[mapserver-users] Rotated TrueType Point Symbols - Issue

[Fawcett, David]

Donald, 

I have seen this same issue using MapServer 5.2.0.  I first noticed it
when I was doing an attribute binding example, but I can confirm that it
also occurs when using a fixed angle.  

When specifying a fixed angle of 0 or 360, the symbol lands correctly on
the point.  An angle of 90 sets it off of the point to the left.  180
puts it to the lower left, 270 is lower center, 359 is almost back on
center.

The more that I look at this, it really looks like it is This related to
the proper centering of symbols.  My square .ttf character appears to be
rotating around the lower left corner of the glyph.  When the symbol is
placed unrotated, it is correctly centered on the point feature.  So,
maybe the rotation point needs to be calculated the same way as the
placement point.

I was quickly trying to pull together a demo, so I abandoned the .ttf
approach and used a pixmap symbol of an arrow.  

I think that it is likely a bug, I should have filed a bug report back
then.  I will do so now if you aren't interested in doing it.

David.

-----Original Message-----
From: mapserver-users-bounces at lists.osgeo.org
[mailto:mapserver-users-bounces at lists.osgeo.org] On Behalf Of Donald
Kerr
Sent: Friday, January 02, 2009 9:58 AM
To: mapserver-users at lists.osgeo.org
Subject: [mapserver-users] Rotated TrueType Point Symbols - Issue


I have created a number of TrueType symbols most of which are good
enough when positioned on the map at their default centre point.

There appears to be an issue when these are then rotated using ANGLE.
They do not appear to rotate around their centre point which can make
them look very odd when placed on the map.

Please see attached images one showing without the ANGLE applied and the
other with. If the symbol rotated around its centre point then the arrow
would simply show pointing the other way (180 degrees).

I tried to make a TTF symbol that had the head of the arrow in the
centre in the hope that it would rotate on this point. It shows
correctly when no ANGLE is applied i.e. the head of the arrow over the
point on the map but, when rotated, the error in placement is
exaggerated.

Here are the CLASS and SYMBOL definitions:

CLASS
	NAME "benchMarkSymbol"
	EXPRESSION "benchMarkSymbol"
	STYLE
		SYMBOL "benchMarkSymbol"
		SIZE 12
		MAXSIZE 12
		MINSIZE 4
		COLOR 0 0 0
		ANTIALIAS TRUE
		ANGLE [orientation]
	END # STYLE
END # CLASS

SYMBOL
	NAME 'benchMarkSymbol'
	TYPE TRUETYPE
	FONT OSMasterMapSymbols
	FILLED false
	ANTIALIAS true
	CHARACTER '7'
END # SYMBOL

The question is, am I doing something wrong or is this a bug? If the
latter, is there a workaround meantime? For me just now, I'd be happy if
the symbol rotated around its centre - I could live with that as a
workaround.

In a previous thread that I started about positioning of LABELs, Thomas
Bonfort said, "I'd be in favor of having a way to specify the anchor
point, either by fixing it on 0,0 , or by specifying it alongside the
symbol, eg ANCHORPOINT 0.3 2". I agree that this would be good for
symbols too but it may present a problem with scaled symbols, especially
TTF. If it's not a fixed size then how can an anchor point be specified?

Many thanks.

Regards,

Donald Kerr