simpler version

It uses some JS to emulate a document - so it'll be simpler to add any JS function in case Zippyshare's devs decide to add deeper instrospection.
It has worked for several months on my side - quite the same code actually.

1 files changed, 22 insertions, 24 deletions

diff --git a/module/plugins/hoster/ZippyshareCom.py b/module/plugins/hoster/ZippyshareCom.py
index b1b084ee8..491bf727a 100644
--- a/module/plugins/hoster/ZippyshareCom.py
+++ b/module/plugins/hoster/ZippyshareCom.py
@@ -12,7 +12,7 @@ from module.plugins.internal.SimpleHoster import SimpleHoster
 class ZippyshareCom(SimpleHoster):
     __name__    = "ZippyshareCom"
     __type__    = "hoster"
-    __version__ = "0.86"
+    __version__ = "0.87"
     __status__  = "testing"
 
     __pattern__ = r'http://www\d{0,3}\.zippyshare\.com/v(/|iew\.jsp.*key=)(?P<KEY>[\w^_]+)'
@@ -66,32 +66,30 @@ class ZippyshareCom(SimpleHoster):
     def get_link(self):
         #: Get all the scripts inside the html body
         soup = BeautifulSoup.BeautifulSoup(self.data)
-        scripts = (s.getText().strip() for s in soup.body.findAll('script', type='text/javascript'))
-
-        #: Meant to be populated with the initialization of all the DOM elements found in the scripts
-        initScripts = set()
-
-        def repl_element_by_id(element):
-            id   = element.group(1)  #: Id might be either 'x' (a real id) or x (a variable)
-            attr = element.group(4)  #: Attr might be None
-
-            varName = re.sub(r'-', '', 'GVAR[%s+"_%s"]' %(id, attr))
-
-            realid = id.strip('"\'')
-            if id is not realid:  #: Id is not a variable, so look for realid.attr in the html
-                initValues = filter(None, [elt.get(attr, None) for elt in soup.findAll(id=realid)])
-                initValue  = '"%s"' % initValues[-1] if initValues else 'null'
-                initScripts.add('%s = %s;' % (varName, initValue))
-
-            return varName
-
-        #: Handle all getElementById
-        reVar = r'document.getElementById\(([\'"\w\-]+)\)(\.)?(getAttribute\([\'"])?(\w+)?([\'"]\))?'
-        scripts = [re.sub(reVar, repl_element_by_id, script) for script in scripts if script]
+        scripts = [s.getText() for s in soup.body.findAll('script', type='text/javascript')]
+
+        #: Emulate a document in JS
+        inits = ['''
+                var document = {}
+                document.getElementById = function(x) {
+                    if (!this.hasOwnProperty(x)) {
+                        this[x] = {getAttribute : function(x) { return this[x] } }
+                    }
+                    return this[x]
+                }
+                ''']
+
+        #: inits is meant to be populated with the initialization of all the DOM elements found in the scripts
+        eltRE = r'getElementById\([\'"](.+?)[\'"]\)(\.)?(getAttribute\([\'"])?(\w+)?([\'"]\))?'
+        for m in re.findall(eltRE, ' '.join(scripts)):
+            JSid, JSattr = m[0], m[3]
+            values = filter(None, (elt.get(JSattr, None) for elt in soup.findAll(id=JSid)))
+            if values:
+                inits.append('document.getElementById("%s")["%s"] = "%s"' %(JSid, JSattr, values[-1]))
 
         #: Add try/catch in JS to handle deliberate errors
         scripts = ['\n'.join(('try{', script, '} catch(err){}')) for script in scripts]
 
         #: Get the file's url by evaluating all the scripts
-        scripts = ["var GVAR = {}"] + list(initScripts)  + scripts + ['GVAR["dlbutton_href"]']
+        scripts = inits + scripts + ['document.dlbutton.href']
         return self.js.eval('\n'.join(scripts))