debian/control: Update list of extractors/supported sites.

[youtubedl] / youtube_dl / extractor / cbc.py

diff --git a/youtube_dl/extractor/cbc.py b/youtube_dl/extractor/cbc.py
index 3be0c646bb8b4e431ea47fa5db4b46e13208dcd6..751a3a8f26c94ecb19c130503593515312bac6c7 100644 (file)
--- a/youtube_dl/extractor/cbc.py
+++ b/youtube_dl/extractor/cbc.py
@@ -5,7 +5,10 @@ import json
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..compat import compat_str
+from ..compat import (
+    compat_str,
+    compat_HTTPError,
+)

 from ..utils import (
     js_to_json,
     smuggle_url,
 from ..utils import (
     js_to_json,
     smuggle_url,


@@ -14,9 +17,11 @@ from ..utils import (
     xpath_element,
     xpath_with_ns,
     find_xpath_attr,
     xpath_element,
     xpath_with_ns,
     find_xpath_attr,

+    orderedSet,

     parse_duration,
     parse_iso8601,
     parse_age_limit,
     parse_duration,
     parse_iso8601,
     parse_age_limit,

+    strip_or_none,

     int_or_none,
     ExtractorError,
 )
     int_or_none,
     ExtractorError,
 )


@@ -126,15 +131,23 @@ class CBCIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)

+        title = self._og_search_title(webpage, default=None) or self._html_search_meta(
+            'twitter:title', webpage, 'title', default=None) or self._html_search_regex(
+                r'<title>([^<]+)</title>', webpage, 'title', fatal=False)

         entries = [
             self._extract_player_init(player_init, display_id)
             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
         entries = [
             self._extract_player_init(player_init, display_id)
             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]

+        media_ids = []
+        for media_id_re in (
+                r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"',
+                r'<div[^>]+\bid=["\']player-(\d+)',
+                r'guid["\']\s*:\s*["\'](\d+)'):
+            media_ids.extend(re.findall(media_id_re, webpage))

         entries.extend([
             self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
         entries.extend([
             self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)

-            for media_id in re.findall(r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"', webpage)])
+            for media_id in orderedSet(media_ids)])

         return self.playlist_result(
         return self.playlist_result(

-            entries, display_id,
-            self._og_search_title(webpage, fatal=False),
+            entries, display_id, strip_or_none(title),

             self._og_search_description(webpage))
 
 
             self._og_search_description(webpage))
 
 


@@ -206,30 +219,48 @@ class CBCWatchBaseIE(InfoExtractor):
 
     def _call_api(self, path, video_id):
         url = path if path.startswith('http') else self._API_BASE_URL + path
 
     def _call_api(self, path, video_id):
         url = path if path.startswith('http') else self._API_BASE_URL + path

-        result = self._download_xml(url, video_id, headers={
-            'X-Clearleap-DeviceId': self._device_id,
-            'X-Clearleap-DeviceToken': self._device_token,
-        })
+        for _ in range(2):
+            try:
+                result = self._download_xml(url, video_id, headers={
+                    'X-Clearleap-DeviceId': self._device_id,
+                    'X-Clearleap-DeviceToken': self._device_token,
+                })
+            except ExtractorError as e:
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                    # Device token has expired, re-acquiring device token
+                    self._register_device()
+                    continue
+                raise

         error_message = xpath_text(result, 'userMessage') or xpath_text(result, 'systemMessage')
         if error_message:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message))
         return result
 
     def _real_initialize(self):
         error_message = xpath_text(result, 'userMessage') or xpath_text(result, 'systemMessage')
         if error_message:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message))
         return result
 
     def _real_initialize(self):

-        if not self._device_id or not self._device_token:
-            device = self._downloader.cache.load('cbcwatch', 'device') or {}
-            self._device_id, self._device_token = device.get('id'), device.get('token')
-            if not self._device_id or not self._device_token:
-                result = self._download_xml(
-                    self._API_BASE_URL + 'device/register',
-                    None, data=b'<device><type>web</type></device>')
-                self._device_id = xpath_text(result, 'deviceId', fatal=True)
-                self._device_token = xpath_text(result, 'deviceToken', fatal=True)
-                self._downloader.cache.store(
-                    'cbcwatch', 'device', {
-                        'id': self._device_id,
-                        'token': self._device_token,
-                    })
+        if self._valid_device_token():
+            return
+        device = self._downloader.cache.load('cbcwatch', 'device') or {}
+        self._device_id, self._device_token = device.get('id'), device.get('token')
+        if self._valid_device_token():
+            return
+        self._register_device()
+
+    def _valid_device_token(self):
+        return self._device_id and self._device_token
+
+    def _register_device(self):
+        self._device_id = self._device_token = None
+        result = self._download_xml(
+            self._API_BASE_URL + 'device/register',
+            None, 'Acquiring device token',
+            data=b'<device><type>web</type></device>')
+        self._device_id = xpath_text(result, 'deviceId', fatal=True)
+        self._device_token = xpath_text(result, 'deviceToken', fatal=True)
+        self._downloader.cache.store(
+            'cbcwatch', 'device', {
+                'id': self._device_id,
+                'token': self._device_token,
+            })

 
     def _parse_rss_feed(self, rss):
         channel = xpath_element(rss, 'channel', fatal=True)
 
     def _parse_rss_feed(self, rss):
         channel = xpath_element(rss, 'channel', fatal=True)


@@ -329,7 +360,7 @@ class CBCWatchVideoIE(CBCWatchBaseIE):
 
 class CBCWatchIE(CBCWatchBaseIE):
     IE_NAME = 'cbc.ca:watch'
 
 class CBCWatchIE(CBCWatchBaseIE):
     IE_NAME = 'cbc.ca:watch'

-    _VALID_URL = r'https?://watch\.cbc\.ca/(?:[^/]+/)+(?P<id>[0-9a-f-]+)'
+    _VALID_URL = r'https?://(?:gem|watch)\.cbc\.ca/(?:[^/]+/)+(?P<id>[0-9a-f-]+)'

     _TESTS = [{
         # geo-restricted to Canada, bypassable
         'url': 'http://watch.cbc.ca/doc-zone/season-6/customer-disservice/38e815a-009e3ab12e4',
     _TESTS = [{
         # geo-restricted to Canada, bypassable
         'url': 'http://watch.cbc.ca/doc-zone/season-6/customer-disservice/38e815a-009e3ab12e4',


@@ -355,6 +386,9 @@ class CBCWatchIE(CBCWatchBaseIE):
             'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.',
         },
         'playlist_mincount': 30,
             'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.',
         },
         'playlist_mincount': 30,

+    }, {
+        'url': 'https://gem.cbc.ca/media/this-hour-has-22-minutes/season-26/episode-20/38e815a-0108c6c6a42',
+        'only_matching': True,

     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):