Re: Channels connected to typebased agents.


Subject: Re: Channels connected to typebased agents.
From: Rick Reed TSE (rickreed#tseng.co.uk)
Date: Mon Jan 22 2001 - 11:33:14 GMT


On a related issue (perhaps also covered by TDC33 - I have not checked), it
seems that there are two ways of showing unconnected gates when there is an
agent reference area or graphical typebased agent definition or inherited
agent definition:

a) an unconnected gate symbol outside the block or process symbol;

b) a gate name inside the block or process symbol with no corresponding
connected channel.

However, it is doubtful whether (b) is allowed because of the sentence:
"The <gate>s are placed near the border of the symbols and associated with
the connection point to channels."

I propose that:
a) whenever there is a <gate> inside the symbol there must be a channel
connected to the symbol;

b) a <gate> inside the symbol is mutually exclusive with a gate symbol with
the same name outside the symbol.

The syntax also allows a channel to be connected directly to a gate symbol.
I can see no reason why this would ever be needed. If there is a channel
leading to an agent the gate is not needed. I propose therefore that
explicit gates are never used when channels are directly connected to an
agent.

--
Rick Reed - rickreed#tseng.co.uk
Tel:+44 1455 55 96 55 Fax:+44 1455 55 96 58 Mob.:+44 7970 50 96 50



This archive was generated by hypermail 2a23 : Mon May 05 2008 - 20:30:55 GMT