I was meaning in the layers view; see in your stack trace this line?
- at net.refractions.udig.ui.graphics.Glyph$6.getImageData(Glyph.java:363)
That Glyph class is only used by the Layers view so it can have something to
show for your layer. I think if your style is so much fun; it should focus on
just showing the correct shape (point, line, polygon). At least for the short
term :-) I could try and ask it to go through your function and return the
first color it sees - in your example that would be #FFBB55. That would at
least allow it to show something different if you have more than one style.
Jody
On 06/08/2010, at 11:14 PM, Lukasz Stawicki wrote:
> Jody,
>
> This error appears even if I do not have 'Legend' layer created.
>
> regards
>
> Lukasz
>
> Cytowanie Jody Garnett <jody.garn...@gmail.com>:
>
>> It is just for the legend graphic I think.
>>
>>
>>
>> On 06/08/2010, at 7:31 PM, Lukasz Stawicki <s...@globema.pl> wrote:
>>
>>> Jody,
>>>
>>> I feel that catching an error and then drawing b&w is not a perfect
>>> solution - as I refer inside the function to attributes of feature I want
>>> to draw, maybe just use the real feature to be drawn, not the fake one?
>>>
>>> Here is problematic part of my SLD. It is for some electrodevice, and I do
>>> not know default uDig dataset at all, so you would need to change property
>>> name and literals for limits.
>>>
>>> Lukasz
>>>
>>>
>>> <sld:LineSymbolizer>
>>> <sld:Stroke>
>>> <sld:CssParameter name="stroke">
>>> <ogc:Function name="if_then_else">
>>> <ogc:Function name="lessThan">
>>>
>>> <ogc:PropertyName>VOLTAGE_ID</ogc:PropertyName>
>>> <ogc:Literal>3</ogc:Literal>
>>> </ogc:Function>
>>> <ogc:Literal>#FFBB55</ogc:Literal>
>>> <ogc:Function name="if_then_else">
>>> <ogc:Function name="lessThan">
>>>
>>> <ogc:PropertyName>VOLTAGE_ID</ogc:PropertyName>
>>> <ogc:Literal>7</ogc:Literal>
>>> </ogc:Function>
>>> <ogc:Literal>#F78484</ogc:Literal>
>>> <ogc:Literal>#5C993C</ogc:Literal>
>>> </ogc:Function>
>>> </ogc:Function>
>>> </sld:CssParameter>
>>> <sld:CssParameter name="stroke-linecap">
>>> <ogc:Literal>butt</ogc:Literal>
>>> </sld:CssParameter>
>>> <sld:CssParameter name="stroke-linejoin">
>>> <ogc:Literal>miter</ogc:Literal>
>>> </sld:CssParameter>
>>> <sld:CssParameter name="stroke-opacity">
>>> <ogc:Literal>1</ogc:Literal>
>>> </sld:CssParameter>
>>> <sld:CssParameter name="stroke-width">
>>> <ogc:Literal>3</ogc:Literal>
>>> </sld:CssParameter>
>>> <sld:CssParameter name="stroke-dasharray">4
>>> 4</sld:CssParameter>
>>> </sld:Stroke>
>>> </sld:LineSymbolizer>
>>>
>>>
>>> Cytowanie Jody Garnett <jody.garn...@gmail.com>:
>>>
>>>> Oh cool.
>>>>
>>>> Yeah the glphy code creates a fake feature in the hopes of drawing
>>>> something "in your style". We should put some catch statements around
>>>> that and perhaps draw the feature in B&W if we were unable to figure
>>>> out the style?
>>>>
>>>> Lukasz can you sort out a way for us to reproduce the problem with the
>>>> default dataset? perhaps just an SLD we can cut and paste into the
>>>> style editor?
>>>>
>>>> Jody
>>>>
>>>> 2010/8/6 Lukasz Stawicki <s...@globema.pl>:
>>>>> Hi,
>>>>>
>>>>> I used <ogc:Function name="Categorize"> instead of if_then_else. Anyway,
>>>>> it
>>>>> crashes in CategorizeFunction.java line 152:
>>>>>
>>>>> String lookupValue = lookupExp.evaluate(object, String.class);
>>>>>
>>>>> Object is null - I debugged it up and finally digged what a object is
>>>>> evaluated. net.refractions.udig.ui.graphics.Glyph.java, line 363.
>>>>>
>>>>>
>>>>> d.drawDirect( image, display,
>>>>> d.feature(d.line(new int[]{0,12, 6,3, 11,12,
>>>>> 15,3})),
>>>>> rule );
>>>>>
>>>>> Line is created for line symbolizer, and as this line is not a feature, it
>>>>> causes side-effects. I also checked that if_then_else comes into problems
>>>>> just in the same place.
>>>>>
>>>>> Any help would be greatly appreciated.
