aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/nytimes.py
blob: 2bb77ab249239163d8318a57e8fd0fdb57d2e32a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
# coding: utf-8
from __future__ import unicode_literals

import hmac
import hashlib
import base64

from .common import InfoExtractor
from ..utils import (
    determine_ext,
    float_or_none,
    int_or_none,
    js_to_json,
    mimetype2ext,
    parse_iso8601,
    remove_start,
)


class NYTimesBaseIE(InfoExtractor):
    _SECRET = b'pX(2MbU2);4N{7J8)>YwKRJ+/pQ3JkiU2Q^V>mFYv6g6gYvt6v'

    def _extract_video_from_id(self, video_id):
        # Authorization generation algorithm is reverse engineered from `signer` in
        # http://graphics8.nytimes.com/video/vhs/vhs-2.x.min.js
        path = '/svc/video/api/v3/video/' + video_id
        hm = hmac.new(self._SECRET, (path + ':vhs').encode(), hashlib.sha512).hexdigest()
        video_data = self._download_json('http://www.nytimes.com' + path, video_id, 'Downloading video JSON', headers={
            'Authorization': 'NYTV ' + base64.b64encode(hm.encode()).decode(),
            'X-NYTV': 'vhs',
        }, fatal=False)
        if not video_data:
            video_data = self._download_json(
                'http://www.nytimes.com/svc/video/api/v2/video/' + video_id,
                video_id, 'Downloading video JSON')

        title = video_data['headline']

        def get_file_size(file_size):
            if isinstance(file_size, int):
                return file_size
            elif isinstance(file_size, dict):
                return int(file_size.get('value', 0))
            else:
                return None

        urls = []
        formats = []
        for video in video_data.get('renditions', []):
            video_url = video.get('url')
            format_id = video.get('type')
            if not video_url or format_id == 'thumbs' or video_url in urls:
                continue
            urls.append(video_url)
            ext = mimetype2ext(video.get('mimetype')) or determine_ext(video_url)
            if ext == 'm3u8':
                formats.extend(self._extract_m3u8_formats(
                    video_url, video_id, 'mp4', 'm3u8_native',
                    m3u8_id=format_id or 'hls', fatal=False))
            elif ext == 'mpd':
                continue
            #     formats.extend(self._extract_mpd_formats(
            #         video_url, video_id, format_id or 'dash', fatal=False))
            else:
                formats.append({
                    'url': video_url,
                    'format_id': format_id,
                    'vcodec': video.get('videoencoding') or video.get('video_codec'),
                    'width': int_or_none(video.get('width')),
                    'height': int_or_none(video.get('height')),
                    'filesize': get_file_size(video.get('file_size') or video.get('fileSize')),
                    'tbr': int_or_none(video.get('bitrate'), 1000),
                    'ext': ext,
                })
        self._sort_formats(formats)

        thumbnails = []
        for image in video_data.get('images', []):
            image_url = image.get('url')
            if not image_url:
                continue
            thumbnails.append({
                'url': 'http://www.nytimes.com/' + image_url,
                'width': int_or_none(image.get('width')),
                'height': int_or_none(image.get('height')),
            })

        publication_date = video_data.get('publication_date')
        timestamp = parse_iso8601(publication_date[:-8]) if publication_date else None

        return {
            'id': video_id,
            'title': title,
            'description': video_data.get('summary'),
            'timestamp': timestamp,
            'uploader': video_data.get('byline'),
            'duration': float_or_none(video_data.get('duration'), 1000),
            'formats': formats,
            'thumbnails': thumbnails,
        }


class NYTimesIE(NYTimesBaseIE):
    _VALID_URL = r'https?://(?:(?:www\.)?nytimes\.com/video/(?:[^/]+/)+?|graphics8\.nytimes\.com/bcvideo/\d+(?:\.\d+)?/iframe/embed\.html\?videoId=)(?P<id>\d+)'

