From 07cf18b9c56bf8ce08404b084672f33e860a805e Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Wed, 29 Nov 2017 14:21:38 +0100 Subject: [PATCH] [br] add support for BR Mediathek videos(fixes #14560)(fixes #14788) --- youtube_dl/extractor/br.py | 146 ++++++++++++++++++++++++++++- youtube_dl/extractor/extractors.py | 5 +- 2 files changed, 147 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/br.py b/youtube_dl/extractor/br.py index 2c32b6ae2..9bde7f2d8 100644 --- a/youtube_dl/extractor/br.py +++ b/youtube_dl/extractor/br.py @@ -1,20 +1,23 @@ # coding: utf-8 from __future__ import unicode_literals +import json import re from .common import InfoExtractor from ..utils import ( + determine_ext, ExtractorError, int_or_none, parse_duration, + parse_iso8601, xpath_element, xpath_text, ) class BRIE(InfoExtractor): - IE_DESC = 'Bayerischer Rundfunk Mediathek' + IE_DESC = 'Bayerischer Rundfunk' _VALID_URL = r'(?Phttps?://(?:www\.)?br(?:-klassik)?\.de)/(?:[a-z0-9\-_]+/)+(?P[a-z0-9\-_]+)\.html' _TESTS = [ @@ -123,10 +126,10 @@ class BRIE(InfoExtractor): for asset in assets.findall('asset'): format_url = xpath_text(asset, ['downloadUrl', 'url']) asset_type = asset.get('type') - if asset_type == 'HDS': + if asset_type.startswith('HDS'): formats.extend(self._extract_f4m_formats( format_url + '?hdcore=3.2.0', media_id, f4m_id='hds', fatal=False)) - elif asset_type == 'HLS': + elif asset_type.startswith('HLS'): formats.extend(self._extract_m3u8_formats( format_url, media_id, 'mp4', 'm3u8_native', m3u8_id='hds', fatal=False)) else: @@ -169,3 +172,140 @@ class BRIE(InfoExtractor): } for variant in variants.findall('variant') if xpath_text(variant, 'url')] thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True) return thumbnails + + +class BRMediathekIE(InfoExtractor): + IE_DESC = 'Bayerischer Rundfunk Mediathek' + _VALID_URL = r'https?://(?:www\.)?br\.de/mediathek/video/[^/?&#]*?-(?Pav:[0-9a-f]{24})' + + _TESTS = [{ + 'url': 'https://www.br.de/mediathek/video/gesundheit-die-sendung-vom-28112017-av:5a1e6a6e8fce6d001871cc8e', + 'md5': 'fdc3d485835966d1622587d08ba632ec', + 'info_dict': { + 'id': 'av:5a1e6a6e8fce6d001871cc8e', + 'ext': 'mp4', + 'title': 'Die Sendung vom 28.11.2017', + 'description': 'md5:6000cdca5912ab2277e5b7339f201ccc', + 'timestamp': 1511942766, + 'upload_date': '20171129', + } + }] + + def _real_extract(self, url): + clip_id = self._match_id(url) + + clip = self._download_json( + 'https://proxy-base.master.mango.express/graphql', + clip_id, data=json.dumps({ + "query": """{ + viewer { + clip(id: "%s") { + title + description + duration + createdAt + ageRestriction + videoFiles { + edges { + node { + publicLocation + fileSize + videoProfile { + width + height + bitrate + encoding + } + } + } + } + captionFiles { + edges { + node { + publicLocation + } + } + } + teaserImages { + edges { + node { + imageFiles { + edges { + node { + publicLocation + width + height + } + } + } + } + } + } + } + } +}""" % clip_id}).encode(), headers={ + 'Content-Type': 'application/json', + })['data']['viewer']['clip'] + title = clip['title'] + + formats = [] + for edge in clip.get('videoFiles', {}).get('edges', []): + node = edge.get('node', {}) + n_url = node.get('publicLocation') + if not n_url: + continue + ext = determine_ext(n_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + n_url, clip_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + else: + video_profile = node.get('videoProfile', {}) + tbr = int_or_none(video_profile.get('bitrate')) + format_id = 'http' + if tbr: + format_id += '-%d' % tbr + formats.append({ + 'format_id': format_id, + 'url': n_url, + 'width': int_or_none(video_profile.get('width')), + 'height': int_or_none(video_profile.get('height')), + 'tbr': tbr, + 'filesize': int_or_none(node.get('fileSize')), + }) + self._sort_formats(formats) + + subtitles = {} + for edge in clip.get('captionFiles', {}).get('edges', []): + node = edge.get('node', {}) + n_url = node.get('publicLocation') + if not n_url: + continue + subtitles.setdefault('de', []).append({ + 'url': n_url, + }) + + thumbnails = [] + for edge in clip.get('teaserImages', {}).get('edges', []): + for image_edge in edge.get('node', {}).get('imageFiles', {}).get('edges', []): + node = image_edge.get('node', {}) + n_url = node.get('publicLocation') + if not n_url: + continue + thumbnails.append({ + 'url': n_url, + 'width': int_or_none(node.get('width')), + 'height': int_or_none(node.get('height')), + }) + + return { + 'id': clip_id, + 'title': title, + 'description': clip.get('description'), + 'duration': int_or_none(clip.get('duration')), + 'timestamp': parse_iso8601(clip.get('createdAt')), + 'age_limit': int_or_none(clip.get('ageRestriction')), + 'formats': formats, + 'subtitles': subtitles, + 'thumbnails': thumbnails, + } diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 6db1297f1..2cc3bc463 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -127,7 +127,10 @@ from .bloomberg import BloombergIE from .bokecc import BokeCCIE from .bostonglobe import BostonGlobeIE from .bpb import BpbIE -from .br import BRIE +from .br import ( + BRIE, + BRMediathekIE, +) from .bravotv import BravoTVIE from .breakcom import BreakIE from .brightcove import (