Opened 17 years ago
Closed 17 years ago
#2233 closed defect (fixed)
Polygon outline thickness keeps growing and growing...
Reported by: | sdlime | Owned by: | tbonfort |
---|---|---|---|
Priority: | normal | Milestone: | |
Component: | AGG | Version: | unspecified |
Severity: | normal | Keywords: | |
Cc: | tbonfort |
Description (last modified by )
Polygon outlines, e.g.
STYLE COLOR 0 153 204 OUTLINECOLOR 102 204 255 END
should always be 1 pixel but with AGG they seem to be being scaled. Here's an example:
The lakes are drawn with style above. I would expect a 1 pixel outline at all scales. Thick outlines require a seperate style since there's no way to specify outline size in a style.
Change History (4)
comment:1 by , 17 years ago
Description: | modified (diff) |
---|
comment:2 by , 17 years ago
comment:3 by , 17 years ago
Owner: | changed from | to
---|
should be fixed with r6598, you can close this once you've confirmed.
By the way, the WIDTH param has multiple meanings depending on context. I'd be quite keen on introducing an OUTLINEWIDTH keyword to disambiguate. (which could take floating point values, thin outlines with agg are cool ;)
comment:4 by , 17 years ago
Resolution: | → fixed |
---|---|
Status: | new → closed |
Scale factors are can be something other than 1 when SYMBOLSCALE is set at the layer level. Both width and size can be scaled, as could a symbol PATTERN or GAP.
Could consider OUTLINEWIDTH for 5.2. Should look at the SLD specs to see what that specifies. There are other params to consider too like miter and cap.
I was thinking that both WIDTH and SIZE should be made doubles.
(BTW, your fix worked, closing...)
Steve
yep I see... actually for a shade symbol I've been treating WIDTH as the outline width drawn around the polygon, which does make it easy to specify that kind of styling. And width is scaled by the scalefactor which seems to be the problem here.
What's the use case of the 'scalefactor' param for all these functions. here's the first time I've actually seen it kicking in at something != 1