For a moment I was thinking you were saying the GPS is not really NMEA compliant, which would surprise me greatly.
toggle quoted messageShow quoted text
But what is happening is that the tty driver is set to squish CFLF into a newline, and then the opposite is not being done on the output side.
So you could also fix this by disentangling the tty driver’s treatment of CR and LF on input and output.
If you had asked me in 1977 I would have said that by now humanity would have this cr, lf, business fully sorted by 2020. But no. Apparently this is going to haunt us forever.
73, Don’t forget to smile and have fun!
On Nov 23, 2020, at 10:06 PM, Ryan - KJ7GIE <firstname.lastname@example.org> wrote:
I exhausted all cable troubleshooting and I was really getting down on myself. But, I had also convinced myself that, yes indeed, the cable must be good. If it wasn't the cable it must be the GPS sentences. The silly radio must simply be throwing away what I was sending it. Why would it do that? Did it not not like the sentences... did it not like... hmmm... wait... is this thing _that_ finicky that any misalignment across a row would cause it to throw away the sentence? Yes, it's that finicky.
The PC looks for GPS sentences which start at "$GP" but that isn't good enough for the spec; or the radio. The NMEA spec is that the sentence must start at the beginning of the line and end with a CR/LF (carriage return / linefeed). And, after just literally staring at the output for almost an hour, I started to realize that this GPS / serial set-up wasn't entirely accurate with starting at the beginning of the line which meant the CR/LF was probably also not landing at the end of the line. PC happy. Radio no.
I'm now taking the GPS sentences off the serial, reformatting them, and pushing them out to the radio. Radio go happy now! Funny thing is that this is assisted by my old unix friend unix2dos:
cat /dev/serial0 | unix2dos -f > /dev/serial0
Now to clean-up the shack of all my little experiments and button this thin up.