The javascript would be there as the first line of defense because the action will
only be called once, and the submission will go through (assuming they do not have
jscript turned off of course). If an error occurs in that submission, the appropriate
error handling will occur.
Contrast this to the situation where I check the token and forward somewhere else. I
don't know what happened in the first action submission. It could have been
successful, it could have caused errors, the user will have no idea. All they will
know is that they double submitted. They could have kicked off a long asynchronous
process, but because they were forwarded to "badToken", they decide to go kick it off
again. Do you see now my issue with the token checking? It seems like it really
doesn't buy you a whole lot because you can't cancel the fact that the first
submission cause an action to execute.
-----Original Message-----
From: David Graham [mailto:[EMAIL PROTECTED]]
Sent: Fri 1/31/2003 3:07 PM
To: [EMAIL PROTECTED]
Cc:
Subject: RE: proper use of back button -- design patterns
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
