Yes, indeed, the EU-DX rules really mix DXCCs and legal countries.
I have updated my branch to reflect this, as far as I could interpret the rules.
Now scoring is also done in the plugin.
One has to specify their (legal) country code (like IT for Italy)
or DX for non-EU stations.
The available
QSO with the same station in CW and SSB can be considered using MIXED.
Updated my branch to reflect this.
73,
Zoli
On Thu, Feb 01, 2024 at 07:19:20AM +0200, Matei Conovici wrote:
> > Check the branch https://github.com/zcsahok/tlf/tree/poc_xmult,
> > the rules are currently in test/rules/eudx.
>
Hi Matei,
The TLF core is somwhat limited when it comes to multiplier handling.
To overcome this one can write contest specific plugins.
The choice of multipliers for EU-DX (country+EU region) can be implemented
using the "several multipliers per QSO" approach.
Hello Herve,
python-config in Debian/Ubuntu is part of python3-dev package. Not sure about
the OS you are using.
For a reference you can check the step "Install dependencies" in
https://github.com/Tlf/tlf/blob/master/.github/workflows/ci-build.yml
Hope this helps.
73,
Zoli
On Wed, Jan 24,
Hi Martin,
The rules are in PR https://github.com/Tlf/tlf/pull/420
(or in my fork: https://github.com/zcsahok/tlf/tree/marconi_day_rules)
You need a TLF built from github master with python plugin support.
Then copy the 3 files from rules/mcd into your rules/ dir and set
RULES=mcd
73 and hny,
L,13;EXCH'
>I get the same result if the first field is QSO instead of QSO_FMT.
>Thanks,
> w8lvn
>On Sun, Dec 31, 2023 at 1:32 PM Csahok Zoltan <[1]ha5...@gmx.com> wrote:
>
> Check that you have set your locator right in logcfg.dat:
>
> MYQRA=AB12CD
&
Check that you have set your locator right in logcfg.dat:
MYQRA=AB12CD
73,
Zoli
On Sun, Dec 31, 2023 at 11:08:21AM -0600, Bill Lederer wrote:
>I ran the december stew perry contest. TLF is working for me well, but I
>haven't worked out how the scoring is done.
>I'm using the
Hi,
I see currently no issue that introducing a DB backend would solve.
Technically for 1k records and holding it in memory no DB is required.
One thing, on the other hand, that is now seriously broken is the networking
support. I requires a careful rework, including probably a redesign
as it
you would need to switch RUN and S mode all the time. But I
> don't want to block development, I can use old versions.
>
> 73 Fred
>
> Am 27.08.23 um 18:16 schrieb Csahok Zoltan:
> > Fred, what is exactly the problem with the normal (ESM) mode in the sprint?
> > And how
Fred, what is exactly the problem with the normal (ESM) mode in the sprint?
And how does the behavior differ from CTCOMPATIBLE mode?
I'd also suggest to drop CTCOMPATIBLE and if this the case then improve ESM.
73,
Zoli
On Sun, Aug 27, 2023 at 01:16:20PM +0200, Fred Siegmund wrote:
> I still see
Hi Marcin,
The ToDo list is somewhat outdated, qso data has been already collected
into qso_t structure. Other items could be also done in the meantime.
Regarding code formatting we use astyle with the 'astylerc' file that
is provided in the tools directory. Ideally after all changes code should
Hi Marcin,
I'm not quite sure which list of issues do you mean? (Or is it the ToDo file?)
The issues on github (https://github.com/Tlf/tlf/issues) are closed
as soon as there is a working fix for them.
Yes, please create a usual pull request for the ToDo list.
All contributions are welcome!
73,
Hi John,
The failed test (rules) requires a supported terminal setting.
For me this line does the trick
TERM=xterm make check
73,
Zoli
On Tue, Jun 27, 2023 at 01:12:12AM +, john kuras wrote:
>I built tlf from the latest github source with no errors. I then did a
>make check and
The rules are meant to be more like templates as they typically require some
customizing for the particular user (setting county code, etc).
