Tactics Plugin

[rgleason]

Ron and Petri,
Yes I've noticed this difference.

Applied angle and speed no waypoint info except when tactics mark is selected. Both tactic’s setting selections are the same.

Note: Boat is late into the water and am working on several fronts,etc, so my computer time will be much curtailed this week. Will try to find some time, but boat has to go in or season will be lost.

[RonSouthworth]

All the best with sorting the boat out for the season Rick.

I will get something more detailed together for you Canne. After installing a lot of plugins !!!.!,testing them today i want to freshen the ini file up and try the import again and see if it works.

Thanks to a plugin Sean has developed I have a clue now on finding decimal points

I tried to finish the message types today but there were too many interruptions so maybe tomorrow. One to go it is doing something unusual

[Canne]

Originally Posted by rgleason

Yes I've noticed this difference.

Originally Posted by RonSouthworth

Applied angle and speed no waypoint info except when tactics mark is selected. Both tactic’s setting selections are the same.

Thanks Ron, yes, please provide more information about this issue while Rick is busy (have a nice time with the real boat!). I can't figure out what's wrong and what is the functional requirement we try to achieve? On the other hand, I've never tested this function with both Tactics and Dashboard-Tactics active simultaneously (I did not envisage such a use case). Do I need you Simulator to reproduce it or is it visible with tom_r's Tactics test NMEA-sentence playbacks (Rick seems to suggest so)?

[SCarns]

Originally Posted by Canne

Is this a VDR-playback or your boat's instruments?

boat instruments in real time...

[Canne]

Originally Posted by Nereus32

boat instruments in real time...

Strange indeed. I looked at the Dashboard code in O https://github.com/OpenCPN/OpenCPN/b.../src/clock.cpp
Moon phase instrument is derivated from clock instrument:

DashboardInstrument_Moon::DashboardInstrument_Moon( wxWindow *parent, wxWindowID id, wxString title ) :
      DashboardInstrument_Clock( parent, id, title, OCPN_DBP_STC_CLK|OCPN_DBP_STC_LAT, _T("%i/4 %c") )

from which it gets the UTC time and lat+lon. From
https://github.com/OpenCPN/OpenCPN/b...shboard_pi.cpp
we can see that both _STC_CLK and _STC_LAT to clock instrument are coming from GPS NMEA sentences. First thing to verify is that does the UTC (instrument "GPS clock" not "local GPS clock") and LAT/LON show correct while your GPS is running, then check the moon phase. If it is still wrong, maybe you want to try with built-in Dashboard, not with Tactics to compare.

[Canne]

Originally Posted by RonSouthworth

I can make it crash by trying to add an additional panel to the dashboard tactics plugin otherwise it is happy. I think you are aware of this issue.

Originally Posted by Canne

No, I am not aware of this issue. But I am willing to fix it. Can you please explain a bit more, perhaps with a screenshot so that I can attempt to reproduce the issue.

Ron, don't bother to prepare a more detailed report for this particulare issue. I was able to reproduce the issue by starting from scratch my testing. Here's the fix https://github.com/canne/dashboard_t...f8028f3f45ef43 . Thank you for drawing my attention to this showstopper!

[Canne]

Originally Posted by Canne

I looked at the Dashboard code in O ... Moon phase instrument is derivated from clock instrument... which it gets the UTC time and lat+lon....First thing to verify is that does the UTC (instrument "GPS clock" not "local GPS clock") and LAT/LON show correct while your GPS is running, then check the moon phase.

I checked with recorded data for which I have the date and time in my logbook. I verified the GPS sentence including date and displayed values (there is no date):

Then I verified using an on-line calculator for the same date and location. 32%,1st quarter in 22 hours. Pass.

[AedanC]

Originally Posted by Canne

Of course, but let's wait if Aedan has time to test the released version - I am worried of not being able to reproduce his crash condition.

I apologise, work got in the way recently, I’m travelling at the moment but I’ll try and get back to it either tonight or tomorrow morning. Sorry again for going so quiet.

Aedan

[Canne]

Hi Aeadan, no worries, and no rush; the boss (Rick) is busy putting his boat in the water so we can take our time Me too, this weekend I can get to my boat and I can test in real!

Anyway, meanwhile, you've been saved from an annoying bug which was found by Ron; the latest and fixed beta-version of Dashboard-Tactics is now here: https://github.com/canne/dashboard_t...s/tag/v0.2.005 (I haven't test yet the just-now-generated *.exe installer file myself, I'll do that this evening.)

Just to make sure: for testing, you are still using the same NMEA recording with STW=0.0, right?

Originally Posted by AedanC

Sample data set attached and a screen photo below, remove the .pdf extension as usual.

