Worth noting that for a web app, if you are going to lock records then
those locks must timeout after a while- otherwise someone can leave
something locked forever, and there is no way for you to detect for
sure that they are gone.
And you do? MySQL does? I don't think so.
The internet is stateless. If you want to build a custom record
locking
solution for web transactions, you can do it and you will face the
same
issues in any system. It only takes a very simple calculation in edit
privileges to accomplish.
uh... I use XML and a little array. when record is accessed, the
arrays (hum... actually is not an array, it's a queue) is checked, if
the record is in the arrays, this meaning it's being edited right now,
then it cannot be accessed, if it's not in the array, then it's
unlocked. This is done before any DB connection function. It works
fine. what do you think?
_______________________________________________
use-revolution mailing list
[EMAIL PROTECTED]
http://lists.runrev.com/mailman/listinfo/use-revolution
