youtube-dl

commit 89fd03079b83e1dfae016e4e28746b00befe8b2d
parent 1c45b7a8a9ac50b87b9b1a540e9d9e875ac22a0b
Author: mrBliss <dewinant@gmail.com>
Date:   Mon, 15 May 2017 16:46:55 +0200

[vier] Improve extraction

+ Add support for authentication
* Bypass authentication when no credentials provded
* Improve extraction robustness
Diffstat:
M
youtube_dl/extractor/vier.py
 | 
74
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++--

1 file changed, 72 insertions(+), 2 deletions(-)
diff --git a/youtube_dl/extractor/vier.py b/youtube_dl/extractor/vier.py
@@ -5,12 +5,17 @@ import re
 import itertools
 
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    urlencode_postdata,
+)
 
 
 class VierIE(InfoExtractor):
     IE_NAME = 'vier'
     IE_DESC = 'vier.be and vijf.be'
     _VALID_URL = r'https?://(?:www\.)?(?P<site>vier|vijf)\.be/(?:[^/]+/videos/(?P<display_id>[^/]+)(?:/(?P<id>\d+))?|video/v3/embed/(?P<embed_id>\d+))'
+    _NETRC_MACHINE = 'vier'
     _TESTS = [{
         'url': 'http://www.vier.be/planb/videos/het-wordt-warm-de-moestuin/16129',
         'info_dict': {
@@ -24,13 +29,14 @@ class VierIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        # 'skip': 'Requires account credentials',
     }, {
         'url': 'http://www.vijf.be/temptationisland/videos/zo-grappig-temptation-island-hosts-moeten-kiezen-tussen-onmogelijke-dilemmas/2561614',
         'info_dict': {
             'id': '2561614',
             'display_id': 'zo-grappig-temptation-island-hosts-moeten-kiezen-tussen-onmogelijke-dilemmas',
             'ext': 'mp4',
-            'title': 'ZO grappig: Temptation Island hosts moeten kiezen tussen onmogelijke dilemma\'s',
+            'title': 'EXTRA: Temptation Island hosts moeten kiezen tussen onmogelijke dilemma\'s',
             'description': 'Het spel is simpel: Annelien Coorevits en Rick Brandsteder krijgen telkens 2 dilemma\'s voorgeschoteld en ze MOETEN een keuze maken.',
         },
         'params': {
@@ -38,6 +44,33 @@ class VierIE(InfoExtractor):
             'skip_download': True,
         },
     }, {
+        'url': 'http://www.vier.be/janigaat/videos/jani-gaat-naar-tokio-aflevering-4/2674839',
+        'info_dict': {
+            'id': '2674839',
+            'display_id': 'jani-gaat-naar-tokio-aflevering-4',
+            'ext': 'mp4',
+            'title': 'Jani gaat naar Tokio - Aflevering 4',
+            'description': 'Bekijk hier de volledige vierde aflevering van het 2de seizoen van Jani gaat...',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Requires account credentials',
+    }, {
+        'url': 'http://www.vier.be/janigaat/videos/jani-gaat-naar-tokio-aflevering-4/2674839',
+        'info_dict': {
+            'id': '2674839',
+            'display_id': 'jani-gaat-naar-tokio-aflevering-4',
+            'ext': 'mp4',
+            'title': 'jani-gaat-naar-tokio-aflevering-4',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'expected_warnings': ['Log in to extract metadata'],
+    }, {
         'url': 'http://www.vier.be/planb/videos/mieren-herders-van-de-bladluizen',
         'only_matching': True,
     }, {
@@ -45,17 +78,54 @@ class VierIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _real_initialize(self):
+        self._logged_in = False
+
+    def _login(self, site):
+        username, password = self._get_login_info()
+        if username is None or password is None:
+            return
+
+        login_page = self._download_webpage(
+            'http://www.%s.be/user/login' % site,
+            None, note='Logging in', errnote='Unable to log in',
+            data=urlencode_postdata({
+                'form_id': 'user_login',
+                'name': username,
+                'pass': password,
+            }),
+            headers={'Content-Type': 'application/x-www-form-urlencoded'})
+
+        login_error = self._html_search_regex(
+            r'(?s)<div class="messages error">\s*<div>\s*<h2.+?</h2>(.+?)<',
+            login_page, 'login error', default=None)
+        if login_error:
+            self.report_warning('Unable to log in: %s' % login_error)
+        else:
+            self._logged_in = True
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         embed_id = mobj.group('embed_id')
         display_id = mobj.group('display_id') or embed_id
+        video_id = mobj.group('id') or embed_id
         site = mobj.group('site')
 
+        if not self._logged_in:
+            self._login(site)
+
         webpage = self._download_webpage(url, display_id)
 
+        if r'id="user-login"' in webpage:
+            self.report_warning(
+                'Log in to extract metadata', video_id=display_id)
+            webpage = self._download_webpage(
+                'http://www.%s.be/video/v3/embed/%s' % (site, video_id),
+                display_id)
+
         video_id = self._search_regex(
             [r'data-nid="(\d+)"', r'"nid"\s*:\s*"(\d+)"'],
-            webpage, 'video id')
+            webpage, 'video id', default=video_id)
         application = self._search_regex(
             [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
             webpage, 'application', default=site + '_vod')