Proposed change to symbols.

[Jason Maestri]

Okay, guys... This'll probably do it... I appreciate all of your help

--Jason

On Thu, 27 Jul 2000, Stephen Lime wrote:

> The simple solution pretty much already exists. Loop through the layers,
> all have to be off to start with, and one-by-one create a legend without
> any legend text (actually a " "). Do a drawLegend and save. There'd be
> some slop around the legend but it'd work.
> 
> You could also create mini maps using a virtual mapfile and one layer for point, 
> line and polygon data. An inline feature could be used to represent the line,
> center point or the polygon. Just loop over the layers and classes and
> use those values to set the class values for the appropriate layer in the
> virtual map, draw and save. 
> 
> You can reference the resulting images at will then in resulting docs.
> 
> s.
> 
> >>> <ender at titan.lab.csuchico.edu> 07/27/00 07:24AM >>>
> 
> I think I know what Jason wants, and I want it myself as well.  I'll
> describe what I want it for, and Jason can confirm if it's the same thing
> he's thinking of.
> 
> Say I want to create a legend similar to the one on DMSolutions page:
> 
> http://www2.dmsolutions.on.ca/gmap/gmap75.phtml 
> 
> I need to be able to get the individual images used in the legend table,
> so I could show the symbols as they actually are.  I could write a program
> to read the mapfile and spit out all of the classification images,
> allowing me to hardcode each symbol with an image representing that
> classification.  The problem with this is that it's static.  There's no
> way to dynamically get just the symbol after a dynamic reclassification
> using a call to mapserver.
> 
> The only solution to this problem that I can think of involves setting up
> sessions and having mapserver start to log hits.  I've thought of setting
> this up for years now, and have never had the time or money to do it, but
> that's a seperate topic...
> 
> I think that a simple sollution that just returned the image for a
> specific map, layer and class would be pretty easy to write, but would
> have very limited use.
> 
> -Aaron
> 
>  On Thu, 27 Jul 2000, Stephen Lime wrote:
> 
> > How do you mean query? You can get at that in MapScript. I'm not sure why you'd want
> > that capability in the C CGI program.
> > 
> > Steve
> > 
> > Stephen Lime
> > Internet Applications Analyst
> > 
> > Minnesota DNR
> > 500 Lafayette Road
> > St. Paul, MN 55155
> > 651-297-2937
> > 
> > >>> Jason Maestri <jmaes at riparia.net> 07/27/00 02:27AM >>>
> > Hey, while we're discussing symbols... As nearly as I can tell, there is no way to query mapserver for the symbol for a specific layer/class (for use in legends, and elsewhere),  is that something is possible, and if not, could it be made possible (Steve)?
> > 
> > --Jason
> > 
> > imap at chesapeake.net wrote:
> > 
> > > Steve/List,
> > >
> > > Being one of the advocates for symbol/marker changes, here is some rationale
> > > behind the changes.  The current symbol files are somewhat static and if the
> > > files become large, there is overhead incurred on every invocation of mapserv.
> > > Also, the current scheme also does not lend itself to dynamic marker placement
> > > very
> > > well.  (that is, using a marker NOT defined in marker.sym)
> > >
> > > We want to be able to use external markers from a PATH filespec to make it
> > > easier
> > > to maintain (no more editing the marker.sym file ;) and (2) make it more
> > > programmer
> > > friendly and simplifies the mapscript a bit.
> > >
> > > As far as functionality, it would be nice to interrogate a $class->{symbol} to
> > > determine it's height/width and some sort of option to set/get control of
> > > collision
> > > avoidance (labelcache).  (I think needed for both POINT and ANNOTATION
> > > layers) And lastly, make sure to preserve the ability to examine to labelcache
> > > to
> > > extract the polygons of the markers and text placed on the map.
> > >
> > > I am all for getting rid of marker.sym 100% although it may break my current
> > > applications...  sometimes change is painful.   Anyway to code an interim
> > > solutions that works both ways?  probably not, but thought I would ask anyway.
> > >
> > > Regards,
> > >
> > > Chris Stuber (mapsurfer)
> > > Silicon Mapping Solutions, Inc.
> > > 410.257.3187
> > >
> > > Stephen Lime wrote:
> > > >
> > > > I'd like to propose a couple of changes to the way symbols are handled. This is to make symbols easier to manage and to support some additional functionality proposed by a couple of other users. The initial change would be this:
> > > >
> > > > Do away with separate marker, line and shade symbol files in favor of a single symbol file. Also, allow symbol definitions in the map file.
> > > >
> > > > Syntax would be the same and access (i.e. symbols referenced in class objects) would be the same. These changes allow for a simpler MapScript symbol interface and additional opportunities for defining symbols on-the-fly. Begining in 3.3.011 you can already specify a GIF/PNG file (relative to the location of the map file) in a class object (i.e. SYMBOL "graphics/marker.gif") and a symbol is dynamically created. MapScript method $map->getSymbolByName("graphics/marker.gif"); will also work.
> > > >
> > > > Any thoughts or can I code it?
> > > >
> > > > Steve
> > 
> > 
> 
>