1 files changed, 88 insertions, 0 deletions
diff --git a/module/plugins/hoster/SmuleCom.py b/module/plugins/hoster/SmuleCom.py
new file mode 100644
index 000000000..865b18698
--- /dev/null
+++ b/module/plugins/hoster/SmuleCom.py
@@ -0,0 +1,88 @@
+# -*- coding: utf-8 -*-
+
+import re
+
+from module.common.JsEngine import JsEngine
+from module.plugins.internal.SimpleHoster import SimpleHoster
+
+
+class SmuleCom(SimpleHoster):
+    __name__    = "SmuleCom"
+    __type__    = "hoster"
+    __version__ = "0.02"
+    __status__  = "testing"
+
+    __pattern__ = r'https?://(?:www\.)?smule\.com/recording/.+'
+
+    __description__ = """SmuleCom hoster plugin"""
+    __license__     = "GPLv3"
+    __authors__     = [("igel"     , None                        ),
+                       ("GammaC0de", "nitzo2001[AT]yahoo[DOT]com")]
+
+
+    MEDIA_URL_PATTERN       = r'initPlayer\(.+?["\']video_media_url["\']:["\'](.+?)["\']'
+    JS_HEADER_PATTERN       = r'(?P<decoder>function \w+\(\w+\){.+?ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789\+.+?}).+?;(?P<initvars>var r=.+?;)'
+    JS_PROCESS_PATTERN      = r'processRecording\s*=\s*function.+?}'
+    JS_SPLIT_WORD           = r'EXIF' # all interesting parts of the javascript function occur before the first occurance of this word
+    NAME_PATTERN            = r'initPlayer\(.+?["\']title["\']:["\'](?P<N>.+?)["\']'
+    COMMUNITY_JS_PATTERN    = r'<script.+?src=["\']/*(\w[^"\']*community.+?js)["\']'
+    OFFLINE_PATTERN         = r'Page Not Found'
+
+
+    @classmethod
+    def get_info(cls, url="", html=""):
+        info = super(SmuleCom, cls).get_info(url, html)
+        # Unfortunately, NAME_PATTERN does not include file extension so we blindly add '.mp4' as an extension.
+        # (hopefully all links are '.mp4' files)
+        info['name'] += ".mp4"
+
+        return info
+
+
+    def handle_free(self, pyfile):
+        # step 1: get essential information: the media URL and the javascript translating the URL
+        m = re.search(self.MEDIA_URL_PATTERN, self.data)
+        if m is None:
+            self.fail(_("Could not find any media URLs"))
+
+        encoded_media_url = m.group(1)
+        self.log_debug(_("Found encoded media URL: %s") % encoded_media_url)
+       
+        m = re.search(self.COMMUNITY_JS_PATTERN, self.data)
+        if m is None:
+            self.fail(_("Could not find necessary javascript script to load"))
+
+        community_js_url = m.group(1)
+        self.log_debug(_("Found community js at %s") % community_js_url)
+
+        community_js_code = self.load(community_js_url)
+        
+        # step 2: from the js code, parse the necessary parts: the decoder function and the headers
+        # as the jscript is fairly long, we'll split it to make parsing easier
+        if self.JS_SPLIT_WORD:
+            community_js_code = community_js_code.partition(self.JS_SPLIT_WORD)[0]
+
+        m = re.search(self.JS_HEADER_PATTERN, community_js_code)
+        if m is None:
+            self.fail(_("Could not parse the necessary parts off the javascript"))
+
+        decoder_function = m.group('decoder')
+        initialization = m.group('initvars')
+
+        m = re.search(self.JS_PROCESS_PATTERN, community_js_code)
+        if m is None:
+            self.fail(_("Could not parse the processing function off the javascript"))
+
+        process_function = m.group(0)
+
+        # step 3: assemble the new js code
+        js = JsEngine()
+
+        new_js_code = decoder_function + '; ' + initialization + '; var ' + process_function + '; processRecording("' + encoded_media_url + '");'
+
+        self.log_debug(_("Running js script: %s") % new_js_code)
+        js_result = js.eval(new_js_code)
+        self.log_debug(_("Result is: %s") % js_result)
+
+        self.link = js_result
+