I am not quite sure if it is worth it without a real speed benefit as it makes the code more complicated.
Opinions?
Oliver
[EMAIL PROTECTED] wrote:
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=30442>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=30442
How to make slide faster - submission
------- Additional Comments From [EMAIL PROTECTED] 2004-08-04 08:35 ------- First of all, your contribution really is appreciated. I am just particular picky about the details as your patch concerns the very core the Slide kernel and the database store. I hope you understand that...
As I am wondering whether the the binding optimization really significantly adds
to performance in the first place, I would propose to restrict ourselves to the
properties first.
I would propose to store the properties in NodeRevisionDescriptor along with the composed namspace/name in a map. The only modification in StandardRDBMSAdapter would then be method storeRevisionDescriptor. There you iterate over the deleted properties and delete from with a delete statement. This should be ok as I would not expect them to be too many. Then you iterate over the updated properties and - instead of doing a delete followed by an insert - you do an UPDATE which of course would be much faster.
How does that sound? I will add a proposal based on yours (Tara's) myself. Could you review it and tell me if you are satisfied with it?
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
