mapscript generic variable substitution syntax

[Stephen Woodbridge]

Rich,

You can't really do what you want this way. In the CGI the mapfile is 
pre-processed to do the substitution before it is handed off to the 
parser. I do not think you have access to that facility in mapscript.

The way other people have done the equivalent to this is to generate a 
template mapfile with generic objects. the then read call

    map = new mapObj(mapfile);

to get the incomplete mapfile obj and then read the variables that they 
need to process and make calls that modify the map object appropriately.

This is the standard way of doing this in mapscript.

-Steve W

rich.fromm wrote:
> I have a map file that I am reading in via Java mapscript:
> 
>          map = new mapObj(mapfile);
> 
> What I do is read the somewhat generic map file in, then substitute some
> variables based on the particular request.
> 
> I'm trying to figure out what is the proper way to specify the variables in
> the map file in this instance.
> 
> The following describes using [SquareBrackets] for HTML templates for CGI
> Mapserver:
> http://mapserver.gis.umn.edu/docs/reference/templatereference/templatereference
> 
> The following describes using %PercentageSigs% for run time substitution,
> again for CGI Mapserver:
> http://mapserver.gis.umn.edu/docs/howto/runsub
> 
> Neither of these precisely pertain to my situation, since I'm using
> Mapscript.  
> 
> The latter URL says:
>   Within MapScript this is easy to do since the developer has complete
> control over how input is handled.
> 
> But that's not entirely the case, since I want some input from map files and
> some to be programmatic.  I tried a few things.
> 
> For variable substitutions of numeric values, neither forms like %var% nor
> [var] works.  For instance, here are pairs of input lines from my map file
> and
> the corresponding output error messages:
> 
>     EXTENT		%minx% %miny% %maxx% %maxy%
> [Mon Jan  7 13:00:21 2008].149387 getDouble(): Symbol definition error.
> Parsing error near (%):(line 34)
> 
>     SIZE  		%width% %height%
> [Mon Jan  7 13:01:08 2008].196475 getInteger(): Symbol definition error.
> Parsing error near (%):(line 45)
> 
>     EXTENT		[minx] [miny] [maxx] [maxy]
> [Mon Jan  7 13:01:59 2008].337825 getDouble(): Symbol definition error.
> Parsing error near (minx] [miny] [maxx] [maxy):(line 34)
> 
>     SIZE  		[width] [height]
> [Mon Jan  7 13:03:24 2008].715273 getInteger(): Symbol definition error.
> Parsing error near (width] [height):(line 47)
> 
> A workaround here is to just specify some bogus values, and then directly
> set
> the keys I want to the values I want in the mapscript code:
> 
>     EXTENT  	-180 -90 180 90
>     SIZE		   1 1
> 
> For variable substitutions within numeric strings, I can get it to work with
> something like the following:
> 
>         FILTER		"the_geom && 'SRID=4326;LINESTRING(%minx% %miny%, %maxx%
> %maxy%)' AND feat_cod=2000123"
> 
> I do end up getting the following error:
> 
> [Mon Jan  7 13:15:35 2008].866125 getSymbol(): Symbol definition error.
> Parsing error near (the_geom && 'SRID=4326;LINESTRING(-122.3019062479681
> 37.83351284250283, -122.28009375203187 37.84648658222045)' AND
> feat_cod=2000123):(line 39)
> 
> But I get that even if I don't use variable substitution here and specify
> the
> values directly, so I think it's probably a separate issue.  But it doesn't
> seem to affect the output, so I think it's a false error msg and I'm not
> worrying too much about it, although I would be happier if it went away.
> 
> I searched the mailing list a bit trying to find an answer, but most of what
> I
> found pertains to using mapserver via CGI, not mapscript.
> 
> Thanks for any advice that anyone might have.
>