Rob ha scritto:
> Removing the dashes element of the style does seem to have improved 
> performance, and no longer brings the app to its knees.
> 
> Unfortunately, some styles do need to be dashed - for example showing 
> tunnels that go underground.
> 
> <Rule>
> <Name>Tunnel</Name>
> <Title>Tunnel</Title>
> <MaxScaleDenominator>1500</MaxScaleDenominator>
> <ogc:Filter>
> <ogc:PropertyIsEqualTo>
> <ogc:PropertyName>TYPE</ogc:PropertyName>
> <ogc:Literal>Tunnel</ogc:Literal>
> </ogc:PropertyIsEqualTo>
> </ogc:Filter>
> <LineSymbolizer>
> <Stroke>
> <CssParameter name="stroke">#333333</CssParameter>
> <CssParameter name="stroke-width">0.2</CssParameter>
> <!-- CssParameter name="dasharray">3.0 1.0</CssParameter -->
> </Stroke>
> </LineSymbolizer>
> </Rule>
> 
> One workaround would be for me to use a dash above a certain scale, and 
> no dash below it.  Any SLD gurus advise on whether this can be done 
> within a single rule, or whether I need a separate rule for each 
> undashed style?

You need a separate rule.

Or you need to modify the rendering code so that it clips the lines
to the current rendering area before passing them to java2d... which
is really not as trivial as it seems, you have to take into 
consideration the stroke widths (the line might be sitting outside
parallel to the map border but the stroke be so think that it actually
shows up) and avoid using JTS intersection functions because they
are just too slow to be usable during rendering...
I know how to fix that, but unfortunately have no spare time to dedicate
to this issue... can you open a ticket on jira.codehaus.org?

Cheers
Andrea

-- 
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.

------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Geoserver-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geoserver-users

Reply via email to