5 from .common
import InfoExtractor
17 class NYTimesBaseIE(InfoExtractor
):
18 _SECRET
= b
'pX(2MbU2);4N{7J8)>YwKRJ+/pQ3JkiU2Q^V>mFYv6g6gYvt6v'
20 def _extract_video_from_id(self
, video_id
):
21 # Authorization generation algorithm is reverse engineered from `signer` in
22 # http://graphics8.nytimes.com/video/vhs/vhs-2.x.min.js
23 path
= '/svc/video/api/v3/video/' + video_id
24 hm
= hmac
.new(self
._SECRET
, (path
+ ':vhs').encode(), hashlib
.sha512
).hexdigest()
25 video_data
= self
._download
_json
('http://www.nytimes.com' + path
, video_id
, 'Downloading video JSON', headers
={
26 'Authorization': 'NYTV ' + base64
.b64encode(hm
.encode()).decode(),
30 video_data
= self
._download
_json
(
31 'http://www.nytimes.com/svc/video/api/v2/video/' + video_id
,
32 video_id
, 'Downloading video JSON')
34 title
= video_data
['headline']
36 def get_file_size(file_size
):
37 if isinstance(file_size
, int):
39 elif isinstance(file_size
, dict):
40 return int(file_size
.get('value', 0))
47 for video
in video_data
.get('renditions', []):
48 video_url
= video
.get('url')
49 format_id
= video
.get('type')
50 if not video_url
or format_id
== 'thumbs' or video_url
in urls
:
52 urls
.append(video_url
)
53 ext
= mimetype2ext(video
.get('mimetype')) or determine_ext(video_url
)
55 m3u8_fmts
, m3u8_subs
= self
._extract
_m
3u8_formats
_and
_subtitles
(
56 video_url
, video_id
, 'mp4', 'm3u8_native',
57 m3u8_id
=format_id
or 'hls', fatal
=False)
58 formats
.extend(m3u8_fmts
)
59 subtitles
= self
._merge
_subtitles
(subtitles
, m3u8_subs
)
62 # formats.extend(self._extract_mpd_formats(
63 # video_url, video_id, format_id or 'dash', fatal=False))
67 'format_id': format_id
,
68 'vcodec': video
.get('videoencoding') or video
.get('video_codec'),
69 'width': int_or_none(video
.get('width')),
70 'height': int_or_none(video
.get('height')),
71 'filesize': get_file_size(video
.get('file_size') or video
.get('fileSize')),
72 'tbr': int_or_none(video
.get('bitrate'), 1000) or None,
75 self
._sort
_formats
(formats
)
78 for image
in video_data
.get('images', []):
79 image_url
= image
.get('url')
83 'url': 'http://www.nytimes.com/' + image_url
,
84 'width': int_or_none(image
.get('width')),
85 'height': int_or_none(image
.get('height')),
88 publication_date
= video_data
.get('publication_date')
89 timestamp
= parse_iso8601(publication_date
[:-8]) if publication_date
else None
94 'description': video_data
.get('summary'),
95 'timestamp': timestamp
,
96 'uploader': video_data
.get('byline'),
97 'duration': float_or_none(video_data
.get('duration'), 1000),
99 'subtitles': subtitles
,
100 'thumbnails': thumbnails
,
104 class NYTimesIE(NYTimesBaseIE
):
105 _VALID_URL
= r
'https?://(?:(?:www\.)?nytimes\.com/video/(?:[^/]+/)+?|graphics8\.nytimes\.com/bcvideo/\d+(?:\.\d+)?/iframe/embed\.html\?videoId=)(?P<id>\d+)'
106 _EMBED_REGEX
= [r
'<iframe[^>]+src=(["\'])(?P
<url
>(?
:https?
:)?
//graphics8\
.nytimes\
.com
/bcvideo
/[^
/]+/iframe
/embed\
.html
.+?
)\
1>']
109 'url
': 'http
://www
.nytimes
.com
/video
/opinion
/100000002847155/verbatim
-what
-is-a
-photocopier
.html?playlistId
=100000001150263',
110 'md5
': 'd665342765db043f7e225cff19df0f2d
',
112 'id': '100000002847155',
114 'title
': 'Verbatim
: What Is a Photocopier?
',
115 'description
': 'md5
:93603dada88ddbda9395632fdc5da260
',
116 'timestamp
': 1398631707,
117 'upload_date
': '20140427',
118 'uploader
': 'Brett Weiner
',
122 'url
': 'http
://www
.nytimes
.com
/video
/travel
/100000003550828/36-hours
-in-dubai
.html
',
123 'only_matching
': True,
126 def _real_extract(self, url):
127 video_id = self._match_id(url)
129 return self._extract_video_from_id(video_id)
132 class NYTimesArticleIE(NYTimesBaseIE):
133 _VALID_URL = r'https?
://(?
:www\
.)?nytimes\
.com
/(.(?
<!video
))*?
/(?
:[^
/]+/)*(?P
<id>[^
.]+)(?
:\
.html
)?