    _TESTS = [{
        'url': 'http://www.nytimes.com/video/opinion/100000002847155/verbatim-what-is-a-photocopier.html?playlistId=100000001150263',
        'md5': 'd665342765db043f7e225cff19df0f2d',
        'info_dict': {
            'id': '100000002847155',
            'ext': 'mov',
            'title': 'Verbatim: What Is a Photocopier?',
            'description': 'md5:93603dada88ddbda9395632fdc5da260',
            'timestamp': 1398631707,
            'upload_date': '20140427',
            'uploader': 'Brett Weiner',
            'duration': 419,
        }
    }, {
        'url': 'http://www.nytimes.com/video/travel/100000003550828/36-hours-in-dubai.html',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        video_id = self._match_id(url)

        return self._extract_video_from_id(video_id)


class NYTimesArticleIE(NYTimesBaseIE):
    _VALID_URL = r'https?://(?:www\.)?nytimes\.com/(.(?<!video))*?/(?:[^/]+/)*(?P<id>[^.]+)(?:\.html)?'
    _TESTS = [{
        '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',
        'md5': 'e2076d58b4da18e6a001d53fd56db3c9',
        'info_dict': {
            'id': '100000003628438',
            'ext': 'mov',
            'title': 'New Minimum Wage: $70,000 a Year',
            '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.',
            'timestamp': 1429033037,
            'upload_date': '20150414',
            'uploader': 'Matthew Williams',
        }
    }, {
        'url': 'http://www.nytimes.com/2016/10/14/podcasts/revelations-from-the-final-weeks.html',
        'md5': 'e0d52040cafb07662acf3c9132db3575',
        'info_dict': {
            'id': '100000004709062',
            'title': 'The Run-Up: ‘He Was Like an Octopus’',
            'ext': 'mp3',
            'description': 'md5:fb5c6b93b12efc51649b4847fe066ee4',
            'series': 'The Run-Up',
            'episode': '‘He Was Like an Octopus’',
            'episode_number': 20,
            'duration': 2130,
        }
    }, {
        'url': 'http://www.nytimes.com/2016/10/16/books/review/inside-the-new-york-times-book-review-the-rise-of-hitler.html',
        'info_dict': {
            'id': '100000004709479',
            'title': 'The Rise of Hitler',
            'ext': 'mp3',
            'description': 'md5:bce877fd9e3444990cb141875fab0028',
            'creator': 'Pamela Paul',
            'duration': 3475,
        },
        'params': {
            'skip_download': True,
        },
    }, {
        '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',
        'only_matching': True,
    }]

    def _extract_podcast_from_json(self, json, page_id, webpage):
        podcast_audio = self._parse_json(
            json, page_id, transform_source=js_to_json)

        audio_data = podcast_audio['data']
        track = audio_data['track']

        episode_title = track['title']
        video_url = track['source']

        description = track.get('description') or self._html_search_meta(
            ['og:description', 'twitter:description'], webpage)

        podcast_title = audio_data.get('podcast', {}).get('title')
        title = ('%s: %s' % (podcast_title, episode_title)
                 if podcast_title else episode_title)

        episode = audio_data.get('podcast', {}).get('episode') or ''
        episode_number = int_or_none(self._search_regex(
            r'[Ee]pisode\s+(\d+)', episode, 'episode number', default=None))

        return {
            'id': remove_start(podcast_audio.get('target'), 'FT') or page_id,
            'url': video_url,
            'title': title,
            'description': description,
            'creator': track.get('credit'),
            'series': podcast_title,
            'episode': episode_title,
            'episode_number': episode_number,
            'duration': int_or_none(track.get('duration')),
        }

    def _real_extract(self, url):
        page_id = self._match_id(url)

        webpage = self._download_webpage(url, page_id)

        video_id = self._search_regex(
            r'data-videoid=["\'](\d+)', webpage, 'video id',
            default=None, fatal=False)
        if video_id is not None:
            return self._extract_video_from_id(video_id)

        podcast_data = self._search_regex(
            (r'NYTD\.FlexTypes\.push\s*\(\s*({.+?})\s*\)\s*;\s*</script',
             r'NYTD\.FlexTypes\.push\s*\(\s*({.+})\s*\)\s*;'),
            webpage, 'podcast data')
        return self._extract_podcast_from_json(podcast_data, page_id, webpage)