,
|
|
|
|
'play_path': src,
|
|
|
|
'ext': 'flv',
|
|
|
|
'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
|
|
|
|
'tbr': bitrate,
|
|
|
|
'filesize': filesize,
|
|
|
|
'width': width,
|
|
|
|
'height': height,
|
|
|
|
})
|
2015-08-19 10:11:25 -05:00
|
|
|
if transform_rtmp_url:
|
|
|
|
streamer, src = transform_rtmp_url(streamer, src)
|
|
|
|
formats[-1].update({
|
|
|
|
'url': streamer,
|
|
|
|
'play_path': src,
|
|
|
|
})
|
2015-08-01 14:13:21 -05:00
|
|
|
continue
|
|
|
|
|
|
|
|
src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src)
|
2016-02-12 10:38:48 -06:00
|
|
|
src_url = src_url.strip()
|
2015-08-01 14:13:21 -05:00
|
|
|
|
|
|
|
if proto == 'm3u8' or src_ext == 'm3u8':
|
2016-01-26 10:44:44 -06:00
|
|
|
m3u8_formats = self._extract_m3u8_formats(
|
|
|
|
src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)
|
|
|
|
if len(m3u8_formats) == 1:
|
|
|
|
m3u8_count += 1
|
|
|
|
m3u8_formats[0].update({
|
|
|
|
'format_id': 'hls-%d' % (m3u8_count if bitrate is None else bitrate),
|
|
|
|
'tbr': bitrate,
|
|
|
|
'width': width,
|
|
|
|
'height': height,
|
|
|
|
})
|
|
|
|
formats.extend(m3u8_formats)
|
2015-08-01 14:13:21 -05:00
|
|
|
continue
|
|
|
|
|
|
|
|
if src_ext == 'f4m':
|
|
|
|
f4m_url = src_url
|
|
|
|
if not f4m_params:
|
|
|
|
f4m_params = {
|
|
|
|
'hdcore': '3.2.0',
|
|
|
|
'plugin': 'flowplayer-3.2.0.1',
|
|
|
|
}
|
|
|
|
f4m_url += '&' if '?' in f4m_url else '?'
|
2015-08-01 14:20:49 -05:00
|
|
|
f4m_url += compat_urllib_parse.urlencode(f4m_params)
|
2015-12-28 12:58:24 -06:00
|
|
|
formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False))
|
2015-08-01 14:13:21 -05:00
|
|
|
continue
|
|
|
|
|
2015-10-01 11:54:54 -05:00
|
|
|
if src_url.startswith('http') and self._is_valid_url(src, video_id):
|
2015-08-01 14:13:21 -05:00
|
|
|
http_count += 1
|
|
|
|
formats.append({
|
|
|
|
'url': src_url,
|
|
|
|
'ext': ext or src_ext or 'flv',
|
|
|
|
'format_id': 'http-%d' % (bitrate or http_count),
|
|
|
|
'tbr': bitrate,
|
|
|
|
'filesize': filesize,
|
|
|
|
'width': width,
|
|
|
|
'height': height,
|
|
|
|
})
|
|
|
|
continue
|
2015-02-22 02:16:51 -06:00
|
|
|
|
2014-12-09 10:28:28 -06:00
|
|
|
self._sort_formats(formats)
|
|
|
|
|
|
|
|
return formats
|
|
|
|
|
2015-08-19 13:56:17 -05:00
|
|
|
def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
|
2016-02-09 10:15:41 -06:00
|
|
|
urls = []
|
2015-08-01 14:13:21 -05:00
|
|
|
subtitles = {}
|
|
|
|
for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))):
|
|
|
|
src = textstream.get('src')
|
2016-02-09 10:15:41 -06:00
|
|
|
if not src or src in urls:
|
2015-08-01 14:13:21 -05:00
|
|
|
continue
|
2016-02-09 10:15:41 -06:00
|
|
|
urls.append(src)
|
2016-02-20 15:02:03 -06:00
|
|
|
ext = textstream.get('ext') or determine_ext(src) or mimetype2ext(textstream.get('type'))
|
2015-08-20 10:18:58 -05:00
|
|
|
lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang
|
2015-08-01 14:13:21 -05:00
|
|
|
subtitles.setdefault(lang, []).append({
|
|
|
|
'url': src,
|
|
|
|
'ext': ext,
|
|
|
|
})
|
|
|
|
return subtitles
|
2015-02-22 02:16:51 -06:00
|
|
|
|
