Experimental Physics and
Industrial Control System

[email protected] (William Lupton) · Wed, 17 Apr 96 16:13:10 HST

----- Begin Included Message -----

>From wlupton Wed Apr 17 15:56:07 1996
To: [email protected]
Subject: Re: rewrite of et_wish
Status: O

Claude,

> I am currently reviewing and reworking the et_wish library (the
> channel access extension to tcl/tk). It will be renamed tclCa...

Sounds good. I have a few suggestions (the first is specific to a problem
that I had in my _very_ limited use of et):

1. In a simple example, I found that a monitor, although it changed the
   specified Tcl variable, did not automatically update a label that
   referenced it via -textvariable. I had to do something like {set v
   $v} in the monitor callback in order to get this to work.

2. It would be good to maximize support for event-driven applications
   (e.g.  as well as supporting connection state change callbacks,
   support get and put completion callbacks rather than having to use
   "pv stat" to poll for completion).

3. What about passing automatic extra arguments to callback routines,
   e.g. the name of the linked Tcl variable, an error code (rather than
   having to use the global errorCode), ... (or specify a different
   routine to call in the event of error?)? This might make it neater to
   write generic callback routines, and should be upwards compatible.

William

----- End Included Message -----

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System