[mapserver-users] mismatches tiling images produced by mapserver
Ed McNierney
ed at topozone.com
Thu Oct 25 17:45:13 PDT 2001
Brian -
As far as I can tell, your images fit together perfectly. Is it
possible the problem is or was in the cell spacing/padding of the IMG
tags inside your HTML table?
While it's certainly possible there are bugs, I have been using tile4ms
on literally tens of thousands of vector and raster files over the last
few months and have yet to find a bug.
- Ed
Ed McNierney
Chief Mapmaker
TopoZone.com
-----Original Message-----
From: Brian Hill [mailto:bhill at nextbus.com]
Sent: Thursday, October 25, 2001 8:09 PM
To: mapserver-users at lists.gis.umn.edu
Subject: [mapserver-users] mismatches tiling images produced by
mapserver
I am still wringing the bugs out of a new 3.3.012 installation, and am
hoping someone on the list will have experience with this one.
Our application needs to tile images produced by the mapserver. I.e.,
if an image is produced with extent,
(lon - delta_lon / 2, lat), (lon + delta_lon / 2, lat),
then it should fit very neatly to the left of an image with extent,
(lon + delta_lon / 2, lat), (lon + 3 * delta_lon / 2, lat).
Here is an example of four tiles which should fit together:
http://skip.nextbus.com/tile-skip.html
The .html file hits a perl wrapper for the map server four times. As
you can see by scrolling to the center of the image, the four panes do
not fit together neatly. The projection is Mercator, so tiles bounded
by lines of constant latitude and constant longitude should fit.
If there are rounding errors or some other problem down in the C
libraries that is partly or completely understood and could explain
this, please let me know.
Wading directly into the sources is a bit daunting, especially given
that any of three layers could be involved (perl/mapscript,
mapserver-3.3.012 or proj-4.4.4), but it is critical for our application
that the tiling come out neatly, so with some pointers, I am motivated
and might be able to track it down. It is possible the bug is in the
perl file that makes all the mapscript calls, but I think it is lower
down, and therefore that fixing it might be of general interest.
Brian Hill
Sr. Software Engineer
NextBus Information Systems
More information about the MapServer-users
mailing list