I have to ask, why is it such a sin to have a stand alone program with a
null modem cable and mini-tester as a testing tool. I have been working
with serial ports off and on since the days of DOS 3.x and have always
tested in this manner. The first piece of code one writes on a project
which needs external inputs is the test tool to generate 1-N pre-defined
inputs. There "should" be one laying around at the shop which was
originally used to validate the product.
If you are dealing with a bug that only bites you ever so many years, I
will leave you with this thought.
Having written a lot of scale software early in my career it was never
the packet which had the problem that actually caused the problem, it
was always a data packet 1-N in front of that packet. This happened
because scale manufacturers would use bounding characters for differing
packet types and they dutifully calculated a checksum or CRC for the
packet. Inevitably that byte would collide with a value chosen for one
of the bounding characters and it would be a seemingly random problem
usually resulting from a truck being much lighter or heavier than any
value they used to bench test their error checking byte with.
No idea what your serial project is, but that was always the problem we
ran into in the field, especially when someone brought in a shiny new
scale vendor or model.
If your system can be dynamically configured in the field, you should
add a feature to keep track of the last 10 packets and when the code
recognizes/traps the error (you obviously have some idea of what and
where it manifests since you know it is a bug) dump the hex value of the
last 10 packets to a log.
Won't fix your problem today or quickly, but it will actually identify
what is causing the problem.
Or perhaps you have already done that and I'm just rambling...
--
Roland Hughes, President
Logikal Solutions
(630)-205-1593
http://www.theminimumyouneedtoknow.com
http://www.infiniteexposure.net
http://www.johnsmith-book.com
http://www.logikalblog.com
http://www.interestingauthors.com/blog
http://lesedi.us/
_______________________________________________
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest
