From c707d2067d649f23f03cf11995d1eeb665d51838 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Wed, 22 Aug 2018 23:18:07 +0700 Subject: [PATCH] [nova:embed] Add extractor (closes #17282) --- youtube_dl/extractor/extractors.py | 5 +- youtube_dl/extractor/nova.py | 138 ++++++++++++++++++++--------- 2 files changed, 101 insertions(+), 42 deletions(-) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 9fc1cfa70..a8b89bcde 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -741,7 +741,10 @@ from .nonktube import NonkTubeIE from .noovo import NoovoIE from .normalboots import NormalbootsIE from .nosvideo import NosVideoIE -from .nova import NovaIE +from .nova import ( + NovaEmbedIE, + NovaIE, +) from .novamov import ( AuroraVidIE, CloudTimeIE, diff --git a/youtube_dl/extractor/nova.py b/youtube_dl/extractor/nova.py index edf8b82c6..80186ec50 100644 --- a/youtube_dl/extractor/nova.py +++ b/youtube_dl/extractor/nova.py @@ -6,30 +6,90 @@ import re from .common import InfoExtractor from ..utils import ( clean_html, + int_or_none, js_to_json, + qualities, unified_strdate, url_or_none, ) +class NovaEmbedIE(InfoExtractor): + _VALID_URL = r'https?://media\.cms\.nova\.cz/embed/(?P[^/?#&]+)' + _TEST = { + 'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1', + 'md5': 'b3834f6de5401baabf31ed57456463f7', + 'info_dict': { + 'id': '8o0n0r', + 'ext': 'mp4', + 'title': '2180. díl', + 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 2578, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + bitrates = self._parse_json( + self._search_regex( + r'(?s)bitrates\s*=\s*({.+?})\s*;', webpage, 'formats'), + video_id, transform_source=js_to_json) + + QUALITIES = ('lq', 'mq', 'hq', 'hd') + quality_key = qualities(QUALITIES) + + formats = [] + for format_id, format_list in bitrates.items(): + if not isinstance(format_list, list): + continue + for format_url in format_list: + format_url = url_or_none(format_url) + if not format_url: + continue + f = { + 'url': format_url, + } + f_id = format_id + for quality in QUALITIES: + if '%s.mp4' % quality in format_url: + f_id += '-%s' % quality + f.update({ + 'quality': quality_key(quality), + 'format_note': quality.upper(), + }) + break + f['format_id'] = f_id + formats.append(f) + self._sort_formats(formats) + + title = self._og_search_title( + webpage, default=None) or self._search_regex( + (r'(?P[^<]+)', + r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage, + 'title', group='value') + thumbnail = self._og_search_thumbnail( + webpage, default=None) or self._search_regex( + r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage, + 'thumbnail', fatal=False, group='value') + duration = int_or_none(self._search_regex( + r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False)) + + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + } + + class NovaIE(InfoExtractor): IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz' _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)' _TESTS = [{ - 'url': 'http://tvnoviny.nova.cz/clanek/novinky/co-na-sebe-sportaci-praskli-vime-jestli-pujde-hrdlicka-na-materskou.html?utm_source=tvnoviny&utm_medium=cpfooter&utm_campaign=novaplus', - 'info_dict': { - 'id': '1608920', - 'display_id': 'co-na-sebe-sportaci-praskli-vime-jestli-pujde-hrdlicka-na-materskou', - 'ext': 'flv', - 'title': 'Duel: Michal Hrdlička a Petr Suchoň', - 'description': 'md5:d0cc509858eee1b1374111c588c6f5d5', - 'thumbnail': r're:^https?://.*\.(?:jpg)', - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { 'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260', 'md5': '1dd7b9d5ea27bc361f110cd855a19bd3', 'info_dict': { @@ -40,33 +100,6 @@ class NovaIE(InfoExtractor): 'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53', 'thumbnail': r're:^https?://.*\.(?:jpg)', } - }, { - 'url': 'http://novaplus.nova.cz/porad/policie-modrava/video/5591-policie-modrava-15-dil-blondynka-na-hrbitove', - 'info_dict': { - 'id': '1756825', - 'display_id': '5591-policie-modrava-15-dil-blondynka-na-hrbitove', - 'ext': 'flv', - 'title': 'Policie Modrava - 15. díl - Blondýnka na hřbitově', - 'description': 'md5:dc24e50be5908df83348e50d1431295e', # Make sure this description is clean of html tags - 'thumbnail': r're:^https?://.*\.(?:jpg)', - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://novaplus.nova.cz/porad/televizni-noviny/video/5585-televizni-noviny-30-5-2015/', - 'info_dict': { - 'id': '1756858', - 'ext': 'flv', - 'title': 'Televizní noviny - 30. 5. 2015', - 'thumbnail': r're:^https?://.*\.(?:jpg)', - 'upload_date': '20150530', - }, - 'params': { - # rtmp download - 'skip_download': True, - } }, { 'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html', 'info_dict': { @@ -81,6 +114,20 @@ class NovaIE(InfoExtractor): # rtmp download 'skip_download': True, } + }, { + # media.cms.nova.cz embed + 'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil', + 'info_dict': { + 'id': '8o0n0r', + 'ext': 'mp4', + 'title': '2180. díl', + 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 2578, + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [NovaEmbedIE.ie_key()], }, { 'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html', 'only_matching': True, @@ -105,6 +152,15 @@ class NovaIE(InfoExtractor): webpage = self._download_webpage(url, display_id) + # novaplus + embed_id = self._search_regex( + r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)', + webpage, 'embed url', default=None) + if embed_id: + return self.url_result( + 'https://media.cms.nova.cz/embed/%s' % embed_id, + ie=NovaEmbedIE.ie_key(), video_id=embed_id) + video_id = self._search_regex( [r"(?:media|video_id)\s*:\s*'(\d+)'", r'media=(\d+)', -- 2.22.2