Improve geo bypass mechanism

[youtube-dl] / youtube_dl / extractor / go.py

diff --git a/youtube_dl/extractor/go.py b/youtube_dl/extractor/go.py
index 6a437c54db87bac54f913b90d466e63b460457d7..b205bfc7c99922bda00326bf4ac3caeab7ac2c95 100644 (file)
--- a/youtube_dl/extractor/go.py
+++ b/youtube_dl/extractor/go.py
@@ -3,23 +3,41 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .adobepass import AdobePassIE
 from ..utils import (
     int_or_none,
     determine_ext,
     parse_age_limit,
+    urlencode_postdata,
+    ExtractorError,
 )
 
 
-class GoIE(InfoExtractor):
-    _BRANDS = {
-        'abc': '001',
-        'freeform': '002',
-        'watchdisneychannel': '004',
-        'watchdisneyjunior': '008',
-        'watchdisneyxd': '009',
+class GoIE(AdobePassIE):
+    _SITE_INFO = {
+        'abc': {
+            'brand': '001',
+            'requestor_id': 'ABC',
+        },
+        'freeform': {
+            'brand': '002',
+            'requestor_id': 'ABCFamily',
+        },
+        'watchdisneychannel': {
+            'brand': '004',
+            'requestor_id': 'Disney',
+        },
+        'watchdisneyjunior': {
+            'brand': '008',
+            'requestor_id': 'DisneyJunior',
+        },
+        'watchdisneyxd': {
+            'brand': '009',
+            'requestor_id': 'DisneyXD',
+        }
     }
-    _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/.*?vdka(?P<id>\w+)' % '|'.join(_BRANDS.keys())
+    _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/(?:[^/]+/)*(?:vdka(?P<id>\w+)|season-\d+/\d+-(?P<display_id>[^/?#]+))' % '|'.join(_SITE_INFO.keys())
+    _GEO_COUNTRIES = ['US']
     _TESTS = [{
         'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
         'info_dict': {
@@ -38,9 +56,17 @@ class GoIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        sub_domain, video_id = re.match(self._VALID_URL, url).groups()
+        sub_domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
+        if not video_id:
+            webpage = self._download_webpage(url, display_id)
+            video_id = self._search_regex(
+                # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
+                # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
+                r'data-video-id=["\']*VDKA(\w+)', webpage, 'video id')
+        site_info = self._SITE_INFO[sub_domain]
+        brand = site_info['brand']
         video_data = self._download_json(
-            'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (self._BRANDS[sub_domain], video_id),
+            'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (brand, video_id),
             video_id)['video'][0]
         title = video_data['title']
 
@@ -52,6 +78,37 @@ class GoIE(InfoExtractor):
             format_id = asset.get('format')
             ext = determine_ext(asset_url)
             if ext == 'm3u8':
+                video_type = video_data.get('type')
+                if video_type == 'lf':
+                    data = {
+                        'video_id': video_data['id'],
+                        'video_type': video_type,
+                        'brand': brand,
+                        'device': '001',
+                    }
+                    if video_data.get('accesslevel') == '1':
+                        requestor_id = site_info['requestor_id']
+                        resource = self._get_mvpd_resource(
+                            requestor_id, title, video_id, None)
+                        auth = self._extract_mvpd_auth(
+                            url, video_id, requestor_id, resource)
+                        data.update({
+                            'token': auth,
+                            'token_type': 'ap',
+                            'adobe_requestor_id': requestor_id,
+                        })
+                    entitlement = self._download_json(
+                        'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
+                        video_id, data=urlencode_postdata(data), headers=self.geo_verification_headers())
+                    errors = entitlement.get('errors', {}).get('errors', [])
+                    if errors:
+                        for error in errors:
+                            if error.get('code') == 1002:
+                                self.raise_geo_restricted(
+                                    error['message'], countries=self._GEO_COUNTRIES)
+                        error_message = ', '.join([error['message'] for error in errors])
+                        raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
+                    asset_url += '?' + entitlement['uplynkData']['sessionKey']
                 formats.extend(self._extract_m3u8_formats(
                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
             else: