[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/hidive.py b/youtube_dl/extractor/hidive.py
index 1c70b6646e3230f40d31e9942123505a3590def7..f26f80265678576217994746d9286482c5184a3d 100644 (file)
--- a/youtube_dl/extractor/hidive.py
+++ b/youtube_dl/extractor/hidive.py
@@ -8,6 +8,7 @@ from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     int_or_none,
+    url_or_none,
     urlencode_postdata,
 )
 
@@ -17,6 +18,8 @@ class HiDiveIE(InfoExtractor):
     # Using X-Forwarded-For results in 403 HTTP error for HLS fragments,
     # so disabling geo bypass completely
     _GEO_BYPASS = False
+    _NETRC_MACHINE = 'hidive'
+    _LOGIN_URL = 'https://www.hidive.com/account/login'
 
     _TESTS = [{
         'url': 'https://www.hidive.com/stream/the-comic-artist-and-his-assistants/s01e001',
@@ -30,10 +33,27 @@ class HiDiveIE(InfoExtractor):
         },
         'params': {
             'skip_download': True,
-            'proxy': '192.99.245.228:3128',
         },
+        'skip': 'Requires Authentication',
     }]
 
+    def _real_initialize(self):
+        email, password = self._get_login_info()
+        if email is None:
+            return
+
+        webpage = self._download_webpage(self._LOGIN_URL, None)
+        form = self._search_regex(
+            r'(?s)<form[^>]+action="/account/login"[^>]*>(.+?)</form>',
+            webpage, 'login form')
+        data = self._hidden_inputs(form)
+        data.update({
+            'Email': email,
+            'Password': password,
+        })
+        self._download_webpage(
+            self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(data))
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         title, key = mobj.group('title', 'key')
@@ -44,6 +64,7 @@ class HiDiveIE(InfoExtractor):
             data=urlencode_postdata({
                 'Title': title,
                 'Key': key,
+                'PlayerId': 'f4f895ce1ca713ba263b91caeb1daa2d08904783',
             }))
 
         restriction = settings.get('restrictionReason')
@@ -60,8 +81,8 @@ class HiDiveIE(InfoExtractor):
             bitrates = rendition.get('bitrates')
             if not isinstance(bitrates, dict):
                 continue
-            m3u8_url = bitrates.get('hls')
-            if not isinstance(m3u8_url, compat_str):
+            m3u8_url = url_or_none(bitrates.get('hls'))
+            if not m3u8_url:
                 continue
             formats.extend(self._extract_m3u8_formats(
                 m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
@@ -73,13 +94,13 @@ class HiDiveIE(InfoExtractor):
                 if not isinstance(cc_file, list) or len(cc_file) < 3:
                     continue
                 cc_lang = cc_file[0]
-                cc_url = cc_file[2]
-                if not isinstance(cc_lang, compat_str) or not isinstance(
-                        cc_url, compat_str):
+                cc_url = url_or_none(cc_file[2])
+                if not isinstance(cc_lang, compat_str) or not cc_url:
                     continue
                 subtitles.setdefault(cc_lang, []).append({
                     'url': cc_url,
                 })
+        self._sort_formats(formats)
 
         season_number = int_or_none(self._search_regex(
             r's(\d+)', key, 'season number', default=None))