logo separator

[mkgmap-dev] [PATCH v1] make DP filter avoid removing points located at nodes

From Felix Hartmann extremecarver at googlemail.com on Sun Nov 22 20:58:17 GMT 2009


Felix Hartmann wrote:
>
>
> Mark Burton wrote:
>> Hi Johann,
>>
>>   
>>> Hi Mark,
>>>
>>> I've not tested the patch, but looked at the diff. Could not see the 
>>> sense in it at a glance. For me it makes no difference to the previous 
>>> state.
>>> What do you intend with this patch?
>>>
>>> Regards,
>>> Johann
>>>     
>>
>> If I understand the problem correctly, if someone (Felix, for example)
>> generates multiple lines from the same OSM way and one of the lines is
>> routable and the other(s) are not, then the original DP code would
>> generate different shaped lines for the routable and non-routable ways
>> because only the routable version of the line had CoordNodes in it. So,
>> my patch just marks those points that are going to be nodes but that's
>> done before the way is duplicated so those marked points will be seen
>> in the DP code for both the routable and non-routable lines that are
>> based on the same set of points. That's the theory, hopefully someone
>> (Felix, for example) will test the patch and let us know if it works as
>> expected.
>>
>> Cheers,
>>
>> Mark
>>   
> okay back...
>
> Well running ant-clean it now doesn't even compile using the patch (of 
> course I know that not ant-clean is the problem here, more that I did 
> not use it before cause I though just another patch on top...)
>
> [javac] Compiling 315 source files to 
> d:\Garmin\mkgmap_svn_trunk\build\classes
> [javac] 
> d:\Garmin\mkgmap_svn_trunk\src\uk\me\parabola\mkgmap\filters\RoundCoordsFilter.java:56: 
> cannot find symbol
> [javac] symbol  : method isBoundary()
> [javac] location: class uk.me.parabola.imgfmt.app.CoordNode
> [javac]                                     newP = new CoordNode(lat, 
> lon, ((CoordNode)p).getId(), ((CoordNode)p).isBoundary());
> [javac]                                                                                                          
> ^
> [javac] 
> d:\Garmin\mkgmap_svn_trunk\src\uk\me\parabola\mkgmap\filters\RoundCoordsFilter.java:56: 
> internal error; cannot instantiate Co
> [javac]                                     newP = new CoordNode(lat, 
> lon, ((CoordNode)p).getId(), ((CoordNode)p).isBoundary());
> [javac]                                            ^
>
> I attach the patches that I am using. - I think the mb-round-coords-v2 
> is causing some problems in combination.
>
>
> -- I'll remove that patch and try again the DP-avoid filter without 
> round cords. More soon.

Okay, works like a treat. See the attached screenshot what difference it 
makes.


vs before (disregard the brown contourlines) - this is both using detail 
lowest, to spot the error easier.



>> _______________________________________________
>> mkgmap-dev mailing list
>> mkgmap-dev at lists.mkgmap.org.uk
>> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>>   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20091122/68c67c9a/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: with patch.png
Type: image/png
Size: 34915 bytes
Desc: not available
Url : http://lists.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20091122/68c67c9a/attachment.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: without patch.png
Type: image/png
Size: 41348 bytes
Desc: not available
Url : http://lists.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20091122/68c67c9a/attachment-0001.png 


More information about the mkgmap-dev mailing list