]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/crunchyroll.py
Initiate new release.
[youtubedl] / youtube_dl / extractor / crunchyroll.py
index 920728e01f1c2b2a6e6b75c3b6740e84b3043897..f1da7d09bc934af86f08aa45f8a1a3de32fa4673 100644 (file)
@@ -1,41 +1,53 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re, base64, zlib
+import re
+import json
+import base64
+import zlib
+import xml.etree.ElementTree
+
 from hashlib import sha1
 from math import pow, sqrt, floor
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
+from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+)
+from ..utils import (
+    ExtractorError,
     bytes_to_intlist,
     intlist_to_bytes,
     unified_strdate,
-    clean_html,
+    urlencode_postdata,
 )
 from ..aes import (
     aes_cbc_decrypt,
     inc,
 )
 
+
 class CrunchyrollIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
-        'file': '645513.flv',
-        #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
         'info_dict': {
+            'id': '645513',
+            'ext': 'flv',
             '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',
             'uploader': 'Yomiuri Telecasting Corporation (YTV)',
             'upload_date': '20131013',
+            'url': 're:(?!.*&amp)',
         },
         'params': {
             # rtmp
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
+        'only_matching': True,
     }]
 
     _FORMAT_IDS = {
@@ -45,6 +57,24 @@ class CrunchyrollIE(InfoExtractor):
         '1080': ('80', '108'),
     }
 
+    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_request = compat_urllib_request.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')
+
+    def _real_initialize(self):
+        self._login()
+
     def _decrypt_subtitles(self, data, iv, id):
         data = bytes_to_intlist(data)
         iv = bytes_to_intlist(iv)
@@ -68,10 +98,12 @@ class CrunchyrollIE(InfoExtractor):
             shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
             # Extend 160 Bit hash to 256 Bit
             return shaHash + [0] * 12
-        
+
         key = obfuscate_key(id)
+
         class Counter:
             __value = iv
+
             def next_value(self):
                 temp = self.__value
                 self.__value = inc(self.__value)
@@ -79,21 +111,114 @@ class CrunchyrollIE(InfoExtractor):
         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
         return zlib.decompress(decrypted_data)
 
-    def _convert_subtitles_to_srt(self, subtitles):
-        i=1
+    def _convert_subtitles_to_srt(self, sub_root):
         output = ''
-        for start, end, text in re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles):
-            start = start.replace('.', ',')
-            end = end.replace('.', ',')
-            text = clean_html(text)
-            text = text.replace('\\N', '\n')
-            if not text:
-                continue
+
+        for i, event in enumerate(sub_root.findall('./events/event'), 1):
+            start = event.attrib['start'].replace('.', ',')
+            end = event.attrib['end'].replace('.', ',')
+            text = event.attrib['text'].replace('\\N', '\n')
             output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
-            i+=1
         return output
 
-    def _real_extract(self,url):
+    def _convert_subtitles_to_ass(self, sub_root):
+        output = ''
+
+        def ass_bool(strvalue):
+            assvalue = '0'
+            if strvalue == '1':
+                assvalue = '-1'
+            return assvalue
+
+        output = '[Script Info]\n'
+        output += 'Title: %s\n' % sub_root.attrib["title"]
+        output += 'ScriptType: v4.00+\n'
+        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
+
+[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
+"""
+        for style in sub_root.findall('./styles/style'):
+            output += 'Style: ' + style.attrib["name"]
+            output += ',' + style.attrib["font_name"]
+            output += ',' + style.attrib["font_size"]
+            output += ',' + style.attrib["primary_colour"]
+            output += ',' + style.attrib["secondary_colour"]
+            output += ',' + style.attrib["outline_colour"]
+            output += ',' + style.attrib["back_colour"]
+            output += ',' + ass_bool(style.attrib["bold"])
+            output += ',' + ass_bool(style.attrib["italic"])
+            output += ',' + ass_bool(style.attrib["underline"])
+            output += ',' + ass_bool(style.attrib["strikeout"])
+            output += ',' + style.attrib["scale_x"]
+            output += ',' + style.attrib["scale_y"]
+            output += ',' + style.attrib["spacing"]
+            output += ',' + style.attrib["angle"]
+            output += ',' + style.attrib["border_style"]
+            output += ',' + style.attrib["outline"]
+            output += ',' + style.attrib["shadow"]
+            output += ',' + style.attrib["alignment"]
+            output += ',' + style.attrib["margin_l"]
+            output += ',' + style.attrib["margin_r"]
+            output += ',' + style.attrib["margin_v"]
+            output += ',' + style.attrib["encoding"]
+            output += '\n'
+
+        output += """
+[Events]
+Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
+"""
+        for event in sub_root.findall('./events/event'):
+            output += 'Dialogue: 0'
+            output += ',' + event.attrib["start"]
+            output += ',' + event.attrib["end"]
+            output += ',' + event.attrib["style"]
+            output += ',' + event.attrib["name"]
+            output += ',' + event.attrib["margin_l"]
+            output += ',' + event.attrib["margin_r"]
+            output += ',' + event.attrib["margin_v"]
+            output += ',' + event.attrib["effect"]
+            output += ',' + event.attrib["text"]
+            output += '\n'
+
+        return output
+
+    def _get_subtitles(self, video_id, webpage):
+        subtitles = {}
+        for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
+            sub_page = self._download_webpage(
+                'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id,
+                video_id, note='Downloading subtitles for ' + sub_name)
+            id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
+            iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
+            data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
+            if not id or not iv or not data:
+                continue
+            id = int(id)
+            iv = base64.b64decode(iv)
+            data = base64.b64decode(data)
+
+            subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
+            lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
+            if not lang_code:
+                continue
+            sub_root = xml.etree.ElementTree.fromstring(subtitle)
+            subtitles[lang_code] = [
+                {
+                    'ext': 'srt',
+                    'data': self._convert_subtitles_to_srt(sub_root),
+                },
+                {
+                    'ext': 'ass',
+                    'data': self._convert_subtitles_to_ass(sub_root),
+                },
+            ]
+        return subtitles
+
+    def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
 
