[mapserver-users] Re: polygon border artifacts

[Lime, Steve D (DNR)]

I'm unable to recreate the artifact with a simple test case though so I'm not exactly sure what's going wrong. This mapfile
defines a couple of overlapping polygons and there is no rendering issue.

MAP
  SIZE 512 512

  IMAGETYPE PNG8

  EXTENT 0 0 1000 1000
  IMAGECOLOR 255 0 0

  LAYER
     NAME "test"
     TYPE POLYGON
     STATUS DEFAULT
     FEATURE
       POINTS 100 100 600 100 600 600 100 600 100 100 END
     END
     FEATURE
       POINTS 300 70 800 70 800 800 300 800 300 70 END
     END
     CLASS
        STYLE
         COLOR 128 255 255
         # OUTLINECOLOR 0 0 0
       END
     END # of class
  END # of layer
END

Online at: http://maps1.dnr.state.mn.us/cgi-bin/mapserv60?map=/usr/local/mapserver/apps/test/ocean/test4.map&mode=map
Your original test case (+labels): http://maps1.dnr.state.mn.us/cgi-bin/mapserv60?map=/usr/local/mapserver/apps/test/ocean/test.map&mode=map

Steve

-----Original Message-----
From: mapserver-users-bounces at lists.osgeo.org [mailto:mapserver-users-bounces at lists.osgeo.org] On Behalf Of Sven Geggus
Sent: Wednesday, May 09, 2012 8:18 AM
To: mapserver-users at lists.osgeo.org
Subject: [mapserver-users] Re: polygon border artifacts

"Lime, Steve D (DNR)" <Steve.Lime at state.mn.us> wrote:

> Do outlines fix things at all zoom levels?

Not really. The Workaround which is working best for now is using a very big polygon Overlap of 100Meters (Talking about Google Mercator here) which I feel is way to much and an outline with of 1. To completely get rid of artifacts I would need an Outline width of 2, but this is not small enough IMO.

> What about the possibility of fixing the data?

The bug is clearly with mapserver here, the data is fine IMO. The tool used for generation of the tiles does already allow for generation of arbitrary overlap widths (https://github.com/joto/osmcoastline/).

> I mean, unioning the ocean polys into one, or better yet deriving land 
> polygons from that data.

No, unfortunately both of them are not an Option!

1. Unioning the polygons:
I presented a small shapefile just to demonstrate the problem.
The original file I want to use contains ocean tiles covering the whole world. So this has been spitted into tiles for a reason!

2. Land polygons:
These are also not an option because I want to use the ocean tiles to cover inaccurate hillshading derived from srtm. This will not work the other way round.

The only thing I could imagine as a solution on the data side would be the clipping of the hillshades by means of the polygon shape.

Regards

Sven

--
TCP/IP: telecommunication protocol for imbibing pilsners
	                             (Man-page uubp(1C) on Debian/GNU Linux)

/me is giggls at ircnet, http://sven.gegg.us/ on the Web _______________________________________________
mapserver-users mailing list
mapserver-users at lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users