So the process of using them is
1) build a current verison of TLF (on a recent Linux, e.g. Ubuntu 22 or Debian
11)
autoreconf -i \
&&
See https://github.com/Tlf/tlf/pull/391
73,
Zoli
On Thu, Apr 27, 2023 at 06:21:09PM +0200, Martin Kratoska wrote:
> Just trying to prepare files for the UKEICC Contest. Rather weird, rather
> complex. Any help/ideas?
>
> Has anybody ready to use, tested files?
>
> Would be nice to see in the
Yes, please add the rules to rules/neqp. I'd be nice to have also a test
for it, check the recently merged PR #388 for an example.
73,
Zoli
On Thu, Apr 13, 2023 at 12:01:16PM -0400, Alan Dove wrote:
> Zoli:
>
> Okay, I created the "neqp" branch, changed that line to
>
> #define
Hi Alan,
As a quick fix increase the max section length defined in tlf.h
#define MAX_SECTION_LENGTH 4
We'll have to check if this causes any issues before making it official.
73,
Zoli
Hi Martin,
Here is a screenshot from today's mix of Oster Contest and MWC.
I'm using xfce4-terminal, it has no issues with readibility.
These are the default colors of the terminal but they can also be easily
customized.
73,
Zoli
On Fri, Mar 31, 2023 at 11:06:58PM +0200, Martin Kratoska
Hi Marcin,
How did you export the log to Cabrillo?
Normally it is done by entering
:WRI
in the call input field.
The file you got is just TLF's internal log file. So either the export was done
differently or you are looking at the wrong file.
The exported Cabrillo file is called .cbr.
73,
Zoli
Hi Alexej,
You probably missed the customizing step. TLF uses by default
the delivered generic configuration files from /usr/share/tlf.
Generally they have to be tailored for the actual user and contest.
This is also the case for CQWW. The default rule file
(/usr/share/tlf/rules/cqww) contains
Thanks for sharing, Drew! It's always nice to see TLF in action.
73,
Zoi
On Sun, Jan 01, 2023 at 05:08:33PM +, Drew Arnett wrote:
> I was delighted to see TLF in the California QSO Party results video.
> https://youtu.be/5OkXZqPYd_4?t=752
>
> Drew
> n7da
Here a minimal working example. All fields are either pre-filled or disabled,
so no questions are asked.
73,
Zoli
sscbr.tgz
Description: application/gtar-compressed
> Is something like 'CABRILLO-EXCHANGE= # A 83 KS' supported?
Yes, exactly.
Regarding the template: could it be that TLF doesn't use it? Due to eg. not
finding it.
It works OK for me.
73,
Zoli
And there is surely neither CABRILLO-EXCHANGE nor SERIAL_EXCHANGE in
the logcfg and rule file?
(stock logcfg.dat has "CABRILLO-EXCHANGE= #" on line 171)
73,
Zoli
On Mon, Nov 21, 2022 at 06:37:33AM -0600, Nate Bargmann wrote:
> * On 2022 21 Nov 06:12 -0600, Csahok Zoltan wrote:
&
of adding it to the template. All configs can be specified
both ways.
73,
Zoli
On Mon, Nov 21, 2022 at 05:19:08AM -0600, Nate Bargmann wrote:
> * On 2022 20 Nov 22:27 -0600, Csahok Zoltan wrote:
> > Hi Nate,
> >
> > How are your CABRILLO* config parameters set?
&
Hi Nate,
How are your CABRILLO* config parameters set?
Did TLF ask for the other fields like category info?
73,
Zoli
On Sun, Nov 20, 2022 at 09:44:09PM -0600, Nate Bargmann wrote:
> Hi All.
