From 117589dfa2ac32566dc705ee7e712139105e6dd8 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Mon, 2 Oct 2017 04:14:36 +0700 Subject: [PATCH] [canvas] Generalize mediazone.vrt.be extractor and rework canvas and een --- youtube_dl/extractor/canvas.py | 142 +++++++++++++++++++---------- youtube_dl/extractor/extractors.py | 5 +- 2 files changed, 99 insertions(+), 48 deletions(-) diff --git a/youtube_dl/extractor/canvas.py b/youtube_dl/extractor/canvas.py index aada02917..6899f8431 100644 --- a/youtube_dl/extractor/canvas.py +++ b/youtube_dl/extractor/canvas.py @@ -3,24 +3,104 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import float_or_none +from ..utils import ( + float_or_none, + strip_or_none, +) class CanvasIE(InfoExtractor): + _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?Pcanvas|een|ketnet)/assets/(?Pm[dz]-ast-[^/?#&]+)' + _TESTS = [{ + 'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', + 'md5': '90139b746a0a9bd7bb631283f6e2a64e', + 'info_dict': { + 'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', + 'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', + 'ext': 'flv', + 'title': 'Nachtwacht: De Greystook', + 'description': 'md5:1db3f5dc4c7109c821261e7512975be7', + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 1468.03, + }, + 'expected_warnings': ['is not a supported codec', 'Unknown MIME type'], + }, { + 'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + site_id, video_id = mobj.group('site_id'), mobj.group('id') + + data = self._download_json( + 'https://mediazone.vrt.be/api/v1/%s/assets/%s' + % (site_id, video_id), video_id) + + title = data['title'] + description = data.get('description') + + formats = [] + for target in data['targetUrls']: + format_url, format_type = target.get('url'), target.get('type') + if not format_url or not format_type: + continue + if format_type == 'HLS': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id=format_type, fatal=False)) + elif format_type == 'HDS': + formats.extend(self._extract_f4m_formats( + format_url, video_id, f4m_id=format_type, fatal=False)) + elif format_type == 'MPEG_DASH': + formats.extend(self._extract_mpd_formats( + format_url, video_id, mpd_id=format_type, fatal=False)) + elif format_type == 'HSS': + formats.extend(self._extract_ism_formats( + format_url, video_id, ism_id='mss', fatal=False)) + else: + formats.append({ + 'format_id': format_type, + 'url': format_url, + }) + self._sort_formats(formats) + + subtitles = {} + subtitle_urls = data.get('subtitleUrls') + if isinstance(subtitle_urls, list): + for subtitle in subtitle_urls: + subtitle_url = subtitle.get('url') + if subtitle_url and subtitle.get('type') == 'CLOSED': + subtitles.setdefault('nl', []).append({'url': subtitle_url}) + + return { + 'id': video_id, + 'display_id': video_id, + 'title': title, + 'description': description, + 'formats': formats, + 'duration': float_or_none(data.get('duration'), 1000), + 'thumbnail': data.get('posterImageUrl'), + 'subtitles': subtitles, + } + + +class CanvasEenIE(InfoExtractor): IE_DESC = 'canvas.be and een.be' _VALID_URL = r'https?://(?:www\.)?(?Pcanvas|een)\.be/(?:[^/]+/)*(?P[^/?#&]+)' _TESTS = [{ 'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week', - 'md5': 'ea838375a547ac787d4064d8c7860a6c', + 'md5': 'ed66976748d12350b118455979cca293', 'info_dict': { 'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e', 'display_id': 'de-afspraak-veilt-voor-de-warmste-week', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'De afspraak veilt voor de Warmste Week', 'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6', 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 49.02, - } + }, + 'expected_warnings': ['is not a supported codec'], }, { # with subtitles 'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167', @@ -40,7 +120,8 @@ class CanvasIE(InfoExtractor): }, 'params': { 'skip_download': True, - } + }, + 'skip': 'Pagina niet gevonden', }, { 'url': 'https://www.een.be/sorry-voor-alles/herbekijk-sorry-voor-alles', 'info_dict': { @@ -54,7 +135,8 @@ class CanvasIE(InfoExtractor): }, 'params': { 'skip_download': True, - } + }, + 'skip': 'Episode no longer available', }, { 'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend', 'only_matching': True, @@ -66,55 +148,21 @@ class CanvasIE(InfoExtractor): webpage = self._download_webpage(url, display_id) - title = (self._search_regex( + title = strip_or_none(self._search_regex( r']+class="video__body__header__title"[^>]*>(.+?)', webpage, 'title', default=None) or self._og_search_title( - webpage)).strip() + webpage, default=None)) video_id = self._html_search_regex( - r'data-video=(["\'])(?P(?:(?!\1).)+)\1', webpage, 'video id', group='id') - - data = self._download_json( - 'https://mediazone.vrt.be/api/v1/%s/assets/%s' - % (site_id, video_id), display_id) - - formats = [] - for target in data['targetUrls']: - format_url, format_type = target.get('url'), target.get('type') - if not format_url or not format_type: - continue - if format_type == 'HLS': - formats.extend(self._extract_m3u8_formats( - format_url, display_id, entry_protocol='m3u8_native', - ext='mp4', preference=0, fatal=False, m3u8_id=format_type)) - elif format_type == 'HDS': - formats.extend(self._extract_f4m_formats( - format_url, display_id, f4m_id=format_type, fatal=False)) - elif format_type == 'MPEG_DASH': - formats.extend(self._extract_mpd_formats( - format_url, display_id, mpd_id=format_type, fatal=False)) - else: - formats.append({ - 'format_id': format_type, - 'url': format_url, - }) - self._sort_formats(formats) - - subtitles = {} - subtitle_urls = data.get('subtitleUrls') - if isinstance(subtitle_urls, list): - for subtitle in subtitle_urls: - subtitle_url = subtitle.get('url') - if subtitle_url and subtitle.get('type') == 'CLOSED': - subtitles.setdefault('nl', []).append({'url': subtitle_url}) + r'data-video=(["\'])(?P(?:(?!\1).)+)\1', webpage, 'video id', + group='id') return { + '_type': 'url_transparent', + 'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id), + 'ie_key': CanvasIE.ie_key(), 'id': video_id, 'display_id': display_id, 'title': title, 'description': self._og_search_description(webpage), - 'formats': formats, - 'duration': float_or_none(data.get('duration'), 1000), - 'thumbnail': data.get('posterImageUrl'), - 'subtitles': subtitles, } diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 4232a4fef..24e9acda6 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -150,7 +150,10 @@ from .camdemy import ( from .camwithher import CamWithHerIE from .canalplus import CanalplusIE from .canalc2 import Canalc2IE -from .canvas import CanvasIE +from .canvas import ( + CanvasIE, + CanvasEenIE, +) from .carambatv import ( CarambaTVIE, CarambaTVPageIE, -- 2.39.5