[Xastir] LaCrosse database program --- open2300db2APRS

[Tom Russo]

On Thu, Nov 26, 2009 at 09:39:13AM -0700, we recorded a bogon-computron collision of the <ka7o at ka7o.net> flavor, containing:
> YES!
> 
> My Lacross station has been offlined for several months now. I would 
> love to get it back up again.
> 
> Thanks!

Fix committed.  Let me know if it fixes the problem for you.

Details of what was wrong are in the commit message.  As far as I can tell,
the broken code was relying (accidentally) on the lucky chance that an internal 
datastructure in the mysql API wasn't changing between queries.  It should
never have worked, but more often than not, it appears, users of the code got
lucky and that internal data wasn't changing.

> On 11/26/2009 01:20 AM, Tom Russo wrote:
> > Is anyone here besides me actually using a (lousy) LaCrosse 23xx weather
> > station and the open2300db2APRS program we have bundled in the Xastir source?
> >
> > I just upgraded the OS on my primary system and discovered that
> > open2300db2APRS started segfaulting afterwards.  Turns out there's a huge
> > error in how that code accesses the mysql database, and I'm amazed it ever
> > worked at all --- but it did for me for several years until tonight's upgrade.
> > Specifically, the code never actually issued the mysql_fetch_row command after
> > querying the database for weather data --- and then started pulling data out
> > of the row datastructure, and apparently got what it was expecting.  This
> > definitely doesn't work in the version of os that I updated to (MySQL 5.1.41
> > on FreeBSD 7.2, previously running MySQL 5.1.41 on FreeBSD 6.4 --- all I did
> > was recompile mysql on the new system, and boom, segfaults from
> > open2300db2APRS).
> >
> > I'm pretty sure that the changes I had to make to my local copy of
> > open2300db2APRS to get it to work after the upgrade would be correct even for
> > older versions of mysql, and that it's likely it only worked because of a
> > fluke.  My change is merely to fetch the row after the query.
> >
> > If anyone else is actually using this code now, lemme know.  I'll commit my
> > changes and work with you to make sure they don't break your currently working
> > build.  I can't imagine it will, but I figure a head's up is in order first.
> >

-- 
Tom Russo    KM5VY   SAR502   DM64ux          http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236        http://kevan.org/brain.cgi?DDTNM
  In some cultures what I do would be considered normal. 
                                  -- Ineffective daily affirmation