2015-08-09 08:41:55 -05:00
|
|
|
def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True):
|
|
|
|
xspf = self._download_xml(
|
2015-08-09 08:07:18 -05:00
|
|
|
playlist_url, playlist_id, 'Downloading xpsf playlist',
|
2015-08-09 08:41:55 -05:00
|
|
|
'Unable to download xspf manifest', fatal=fatal)
|
|
|
|
if xspf is False:
|
|
|
|
return []
|
|
|
|
return self._parse_xspf(xspf, playlist_id)
|
2015-08-09 08:07:18 -05:00
|
|
|
|
2015-08-09 08:41:55 -05:00
|
|
|
def _parse_xspf(self, playlist, playlist_id):
|
2015-08-09 08:07:18 -05:00
|
|
|
NS_MAP = {
|
|
|
|
'xspf': 'http://xspf.org/ns/0/',
|
|
|
|
's1': 'http://static.streamone.nl/player/ns/0',
|
|
|
|
}
|
|
|
|
|
|
|
|
entries = []
|
|
|
|
for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)):
|
|
|
|
title = xpath_text(
|
2015-08-09 08:18:50 -05:00
|
|
|
track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id)
|
2015-08-09 08:07:18 -05:00
|
|
|
description = xpath_text(
|
|
|
|
track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description')
|
|
|
|
thumbnail = xpath_text(
|
|
|
|
track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail')
|
|
|
|
duration = float_or_none(
|
|
|
|
xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000)
|
|
|
|
|
|
|
|
formats = [{
|
|
|
|
'url': location.text,
|
|
|
|
'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)),
|
|
|
|
'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))),
|
|
|
|
'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))),
|
|
|
|
} for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))]
|
|
|
|
self._sort_formats(formats)
|
|
|
|
|
|
|
|
entries.append({
|
|
|
|
'id': playlist_id,
|
|
|
|
'title': title,
|
|
|
|
'description': description,
|
|
|
|
'thumbnail': thumbnail,
|
|
|
|
'duration': duration,
|
|
|
|
'formats': formats,
|
|
|
|
})
|
|
|
|
return entries
|
|
|
|
|
2016-02-02 11:07:07 -06:00
|
|
|
def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}):
|
|
|
|
res = self._download_webpage_handle(
|
|
|
|
mpd_url, video_id,
|
|
|
|
note=note or 'Downloading MPD manifest',
|
|
|
|
errnote=errnote or 'Failed to download MPD manifest',
|
2016-01-30 08:52:23 -06:00
|
|
|
fatal=fatal)
|
2016-02-02 11:07:07 -06:00
|
|
|
if res is False:
|
2016-01-30 08:52:23 -06:00
|
|
|
return []
|
2016-02-02 11:07:07 -06:00
|
|
|
mpd, urlh = res
|
|
|
|
mpd_base_url = re.match(r'https?://.+/', urlh.geturl()).group()
|
|
|
|
|
2016-02-06 07:03:48 -06:00
|
|
|
return self._parse_mpd_formats(
|
2016-02-02 11:07:07 -06:00
|
|
|
compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict)
|
2016-01-30 08:52:23 -06:00
|
|
|
|
2016-02-06 07:03:48 -06:00
|
|
|
def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}):
|
2016-02-02 11:07:07 -06:00
|
|
|
if mpd_doc.get('type') == 'dynamic':
|
|
|
|
return []
|
2016-01-30 08:52:23 -06:00
|
|
|
|
2016-02-06 07:03:48 -06:00
|
|
|
namespace = self._search_regex(r'(?i)^{([^}]+)?}MPD$', mpd_doc.tag, 'namespace', default=None)
|
2016-02-02 15:02:08 -06:00
|
|
|
|
|
|
|
def _add_ns(path):
|
|
|
|
return self._xpath_ns(path, namespace)
|
|
|
|
|
