fix for pdf that has %PDF in first 1024 bytes instead of at very begi…

…nning and better error message in case of missing js

jsunpackn.py

@@ -385,7 +385,7 @@ def decodeJShelper(self,to_write,fixErrors=3):
                     else:
                         self.rooturl[self.url].log(self.OPTIONS.veryverbose,-1,'%s' % re.sub('\n','\n\terror: ',errors))
         except Exception, e:
-            self.rooturl[self.url].log(self.OPTIONS.veryverbose,-1,'Error: Fatal error in decodeJS: %s' % e)
+            self.rooturl[self.url].log(self.OPTIONS.veryverbose,-1,'Error: Fatal error in decodeJS: %s (probably you are missing "js" in your path)' % e)
             return '',(time.time()-begin)
 
         return decoded,(time.time()-begin)
@@ -966,7 +966,7 @@ def main_decoder(self, data, url, tcpaddr = [], lastModified = ''):
 
 
         pdfjs,pdfjs_header = '',''
-        if data.startswith('%PDF-') or data.startswith('%%PDF-'):
+        if 0 <= data[0:1024].find('%PDF-') <= 1024:
             isPDF = True
             self.rooturl[self.url].filetype = 'PDF'
             mypdf = pdf.pdf(data,'PDF-'+self.url)

pdf.py

@@ -510,7 +510,7 @@ def decryptRC4(self,data,key):
 
 
     def is_valid(self):
-        if 0 <= self.indata.find('%PDF-') <= 1024:
+        if 0 <= self.indata[0:1024].find('%PDF-') <= 1024:
             return True
         return False