[Mapserver-users] Projection woes, Revisited.
Hankley, Chip
Chip.Hankley at GASAI.Com
Tue Feb 4 11:51:41 PST 2003
OK,
I've added another demo onto the page:
http://207.250.243.23/projtest/projtest.html
This last one has only the two 'boxes', and the output projection is WI
State Plane South, NAD83, meters (epsg:32154). Note that this necessitated
me creating a new data set, since my other SPS83 box was FEET. The epsg code
assumes the data is meters.
I also changed the NAD27 box to be transparent with a red outline for the
last two examples (just to show for sure that it is not reprojecting).
Also, just in case anyone was wondering... those are in-line images being
created on the fly using MS (look at the source) so.. you are seeing the
real deal - i.e. I'm not making this up ;)
Thanks for the interest and help!
Chip
> -----Original Message-----
> From: Ed McNierney [mailto:ed at topozone.com]
> Sent: Tuesday, February 04, 2003 1:07 PM
> To: Hankley, Chip; MapServer List (E-mail)
> Subject: RE: [Mapserver-users] Projection woes, Revisited.
>
>
> Chip -
>
> Thanks for posting all the data and samples. I think it
> might be helpful to simplify things a bit.
>
> Here's what I see. You have two boxes, of the same extents,
> one in Wisconsin South NAD27 Feet, the other in Wisconsin
> South NAD83 Feet (both LCC projections). Let's ignore
> everything else and try to get those two to overlap.
>
> First, you might want to use the EPSG codes for these
> projections. They're more obtuse, but they protect against
> typos when spelling out all the projection parameters.
>
> For Wisconsin State Plane South NAD27, the code is 32054, and
> for NAD83 it's 32154. So each of your projection blocks can become:
>
> # NAD27
>
> PROJECTION
> "init=epsg:32054"
> END
>
> # NAD83
>
> PROJECTION
> "init=epsg:32154"
> END
>
> This is similar to your last example. In looking at your
> last example, however, one wonders whether the NAD83 box is
> right there UNDERNEATH the red NAD27 box - if it were
> projecting properly, how could you tell?
>
> Can you set up another example that does the following:
>
> 1. Uses the EPSG codes for projections
> 2. Defines the output projection to be Wisconsin State Plane
> South NAD83
> 3. Defines only two input layers
> 4. Makes the OUTLINECOLOR for the NAD27 box be "255 0 0"
> (red) and removes the COLOR statement, so it's hollow?
>
> Then let us take a look. Thanks!
>
> - Ed
>
>
> -----Original Message-----
> From: Hankley, Chip [mailto:Chip.Hankley at GASAI.Com]
> Sent: Tuesday, February 04, 2003 1:18 PM
> To: Ed McNierney; MapServer List (E-mail)
> Subject: RE: [Mapserver-users] Projection woes, Revisited.
>
>
> Ed (and everyone) -
>
> OK... showing my ignorance here. I ALWAYS use pre-compiled
> binaries (I'm
> trying to overcome this personal defect...). Anyway, I'm only
> beginning to
> understand the compilation sequence, and how all the parts
> fits together...
> so, patience please!
>
> When I download the binaries from the Mapserver site, the
> only PROJ related
> file that comes with the distribution is the epsg file. I have this in
> C:\PROJ\NAD.
>
> I did download the datum shift files (from remotesensing.org)
> and simply put
> them in the NAD subdirectory, but this doesn't seem to make
> any difference
> (not suprisingly). I'm assuming that when you say that I
> "need to have these
> datum shift files set up and PROJ pointed to them in order to
> do the shift"
> that this is an operation that would occur when compiling the
> PROJ4 library
> prior to compiling MapServer? Have I got that sequence right?
>
> If this is right, why don't the pre-compiled binaries that
> exist already do
> this (just curious)? Seems like datum shifts would be a fairly common
> task...
>
> Chip
>
>
>
> > -----Original Message-----
> > From: Ed McNierney [mailto:ed at topozone.com]
> > Sent: Tuesday, February 04, 2003 11:36 AM
> > To: Hankley, Chip; MapServer List (E-mail)
> > Subject: RE: [Mapserver-users] Projection woes, Revisited.
> >
> >
> > Chip -
> >
> > The PROJ4 FAQ refers to the work you need to do to have the
> > NAD datum shift files built - the code doesn't change. Do
> > you have your PROJ installation in C:\PROJ? Whereever it is,
> > look in that directory for a NAD subdirectory and files named
> > CONUS, NAD27, NAD83, ALASKA, etc. (no file extensions). You
> > need to have these datum shift files set up and PROJ pointed
> > to them in order to do the shift.
> >
> > - Ed
> >
> > Ed McNierney
> > President and Chief Mapmaker
> > TopoZone.com / Maps a la carte, Inc.
> > 73 Princeton Street, Suite 305
> > North Chelmsford, MA 01863
> > ed at topozone.com
> > (978) 251-4242
> >
> >
> > -----Original Message-----
> > From: Hankley, Chip [mailto:Chip.Hankley at gasai.com]
> > Sent: Tuesday, February 04, 2003 12:17 PM
> > To: MapServer List (E-mail)
> > Subject: [Mapserver-users] Projection woes, Revisited.
> >
> >
> > OK, I'm still unable to do a datum shift between NAD27 and NAD83.
> >
> > I've tried to strip out all of the non-essentials and come
> up with an
> > example. You can view it on-line at:
> >
> > http://207.250.243.23/projtest/projtest.html
> >
> > Basically, I can reproject data FINE when it's in the same
> > datum. It's when
> > I try to do a datum shift that things don't work. I'm using
> one of the
> > pre-compiled binaries from the MS site. Here's my version info:
> >
> > MapServer version 3.6.3 OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
> > SUPPORTS=PROJ
> > SUPPORTS=TTF SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
> > INPUT=TIFF INPUT=EPPL7
> > INPUT=JPEG INPUT=OGR INPUT=SHAPEFILE
> >
> > Could the problem be with the way PROJ4 was compiled into
> > this binary? The
> > PROJ4 FAQ has some specific instructions for building PROJ4
> > to support datum
> > shifts... (see http://www.remotesensing.org/proj/faq.html). I
> > don't know how
> > to check this on the binary...
> >
> > TIA!
> >
> > Chip
> > _______________________________________________
> > Mapserver-users mailing list
> > Mapserver-users at lists.gis.umn.edu
> > http://lists.gis.umn.edu/mailman/listinfo/mapserver-users
> >
>
More information about the MapServer-users
mailing list