2016-02-03 11:44:43 -06:00
|
|
|
def is_drm_protected(element):
|
|
|
|
return element.find(_add_ns('ContentProtection')) is not None
|
|
|
|
|
2016-02-02 11:07:07 -06:00
|
|
|
def extract_multisegment_info(element, ms_parent_info):
|
|
|
|
ms_info = ms_parent_info.copy()
|
2016-02-02 15:02:08 -06:00
|
|
|
segment_list = element.find(_add_ns('SegmentList'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if segment_list is not None:
|
2016-02-02 15:02:08 -06:00
|
|
|
segment_urls_e = segment_list.findall(_add_ns('SegmentURL'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if segment_urls_e:
|
|
|
|
ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e]
|
2016-02-02 15:02:08 -06:00
|
|
|
initialization = segment_list.find(_add_ns('Initialization'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if initialization is not None:
|
|
|
|
ms_info['initialization_url'] = initialization.attrib['sourceURL']
|
|
|
|
else:
|
2016-02-02 15:02:08 -06:00
|
|
|
segment_template = element.find(_add_ns('SegmentTemplate'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if segment_template is not None:
|
|
|
|
start_number = segment_template.get('startNumber')
|
|
|
|
if start_number:
|
|
|
|
ms_info['start_number'] = int(start_number)
|
2016-02-02 15:02:08 -06:00
|
|
|
segment_timeline = segment_template.find(_add_ns('SegmentTimeline'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if segment_timeline is not None:
|
2016-02-02 15:02:08 -06:00
|
|
|
s_e = segment_timeline.findall(_add_ns('S'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if s_e:
|
|
|
|
ms_info['total_number'] = 0
|
|
|
|
for s in s_e:
|
|
|
|
ms_info['total_number'] += 1 + int(s.get('r', '0'))
|
|
|
|
else:
|
|
|
|
timescale = segment_template.get('timescale')
|
|
|
|
if timescale:
|
|
|
|
ms_info['timescale'] = int(timescale)
|
|
|
|
segment_duration = segment_template.get('duration')
|
|
|
|
if segment_duration:
|
|
|
|
ms_info['segment_duration'] = int(segment_duration)
|
|
|
|
media_template = segment_template.get('media')
|
|
|
|
if media_template:
|
|
|
|
ms_info['media_template'] = media_template
|
|
|
|
initialization = segment_template.get('initialization')
|
|
|
|
if initialization:
|
|
|
|
ms_info['initialization_url'] = initialization
|
|
|
|
else:
|
2016-02-02 15:02:08 -06:00
|
|
|
initialization = segment_template.find(_add_ns('Initialization'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if initialization is not None:
|
|
|
|
ms_info['initialization_url'] = initialization.attrib['sourceURL']
|
|
|
|
return ms_info
|
2016-01-30 07:27:43 -06:00
|
|
|
|
2016-02-02 11:07:07 -06:00
|
|
|
mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration'))
|
2016-01-30 07:05:55 -06:00
|
|
|
formats = []
|
2016-02-02 15:02:08 -06:00
|
|
|
for period in mpd_doc.findall(_add_ns('Period')):
|
2016-02-02 11:07:07 -06:00
|
|
|
period_duration = parse_duration(period.get('duration')) or mpd_duration
|
|
|
|
period_ms_info = extract_multisegment_info(period, {
|
|
|
|
'start_number': 1,
|
|
|
|
'timescale': 1,
|
|
|
|
})
|
2016-02-02 15:02:08 -06:00
|
|
|
