[roosterteeth] Improve (Closes #9864)
authorSergey M․ <dstftw@gmail.com>
Sat, 9 Jul 2016 18:28:28 +0000 (01:28 +0700)
committerSergey M․ <dstftw@gmail.com>
Sat, 9 Jul 2016 18:30:12 +0000 (01:30 +0700)
youtube_dl/extractor/roosterteeth.py

index 4053747d3a994ccbea5165b7746775ebe5b97e60..f5b2f560c7f70c4e341aaf38f9718ea4994b811b 100644 (file)
@@ -1,9 +1,14 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    int_or_none,
+    strip_or_none,
+    unescapeHTML,
     urlencode_postdata,
 )
 
@@ -14,19 +19,17 @@ class RoosterTeethIE(InfoExtractor):
     _NETRC_MACHINE = 'roosterteeth'
     _TESTS = [{
         'url': 'http://roosterteeth.com/episode/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
+        'md5': 'e2bd7764732d785ef797700a2489f212',
         'info_dict': {
             'id': '26576',
+            'display_id': 'million-dollars-but-season-2-million-dollars-but-the-game-announcement',
             'ext': 'mp4',
-            'title': 'Million Dollars, But... The Game Announcement',
+            'title': 'Million Dollars, But...: Million Dollars, But... The Game Announcement',
+            'description': 'md5:0cc3b21986d54ed815f5faeccd9a9ca5',
             'thumbnail': 're:^https?://.*\.png$',
-            'description': 'Introducing Million Dollars, But... The Game! Available for pre-order now at www.MDBGame.com ',
-            'creator': 'Rooster Teeth',
             'series': 'Million Dollars, But...',
             'episode': 'Million Dollars, But... The Game Announcement',
-            'episode_id': '26576',
-        },
-        'params': {
-            'skip_download': True,  # m3u8 downloads
+            'comment_count': int,
         },
     }, {
         'url': 'http://achievementhunter.roosterteeth.com/episode/off-topic-the-achievement-hunter-podcast-2016-i-didn-t-think-it-would-pass-31',
@@ -40,73 +43,106 @@ class RoosterTeethIE(InfoExtractor):
     }, {
         'url': 'http://theknow.roosterteeth.com/episode/the-know-game-news-season-1-boring-steam-sales-are-better',
         'only_matching': True,
+    }, {
+        # only available for FIRST members
+        'url': 'http://roosterteeth.com/episode/rt-docs-the-world-s-greatest-head-massage-the-world-s-greatest-head-massage-an-asmr-journey-part-one',
+        'only_matching': True,
     }]
 
     def _login(self):
         (username, password) = self._get_login_info()
-        if username is None or password is None:
-            return False
+        if username is None:
+            return
 
-        # token is required to authenticate request
-        login_page = self._download_webpage(self._LOGIN_URL, None, 'Getting login token', 'Unable to get login token')
+        login_page = self._download_webpage(
+            self._LOGIN_URL, None,
+            note='Downloading login page',
+            errnote='Unable to download login page')
 
         login_form = self._hidden_inputs(login_page)
+
         login_form.update({
             'username': username,
             'password': password,
         })
-        login_payload = urlencode_postdata(login_form)
-
-        # required for proper responses
-        login_headers = {
-            'Referer': self._LOGIN_URL,
-        }
 
         login_request = self._download_webpage(
             self._LOGIN_URL, None,
             note='Logging in as %s' % username,
-            data=login_payload,
-            headers=login_headers)
-
-        if 'Authentication failed' in login_request:
-            raise ExtractorError(
-                'Login failed (invalid username/password)', expected=True)
+            data=urlencode_postdata(login_form),
+            headers={
+                'Referer': self._LOGIN_URL,
+            })
+
+        if not any(re.search(p, login_request) for p in (
+                r'href=["\']https?://(?:www\.)?roosterteeth\.com/logout"',
+                r'>Sign Out<')):
+            error = self._html_search_regex(
+                r'(?s)<div[^>]+class=(["\']).*?\balert-danger\b.*?\1[^>]*>(?:\s*<button[^>]*>.*?</button>)?(?P<error>.+?)</div>',
+                login_request, 'alert', default=None, group='error')
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
+            raise ExtractorError('Unable to log in')
 
     def _real_initialize(self):
         self._login()
 
     def _real_extract(self, url):
-        match_id = self._match_id(url)
-        webpage = self._download_webpage(url, match_id)
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
 
-        episode_id = self._html_search_regex(r"commentControls\('#comment-([0-9]+)'\)", webpage, 'episode id', match_id, False)
+        episode = strip_or_none(unescapeHTML(self._search_regex(
+            (r'videoTitle\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'<title>(?P<title>[^<]+)</title>'), webpage, 'title',
+            default=None, group='title')))
 
-        self.report_extraction(episode_id)
+        title = strip_or_none(self._og_search_title(
+            webpage, default=None)) or episode
 
-        title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'episode title', self._og_search_title(webpage), False)
-        thumbnail = self._og_search_thumbnail(webpage)
-        description = self._og_search_description(webpage)
-        creator = self._html_search_regex(r'<h3>Latest (.+) Gear</h3>', webpage, 'site', 'Rooster Teeth', False)
-        series = self._html_search_regex(r'<h2>More ([^<]+)</h2>', webpage, 'series', fatal=False)
-        episode = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'episode title', fatal=False)
+        m3u8_url = self._search_regex(
+            r'file\s*:\s*(["\'])(?P<url>http.+?\.m3u8.*?)\1',
+            webpage, 'm3u8 url', default=None, group='url')
 
-        if '<div class="non-sponsor">' in webpage:
-            self.raise_login_required('%s is only available for FIRST members' % title)
+        if not m3u8_url:
+            if re.search(r'<div[^>]+class=["\']non-sponsor', webpage):
+                self.raise_login_required(
+                    '%s is only available for FIRST members' % display_id)
 
-        if '<div class="golive-gate">' in webpage:
-            self.raise_login_required('%s is not available yet' % title)
+            if re.search(r'<div[^>]+class=["\']golive-gate', webpage):
+                self.raise_login_required('%s is not available yet' % display_id)
 
-        formats = self._extract_m3u8_formats(self._html_search_regex(r"file: '(.+?)m3u8'", webpage, 'm3u8 url') + 'm3u8', episode_id, ext='mp4')
+            raise ExtractorError('Unable to extract m3u8 URL')
+
+        formats = self._extract_m3u8_formats(
+            m3u8_url, display_id, ext='mp4',
+            entry_protocol='m3u8_native', m3u8_id='hls')
         self._sort_formats(formats)
 
+        description = strip_or_none(self._og_search_description(webpage))
+        thumbnail = self._proto_relative_url(self._og_search_thumbnail(webpage))
+
+        series = self._search_regex(
+            (r'<h2>More ([^<]+)</h2>', r'<a[^>]+>See All ([^<]+) Videos<'),
+            webpage, 'series', fatal=False)
+
+        comment_count = int_or_none(self._search_regex(
+            r'>Comments \((\d+)\)<', webpage,
+            'comment count', fatal=False))
+
+        video_id = self._search_regex(
+            (r'containerId\s*=\s*["\']episode-(\d+)\1',
+             r'<div[^<]+id=["\']episode-(\d+)'), webpage,
+            'video id', default=display_id)
+
         return {
-            'id': episode_id,
+            'id': video_id,
+            'display_id': display_id,
             'title': title,
-            'formats': formats,
-            'thumbnail': thumbnail,
             'description': description,
-            'creator': creator,
+            'thumbnail': thumbnail,
             'series': series,
             'episode': episode,
-            'episode_id': episode_id,
+            'comment_count': comment_count,
+            'formats': formats,
         }