>
> I just completed working some of the ARRL SS this weekend. As the logs
> must be submitted within
Hi Alan,
Is this the stock TLF from Ubuntu repo or a locally compiled one?
If compiled then was it compiled on Ubuntu 20 or 22?
If it was compiled on 20 then recompile it.
Ubuntu 22 has Hamlib v4, whereas Ubuntu 20 uses Hamlib v3 packaged
bit confusingly into libhamlib2. So it looks like your
Hi Martin,
You have probably experienced the Pyhton 2-to-3 switchover that caused a number
of troubles.
Since 2020 version 3 is the only active one and as far as see it's reasonable
stable.
Of course things can change, but I see no future issues at the moment.
Regarding the development of TLF
We are working on adding Python plugins to TLF. This will allow
defining a wide range of contests not falling into the currently
available categories.
One of such contests is the MWC were I take place semi-regularly.
My prototype plugin supports correct scoring inluding the "last letter"
Hi Martin,
Glad to hear that you were able to sort out the TLF-related issues.
Regading Winkeydaemon: you could try starting it with the option
-q $'\x09\x04'
This sets the initial message to disable PTT on Pin 5. I do not own
a Winkey device, so this is just a guess based on the docs.
Check your libcw6 version. There was a fix in 3.5.1 for something similar.
(see also http://unixcw.sourceforge.net/news.html )
73, Zoli
On Sun, Aug 28, 2022 at 07:19:04PM +, Marcin SP6MI wrote:
> It was Slackware, it doesn't have by default option to install tlf, it has to
> be compiled and
Hi Nate,
Wow, you brought a lot of stuff to check!
I suggest to tackle the issues one by one on github.
It seems the "random mult" issue (#273) hit again after one year.
Could you please open issues for the input clearing on Esc
and the scoring anomaly?
If neither the log file nor the relevant
It could also be that your distro's (older) TLF was started instead
of the one you compiled due to PATH issues.
73,
Zoli
On Sun, Aug 28, 2022 at 07:02:33PM +, Marcin SP6MI wrote:
> Hi,
> I was running in from subfolder in my $HOME, and tlf was built from latest
> git version. Ok, I'll
Hi Marcin,
Thanks for reporting this issue.
I assume that you were running TLF from a directory that was not
writeable by your user (e.g. / or /usr/bin).
Could you check if this issue goes away when starting TLF
from a writeable directory? You could test writetability
by executing
touch .paras
winkeydaemon doesn't write to /var/log.
Here's how I used it.
1) stop cwdaemon to free up the UDP port
sudo systemctl stop cwdaemon
2) start winkeydaemon in foreground mode (not as a daemon)
./winkeydaemon -n -d /dev/ttyUSBx
where ttyUSBx is the USB device of your winkeyer.
Pressing F9
Yes, underscore (_) is unfortunately hardcoded to send
no matter how F3 or S_TU_MSG is configured.
One more issue to be fixed...
Maybe sending the last F3/S_TU_MSG except its first word (normally @ or TU)
could work.
73,
Zoli
On Sat, May 21, 2022 at 11:00:25PM +0100, Nick Craig-Wood wrote:
Hi Zilvis,
Tried with the same cwdaemon (0.10.2) and libcw6 3.5.1-4 on Debian 11
(arch x86_64) + a fresh tlf build, and it had no issues.
cwdaemon started as
/usr/sbin/cwdaemon -d ttyS2 -n -xn -iii | tee cw.log
(I use a physical serial port)
Sent a ? using netcat:
echo '?' | nc -u -w0
Hi Gerard,
The fix is available as a pull request. May I ask you to check if it
solves the issue?
Here are the steps to build it in the /tmp/tlf directory:
/tmp$ git clone https://github.com/zcsahok/tlf --branch fix_countrylist_parsing
/tmp$ cd tlf
/tmp/tlf$ autoreconf --install --force
Hi Gerard,
Even if it's not easy to trigger the buffer overflow, things do not work
as expected. I've opened an issue: https://github.com/Tlf/tlf/issues/319
Thanks for reporting!
