Building the Right Environment to Support AI, Machine Learning and Deep Learning
To use COMMSPY, you'll need a "break-out" cable. Connect two of its ends between the two devices communicating, and the third end to either COMM1 or COMM2 of the PC on which COMMSPY is running. Then update the communication settings, setting the port to either COMM1 or COMM2, whichever you hooked the third end of the cable to. This is the COMMSPY main window:
It is intentionally laid out to look like an oscilloscope. The window on the left is the serial port output, in the middle the serial port settings, on the right the display settings, and on the bottom the command buttons. To start monitoring the serial port, simply select "Start!":
Figure 2: COMMSPY main window while monitoring
The serial port output window changes to full dialog mode, hiding unused controls. To stop monitoring, select "Stop!".
I ran into some problems while developing COMMSPY. First, I had problems
getting the thread to terminate. Since its sitting inside WaitForCommEvent(),
the only way to communicate with the thread is through a comm event. It turns
out that WaitForCommEvent() will instantly return if the comm mask is changed with
SetCommMask(). I was doing that, but it wasn't working. Then I noticed that an OVERLAPPED structure must be passed to WaitForCommEvent() in order for that to work correctly. Second, the controls on the dialog were slow to repaint when monitoring started. This was because I was originally connecting to the COM port AND starting the thread, then rearranging controls. When the second threads starts, it and the main thread start competing for CPU cycles, which slows down the repaint of the controls. When I split the monitoring functionality (GCommMonitor) up (into Connect() and Monitor()), and rearranged the controls in-between calling those two functions, everything worked smoothly.
Below is a few diagrams which will demonstrate a few possible configurations of the "break-out" cable. The below samples assume you have one or more PCs, and some external device with which it is (the are) communicating:
Figure 3: PC#1 communicating to external device via COMM2, COMMSPY monitoring COMM1
Figure 4: PC#1 communicating to external device via COMM1, COMMSPY, on PC#2, is monitoring COMM1
These are the features I really wanted to add to this version, but never had the time:
- Resume scrolling - as new data is monitored, the output window is automatically scrolled. To stop scrolling, the user should be able to simply click anywhere in the output window. To resume scrolling, they would select the command button "Continue Scrolling"
- Trigger - when a user specified stream of data comes across the port, alert the user.
- Print - print all or a selection of the output window.
- Resizable dialog - make the dialog resizable, adjusting the controls as necessary.