youtube-dl

commit 223b27f46cdf3cde9c9556dac19055ad5d8bff5f
parent 425142be601adfb4218f78c10aedbed14ad1facd
Author: Naglis Jonaitis <njonaitis@gmail.com>
Date:   Thu, 26 Mar 2015 19:48:22 +0200

[vessel] Add new extractor (Closes #5275)

Diffstat:
M
youtube_dl/extractor/__init__.py
 | 
1
+
A
youtube_dl/extractor/vessel.py
 | 
127
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

2 files changed, 128 insertions(+), 0 deletions(-)
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
@@ -551,6 +551,7 @@ from .ustream import UstreamIE, UstreamChannelIE
 from .vbox7 import Vbox7IE
 from .veehd import VeeHDIE
 from .veoh import VeohIE
+from .vessel import VesselIE
 from .vesti import VestiIE
 from .vevo import VevoIE
 from .vgtv import VGTVIE
diff --git a/youtube_dl/extractor/vessel.py b/youtube_dl/extractor/vessel.py
@@ -0,0 +1,127 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_request
+from ..utils import (
+    ExtractorError,
+    parse_iso8601,
+)
+
+
+class VesselIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?vessel\.com/videos/(?P<id>[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 = {
+        'url': 'https://www.vessel.com/videos/HDN7G5UMs',
+        'md5': '455cdf8beb71c6dd797fd2f3818d05c4',
+        'info_dict': {
+            'id': 'HDN7G5UMs',
+            'ext': 'mp4',
+            'title': 'Nvidia GeForce GTX Titan X - The Best Video Card on the Market?',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'upload_date': '20150317',
+            'description': 'Did Nvidia pull out all the stops on the Titan X, or does its performance leave something to be desired?',
+            'timestamp': int,
+        },
+    }
+
+    @staticmethod
+    def make_json_request(url, data):
+        payload = json.dumps(data).encode('utf-8')
+        req = compat_urllib_request.Request(url, payload)
+        req.add_header('Content-Type', 'application/json; charset=utf-8')
+        return req
+
+    @staticmethod
+    def find_assets(data, asset_type):
+        for asset in data.get('assets', []):
+            if asset.get('type') == asset_type:
+                yield asset
+
+    def _check_access_rights(self, data):
+        access_info = data.get('__view', {})
+        if access_info.get('allow_access') == False:
+            err_code = access_info.get('error_code') or ''
+            if err_code == 'ITEM_PAID_ONLY':
+                raise ExtractorError(
+                    'This video requires subscription.', expected=True)
+            else:
+                raise ExtractorError(
+                    'Access to this content is restricted. (%s said: %s)' % (self.IE_NAME, err_code), expected=True)
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+        self.report_login()
+        data = {
+            'client_id': 'web',
+            'type': 'password',
+            'user_key': username,
+            'password': password,
+        }
+        login_request = VesselIE.make_json_request(self._LOGIN_URL, data)
+        self._download_webpage(login_request, None, False, 'Wrong login info')
+
+    def _real_initialize(self):
+        self._login()
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+        data = self._parse_json(self._search_regex(
+            r'App\.bootstrapData\((.*?)\);', webpage, 'data'), video_id)
+        asset_id = data['model']['data']['id']
+
+        req = VesselIE.make_json_request(
+            self._API_URL_TEMPLATE % asset_id, {'client': 'web'})
+        data = self._download_json(req, video_id)
+
+        self._check_access_rights(data)
+
+        try:
+            video_asset = next(VesselIE.find_assets(data, 'video'))
+        except StopIteration:
+            raise ExtractorError('No video assets found')
+
+        formats = []
+        for f in video_asset.get('sources', []):
+            if f['name'] == 'hls-index':
+                formats.extend(self._extract_m3u8_formats(
+                    f['location'], video_id, ext='mp4', m3u8_id='m3u8'))
+            else:
+                formats.append({
+                    'format_id': f['name'],
+                    'tbr': f.get('bitrate'),
+                    'height': f.get('height'),
+                    'width': f.get('width'),
+                    'url': f['location'],
+                })
+        self._sort_formats(formats)
+
+        thumbnails = []
+        for im_asset in VesselIE.find_assets(data, 'image'):
+            thumbnails.append({
+                'url': im_asset['location'],
+                'width': im_asset.get('width', 0),
+                'height': im_asset.get('height', 0),
+            })
+
+        return {
+            'id': video_id,
+            'title': data['title'],
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'description': data.get('short_description'),
+            'duration': data.get('duration'),
+            'comment_count': data.get('comment_count'),
+            'like_count': data.get('like_count'),
+            'view_count': data.get('view_count'),
+            'timestamp': parse_iso8601(data.get('released_at')),
+        }