On Fri, 12 Dec 2003, Khoa Vo wrote: > My purpose is to have the server inform me only new changes. For example, in > my mail box on server there is one new mail just arrived, and three among > 1000 email have changed status to 'read'. Using the IDLE command, and keep > the connection alive, the server will inform any change to that mailbox > using the current connection.
The procedure which I outlined, in which you do a NOOP every so often, is functionally identical to IDLE. IDLE does not do anything different. It just saves you from having to send a NOOP every so often. Instead of sending a NOOP every 2 minutes, you send an IDLE every 29 minutes. > I don't want to scan folder-by-folder, mail-by-mail to detect changes made > to mail items on the server, and update to the mail items on client machine. Even with IDLE, you are never informed about changes to a mailbox that is not SELECTED. So I do not understand your "scan folder-by-folder" argument. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Si vis pacem, para bellum.