for adaptation_set in period.findall(_add_ns('AdaptationSet')):
|
2016-02-03 11:44:43 -06:00
|
|
|
if is_drm_protected(adaptation_set):
|
|
|
|
continue
|
2016-02-02 11:07:07 -06:00
|
|
|
adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info)
|
2016-02-02 15:02:08 -06:00
|
|
|
for representation in adaptation_set.findall(_add_ns('Representation')):
|
2016-02-03 11:44:43 -06:00
|
|
|
if is_drm_protected(representation):
|
|
|
|
continue
|
2016-02-02 11:07:07 -06:00
|
|
|
representation_attrib = adaptation_set.attrib.copy()
|
|
|
|
representation_attrib.update(representation.attrib)
|
|
|
|
mime_type = representation_attrib.get('mimeType')
|
|
|
|
content_type = mime_type.split('/')[0] if mime_type else representation_attrib.get('contentType')
|
|
|
|
if content_type == 'text':
|
|
|
|
# TODO implement WebVTT downloading
|
|
|
|
pass
|
|
|
|
elif content_type == 'video' or content_type == 'audio':
|
|
|
|
base_url = ''
|
|
|
|
for element in (representation, adaptation_set, period, mpd_doc):
|
2016-02-02 15:02:08 -06:00
|
|
|
base_url_e = element.find(_add_ns('BaseURL'))
|
2016-02-02 11:07:07 -06:00
|
|
|
if base_url_e is not None:
|
|
|
|
base_url = base_url_e.text + base_url
|
|
|
|
if re.match(r'^https?://', base_url):
|
|
|
|
break
|
2016-02-12 12:13:56 -06:00
|
|
|
if mpd_base_url and not re.match(r'^https?://', base_url):
|
|
|
|
if not mpd_base_url.endswith('/') and not base_url.startswith('/'):
|
|
|
|
mpd_base_url += '/'
|
2016-02-02 11:07:07 -06:00
|
|
|
base_url = mpd_base_url + base_url
|
|
|
|
representation_id = representation_attrib.get('id')
|
2016-02-03 06:24:07 -06:00
|
|
|
lang = representation_attrib.get('lang')
|
2016-02-09 13:05:39 -06:00
|
|
|
url_el = representation.find(_add_ns('BaseURL'))
|
|
|
|
filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
|
2016-02-02 11:07:07 -06:00
|
|
|
f = {
|
2016-02-11 03:33:26 -06:00
|
|
|
'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
|
2016-02-02 11:07:07 -06:00
|
|
|
'url': base_url,
|
|
|
|
'width': int_or_none(representation_attrib.get('width')),
|
|
|
|
'height': int_or_none(representation_attrib.get('height')),
|
|
|
|
'tbr': int_or_none(representation_attrib.get('bandwidth'), 1000),
|
|
|
|
'asr': int_or_none(representation_attrib.get('audioSamplingRate')),
|
|
|
|
'fps': int_or_none(representation_attrib.get('frameRate')),
|
|
|
|
'vcodec': 'none' if content_type == 'audio' else representation_attrib.get('codecs'),
|
|
|
|
'acodec': 'none' if content_type == 'video' else representation_attrib.get('codecs'),
|
2016-02-03 06:24:07 -06:00
|
|
|
'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None,
|
2016-02-02 11:07:07 -06:00
|
|
|
'format_note': 'DASH %s' % content_type,
|
2016-02-09 13:05:39 -06:00
|
|
|
'filesize': filesize,
|
2016-02-02 11:07:07 -06:00
|
|
|
}
|
|
|
|
representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
|
|
|
|
if 'segment_urls' not in representation_ms_info and 'media_template' in representation_ms_info:
|
|
|
|
if 'total_number' not in representation_ms_info and 'segment_duration':
|
2016-02-06 07:27:04 -06:00
|
|
|
