Mapscript CGI/WMS Server call discrepancies

[Frank Warmerdam]

On 9/12/05, Adam Quiney <aquiney at refractions.net> wrote:
> Hello,
> 
> I've noticed a discrepancy between calls I'm making to our mapserver
> instance using WMS server calls and mapscript CGI calls.
> 
> The two calls I'm making are below:
> 
> WMS Server call:
> 
> http://turtle/cgi-bin/mapserv?map=/data/Refractions/BMGS/CWB/Web/WMS/SMT.map&request=getcapabilities&service=wms&version=1.1.1&request=GetMap&layers=SingleLineRiversCanalsMin&BBOX=1186497,398241,1190659,401758
> 
> Mapscript CGI call:
> 
> http://turtle/cgi-bin/mapserv?map=/data/Refractions/BMGS/CWB/Web/WMS/SMT.map&MODE=MAP&LAYER=SingleLineRiversCanalsMin&MINX=1186497&MINY=398241&MAXX=1190659&MAXY=401758
> 
> In both cases, I am requesting the same bounding box, and the same layer.
> However, the CGI call includes a fork at the bottom of the right-most
> segment, and is missing a portion of the segment at the top, when compared
> to the WMS server call.

Adam,

WMS requests will by default return non-square pixels if requested
while "old style" cgi requests will grow the bbox in one direction or the
other if needed in order to ensure square pixels.  You don't select 
explicit sizes in the requests above so I don't know if there is an 
aspect ratio issue or not, but it seems like the most likely cause. 

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent