[aprssig] BEACON - APRS or Not?

[Shawn Stoddard]

Lynn W. Deffenbaugh (Mr) wrote:
> Shawn Stoddard wrote:
>> The two stations in question are W4LAW-5 and W4LAW-6. You can see the 
>> packets on the beacons at http://aprs.fi/?c=beacon&call=W4l*&limit=50.
>>
>> Steve responded back with good advice: be prepared for the 
>> unexpected. As my understanding of APRS improves I have a much better 
>> idea of right and wrong. So should my parser.
>>
>> The question becomes do you agree or disagree with Steve's response. 
>> If you disagree, why?
> I completely agree with Steve's response.  In my experience with my 
> own parser (adopted from the Open Source T2-135 source code), not only 
> do you have to ignore anything that may not be in the spec, but you 
> also have to be ready for some truly bizarre packets that DO have an 
> expected datatype.  There are still some Digi's out in the wild that 
> seem to be running with PASSALL (or whatever the equivalent is).  In 
> this mode, the digi/IGate doesn't care about the checksum on the 
> original packet, but rebroadcasts the corrupted contents with a new 
> checksum so later recipients can't even tell it's been mangled.
>
> You'll see fixed coordinates that are not even in the range of 0-90 or 
> 0-180.  So far I haven't seen one (that I know of) with the NESW 
> messed up, but I'm sure it's happened.
>
> For my parser, I actually log packets received from APRS-IS to one of 
> several directories.  GOOD gets the ones I'm currently handling, BAD 
> gets the ones that I'm not handling, and NOPOS gets the ones that I 
> handle, but do not have a position (like weather, messages, status, ...).
>
> May I ask what you'll be doing with the parsed data?
>
> Lynn (D) - KJ4ERJ - (No, I haven't check out the W4LAW-* stations yet)
>
>
>
> _______________________________________________
> aprssig mailing list
> aprssig at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/aprssig
>
I have an interest in writing some APRS software. A good parser is a key 
component in practically any component. It seemed like a good place to 
start to build up a base of knowledge. This group has proved to be a 
good place to validate what I understand versus what the specification 
says versus what is really out there.

Shawn
KG4OVQ