@@ -108,6 +233,12 @@ class CrunchyrollIE(InfoExtractor):
         if note_m:
             raise ExtractorError(note_m)
 
+        mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
+        if mobj:
+            msg = json.loads(mobj.group('msg'))
+            if msg.get('type') == 'error':
+                raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
+
         video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
         video_title = re.sub(r' {2,}', ' ', video_title)
         video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
@@ -123,56 +254,78 @@ class CrunchyrollIE(InfoExtractor):
         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
         playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
-        
+
         stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
         video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
 
         formats = []
-        for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
+        for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
-            video_format = fmt+'p'
+            video_format = fmt + 'p'
             streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
             # urlencode doesn't work!
-            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
+            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality=' + stream_quality + '&media%5Fid=' + stream_id + '&video%5Fformat=' + stream_format
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
             streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
-            streamdata = self._download_webpage(streamdata_req, video_id, note='Downloading media info for '+video_format)
-            video_url = self._search_regex(r'<host>([^<]+)', streamdata, 'video_url')
-            video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, 'video_play_path')
+            streamdata = self._download_xml(
+                streamdata_req, video_id,
+                note='Downloading media info for %s' % video_format)
+            video_url = streamdata.find('.//host').text
+            video_play_path = streamdata.find('.//file').text
             formats.append({
                 'url': video_url,
-                'play_path':   video_play_path,
+                'play_path': video_play_path,
                 'ext': 'flv',
                 'format': video_format,
                 'format_id': video_format,
             })
 
-        subtitles = {}
-        for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
-            sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
-                                              video_id, note='Downloading subtitles for '+sub_name)
-            id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
-            iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
-            data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
-            if not id or not iv or not data:
-                continue
-            id = int(id)
-            iv = base64.b64decode(iv)
-            data = base64.b64decode(data)
-
-            subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
-            lang_code = self._search_regex(r'lang_code=\'([^\']+)', subtitle, 'subtitle_lang_code', fatal=False)
-            if not lang_code:
-                continue
-            subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
+        subtitles = self.extract_subtitles(video_id, webpage)
 
         return {
-            'id':          video_id,
-            'title':       video_title,
+            'id': video_id,
+            'title': video_title,
             'description': video_description,
-            'thumbnail':   video_thumbnail,
-            'uploader':    video_uploader,
+            'thumbnail': video_thumbnail,
+            'uploader': video_uploader,
             'upload_date': video_upload_date,
-            'subtitles':   subtitles,
-            'formats':     formats,
+            'subtitles': subtitles,
+            'formats': formats,
+        }
+
+
+class CrunchyrollShowPlaylistIE(InfoExtractor):
+    IE_NAME = "crunchyroll:playlist"
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?$'
+
+    _TESTS = [{
+        'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
+        'info_dict': {
+            'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
+            'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
+        },
+        'playlist_count': 13,
+    }]
+
+    def _real_extract(self, url):
+        show_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, show_id)
+        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="([^"]+)"',
+            webpage)
+        entries = [
+            self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
+            for ep in episode_paths
+        ]
+        entries.reverse()
+
+        return {
+            '_type': 'playlist',
+            'id': show_id,
+            'title': title,
+            'entries': entries,
         }