From 11380753b5aa9d8128ef28a968ab325973276fa5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 9 Jun 2016 04:00:47 +0700 Subject: [PATCH] [vessel] Add support for embed urls and improve extraction --- youtube_dl/extractor/vessel.py | 27 ++++++++++++++++++++------- 1 file changed, 20 insertions(+), 7 deletions(-) diff --git a/youtube_dl/extractor/vessel.py b/youtube_dl/extractor/vessel.py index 1a0ff3395..e027c018b 100644 --- a/youtube_dl/extractor/vessel.py +++ b/youtube_dl/extractor/vessel.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import json +import re from .common import InfoExtractor from ..utils import ( @@ -12,11 +13,11 @@ from ..utils import ( class VesselIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vessel\.com/videos/(?P[0-9a-zA-Z]+)' + _VALID_URL = r'https?://(?:www\.)?vessel\.com/(?:videos|embed)/(?P[0-9a-zA-Z]+)' _API_URL_TEMPLATE = 'https://www.vessel.com/api/view/items/%s' _LOGIN_URL = 'https://www.vessel.com/api/account/login' _NETRC_MACHINE = 'vessel' - _TEST = { + _TESTS = [{ 'url': 'https://www.vessel.com/videos/HDN7G5UMs', 'md5': '455cdf8beb71c6dd797fd2f3818d05c4', 'info_dict': { @@ -28,7 +29,16 @@ class VesselIE(InfoExtractor): 'description': 'Did Nvidia pull out all the stops on the Titan X, or does its performance leave something to be desired?', 'timestamp': int, }, - } + }, { + 'url': 'https://www.vessel.com/embed/G4U7gUJ6a?w=615&h=346', + 'only_matching': True, + }] + + @staticmethod + def _extract_urls(webpage): + return [url for _, url in re.findall( + r']+src=(["\'])((?:https?:)?//(?:www\.)?vessel\.com/embed/[0-9a-zA-Z]+.*?)\1', + webpage)] @staticmethod def make_json_request(url, data): @@ -98,16 +108,19 @@ class VesselIE(InfoExtractor): formats = [] for f in video_asset.get('sources', []): - if f['name'] == 'hls-index': + location = f.get('location') + if not location: + continue + if f.get('name') == 'hls-index': formats.extend(self._extract_m3u8_formats( - f['location'], video_id, ext='mp4', m3u8_id='m3u8')) + location, video_id, ext='mp4', m3u8_id='m3u8')) else: formats.append({ - 'format_id': f['name'], + 'format_id': f.get('name'), 'tbr': f.get('bitrate'), 'height': f.get('height'), 'width': f.get('width'), - 'url': f['location'], + 'url': location, }) self._sort_formats(formats)