I suggest generally to use the latest code (master branch from github)
if possible, as it contains a number of fixes
Hi Alan,
Ctrl-A sends the spot only to other networked TLF nodes,
hence the "de TLF-A" where "A" is the node id (1st node).
man page says:
Ctrl-A Add a spot to bandmap and broadcast it on the local network.
To send to the actual cluster network Ctrl-B can be used.
For actual spotting purpose
It's the auto CQ delay time in 500 ms units.
It can be set through CQDELAY= and changed by Ctl-PgUp/PgDn.
73,
Zoli
Glad to hear that it worked! Current Tlf version supports specifying
Cabrillo header value through CABRILLO-* configuration parameters.
This avoids the need to enter all the values each time an export is made.
Have a look at share/logcfg.dat in the repo.
73,
Zoli
Hi Alan,
What Tlf version are you using?
Is this the expected format?
QSO: 14038 CW 2022-02-19 1732 AB1XW599 MAHA8VV599 KW
73,
Zoli
On Mon, Feb 21, 2022 at 09:29:58PM +, Alan Dove wrote:
> Hey, folks:
>
> I spoke too soon, here's one more issue. In trying to submit my log for
>
Hi Jack,
Tlf surely does not create or move any files into a trash folder.
Where is your trash folder located? Could you give the list of most recent
files from it?
(I mean "ls -ltr trash | tail")
The way Xplanet is called saves the current view in a JPG file.
I think the typical use case would
Hi Lukasz,
Even though Tlf can be started without rig control but some of its
functionality will be limited as you have exprienced it.
To overcome this you can use rigctld in dummy mode.
Here is how:
1) start rigctld for Dummy radio (ignore the warnings)
$ rigctld -m 1 &
2) check
Hi Phil,
Reviewing the commit history I found that your issue was fixed around
this time last year by a fellow FD participant:
https://github.com/Tlf/tlf/pull/177
As 1.4.1 dates from April 2020 this fix is only part of the current master.
When using 1.5 and specifying '1D WV' as 'Your exchange'
I'll check this.
73,
Zoli
On Sun, Jan 17, 2021 at 12:47:37PM -0600, Nate Bargmann wrote:
>
> I think the problem lies in the declaration of int_p in parse_logcfg.h,
> line 61 where its type is int and rig_model_t has a type of uint32_t.
>
> Perhaps adding another member to that union that is
Hi Matthew,
I did run into the issue with cwdaemon: after an upgrade its config file
was reset and it produced an erratic keying. Changing sound back to 'n'
solved the issue and my conclusion was that cwdaemon is meant for either
keying or side tone generation, but not for both. Which was just
Hi Matthew,
I did run into this issue with cwdaemon: after an upgrade its config file
was reset and it produced an erratic keying. Changing sound back to 'n'
solved the issue and my conclusion was that cwdaemon is meant for either
keying or side tone generation, but not for both. Which was just
While working on improving the handling of WWV messages
I did some digital archeology and found that the propagation
display (Ctrl-P) is a straight port of MICROMUF.PAS, a MUF/LUF tool
from the 80s. It was originally written in Basic then translated
to Pascal. Here one can still find both:
> > Tap Page-Dn until '1' is reached and then like an odometer the first
> > numeral rolls down to '4', the next tap rolls the first numeral down
> > to '3' and so on while leaving the second numeral at '1', then tap
> > Page-Up to set the second numeral. That would require a few taps of
> >
Hi Nate,
Everything is possible. :-)
An option I see could be making the exchange box wider to span the RST fields
and let the user input the values. Optionally pre-fill with 599/59.
Then upon storing the qso we would parse the first 2 words and store
them as RSTs.
A bit tedious for the user to
Hi Nate,
Personally I always use the default TLF mode and quite happy with it.
