SCP Template Reset Conformance

, ,
1

Imported from previous forum

2

[ original email was from Steven Grossman - steven.grossman@transacttools.net ]
The SCP 1.1 draft 5 specification states that an implementation “must not use any other reset method than through a message with a template having the reset property set to true”. This correlates with a thread on this forum which dismisses the practice of implicit resets between datagrams, favoring an explicit reset notice from the sender.

This rule would also imply that when using SCP over a stream-based protocol (i.e. TCP), dictionaries should not be reset on session begin or end – only when explicitly instructed to do so by the sender. However, in the last working group conference, it was pointed out that this is not the case.

If SCP insists upon explicit resets in order to keep decoders protocol-agnostic, why do stream-based protocols benefit from implicit resets?

Steven

3

I think it is a good point and for consistency I think any level of conformant SCP usage should start processing with a resetting template.

Further I suggest that the reset property is enabled for the Hello message so that it still is a valid first message. Also, a conformant processor should report an error if the first template encountered was indeed not resetting.

/David

The SCP 1.1 draft 5 specification states that an implementation “must
not use any other reset method than through a message with a template
having the reset property set to true”. This correlates with a thread on
this forum which dismisses the practice of implicit resets between
datagrams, favoring an explicit reset notice from the sender.

This rule would also imply that when using SCP over a stream-based
protocol (i.e. TCP), dictionaries should not be reset on session begin
or end – only when explicitly instructed to do so by the sender.
However, in the last working group conference, it was pointed out that
this is not the case.

If SCP insists upon explicit resets in order to keep decoders protocol-
agnostic, why do stream-based protocols benefit from implicit resets?

Steven