OpenCPN Beta test 4.1.1108 Release

[bdbcat]

Gilletarom...

re:
What difference betwen :
- Scale Less Important Targets
- Scale Less Important AIS Targets

These two strings are redundant. We will drop " Scale Less Important Targets".
But, please note that we may change the wording again, since the term "less important" is not very precise. See above discussion.

And..
2. "AISShowScaled" is an error, will not be present in Release.

Thanks
Dave

[rooiedirk]

Originally Posted by bdbcat

Dirk...

1. Maybe we should say "non-critical targets"

2. I think we definitely need to be able to mask moored targets.

3. Maybe masking of moored targets should not participate in the tool cycling, but be simply an independent option.

4. What exactly is the algortihm to determine "unimportant targets" in the current code, please?

Thanks
Dave

1. I'm a non native English speaker, so you better don't ask me.
4.

            double temp_importance, So, Tcpa, Cpa, Rang, Siz = 0.; //calc the importance of target
            So = g_ScaledNumWeightSOG/12 * td->SOG; //0 - 12 knts gives 0 - g_ScaledNumWeightSOG weight
            if (So > g_ScaledNumWeightSOG) So = g_ScaledNumWeightSOG; 
                       
            if (td->bCPA_Valid){
                Cpa=g_ScaledNumWeightCPA - g_ScaledNumWeightCPA/4 * td->CPA;
                //if TCPA is positief (target is coming closer), make weight of CPA bigger
                if (td->TCPA > .0) Cpa = Cpa + Cpa * g_ScaledNumWeightTCPA/100;
                if ( Cpa < .0 ) Cpa = .0; //if CPA is > 4
            }
            else Cpa = .0;
                        Rang = g_ScaledNumWeightRange / 10 * td->Range_NM;
            if ( Rang > g_ScaledNumWeightRange ) Rang = g_ScaledNumWeightRange;
            Rang = g_ScaledNumWeightRange - Rang;
                                                   
            Siz = g_ScaledNumWeightSizeOfT/30*( td->DimA + td->DimB);
            if ( Siz > g_ScaledNumWeightSizeOfT ) Siz = g_ScaledNumWeightSizeOfT;
            temp_importance = So + Cpa + Rang + Siz;
            
            td->importance=(int)temp_importance;
            int calc_scale = 0;
            if ( td->importance  > ImportanceSwitchPoint ) calc_scale=100; else calc_scale = 50; //50% is minium scale for target
            //with one tick per second targets gan slink from 100 to g_ScaledSizeMinimal% in 25 seconds
            if ( td->importance  < ImportanceSwitchPoint ) targetscale = td->last_scale -2;
            //growing from g_ScaledSizeMinimal till 100% goes faster in 10 seconds
            if ( td->importance  > ImportanceSwitchPoint ) targetscale = td->last_scale +5; 
            if ( targetscale > 100 ) targetscale = 100;
            if ( targetscale < 50 ) targetscale = 50;//g_ScaledSizeMinimal;
            td->last_scale = targetscale;

Or for each factor a number is calculated according the weight setting in the config file, and then ale the factors are added. The scaling of the targets is then calculated and only slowly changed, to avoid an uneasy picture.

[cagney]

Linux data ports.
Options->Connections -> Data Port does no longer show correctly, the available
data ports, if they change in the current session.

For example on start my AIS port is /dev/ttyACM0, which is listed correctly.
If I disconnect the AIS usb cable, and then connect again, the port changes to
/dev/ttyACM1. AIS works correctly, if I manually enter this port, but /dev/ttyACM0 is still listed as a "Data Port", instead of /dev/ttyACM1.

O v4.0 worked fine.

Thomas

[transmitterdan]

Crash report:

Look at attached screen shot. This is using the NOAA ENC chart set. Object Query at this coordinate causes instant crash at line 5654 in s57chart.cpp.

The issue appears to be that the number of points (ntp) returned by:

 int ntp = GetLineFeaturePointArray(obj, (void **) &ptest);

is never used for anything. The for loop

                for( int ip = 1; ip < npt; ip++ ) {

at line 5652 iterates on obj->npt which is not the same value as ntp. I suspect this is a dyslexic error and the for loop should iterate on "ntp" not "npt". If I change the for loop to stop at ntp instead of npt then it never crashes so far as I have been able to test.