segment_duration = float(representation_ms_info['segment_duration']) / float(representation_ms_info['timescale'])
|
|
|
|
representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
|
2016-02-02 11:07:07 -06:00
|
|
|
media_template = representation_ms_info['media_template']
|
|
|
|
media_template = media_template.replace('$RepresentationID$', representation_id)
|
2016-02-03 16:54:34 -06:00
|
|
|
media_template = re.sub(r'\$(Number|Bandwidth)(?:%(0\d+)d)?\$', r'%(\1)\2d', media_template)
|
2016-02-02 11:07:07 -06:00
|
|
|
media_template.replace('$$', '$')
|
2016-02-03 16:54:34 -06:00
|
|
|
representation_ms_info['segment_urls'] = [media_template % {'Number': segment_number, 'Bandwidth': representation_attrib.get('bandwidth')} for segment_number in range(representation_ms_info['start_number'], representation_ms_info['total_number'] + representation_ms_info['start_number'])]
|
2016-02-02 11:07:07 -06:00
|
|
|
if 'segment_urls' in representation_ms_info:
|
|
|
|
f.update({
|
|
|
|
'segment_urls': representation_ms_info['segment_urls'],
|
|
|
|
'protocol': 'http_dash_segments',
|
2016-01-30 07:42:27 -06:00
|
|
|
})
|
2016-02-02 11:07:07 -06:00
|
|
|
if 'initialization_url' in representation_ms_info:
|
|
|
|
initialization_url = representation_ms_info['initialization_url'].replace('$RepresentationID$', representation_id)
|
|
|
|
f.update({
|
|
|
|
'initialization_url': initialization_url,
|
|
|
|
})
|
|
|
|
if not f.get('url'):
|
|
|
|
f['url'] = initialization_url
|
|
|
|
try:
|
|
|
|
existing_format = next(
|
|
|
|
fo for fo in formats
|
|
|
|
if fo['format_id'] == representation_id)
|
|
|
|
except StopIteration:
|
|
|
|
full_info = formats_dict.get(representation_id, {}).copy()
|
|
|
|
full_info.update(f)
|
|
|
|
formats.append(full_info)
|
|
|
|
else:
|
|
|
|
existing_format.update(f)
|
2016-01-30 07:05:55 -06:00
|
|
|
else:
|
2016-02-02 11:07:07 -06:00
|
|
|
self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
|
2016-02-05 23:52:48 -06:00
|
|
|
self._sort_formats(formats)
|
2016-01-30 07:05:55 -06:00
|
|
|
return formats
|
|
|
|
|
2014-09-28 01:53:52 -05:00
|
|
|
def _live_title(self, name):
|
|
|
|
""" Generate the title for a live video """
|
|
|
|
now = datetime.datetime.now()
|
2016-02-14 03:37:17 -06:00
|
|
|
now_str = now.strftime('%Y-%m-%d %H:%M')
|
2014-09-28 01:53:52 -05:00
|
|
|
return name + ' ' + now_str
|
|
|
|
|
2014-09-28 03:34:55 -05:00
|
|
|
def _int(self, v, name, fatal=False, **kwargs):
|
|
|
|
res = int_or_none(v, **kwargs)
|
|
|
|
if 'get_attr' in kwargs:
|
|
|
|
print(getattr(v, kwargs['get_attr']))
|
|
|
|
if res is None:
|
|
|
|
msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
|
|
|
|
if fatal:
|
|
|
|
raise ExtractorError(msg)
|
|
|
|
else:
|
|
|
|
self._downloader.report_warning(msg)
|
|
|
|
return res
|
|
|
|
|
|
|
|
def _float(self, v, name, fatal=False, **kwargs):
|
|
|
|
res = float_or_none(v, **kwargs)
|
|
|
|
if res is None:
|
|
|
|
msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
|
|
|
|
if fatal:
|
|
|
|
raise ExtractorError(msg)
|
|
|
|
else:
|
|
|
|
self._downloader.report_warning(msg)
|
|
|
|
