From 6ef3e65a7b244d5e432e764772177c7d48cab237 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 5 Feb 2017 13:37:27 +0700 Subject: [PATCH] [videopress] Add extractor --- youtube_dl/extractor/extractors.py | 1 + youtube_dl/extractor/generic.py | 22 +++++++ youtube_dl/extractor/videopress.py | 99 ++++++++++++++++++++++++++++++ 3 files changed, 122 insertions(+) create mode 100644 youtube_dl/extractor/videopress.py diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 32420937c..cf608faee 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -1095,6 +1095,7 @@ VideomoreSeasonIE, ) from .videopremium import VideoPremiumIE +from .videopress import VideoPressIE from .vidio import VidioIE from .vidme import ( VidmeIE, diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index a23486620..4156cf27d 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -81,6 +81,7 @@ from .twentymin import TwentyMinutenIE from .ustream import UstreamIE from .openload import OpenloadIE +from .videopress import VideoPressIE class GenericIE(InfoExtractor): @@ -1473,6 +1474,21 @@ class GenericIE(InfoExtractor): 'skip_download': True, }, 'add_ie': [TwentyMinutenIE.ie_key()], + }, + { + # VideoPress embed + 'url': 'https://en.support.wordpress.com/videopress/', + 'info_dict': { + 'id': 'OcobLTqC', + 'ext': 'm4v', + 'title': 'IMG_5786', + 'timestamp': 1435711927, + 'upload_date': '20150701', + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [VideoPressIE.ie_key()], } # { # # TODO: find another test @@ -2438,6 +2454,12 @@ def _playlist_from_matches(matches, getter=None, ie=None): return _playlist_from_matches( openload_urls, ie=OpenloadIE.ie_key()) + # Look for VideoPress embeds + videopress_urls = VideoPressIE._extract_urls(webpage) + if videopress_urls: + return _playlist_from_matches( + videopress_urls, ie=VideoPressIE.ie_key()) + # Looking for http://schema.org/VideoObject json_ld = self._search_json_ld( webpage, video_id, default={}, expected_type='VideoObject') diff --git a/youtube_dl/extractor/videopress.py b/youtube_dl/extractor/videopress.py new file mode 100644 index 000000000..049db25a5 --- /dev/null +++ b/youtube_dl/extractor/videopress.py @@ -0,0 +1,99 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import random +import re + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + determine_ext, + float_or_none, + parse_age_limit, + qualities, + try_get, + unified_timestamp, + urljoin, +) + + +class VideoPressIE(InfoExtractor): + _VALID_URL = r'https?://videopress\.com/embed/(?P[\da-zA-Z]+)' + _TESTS = [{ + 'url': 'https://videopress.com/embed/kUJmAcSf', + 'md5': '706956a6c875873d51010921310e4bc6', + 'info_dict': { + 'id': 'kUJmAcSf', + 'ext': 'mp4', + 'title': 'VideoPress Demo', + 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 634.6, + 'timestamp': 1434983935, + 'upload_date': '20150622', + 'age_limit': 0, + }, + }, { + # 17+, requires birth_* params + 'url': 'https://videopress.com/embed/iH3gstfZ', + 'only_matching': True, + }] + + @staticmethod + def _extract_urls(webpage): + return re.findall( + r']+src=["\']((?:https?://)?videopress\.com/embed/[\da-zA-Z]+)', + webpage) + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + 'https://public-api.wordpress.com/rest/v1.1/videos/%s' % video_id, + video_id, query={ + 'birth_month': random.randint(1, 12), + 'birth_day': random.randint(1, 31), + 'birth_year': random.randint(1950, 1995), + }) + + title = video['title'] + + def base_url(scheme): + return try_get( + video, lambda x: x['file_url_base'][scheme], compat_str) + + base_url = base_url('https') or base_url('http') + + QUALITIES = ('std', 'dvd', 'hd') + quality = qualities(QUALITIES) + + formats = [] + for format_id, f in video['files'].items(): + if not isinstance(f, dict): + continue + for ext, path in f.items(): + if ext in ('mp4', 'ogg'): + formats.append({ + 'url': urljoin(base_url, path), + 'format_id': '%s-%s' % (format_id, ext), + 'ext': determine_ext(path, ext), + 'quality': quality(format_id), + }) + original_url = try_get(video, lambda x: x['original'], compat_str) + if original_url: + formats.append({ + 'url': original_url, + 'format_id': 'original', + 'quality': len(QUALITIES), + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': video.get('description'), + 'thumbnail': video.get('poster'), + 'duration': float_or_none(video.get('duration'), 1000), + 'timestamp': unified_timestamp(video.get('upload_date')), + 'age_limit': parse_age_limit(video.get('rating')), + 'formats': formats, + }