Removing CT compatibility is fine with me.
I didn't quite get the difference between the current and the optional new mode,
though. (I'm not a regular N1MM user)
73,
Zoli
On Tue, Jan 07, 2020 at 08:27:55PM -0600,
Hi Drew,
Trailing zero and newline have most likely historical origin.
The zero was introduced abt 9 years ago in this commit:
https://github.com/Tlf/tlf/commit/5a2be34d01b296bc8b8d8b362407cc3077e3c6d8#diff-8be0c8c50997da170641d3392ba81b5a
Apparently cwdaemon needed it. Now it does not for sure,
Now I remember that actually our FD is setup is just like this:
using a single USB-serial converter with RX/TX controlling the rig via hamlib
and DTR doing keying via cwdaemon. We had no issues with it.
The only thing is that at Tlf start-up the key is down for a second.
I'll check hamlib config
My 2c: try with "export EDITOR=..."
73,
Zoli
On Fri, Nov 22, 2019 at 07:18:35PM -0600, Nate Bargmann wrote:
> My first test with EDITOR=gedit resulted in nothing happening. I ran
> gedit from another terminal and it opened fine. Then I tried emacs
Hi Christian,
Yes, keying does work with the script. But as far I can see
there are two issues with hamlib keying:
- in-band speed changes are not supported. one can't send "+++TEST---"
- sending a large message could block rig interface until it gets queued.
Given the rig protocol (KY
ent settings.
BTW in debian Tlf recommends sox and xplanet, but not cwdaemon.
Could this also be improved?
73,
Zoli
On Thu, Nov 21, 2019 at 06:53:49AM +0100, Thomas Beierlein wrote:
> Hi Zoli,
>
> Am Wed, 20 Nov 2019 19:48:12 +0100
> schrieb Csahok Zoltan :
>
> > I would like to
I would like to use GLib version >=2.40 for set-related hash table functions
(g_hash_table_add).
A quick check of Debian buster shows that it contains GLib 2.58.
On the other hand, tlf in that release (1.3.2) requires only GLib 2.35.9.
Similar (or even worse) for sid: tlf 1.4.0 requires 2.30.0
Thanks, Tom!
Now it's time for feature requests and bug reports. :-)
73,
Zoli
On Wed, Nov 13, 2019 at 03:06:33PM +0100, Thomas Beierlein wrote:
>
> After a lot of work by all people contributing to TLF I would like to
> announce the release of the new TLF-1.4.0 version.
>
> You can find it at
Hi Tom,
Made a quick test with the slowest machine that I use, a Pentium M 1.7GHz.
Execution of readcalls() took about 150 ms per 1k QSOs (data are below).
The code scales mostly linearly except dupe checking that is using
a slow linear search. Latter makes it O(N^2) but its contribution
is quite
Hi Tom,
Nothing is pending from my side. Revert-after-grab will come later this year.
73,
Zoli
On Mon, Oct 28, 2019 at 07:09:27AM +0100, Thomas Beierlein wrote:
> Hi all,
>
> time for a new TLF-1.4.0 release is overdue. I plan to do finish it in
> next days. If there are some late
Thanks for promoting TLF, Koos! Also thanks for sharing.
73,
Zoli
On Thu, Aug 08, 2019 at 09:11:19PM +0200, Koos van den Hout PE4KH wrote:
>
> Linux in the Ham Shack is a podcast (audio series) on using Linux in
> amateur radio.
>
> They did an episode on TLF, based on input by another amateur
Hi Ervin,
Yes: if you enter a frequency in kHz as callsign, then the rig switches to it.
A nice but maybe undocumented feature.
73,
Zoli
> Another note to Thomas/Zoli HA5CQZ: do you know about this line?
>
> https://github.com/Tlf/tlf/blob/tlf-1.3.2/src/callinput.c#L648
>
> if
This can be quite easily done in recall_exchange without disturbing
the contest state machine (i.e. keeping Enter and Tab functions as-is).
