GeoServer

Various issues converting a CSS style with scale dependencies

Details

  • Type: Bug Bug
  • Status: Closed Closed
  • Priority: Major Major
  • Resolution: Fixed
  • Affects Version/s: None
  • Fix Version/s: 2.0.3, 2.1-beta1
  • Component/s: None
  • Labels:
    None
  • Number of attachments :
    0

Description

Trying to convert this style with scale dependencies:

[PERSONS < 2000000] {
  fill : #4DFF4D;
}

[PERSONS >= 2000000] [PERSONS < 4000000] {
  fill: #FF4D4D;
}

[PERSONS >= 4000000] {
  fill: #4D4DFF;
}


* {
  stroke: black;
  stroke-width: 0.2;
  fill-opacity: 0.8;
  font-family: "Times New Roman";
  font-size: 14;
}

[@scale > 40000000] {
    label: [STATE_ABBR];
}

[@scale < 40000000] {
    label: [STATE_NAME];
}

I'm getting quite a bit of duplicated rules, but also colors that look like:

<sld:CssParameter name="fill">  <ogc:Literal>java.awt.Color[r=128,g=128,b=128]</ogc:Literal>  </sld:CssParameter>

My guess is that the code building the style is setting in a java.awt.Color instead of a string or something like that... maybe it's a bug in the encoder, but it's a little odd that the first symbolizers have the colors properly translated, just the last ones have this issue

Issue Links

depends upon

Bug - A problem which impairs or prevents the functions of the product. GEOT-3036 Default fill parameters use Color object instead of hex string

  • Minor - Minor loss of function, or other problem where easy workaround is present.
  • Closed - The issue is considered finished, the resolution is correct. Issues which are not closed can be reopened.

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
David Winslow added a comment -

I've noticed this before; it seems to happen only on the fills for TextSymbolizers, so I suspect a bug in the GeoTools encoder. If you add a 

font-fill: black;

Then the SLD should have regular hex codes instead of java.awt.Color...

As for the duplicated rules, well, it used to be worse Eliminating repetition for rules that don't change between scale levels is something I'd like to do in the future, but as long as you don't need a legend it shouldn't cause any visible problems at rendering time, so it is a little lower on the priority list than some other issues with the CSS converter.

Show
David Winslow added a comment - I've noticed this before; it seems to happen only on the fills for TextSymbolizers, so I suspect a bug in the GeoTools encoder. If you add a font-fill: black; Then the SLD should have regular hex codes instead of java.awt.Color... As for the duplicated rules, well, it used to be worse Eliminating repetition for rules that don't change between scale levels is something I'd like to do in the future, but as long as you don't need a legend it shouldn't cause any visible problems at rendering time, so it is a little lower on the priority list than some other issues with the CSS converter.
Hide
Permalink
David Winslow added a comment -

r14292 cleans this up a fair bit (although the problem with Color objects leaking out is still there). Andrea, if you could take a look sometime and let me know how it works for you that would be great.

Show
David Winslow added a comment - r14292 cleans this up a fair bit (although the problem with Color objects leaking out is still there). Andrea, if you could take a look sometime and let me know how it works for you that would be great.
Hide
Permalink
Jody Garnett added a comment -

Please use expr.evaluate( feature, String.class ) if you are wanting strings back; even with a Literal expression such as a ConstantExpression (as used by Fill) you can do this.

rather then use
if( expr instanceof Literal) return expr.getLiteral().toString();

if( expr instanceof Literal) return expr.evaualte( null, String.class );

This will allow the use of GeoTools converters which will handle the Color -> String conversation for you.

Show
Jody Garnett added a comment - Please use expr.evaluate( feature, String.class ) if you are wanting strings back; even with a Literal expression such as a ConstantExpression (as used by Fill) you can do this. rather then use if( expr instanceof Literal) return expr.getLiteral().toString(); if( expr instanceof Literal) return expr.evaualte( null, String.class ); This will allow the use of GeoTools converters which will handle the Color -> String conversation for you.
Hide
Permalink
David Winslow added a comment -

I'm not evaluating expressions anywhere; the CSS file gets converted to an instance of org.geotools.styling.Style and passed to GeoTools' SLDTransformer.transform(). When no color is specified for the label, the corresponding SLD Fill is created with StyleFactory.createFill(null, null, null, null) which seems to be the root of the problem. Anyway with the patch for GEOT-3036 applied all should be well.

Show
David Winslow added a comment - I'm not evaluating expressions anywhere; the CSS file gets converted to an instance of org.geotools.styling.Style and passed to GeoTools' SLDTransformer.transform(). When no color is specified for the label, the corresponding SLD Fill is created with StyleFactory.createFill(null, null, null, null) which seems to be the root of the problem. Anyway with the patch for GEOT-3036 applied all should be well.
Hide
Permalink
Andrea Aime added a comment -

I see the GeoTools one got closed as fixed, but in fact no code was changed (besides some Jody's reformats in the code and a new test case).
Anything else to do for this one?

Show
Andrea Aime added a comment - I see the GeoTools one got closed as fixed, but in fact no code was changed (besides some Jody's reformats in the code and a new test case). Anything else to do for this one?
Hide
Permalink
Andrea Aime added a comment -

Hem... yeah, I guess... me trying it out again

Show
Andrea Aime added a comment - Hem... yeah, I guess... me trying it out again
Hide
Permalink
Andrea Aime added a comment -

Mass closing all issues that have been in "resolved" state for 2 months or more without any feedback or update

Show
Andrea Aime added a comment - Mass closing all issues that have been in "resolved" state for 2 months or more without any feedback or update

People

  • Assignee:
    David Winslow
    Reporter:
    Andrea Aime
Vote (0)
Watch (0)

Dates

  • Created:
    Updated:
    Resolved: