Prepare for release

[youtubedl] / youtube_dl / extractor / crunchyroll.py

diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index 90a64303d3282c925d1620787922622e99a8b612..8bdaf0c2c5af7aa1a562033487a16867c3304f19 100644 (file)
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -1,4 +1,4 @@
-# encoding: utf-8
+# coding: utf-8

 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re


@@ -34,22 +34,58 @@ from ..aes import (
 
 
 class CrunchyrollBaseIE(InfoExtractor):
 
 
 class CrunchyrollBaseIE(InfoExtractor):

+    _LOGIN_URL = 'https://www.crunchyroll.com/login'
+    _LOGIN_FORM = 'login_form'

     _NETRC_MACHINE = 'crunchyroll'
 
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
             return
     _NETRC_MACHINE = 'crunchyroll'
 
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
             return

-        self.report_login()
-        login_url = 'https://www.crunchyroll.com/?a=formhandler'
-        data = urlencode_postdata({
-            'formname': 'RpcApiUser_Login',
-            'name': username,
-            'password': password,
+
+        login_page = self._download_webpage(
+            self._LOGIN_URL, None, 'Downloading login page')
+
+        def is_logged(webpage):
+            return '<title>Redirecting' in webpage
+
+        # Already logged in
+        if is_logged(login_page):
+            return
+
+        login_form_str = self._search_regex(
+            r'(?P<form><form[^>]+?id=(["\'])%s\2[^>]*>)' % self._LOGIN_FORM,
+            login_page, 'login form', group='form')
+
+        post_url = extract_attributes(login_form_str).get('action')
+        if not post_url:
+            post_url = self._LOGIN_URL
+        elif not post_url.startswith('http'):
+            post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
+
+        login_form = self._form_hidden_inputs(self._LOGIN_FORM, login_page)
+
+        login_form.update({
+            'login_form[name]': username,
+            'login_form[password]': password,

         })
         })

-        login_request = sanitized_Request(login_url, data)
-        login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        self._download_webpage(login_request, None, False, 'Wrong login info')
+
+        response = self._download_webpage(
+            post_url, None, 'Logging in', 'Wrong login info',
+            data=urlencode_postdata(login_form),
+            headers={'Content-Type': 'application/x-www-form-urlencoded'})
+
+        # Successful login
+        if is_logged(response):
+            return
+
+        error = self._html_search_regex(
+            '(?s)<ul[^>]+class=["\']messages["\'][^>]*>(.+?)</ul>',
+            response, 'error message', default=None)
+        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_initialize(self):
         self._login()


@@ -87,7 +123,7 @@ class CrunchyrollIE(CrunchyrollBaseIE):
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
         'info_dict': {
             'id': '645513',
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
         'info_dict': {
             'id': '645513',

-            'ext': 'flv',
+            'ext': 'mp4',

             'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
             'description': 'md5:2d17137920c64f2f49981a7797d275ef',
             'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
             'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
             'description': 'md5:2d17137920c64f2f49981a7797d275ef',
             'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',


@@ -106,7 +142,7 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             'ext': 'flv',
             'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
             'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
             'ext': 'flv',
             'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
             'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',

-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',

             'uploader': 'Danny Choo Network',
             'upload_date': '20120213',
         },
             'uploader': 'Danny Choo Network',
             'upload_date': '20120213',
         },


@@ -114,6 +150,42 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             # rtmp
             'skip_download': True,
         },
             # rtmp
             'skip_download': True,
         },

+        'skip': 'Video gone',
+    }, {
+        'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409',
+        'info_dict': {
+            'id': '702409',
+            'ext': 'mp4',
+            'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant',
+            'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'TV TOKYO',
+            'upload_date': '20160508',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.crunchyroll.com/konosuba-gods-blessing-on-this-wonderful-world/episode-1-give-me-deliverance-from-this-judicial-injustice-727589',
+        'info_dict': {
+            'id': '727589',
+            'ext': 'mp4',
+            'title': "KONOSUBA -God's blessing on this wonderful world! 2 Episode 1 – Give Me Deliverance From This Judicial Injustice!",
+            'description': 'md5:cbcf05e528124b0f3a0a419fc805ea7d',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Kadokawa Pictures Inc.',
+            'upload_date': '20170118',
+            'series': "KONOSUBA -God's blessing on this wonderful world!",
+            'season': "KONOSUBA -God's blessing on this wonderful world! 2",
+            'season_number': 2,
+            'episode': 'Give Me Deliverance From This Judicial Injustice!',
+            'episode_number': 1,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },

     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,
     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,


@@ -121,6 +193,53 @@ class CrunchyrollIE(CrunchyrollBaseIE):
         # geo-restricted (US), 18+ maturity wall, non-premium available
         'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
         'only_matching': True,
         # geo-restricted (US), 18+ maturity wall, non-premium available
         'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
         'only_matching': True,

+    }, {
+        # A description with double quotes
+        'url': 'http://www.crunchyroll.com/11eyes/episode-1-piros-jszaka-red-night-535080',
+        'info_dict': {
+            'id': '535080',
+            'ext': 'mp4',
+            'title': '11eyes Episode 1 – Piros éjszaka - Red Night',
+            'description': 'Kakeru and Yuka are thrown into an alternate nightmarish world they call "Red Night".',
+            'uploader': 'Marvelous AQL Inc.',
+            'upload_date': '20091021',
+        },
+        'params': {
+            # Just test metadata extraction
+            'skip_download': True,
+        },
+    }, {
+        # make sure we can extract an uploader name that's not a link
+        'url': 'http://www.crunchyroll.com/hakuoki-reimeiroku/episode-1-dawn-of-the-divine-warriors-606899',
+        'info_dict': {
+            'id': '606899',
+            'ext': 'mp4',
+            'title': 'Hakuoki Reimeiroku Episode 1 – Dawn of the Divine Warriors',
+            'description': 'Ryunosuke was left to die, but Serizawa-san asked him a simple question "Do you want to live?"',
+            'uploader': 'Geneon Entertainment',
+            'upload_date': '20120717',
+        },
+        'params': {
+            # just test metadata extraction
+            'skip_download': True,
+        },
+    }, {
+        # A video with a vastly different season name compared to the series name
+        'url': 'http://www.crunchyroll.com/nyarko-san-another-crawling-chaos/episode-1-test-590532',
+        'info_dict': {
+            'id': '590532',
+            'ext': 'mp4',
+            'title': 'Haiyoru! Nyaruani (ONA) Episode 1 – Test',
+            'description': 'Mahiro and Nyaruko talk about official certification.',
+            'uploader': 'TV TOKYO',
+            'upload_date': '20120305',
+            'series': 'Nyarko-san: Another Crawling Chaos',
+            'season': 'Haiyoru! Nyaruani (ONA)',
+        },
+        'params': {
+            # Just test metadata extraction
+            'skip_download': True,
+        },

     }]
 
     _FORMAT_IDS = {
     }]
 
     _FORMAT_IDS = {


@@ -184,8 +303,7 @@ class CrunchyrollIE(CrunchyrollBaseIE):
         output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style']
         output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x']
         output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y']
         output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style']
         output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x']
         output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y']

-        output += """ScaledBorderAndShadow: yes
-
+        output += """

 [V4+ Styles]
 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 """
 [V4+ Styles]
 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 """


@@ -272,7 +390,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         else:
             webpage_url = 'http://www.' + mobj.group('url')
 
         else:
             webpage_url = 'http://www.' + mobj.group('url')
 

-        webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage')
+        webpage = self._download_webpage(
+            self._add_skip_wall(webpage_url), video_id,
+            headers=self.geo_verification_headers())

         note_m = self._html_search_regex(
             r'<div class="showmedia-trailer-notice">(.+?)</div>',
             webpage, 'trailer-notice', default='')
         note_m = self._html_search_regex(
             r'<div class="showmedia-trailer-notice">(.+?)</div>',
             webpage, 'trailer-notice', default='')


@@ -292,9 +412,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
             webpage, 'video_title')
         video_title = re.sub(r' {2,}', ' ', video_title)
             r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
             webpage, 'video_title')
         video_title = re.sub(r' {2,}', ' ', video_title)

-        video_description = self._html_search_regex(
-            r'<script[^>]*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id,
-            webpage, 'description', default=None)
+        video_description = self._parse_json(self._html_search_regex(
+            r'<script[^>]*>\s*.+?\[media_id=%s\].+?({.+?"description"\s*:.+?})\);' % video_id,
+            webpage, 'description', default='{}'), video_id).get('description')

         if video_description:
             video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
         video_upload_date = self._html_search_regex(
         if video_description:
             video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
         video_upload_date = self._html_search_regex(


@@ -303,8 +423,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         if video_upload_date:
             video_upload_date = unified_strdate(video_upload_date)
         video_uploader = self._html_search_regex(
         if video_upload_date:
             video_upload_date = unified_strdate(video_upload_date)
         video_uploader = self._html_search_regex(

