[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Tlf-devel] Issues with contest scoring w/git code, OK w/126.96.36.199 ver
Re: [Tlf-devel] Issues with contest scoring w/git code, OK w/188.8.131.52 version in opensuse
Sat, 6 Jul 2019 11:51:38 +0200
Am Wed, 03 Jul 2019 08:58:39 -0700
schrieb Adam Clark <address@hidden>:
> > > When I loaded up the old stable, the multipliers seemed fine and
> > > the score seemed fine.
> > You mean the 1.2.4 version?
> That's correct, the version I'm comparing against is 1.2.4, the full
> package name and version is tlf-184.108.40.206-lp150.3.3.x86_64
I found time to look into the problem. Here is the result.
TLF from git does NOT score multis wrong.
The problem lies in reading (and interpreting) old log files in new TLF
There was a change in interpreting the log file in last months. Old
version (until 1.3.2) had an inconsistent scheme of recording the
worked multipliers depending on the logconf.dat settings. Sometimes it
used simply the exchange field, sometimes it used the additional field
just before the recorded points.
That was changed to always use the multi field before the points, but
old logs did not always set that field right.
You can check that it counts correct by reinserting the QSOs into a
fresh logfile or better by exporting the actual one to cabrillo and
reimporting the cabrillo file with 'tlf -i'. The only downside of the
latter is that it does not keep any point scoring you changed by hand.
So long story short. The new version counts the mults correctly but
interprets old logfiles wrong.
If that is a problem in everyday use we have to think of a way to
automatically convert to the new format.
73, de Tom DL1JBE
> > Other distros have 1.3.2 stable in meantime.
> It's funny how some things are maintained - the Opensuse HAM
> repository seems a touch behind on some apps, and pretty up to date
> on others.
> Thanks again!
> 73, Adam.
"Do what is needful!"
Ursula LeGuin: Earthsea