Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
09/24/05 16:50
Read: times


 
#101491 - What I am doing with it
Responding to: ???'s previous message

First of all, thanks for the feedback so far. As most seem to indicate, its technically doable, but not considered a standard practice. Just to make sure I am well understood, here is what I wanted to do with it:

I have a 8752 that is polling a communicating thermostat over a RS-485 link using the 8752's uart. The interrupt routine that receives the response stores the received byte in a circular buffer and also checks if its a carriage return. If it is, it then sets a bit flag to indicate that a complete response has now been received.

The main loop checks this "complete response flag" on every loop and if it is set, then starts to parse the buffer contents, looking for fields such as "SPH=73 SPC=78" (setpoint heat = 73, etc). But there is a remote possibility that the response gets garbled and that a carriage return appears earlier then expected because some characters are missing. The idea was to create a short called routine to get the next byte for comparing/parsing and also check if its a carriage return and if it is, then immediately "return" to the end of the parsing routine and abandon parsing this command because it is invalid. This is the only routine (a long list of compares) that would be using this "called routine with alternate return capability".

List of 24 messages in thread
TopicAuthorDate
RET to a different address            01/01/70 00:00      
   here is how pseudocode            01/01/70 00:00      
      RET to a different address            01/01/70 00:00      
         no flaw, but 1.000.000 gotchas            01/01/70 00:00      
            That's what I wanted to know            01/01/70 00:00      
               Yes            01/01/70 00:00      
               "clever"            01/01/70 00:00      
                  OT: my wife            01/01/70 00:00      
         no flaw, but seriously not recommended            01/01/70 00:00      
            experience            01/01/70 00:00      
               reload SP            01/01/70 00:00      
                  restoring stack            01/01/70 00:00      
         Recognisable string            01/01/70 00:00      
      named return value            01/01/70 00:00      
   Bad Practice            01/01/70 00:00      
      Well phrased            01/01/70 00:00      
   What I am doing with it            01/01/70 00:00      
      try...catch            01/01/70 00:00      
         setjmp / longjmp            01/01/70 00:00      
         when to try ... catch            01/01/70 00:00      
            the borderline            01/01/70 00:00      
               Promises            01/01/70 00:00      
                  who cares if an exceptiom is "acceptable            01/01/70 00:00      
      Parsing input data            01/01/70 00:00      

Back to Subject List