-            r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
-            'video_uploader', fatal=False)
+            # try looking for both an uploader that's a link and one that's not
+            [r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', r'<div>\s*Publisher:\s*<span>\s*(.+?)\s*</span>\s*</div>'],
+            webpage, 'video_uploader', fatal=False)

 
         available_fmts = []
         for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
 
         available_fmts = []
         for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):


@@ -336,9 +457,18 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             if video_encode_id in video_encode_ids:
                 continue
             video_encode_ids.append(video_encode_id)
             if video_encode_id in video_encode_ids:
                 continue
             video_encode_ids.append(video_encode_id)

+
+            video_file = xpath_text(stream_info, './file')
+            if not video_file:
+                continue
+            if video_file.startswith('http'):
+                formats.extend(self._extract_m3u8_formats(
+                    video_file, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+                continue
+

             video_url = xpath_text(stream_info, './host')
             video_url = xpath_text(stream_info, './host')

-            video_play_path = xpath_text(stream_info, './file')
-            if not video_url or not video_play_path:
+            if not video_url:

                 continue
             metadata = stream_info.find('./metadata')
             format_info = {
                 continue
             metadata = stream_info.find('./metadata')
             format_info = {


@@ -353,7 +483,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 parsed_video_url = compat_urlparse.urlparse(video_url)
                 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
                     netloc='v.lvlt.crcdn.net',
                 parsed_video_url = compat_urlparse.urlparse(video_url)
                 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
                     netloc='v.lvlt.crcdn.net',

-                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1])))
+                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))

                 if self._is_valid_url(direct_video_url, video_id, video_format):
                     format_info.update({
                         'url': direct_video_url,
                 if self._is_valid_url(direct_video_url, video_id, video_format):
                     format_info.update({
                         'url': direct_video_url,


@@ -363,7 +493,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
             format_info.update({
                 'url': video_url,
 
             format_info.update({
                 'url': video_url,

-                'play_path': video_play_path,
+                'play_path': video_file,

                 'ext': 'flv',
             })
             formats.append(format_info)
                 'ext': 'flv',
             })
             formats.append(format_info)


@@ -378,6 +508,19 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         subtitles = self.extract_subtitles(video_id, webpage)
 
 
         subtitles = self.extract_subtitles(video_id, webpage)
 

+        # webpage provide more accurate data than series_title from XML
+        series = self._html_search_regex(
+            r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
+            webpage, 'series', fatal=False)
+        season = xpath_text(metadata, 'series_title')
+
+        episode = xpath_text(metadata, 'episode_title')
+        episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
+
+        season_number = int_or_none(self._search_regex(
+            r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
+            webpage, 'season number', default=None))
+

         return {
             'id': video_id,
             'title': video_title,
         return {
             'id': video_id,
             'title': video_title,


@@ -385,9 +528,11 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             'thumbnail': xpath_text(metadata, 'episode_image_url'),
             'uploader': video_uploader,
             'upload_date': video_upload_date,
             'thumbnail': xpath_text(metadata, 'episode_image_url'),
             'uploader': video_uploader,
             'upload_date': video_upload_date,

-            'series': xpath_text(metadata, 'series_title'),
-            'episode': xpath_text(metadata, 'episode_title'),
-            'episode_number': int_or_none(xpath_text(metadata, 'episode_number')),
+            'series': series,
+            'season': season,
+            'season_number': season_number,
+            'episode': episode,
+            'episode_number': episode_number,

             'subtitles': subtitles,
             'formats': formats,
         }
             'subtitles': subtitles,
             'formats': formats,
         }


@@ -422,16 +567,18 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
     def _real_extract(self, url):
         show_id = self._match_id(url)
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
 

-        webpage = self._download_webpage(self._add_skip_wall(url), show_id)
+        webpage = self._download_webpage(
+            self._add_skip_wall(url), show_id,
+            headers=self.geo_verification_headers())

         title = self._html_search_regex(
             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
             webpage, 'title')
         episode_paths = re.findall(
         title = self._html_search_regex(
             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
             webpage, 'title')
         episode_paths = re.findall(

-            r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
+            r'(?s)<li id="showview_videos_media_(\d+)"[^>]+>.*?<a href="([^"]+)"',

             webpage)
         entries = [
             webpage)
         entries = [

-            self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
-            for ep in episode_paths
+            self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll', ep_id)
+            for ep_id, ep in episode_paths

         ]
         entries.reverse()
 
         ]
         entries.reverse()