[rgleason]

Dear, Canne.
I am certainly no boss! Just trying to help keep things together on the plugin front. You are the boss! We are all very appreciative of your work too.



Right now I am trying to find a Sporlan dryer filter, so many things to do, but I keep getting draw back to O. I am sure you have the same thing going on. Best, Rick

[Canne]

Thanks Rick, for your encouragement! These two plugins together are quite a handful and without a full commitment the quality would not be good enough for our O sailors out there. (I am still struggling with some side-effects caused the workaround to fix the wxResizer issue you found - I am glad I did since no wxWidgets expert has taken up your bug report in Dashboard that I know.) I hope you get some sailing time too. My FTE this year is still positive in favor of my boat, so all is good...

Progress: I managed to get out from Travis build a OSX package! But I have no way to test it, see here:
https://github.com/canne/dashboard_t...s/tag/v0.2.013
All comments welcome from Mac owners!

"Travis CI uses macOS 10.13 and Xcode 9.4.1" by default. Wishes for changes taken, see here what's available :
https://docs.travis-ci.com/user/reference/osx/#macos-version

[AedanC]

Originally Posted by Canne

Hi Aeadan, no worries, and no rush; the boss (Rick) is busy putting his boat in the water so we can take our time Me too, this weekend I can get to my boat and I can test in real!

Anyway, meanwhile, you've been saved from an annoying bug which was found by Ron; the latest and fixed beta-version of Dashboard-Tactics is now here: https://github.com/canne/dashboard_t...s/tag/v0.2.005 (I haven't test yet the just-now-generated *.exe installer file myself, I'll do that this evening.)

Just to make sure: for testing, you are still using the same NMEA recording with STW=0.0, right?

Result no 1 and it's not good I'm afraid. I removed Dashboard-Tactics andi installed from the above link, it still crashes as soon as I tick the "Use SOG" box with the same test data, ie the test case I posted a few days ago with STW=0.0.


I'm now going to completely remove OpenCpn from my system, reboot, reinstall and try again. I'll report back as soon as I have success or failure.

[AedanC]

I have removed OpenCpn completely from my laptop and searched for any setup/initialisation files everywhere I can think of and deleted anything I found.


I then downloaded and installed the official OpenCpn version, the Official VDR player and Dashboard-Tactics 2.005 as per the link above. I did not install any other plugins.


No difference I'm afraid, the sample file plays perfectly until I tick "Use SOG..." and then OpenCpn crashes with no error messages.



Has anyone any bright ideas for further things for me to try?


Thanks for your patience,


Aedan

[Canne]

Originally Posted by AedanC

Result no 1 and it's not good I'm afraid...it still crashes as soon as I tick the "Use SOG" box with the same test data

Thanks, Aedan. That's actually good! (my guts feeling that I better wait until you test again.)


Now we will debug this: we have the same test data, that's a good start. For me it looks like this, do you confirm you got the same?

When you say "as soon as I tick the "Use SOG" box" do you mean that you start the VDR playback without the SOG box, then you go to Tactics Preferences, tick the SOG box only, then click OK, then it crashes? Like this:

(it does not crash for me, but I'm trying to figure out what can happen).


Thank you for your patience, sorry for all the trouble, I do appreciate!

[AedanC]

Originally Posted by Canne

Thanks, Aedan. That's actually good! (my guts feeling that I better wait until you test again.)

I suppose it is good really, it's better to discover and understand the cause, so you know the fix makes sense.

Originally Posted by Canne

Now we will debug this: we have the same test data, that's a good start. For me it looks like this, do you confirm you got the same?
Attachment 194281

Yes I do get the same.

Originally Posted by Canne

When you say "as soon as I tick the "Use SOG" box" do you mean that you start the VDR playback without the SOG box, then you go to Tactics Preferences, tick the SOG box only, then click OK, then it crashes? Like this:
Attachment 194282

Yes, that's all correct. Note that it crashes whether if I tick the "Use SOG..." box when vdr is running or before I start VDR playing. In the second case it crashes as soon as I start VDR.

Originally Posted by Canne

Thank you for your patience, sorry for all the trouble, I do appreciate!

It's me who should so the appreciating, I wish my programming skills were good enough to contribute more.


I made a small experiment a short while ago, I took every STW reading out of the NMEA file, by removing all the IIVHW sentences. With that data file it does not crash but the true wind direction and speed remain at zero, regardless of the "Use SOG.... " setting. The modified file is attached, with the unmodified version just for completeness.


I have also a picture of it running with the modified file.


Unfortunately I have to go now, I'll try and get back tomorrow.


Aedan