A sensible correction.
73,
Zoli
> > >
> > > As a sidenote: Maybe we should rethink the behaviour of the
> > > RECALL_MULTS and it should only pick up old
Thanks Tom for the clarification! It's indeed RECALL_MULTS that interferes
when used with Tab instead of Enter. I was able to reproduce it with Adam's
setup, while in my case it worked as I wasn't using this feature.
(and was using the Enter+Enter sequence)
btw while testing I realized that
tlf
Hi Adam,
Thanks for the update. I also tried it with Konsole (v18.04.0, KDE 5.54)
and the cursor keys worked.
For the FD issue: is this the sequence that you use?
- input call
- Enter << cursor moves to exchange field
- input exchange
- Enter << here your exchange gets erased, QSO is not logged
Hi Adam,
Glad to hear that you used TLF during the FD. We strive improving it,
so any feedback (including bugs reports, of course) is welcome.
The issue (or at least something similar) in SSB mode I've also experienced:
one had press Enter twice to get the QSO logged. Now, however, I was not
Managed to install a Sparc64 system and reproduce the issue.
It is a test-only error due to only storing a pointer instead of
saving the actual message string during the tests.
Some of these strings are generated in local variables
that may "disappear" before it comes to actually check them.
Hi Tom,
Some topics from my backlog:
* improve the "grab a spot while in run mode" use case:
- while CQing user grabs a spot using Alt-G
- mode switches to Log, freq is saved
- change: as soon as the QSO is logged (or Esc'ed) mode+freq
switch back automatically
(i.e.
Hi Tom,
Thanks for the information.
Once the release is out I'll start with the changes for making
hamlib mandatory. (i.e. getting rid of a bunch of conditional stuff)
73,
Zoli
On Tue, Sep 11, 2018 at 07:07:24AM +0200, Thomas Beierlein wrote:
> Hi all,
>
> after a busy spring time and summer
Hi Nate,
Sorry, using vim here.
But we could (at least in the future) set up astyle to expand tabs.
73,
Zoli
On Tue, Mar 27, 2018 at 07:58:09PM -0500, Nate Bargmann wrote:
> Well, it's been some time since I hacked on Tlf and I'm getting back to
> it. As I've converted to using Emacs over the
Hi Tom,
Nice to hear that it works well.
In fact I have applied astyle formatting to the *.c files in test/,
and thus was assuming that all whitespaces are "correct".
Anyway, no problem making the patterns more relaxed.
Verbose mode is not that trivial. One can execute the script manually
and
that coupling to make testing easier.
>
> 73, de Tom DL1JBE
>
>
> Am Fri, 26 Jan 2018 20:31:00 +0100
> schrieb Csahok Zoltan <ha5...@freemail.hu>:
>
> > Tom, cloned your repo and Travis CI works quite well.
> > I'll try to add a CUnit test as per docs. Not al
Tom, cloned your repo and Travis CI works quite well.
I'll try to add a CUnit test as per docs. Not all of Tlf
can be tested that way, but let's see how far we can get.
The .astylerc you shared looks OK to me.
(yes, I meant indent of 4 spaces)
73,
Zoli
On Fri, Jan 26, 2018 at 02:42:41PM +0100,
Hi,
Another topic: as the code has grown over the time and many authors
contributed to it the formatting is inherently not consistent.
GNU indent is a powerful tool for C source formatting and present in any
modern Linux distro. We could define a common style simply by setting up
an .indent.pro
, Ervin Hegedüs wrote:
> Hi Zoli,
>
> On Tue, Jan 23, 2018 at 06:52:33PM +0100, Csahok Zoltan wrote:
> > Hi,
> >
> > Currently tlf has an optional hamlib support. I guess it's optional due
> > to historical reasons: hamlib may have been not always available or
77 matches
Mail list logo