return res
|
|
|
|
|
2014-11-29 17:03:59 -06:00
|
|
|
def _set_cookie(self, domain, name, value, expire_time=None):
|
2014-12-04 01:27:40 -06:00
|
|
|
cookie = compat_cookiejar.Cookie(
|
|
|
|
0, name, value, None, None, domain, None,
|
2014-11-29 17:03:59 -06:00
|
|
|
None, '/', True, False, expire_time, '', None, None, None)
|
|
|
|
self._downloader.cookiejar.set_cookie(cookie)
|
|
|
|
|
2015-07-29 17:20:37 -05:00
|
|
|
def _get_cookies(self, url):
|
|
|
|
""" Return a compat_cookies.SimpleCookie with the cookies for the url """
|
2015-11-21 10:18:17 -06:00
|
|
|
req = sanitized_Request(url)
|
2015-07-29 17:20:37 -05:00
|
|
|
self._downloader.cookiejar.add_cookie_header(req)
|
|
|
|
return compat_cookies.SimpleCookie(req.get_header('Cookie'))
|
|
|
|
|
2015-01-07 00:20:20 -06:00
|
|
|
def get_testcases(self, include_onlymatching=False):
|
|
|
|
t = getattr(self, '_TEST', None)
|
|
|
|
if t:
|
|
|
|
assert not hasattr(self, '_TESTS'), \
|
|
|
|
'%s has _TEST and _TESTS' % type(self).__name__
|
|
|
|
tests = [t]
|
|
|
|
else:
|
|
|
|
tests = getattr(self, '_TESTS', [])
|
|
|
|
for t in tests:
|
|
|
|
if not include_onlymatching and t.get('only_matching', False):
|
|
|
|
continue
|
|
|
|
t['name'] = type(self).__name__[:-len('IE')]
|
|
|
|
yield t
|
|
|
|
|
|
|
|
def is_suitable(self, age_limit):
|
|
|
|
""" Test whether the extractor is generally suitable for the given
|
|
|
|
age limit (i.e. pornographic sites are not, all others usually are) """
|
|
|
|
|
|
|
|
any_restricted = False
|
|
|
|
for tc in self.get_testcases(include_onlymatching=False):
|
|
|
|
if 'playlist' in tc:
|
|
|
|
tc = tc['playlist'][0]
|
|
|
|
is_restricted = age_restricted(
|
|
|
|
tc.get('info_dict', {}).get('age_limit'), age_limit)
|
|
|
|
if not is_restricted:
|
|
|
|
return True
|
|
|
|
any_restricted = any_restricted or is_restricted
|
|
|
|
return not any_restricted
|
|
|
|
|
2015-02-15 11:03:41 -06:00
|
|
|
def extract_subtitles(self, *args, **kwargs):
|
2015-02-17 15:16:29 -06:00
|
|
|
if (self._downloader.params.get('writesubtitles', False) or
|
|
|
|
self._downloader.params.get('listsubtitles')):
|
|
|
|
return self._get_subtitles(*args, **kwargs)
|
|
|
|
return {}
|
2015-02-15 11:03:41 -06:00
|
|
|
|
|
|
|
def _get_subtitles(self, *args, **kwargs):
|
2016-02-14 03:37:17 -06:00
|
|
|
raise NotImplementedError('This method must be implemented by subclasses')
|
2015-02-15 11:03:41 -06:00
|
|
|
|
2015-08-20 12:37:07 -05:00
|
|
|
@staticmethod
|
|
|
|
def _merge_subtitle_items(subtitle_list1, subtitle_list2):
|
|
|
|
""" Merge subtitle items for one language. Items with duplicated URLs
|
|
|
|
will be dropped. """
|
|
|
|
list1_urls = set([item['url'] for item in subtitle_list1])
|
|
|
|
ret = list(subtitle_list1)
|
|
|
|
ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls])
|
|
|
|
return ret
|
|
|
|
|
|
|
|
@classmethod
|
2015-08-21 04:35:51 -05:00
|
|
|
def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2):
|
2015-08-20 12:37:07 -05:00
|
|
|
""" Merge two subtitle dictionaries, language by language. """
|
|
|
|
ret = dict(subtitle_dict1)
|
|
|
|
for lang in subtitle_dict2:
|
2015-08-21 04:35:51 -05:00
|
|
|
ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang])
|
2015-08-20 12:37:07 -05:00
|
|
|
return ret
|
|
|
|
|
2015-02-16 14:44:17 -06:00
|
|
|
def extract_automatic_captions(self, *args, **kwargs):
|
2015-02-17 15:16:29 -06:00
|
|
|
if (self._downloader.params.get('writeautomaticsub', False) or
|
|
|
|
self._downloader.params.get('listsubtitles')):
|
|
|
|
return self._get_automatic_captions(*args, **kwargs)
|
|
|
|
return {}
|
2015-02-16 14:44:17 -06:00
|
|
|
|
|
|
|
def _get_automatic_captions(self, *args, **kwargs):
|
2016-02-14 03:37:17 -06:00
|
|
|
raise NotImplementedError('This method must be implemented by subclasses')
|
2015-02-16 14:44:17 -06:00
|
|
|
|
2016-02-29 13:01:33 -06:00
|
|
|
def mark_watched(self, *args, **kwargs):
|
|
|
|
if (self._downloader.params.get('mark_watched', False) and
|
|
|
|
(self._get_login_info()[0] is not None or
|
|
|
|
self._downloader.params.get('cookiefile') is not None)):
|
|
|
|
self._mark_watched(*args, **kwargs)
|
|
|
|
|
|
|
|
def _mark_watched(self, *args, **kwargs):
|
|
|
|
raise NotImplementedError('This method must be implemented by subclasses')
|
|
|
|
|
2013-10-05 23:06:30 -05:00
|
|
|
|
2013-06-23 12:57:38 -05:00
|
|
|
class SearchInfoExtractor(InfoExtractor):
|
|
|
|
"""
|
|
|
|
Base class for paged search queries extractors.
|
2015-07-23 12:37:45 -05:00
|
|
|
They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query}
|
2013-06-23 12:57:38 -05:00
|
|
|
Instances should define _SEARCH_KEY and _MAX_RESULTS.
|
|
|
|
"""
|
|
|
|
|
|
|
|
@classmethod
|
|
|
|
def _make_valid_url(cls):
|
|
|
|
return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
|
|
|
|
|
|
|
|
@classmethod
|
|
|
|
def suitable(cls, url):
|
|
|
|
return re.match(cls._make_valid_url(), url) is not None
|
|
|
|
|
|
|
|
def _real_extract(self, query):
|
|
|
|
mobj = re.match(self._make_valid_url(), query)
|
|
|
|
if mobj is None:
|
2014-08-27 18:04:43 -05:00
|
|
|
raise ExtractorError('Invalid search query "%s"' % query)
|
2013-06-23 12:57:38 -05:00
|
|
|
|
|
|
|
prefix = mobj.group('prefix')
|
|
|
|
query = mobj.group('query')
|
|
|
|
if prefix == '':
|
|
|
|
return self._get_n_results(query, 1)
|
|
|
|
elif prefix == 'all':
|
|
|
|
return self._get_n_results(query, self._MAX_RESULTS)
|
|
|
|
else:
|
|
|
|
n = int(prefix)
|
|
|
|
if n <= 0:
|
2014-08-27 18:04:43 -05:00
|
|
|
raise ExtractorError('invalid download number %s for query "%s"' % (n, query))
|
2013-06-23 12:57:38 -05:00
|
|
|
elif n > self._MAX_RESULTS:
|
2014-08-27 18:04:43 -05:00
|
|
|
self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
|
2013-06-23 12:57:38 -05:00
|
|
|
n = self._MAX_RESULTS
|
|
|
|
return self._get_n_results(query, n)
|
|
|
|
|
|
|
|
def _get_n_results(self, query, n):
|
|
|
|
"""Get a specified number of results for a query"""
|
2016-02-14 03:37:17 -06:00
|
|
|
raise NotImplementedError('This method must be implemented by subclasses')
|
2013-07-01 11:52:19 -05:00
|
|
|
|
|
|
|
@property
|
|
|
|
def SEARCH_KEY(self):
|
|
|
|
return self._SEARCH_KEY
|