'
135 'url
': 'http
://www
.nytimes
.com
/2015/04/14/business
/owner
-of
-gravity
-payments
-a
-credit
-card
-processor
-is-setting
-a
-new
-minimum
-wage
-70000-a
-year
.html?_r
=0',
136 'md5
': 'e2076d58b4da18e6a001d53fd56db3c9
',
138 'id': '100000003628438',
140 'title
': 'New Minimum Wage
: $
70,000 a Year
',
141 'description
': 'Dan Price
, C
.E
.O
. of Gravity Payments
, surprised his
120-person staff by announcing that he planned over the next three years to
raise the salary of every employee to $
70,000 a year
.',
142 'timestamp
': 1429033037,
143 'upload_date
': '20150414',
144 'uploader
': 'Matthew Williams
',
147 'url
': 'http
://www
.nytimes
.com
/2016/10/14/podcasts
/revelations
-from-the
-final
-weeks
.html
',
148 'md5
': 'e0d52040cafb07662acf3c9132db3575
',
150 'id': '100000004709062',
151 'title
': 'The Run
-Up
: ‘He Was Like an Octopus’
',
153 'description
': 'md5
:fb5c6b93b12efc51649b4847fe066ee4
',
154 'series
': 'The Run
-Up
',
155 'episode
': '‘He Was Like an Octopus’
',
156 'episode_number
': 20,
160 'url
': 'http
://www
.nytimes
.com
/2016/10/16/books
/review
/inside
-the
-new
-york
-times
-book
-review
-the
-rise
-of
-hitler
.html
',
162 'id': '100000004709479',
163 'title
': 'The Rise of Hitler
',
165 'description
': 'md5
:bce877fd9e3444990cb141875fab0028
',
166 'creator
': 'Pamela Paul
',
170 'skip_download
': True,
173 'url
': 'http
://www
.nytimes
.com
/news
/minute
/2014/03/17/times
-minute
-whats
-next
-in-crimea
/?_php
=true
&_type
=blogs
&_php
=true
&_type
=blogs
&_r
=1',
174 'only_matching
': True,
177 def _extract_podcast_from_json(self, json, page_id, webpage):
178 podcast_audio = self._parse_json(
179 json, page_id, transform_source=js_to_json)
181 audio_data = podcast_audio['data
']
182 track = audio_data['track
']
184 episode_title = track['title
']
185 video_url = track['source
']
187 description = track.get('description
') or self._html_search_meta(
188 ['og
:description
', 'twitter
:description
'], webpage)
190 podcast_title = audio_data.get('podcast
', {}).get('title
')
191 title = ('%s: %s' % (podcast_title, episode_title)
192 if podcast_title else episode_title)
194 episode = audio_data.get('podcast
', {}).get('episode
') or ''
195 episode_number = int_or_none(self._search_regex(
196 r'[Ee
]pisode\s
+(\d
+)', episode, 'episode number
', default=None))
199 'id': remove_start(podcast_audio.get('target
'), 'FT
') or page_id,
202 'description
': description,
203 'creator
': track.get('credit
'),
204 'series
': podcast_title,
205 'episode
': episode_title,
206 'episode_number
': episode_number,
207 'duration
': int_or_none(track.get('duration
')),
210 def _real_extract(self, url):
211 page_id = self._match_id(url)
213 webpage = self._download_webpage(url, page_id)
215 video_id = self._search_regex(
216 r'data
-videoid
=["\'](\d+)', webpage, 'video id',
217 default=None, fatal=False)
218 if video_id is not None:
219 return self._extract_video_from_id(video_id)
221 podcast_data = self._search_regex(
222 (r'NYTD\.FlexTypes\.push\s*\(\s*({.+?})\s*\)\s*;\s*</script',
223 r'NYTD\.FlexTypes\.push\s*\(\s*({.+})\s*\)\s*;'),
224 webpage, 'podcast data')
225 return self._extract_podcast_from_json(podcast_data, page_id, webpage)
228 class NYTimesCookingIE(NYTimesBaseIE):
229 _VALID_URL = r'https?://cooking\.nytimes\.com/(?:guid|recip)es/(?P<id>\d+)'
231 'url': 'https://cooking.nytimes.com/recipes/1017817-cranberry-curd-tart',
232 'md5': 'dab81fa2eaeb3f9ed47498bdcfcdc1d3',
234 'id': '100000004756089',
236 'timestamp': 1479383008,
237 'uploader': 'By SHAW LASH, ADAM SAEWITZ and JAMES HERRON',
238 'title': 'Cranberry Tart',
239 'upload_date': '20161117',
240 'description': 'If you are a fan of lemon curd or the classic French tarte au citron, you will love this cranberry version.',
243 'url': 'https://cooking.nytimes.com/guides/13-how-to-cook-a-turkey',
244 'md5': '4b2e8c70530a89b8d905a2b572316eb8',
246 'id': '100000003951728',
248 'timestamp': 1445509539,
249 'description': 'Turkey guide',
250 'upload_date': '20151022',
255 def _real_extract(self, url):
256 page_id = self._match_id(url)
258 webpage = self._download_webpage(url, page_id)
260 video_id = self._search_regex(
261 r'data-video-id=["\'](\d
+)', webpage, 'video
id')
263 return self._extract_video_from_id(video_id)