[Xastir] Bad Path warning problem

[Tom Russo]

On Wed, Feb 16, 2005 at 10:39:35AM -0800, we recorded a bogon-computron collision of the <archer at eskimo.com> flavor, containing:
> On Wed, 16 Feb 2005, Tom Russo wrote:
> 
> > On Wed, Feb 16, 2005 at 12:55:00PM -0500, we recorded a bogon-computron collision of the <gdwrbw at ftc-i.net> flavor, containing:
> > > Just noticed a little glitch in the warning that pops up for "bad"
> > > patchs.  I put in a path of Wide2-1 and it complains about the path
> > > being bad.  The last time I looked Wide2-1 was less than Wide2-2!
> >
> > Xastir expects "WIDEN-n" to have N==n.
> >
> > WIDE2-1 is intended to be what a WIDEN-n digi turns WIDE2-2 into after
> > its first digipeat.  If you only want one hop, WIDE1-1 is expected.
> 
> All true, but WIDE2-1 is also legal.  It's an unusual thing to need
> to use, but it is a legal path entry.  About the only time I'd want
> to use it would be to get around some weak filtering rules for
> digipeaters (in very special cases), or to test digipeaters/other
> software.

I admit I was surprised when that code was added (back in October, I think), 
as part of a commit that tried to prevent excessively long paths,
but it seemed reasonable enough.  The offending message was from the 
check_unproto_path routine in util.c, down near a comment saying "Whoa, n!=n"

> We really should accept it as valid.
> 
> Please add it to the bug list on SourceForge.

No need.  I have just changed the "WIDEN-n path is invalid if N!=n"
test to "WIDEN-n is invalid if N<n" --- so WIDE2-4 or WIDE1-5 will be
flagged, but not WIDE2-1, WIDE9-1 or any other oddity.

-- 
Tom Russo    KM5VY     SAR502  DM64ux         http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 
 "When life gives you lemons, find someone with a paper cut."