>>>>>
>>>>> regards
>>>>>
>>>>> Lukasz
>>>>>
>>>>> Cytowanie Jody Garnett <jody.garn...@gmail.com>:
>>>>>
>>>>>> Please keep discussion on the list (that way I can be corrected by others
>>>>>> :-D ).
>>>>>>
>>>>>> Sorry for the quick message before; was focused on getting a bug report
>>>>>> in. I was not being intentionally opaque - the geotools website is down
>>>>>> for
>>>>>> me as well.
>>>>>>
>>>>>> Looking at the code the functions are:
>>>>>> - CategorizeFunction
>>>>>> - InterpolateFunction
>>>>>> - RecodeFunction
>>>>>>
>>>>>> Example docs for the first one:
>>>>>> Implementation of "Categorize" as a normal function.
>>>>>>
>>>>>> This implementation is compatible with the Function interface; the
>>>>>> parameter list can be used to set the threshold values etc...
>>>>>>
>>>>>> This function expects:
>>>>>>
>>>>>> PropertyName; use "Rasterdata" to indicate this is a color map
>>>>>> Literal: lookup value
>>>>>> Literal: threshold 1
>>>>>> Literal: value 1
>>>>>> Literal: threshold 2
>>>>>> Literal: value 2
>>>>>> Literal: (Optional) succeeding or preceding
>>>>>> For more details please look at the Symbology Encoding specification
>>>>>> (which provides the definition of these functions and examples.
>>>>>>
>>>>>> Best thing to do is look at test cases:
>>>>>> -
>>>>>> http://svn.osgeo.org/geotools/trunk/modules/library/render/src/test/java/org/geotools/filter/
>>>>>>
>>>>>> Here is an example from the test cases:
>>>>>>
>>>>>> final String[] categories = {"low", "mid", "high", "super"};
>>>>>> final Double[] thresholds = {0.0, 50.0, 100.0};
>>>>>> setupParameters(categories, thresholds);
>>>>>>
>>>>>> Function fn = finder.findFunction("categorize", parameters);
>>>>>>
>>>>>> Cheers,
>>>>>> Jody
>>>>>>
>>>>>>
>>>>>> On 05/08/2010, at 7:00 PM, Lukasz Stawicki wrote:
>>>>>>
>>>>>>> Jody,
>>>>>>>
>>>>>>> I know I can use more <sld:Rule...> entries instead of using
>>>>>>> <ogc:Function name="if_then_else"> but this will make my SLD file 10
>>>>>>> times
>>>>>>> longer (and unable to maintain it). What are the other options for
>>>>>>> managing
>>>>>>> colors?
>>>>>>>
>>>>>>> regards
>>>>>>>
>>>>>>> Lukasz
>>>>>>>
>>>>>>>
>>>>>>> Cytowanie Jody Garnett <jody.garn...@gmail.com>:
>>>>>>>
>>>>>>>> The GeoTools website was hacked last month; I think the DNS is being
>>>>>>>> updated now to point to a new server.
>>>>>>>>
>>>>>>>> As for your example; you are doing a lot of work to map colours? There
>>>>>>>> is a couple of color lookup functions available now
>>>>>>>> which should make it easier.
>>>>>>>>
>>>>>>>> I have reported your issue here:
>>>>>>>> - https://jira.codehaus.org/browse/GEOT-3234
>>>>>>>>
>>>>>>>> On 05/08/2010, at 4:14 PM, Lukasz Stawicki wrote:
>>>>>>>>
>>>>>>>>>
>>>>>>>>> java.lang.IllegalArgumentException: Filter Function problem for
>>>>>>>>> function if_then_else argument #0 - expected type boolean
>>>>>>>>> at
>>>>>>>>> org.geotools.filter.function.FilterFunction_if_then_else.evaluate(FilterFunction_if_then_else.java:45)
>>>>>>>>> at
>>>>>>>>> org.geotools.filter.expression.ExpressionAbstract.evaluate(ExpressionAbstract.java:63)
>>>>>>>>> at org.geotools.styling.SLD.color(SLD.java:1312)
>>>>>>>>> at org.geotools.styling.SLD.color(SLD.java:96)
>>>>>>>>> at org.geotools.styling.SLD.lineColor(SLD.java:84)
>>>>>>>>> at org.geotools.styling.SLD.color(SLD.java:188)
>>>>>>>>> at net.refractions.udig.ui.Drawing.paint(Drawing.java:308)
>>>>>>>>> at net.refractions.udig.ui.Drawing.drawFeature(Drawing.java:246)
>>>>>>>>> at net.refractions.udig.ui.Drawing.drawFeature(Drawing.java:223)
>>>>>>>>> at net.refractions.udig.ui.Drawing.drawDirect(Drawing.java:150)
>>>>>>>>> at
>>>>>>>>> net.refractions.udig.ui.graphics.Glyph$6.getImageData(Glyph.java:363)
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
>
>
_______________________________________________
User-friendly Desktop Internet GIS (uDig)
http://udig.refractions.net
http://lists.refractions.net/mailman/listinfo/udig-devel
