Experimental Physics and
Industrial Control System

Mark Rivers <[email protected]> · Wed, 19 Jul 2017 16:58:55 +0000

It's not that easy.  In order to use the asynTrace interface in asyn one needs to pass it the pasynUser that is connected to the current asyn port and address.  However, that is hidden as a private member variable of the AsynDriverInterface class in StreamDevice.  Each record has a different AsynDriverInterface object.  The StreamError function which is used to print StreamDevice error messages has no knowledge of the AsynDriverInterface object.  It is not clear to me that this information could be made available to the StreamError function without a lot of refactoring.

Mark

-----Original Message-----
From: J. Lewis Muir [mailto:[email protected]] 
Sent: Wednesday, July 19, 2017 10:52 AM
To: Mark Rivers
Cc: Paduan Donadio, Marcio; [email protected]
Subject: Re: Time interval for Asyn auto reconnect

On 07/19, Mark Rivers wrote:
> streamDevice uses asyn for I/O but it does not use asyn for printing
> error messages. That is unfortunate because if it did you would have
> the ability to turn off the messages using asynTrace control.  I have
> suggested this to Dirk but it would introduce asyn dependencies into
> parts of streamDevice code that are currently independent of asyn,
> which he is reluctant to do.

Sounds like a good candidate for the Bridge design pattern.  Create
an interface for the logging, and provide two implementations: the
StreamDevice built-in implementation and the Asyn implementation.  Those
parts of StreamDevice that know nothing about Asyn will continue to know
nothing about Asyn, yet the Asyn implementation could be used.

Lewis

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System