[ka-Map-users] Technical alternatives for SVG on IE

[Paul Spencer]

There is also <canvas> for firefox/safari and a canvas-in-ie that  
wraps canvas syntax around vml for ie (http://me.eae.net/archive/ 
2005/12/29/canvas-in-ie/).  And reading the comments there is also an  
SVG for IE wrapper (http://starkravingfinkle.org/blog/2006/03/svg-in- 
ie/)

Google has also released an open source version of canvas in ie, but  
I haven't tried it (http://excanvas.sourceforge.net/)

Cheers

Paul

On 28-Apr-06, at 10:27 AM, Robert Thau wrote:

> Hi... I've noticed the recent conversation here on SVG overlays.  As a
> possible point of interest, Internet Explorer can display vector
> graphics *without* a plugin --- but not in SVG.  Instead, it supports
> its own, effectively proprietary, XML graphics description language
> called VML.  (Some Microsoft documentation refers people to a copy of
> the spec at http://www.w3.org/TR/NOTE-VML --- but that's a discussion
> draft that the W3C more or less ignored).
>
> The advantage of this is, obviously, you aren't requiring users to
> download a plugin.  The disadvantages are two: you're supporting two
> vector graphics formats, one of which (VML) does *not* have native
> support in, say, PostGIS --- and also, the VML engine actually gets
> kind of slow when there are large numbers of mouse-sensitive overlays
> on the screen.  But for just, say, putting a few lines or polygons
> on a map, it's fine.
>
> One other point --- I never tried <embed>, but I did try embedded SVG,
> using <object> tags in Firefox.  The result was behavior which I
> *think* I've seen described here: the SVG objects seem to "soak up"
> mouse events, so clicking on them (or dragging over them) doesn't work
> as you expect.  The reason for this, near as I could tell, was that if
> you do that, the SVG is considered to be part of a logically separate
> document, as if it had been loaded into an <iframe>, so mouse events
> do *not* bubble up from the SVG elements to the <object> tag and its
> enclosing HTML elements.  To make it work at all, I had to add
> handlers on the SVG nodes which, in effect, forwarded the mouse events
> up the chain --- and even then, you got messy visual artifacts while
> dragging the map.
>
> I found inline SVG (just including <svg> tags directly in the body of
> an HTML document) to work a whole lot better --- event propagation  
> works
> as you'd expect, and the artifacts don't show up.  However, in Firefox
> 1.5, this requires the document you're serving to be strict XHTML
> (and declared as such --- content-type "application/xhtml+xml");
> in this mode, close tags are mandatory, <br> *must* be typed as <br/>,
> uppercasing of tag names doesn't work, and so forth.
>
> (Brief plug: I've used a mixed inline-SVG/VML approach to graphics
> overlays in a GPLed project of my own, IvyGIS, which can be loosely
> described as "PostGIS and Mapserver on Rails and Scriptaculous"; if
> interested, see http://ivygis.justec.co.in/ for more.  Unfortunately,
> the demo is a bit slow at the moment due to network difficulties in
> Tokyo --- however, I'm hosting up the FGS modules containing the code
> on a server in New York, and installing that goes pretty quick).
>
> Robert Thau
> rst at alum.mit.edu
> _______________________________________________
> ka-Map-users mailing list
> ka-Map-users at lists.maptools.org
> http://lists.maptools.org/mailman/listinfo/ka-map-users

+-----------------------------------------------------------------+
|Paul Spencer                           pspencer at dmsolutions.ca   |
+-----------------------------------------------------------------+
|Applications & Software Development                              |
|DM Solutions Group Inc                 http://www.dmsolutions.ca/|
+-----------------------------------------------------------------+