While doing some profiling in a Java application accessing MarkLogic I noticed 
that the XCC classes don't use the Java URLEncoder class to encode a string but 
instead use a custom IOHelper class.  This class will use the URLEncoder class 
if the string to be encoded contains non-ascii characters, otherwise it follows 
similar logic to that of the URLEncoder class.

The custom encoding takes a good deal longer to process than the URLEncoder 
class does.  So I ask, is there a particular reason for the custom encoding?  
Is this just code left over from when either URLEncoder didn't exist or wasn't 
quite as fast?  I ask because this is my performance bottleneck, oddly enough 
(a large simple query/page taking 5 seconds, where 98% of that time is the 
encoding, which doesn't seem like it would scale too well).

Thanks.

Chris Cieslinski



 NOTICE: This email message is for the sole use of the intended recipient(s) 
and may contain confidential and privileged information. Any unauthorized 
review, use, disclosure or distribution is prohibited. If you are not the 
intended recipient, please contact the sender by reply email and destroy all 
copies of the original message.


_______________________________________________
General mailing list
[email protected]
http://xqzone.com/mailman/listinfo/general

Reply via email to