[transmitterdan]

After looking over the code in

int s57chart::GetLineFeaturePointArray(S57Obj *obj, void **ret_array)

I see a memory leak. Have submitted a PR to fix the leak and stop the crash.

[CarCode]

Originally Posted by transmitterdan

After looking over the code in

int s57chart::GetLineFeaturePointArray(S57Obj *obj, void **ret_array)

I see a memory leak. Have submitted a PR to fix the leak and stop the crash.

I don't see a memory leak in this function but 2 dead variables: float a and float b in the while loop.

Gerhard

[transmitterdan]

Originally Posted by CarCode

I don't see a memory leak in this function but 2 dead variables: float a and float b in the while loop.

Gerhard

The variable ret_array is populated with a pointer that was created by malloc() in this function but freed nowhere. It needs to be freed in the calling function after it finishes looking at the array.

[transmitterdan]

Originally Posted by rooiedirk

1. I'm a non native English speaker, so you better don't ask me.

Maybe the phrase you guys are looking for is "low priority targets"?

[rgleason]

I like "low priority targets". And would vote for turn off of moored to be separate action, not included in the rotation.

[bdbcat]

Folks...

I have made some changes to the AIS scaling logic, as follows:

1. Hiding of moored targets is completely independent of the non-critical target scaling logic. The relevant options->Ships->AIS "Suppress moored..." checkbox and the menubar checkbox are synchronized.

2. The AIS toolbar tool cycles as follows:

a. AIS show all targets (normal, default)
b. AIS Scale non-critical targets.
c. AIS Hide all targets
a. ...and back to AIS Show All targets.

3. I changed the (US) messages slightly.

4. The AIS toolbar tool shows the current state:
a. Normal
b. Suppressed, for both hidden moored targets and scaled targets.
c. Disabled, grey, all targets hidden.

5. The Options->Ships->AIS->Enabled scaled AIS targets checkbox should be considered as an "enabler" for the scaling function. If disabled, then the toolbar tool will skip the scaling option on cycling.

Short of serious disagreement or implementation errors, this will be the Release functionality for this feature in O4.2.

Code is in github now.

Thanks
Dave

[bdbcat]

@devs...

Last call...
Plan on a Beta this weekend. This Beta will be Feature Freeze for O4.2

Please tidy up and finalize any outstanding Pull Requests for this Release.

Thanks
Dave

Oh, and have a very Happy and Peaceful Christmas.

[Gilletarom]

Ok Dave,

Originally Posted by bdbcat

Folks...
I have made some changes ...

3. I changed the (US) messages slightly.
..
Dave

But no change in crowdin ?

[Hakan]

AIS scaling:
Serious or not I still would prefer the last mode of scaled/not scaled to be remembered on next OCPN start. Now it's always set to not scaled.

This feature now seems to be working very good.

Thanks
Håkan

[petter5]

Maybe a "remember last scaled" setting ??

[cagney]

Originally Posted by bdbcat

Folks...

I have made some changes to the AIS scaling logic, as follows:

1. Hiding of moored targets is completely independent of the non-critical target scaling logic. The relevant options->Ships->AIS "Suppress moored..." checkbox and the menubar checkbox are synchronized.

2. The AIS toolbar tool cycles as follows:

a. AIS show all targets (normal, default)
b. AIS Scale non-critical targets.
c. AIS Hide all targets
a. ...and back to AIS Show All targets.

3. I changed the (US) messages slightly.

4. The AIS toolbar tool shows the current state:
a. Normal
b. Suppressed, for both hidden moored targets and scaled targets.
c. Disabled, grey, all targets hidden.

5. The Options->Ships->AIS->Enabled scaled AIS targets checkbox should be considered as an "enabler" for the scaling function. If disabled, then the toolbar tool will skip the scaling option on cycling.

Short of serious disagreement or implementation errors, this will be the Release functionality for this feature in O4.2.

Code is in github now.

Thanks
Dave

Dave
When Options->Ships->AIS->Allow scaled AIS targets checkbox is not checked.... I get this:


2. The AIS toolbar tool cycles as follows:

a. AIS show all targets (normal, default)
b. AIS Scale non-critical targets.
(c. AIS Hide all targets) This option never shows!
a. ...and back to AIS Show All targets.

But....as scaling is not on, the button should really just toggle between
a & c if I understand you correctly.

Thomas