[Xastir] findy trails

[soldering fun]

>> On Wed, 9 Jun 2004, Curt, WE7U wrote:
>>
>>> N4YLC-1>APT311,W6CX-3*,wIDE2-2,qAR,AA6VN-5:/230323h3746.27N/ 
>>> 12223.84Wk349/042/A=000052/soldering fun!
>>
>>> I ran it through Xastir as a log file in debug mode.  Got this:
>>>
>>> decode_ax25_line: start parsing
>>> N4YLC-1>APT311,W6CX-3*,wIDE2-2,qAR,AA6VN-5:/230323h3746.27N/ 
>>> 12223.84Wk349/042/A=000052/soldering fun!
>>> valid_path: Bad Path: Anti-loop stuff from aprsd
>>> decode_ax25_line: invalid path: APT311,W6CX-3*,wIDE2-2,qAR,AA6VN-5
>>> decode_ax25_line: exiting
>>>
>>> So..  There's a problem with the path, not the payload.  Change the
>>> small 'w' to a capital 'W' made Xastir parse it properly.  Since the
>>> AX.25 packet cannot contain lower-case charactes in these positions
>>> in the header, perhaps it was an igate that corrupted it?
>>
>> This is done by util.c:valid_path()
>>
>> Anyone know why this should be dropped?  The anti-loop code that is
>> catching it, but it's the 'w' in "wIDE2-2" that is the problem, not
>> the Q-construct stuff.
>>
>> I have no idea why AA6VN-5 changed that to a small 'w' when it
>> injected the packet, but that looks to be the source of the problem.
>> Whether we should allow it through our decoding is the next question
>> of course.
>>
>> --
>> Curt, WE7U			         http://www.eskimo.com/~archer
>> "Lotto:    A tax on people who are bad at math." -- unknown
>> "Windows:  Microsoft's tax on computer illiterates." -- WE7U
>> "The world DOES revolve around me:  I picked the coordinate system!"

Interesting.  Sorry it took me so long to reply (work insane right  
now.. playing
with aprs every free moment i have!)

Thoughts:

- I've done some tinytrak3 flashing+testing and still having the  
problem. today's data:
http://www.findu.com/cgi-bin/raw.cgi?call=N4YLC-1&time=24&start=24

in looking at the data, the 2 trends i notice about 'w' vs 'W' in  
WIDE2-2:
'W' appears only when message contains qAo.... qAR always appears with  
'W'
It seems like multiple intermediaries pass the 'w' instead of 'W'..  
which makes me
wonder if it isn't a problem in my hardware/setup.

- Seems like others in my area use similar 'hops' but do not have this  
problem.

- Definitely could be a problem with my tinytrak3 .  Who creates this  
'wIDE2-2' ?
Is it passed through from originator to destination or is it  
potentially modified
by intermediaries?

- How did you create and open your debug log file to find this info out?
- which spec are you referencing? i should read this.

- I can probably mod xastir locally to ignore case.  What do you think  
I should
do?


Thanks!
John
ps-> i must be blind.. this whole time i thought it was "findy trail"  
not "findu trail"...
that explains alot :)