Why is the bounding box/resolution from the original request not honored?

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|

Why is the bounding box/resolution from the original request not honored?

chris marx
Hi,
 I just got tilecache working on xp and win 2003, and with my wms (using manifold gis) the open layers example works great. However, I configured tilecache in the hopes of using it in conjunction with google maps. However, it appears that tilecache is not honoring the wms requests that are being generated from the script in google maps (see below). The requests are valid, if I point the script directly at the wms, the tiles come back just fine. But when i put tilecache in between, I get 

An error occurred: Current y value 37.718590 is too far from tile corner y 37.968750

You can see the working google map with tiles coming back directly from the wms here:


http://warbler2.cit.cornell.edu/pfw-maps/wms-without-cache.html


the entry in tilecache.cfg looks like this

[google-tiles]
type=WMS
url=http://mycomputer.com/local-wms/wms.asp
layers=Active_Year_2005
tms_type=google


I use the following script to convert the mercator coords to lat lng,  which i modified from here (http://johndeck.blogspot.com/#114071052432996324)

/**
 * GoogleMapsWmsOverlays
 * @param {Object} opts
 *     @param {String} myLayers The layers requested from wms
 *  @param {String} myFormat The image format
 *  @param {String} myBaseURL The url of the wms server
 *  @param {Number} myOpacity opacity
 *  @param {String} myStyles wms styles
 *  @param {Number} myMercLevel The level at which switch to lat/lng is made??
 */
function GoogleMapsWmsOverlay(opts){
    var me = this;
    me.MAGIC_NUMBER=6356752.3142;
    me.WGS84_SEMI_MAJOR_AXIS = 6378137.0;
    me.WGS84_ECCENTRICITY = 0.0818191913108718138;
    me.DEG2RAD=0.0174532922519943;
    me.PI=3.14159267;
    me.FORMAT_DEFAULT="image/png"; //Default image format, used if none is specified
    me.MERC_ZOOM_DEFAULT = 15; //Google Maps Zoom level at which we switch from Mercator to Lat/Long.
   
    me.Options = {
        myLayers: opts.myLayers || "error",
        myFormat: opts.myFormat || me.FORMAT_DEFAULT,
        myBaseURL: opts.myBaseURL || "error",
        myStyles: opts.myStyles || ""
    }
}

GoogleMapsWmsOverlay.prototype.dd2MercMetersLng = function(p_lng) {
    var me = this;
    return me.WGS84_SEMI_MAJOR_AXIS * (p_lng * me.DEG2RAD);
}

GoogleMapsWmsOverlay.prototype.dd2MercMetersLat = function(p_lat) {
    var me = this;
    var lat_rad = p_lat * me.DEG2RAD;
    return me.WGS84_SEMI_MAJOR_AXIS *
                    Math.log(Math.tan((lat_rad + me.PI / 2) / 2) *
                         Math.pow( ((1 - me.WGS84_ECCENTRICITY * Math.sin(lat_rad)) /
                              (1 + me.WGS84_ECCENTRICITY * Math.sin(lat_rad))), (me.WGS84_ECCENTRICITY/2)));
}

/**
 * use?
 */
GoogleMapsWmsOverlay.prototype.customOpacity = function() {
     
     return this.myOpacity;
}

GoogleMapsWmsOverlay.prototype.customGetTileUrl = function(a,b,c) {
    var opts = this.Options;
    /*if (opts.myMercZoomLevel == undefined) {
        opts.myMercZoomLevel = MERC_ZOOM_DEFAULT;
    }*/
   
    var lULP = new GPoint(a.x*256,(a.y+1)*256);
    var lLRP = new GPoint((a.x+1)*256,a.y*256);
    var lUL = G_NORMAL_MAP.getProjection().fromPixelToLatLng(lULP,b,c);
    var lLR = G_NORMAL_MAP.getProjection().fromPixelToLatLng(lLRP,b,c);

    // switch between Mercator and DD if merczoomlevel is set
    // NOTE -it is now safe to use Mercator exclusively for all zoom levels (if your WMS supports it)
    // so you can just use the two lines of code below the IF (& delete the ELSE)
    // drg & doq are topozone layers--- they don't work with epsg:54004
     /*if (opt.myLayers!="drg" && opt.myLayers!="doq") {
        var lBbox=dd2MercMetersLng(lUL.x)+","+dd2MercMetersLat(lUL.y)+","+dd2MercMetersLng(lLR.x)+","+dd2MercMetersLat(lLR.y);
        //Change for GeoServer - 41001 is mercator and installed by default.
        var lSRS="EPSG:54004";
    } else {*/
        //var lBbox=lUL.x.toFixed(6)+","+lUL.y.toFixed(6)+","+lLR.x.toFixed(6)+","+lLR.y.toFixed(6);
        var lBbox=lUL.x+","+lUL.y+","+lLR.x+","+lLR.y;
        var lSRS="EPSG:4326";
    //}
    var lURL=opts.myBaseURL;
    lURL+="?REQUEST=GetMap";
    lURL+="&SERVICE=WMS";
    lURL+="&VERSION=1.1.1";
    lURL+="&EXCEPTIONS=application/vnd.ogc.se_inimage";
    lURL+="&LAYERS="+opts.myLayers;
    lURL+="&STYLES="+opts.myStyles;
    lURL+="&FORMAT="+opts.myFormat;
    lURL+="&BGCOLOR=0xFFFFFF";
    lURL+="&TRANSPARENT=TRUE";
    lURL+="&SRS="+lSRS; //use CRS for v1.3.0
    lURL+="&BBOX="+lBbox;
    lURL+="&WIDTH=256";
    lURL+="&HEIGHT=256";
    //lURL+="&reaspect=false";
    //document.write(lURL + "<br/>")
    //alert(" url is " + lURL);
    return lURL;
}

GoogleMapsWmsOverlay.prototype.getTileLayer = function(){
    var me = this;
    var tileLayer = new GTileLayer(new GCopyrightCollection(""), 0, 17);
    tileLayer.Options = me.Options;
      tileLayer.getTileUrl = me.customGetTileUrl;
    return tileLayer;
}


My maps in manifold are in mercator, but it has no problem accpeting the lat/lng requests, and spitting back the correct mercator tiles. Any help would be greatly appreciated!
chris-

--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/

_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

Christopher Schmidt-2
On Sun, Nov 09, 2008 at 12:36:09PM -0500, chris marx wrote:

> Hi,
>  I just got tilecache working on xp and win 2003, and with my wms (using
> manifold gis) the open layers example works great. However, I configured
> tilecache in the hopes of using it in conjunction with google maps. However,
> it appears that tilecache is not honoring the wms requests that are being
> generated from the script in google maps (see below). The requests are
> valid, if I point the script directly at the wms, the tiles come back just
> fine. But when i put tilecache in between, I get
>
> An error occurred: Current y value 37.718590 is too far from tile
> corner y 37.968750
>
> You can see the working google map with tiles coming back directly from the
> wms here:

You must use Spherical Mercator to overlay tiles on Google Maps.

http://trac.openlayers.org/wiki/SphericalMercator


Regards,
--
Christopher Schmidt
MetaCarta
_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

chris marx
Hi,
 Sorry about the bad link everyone, the site is now up

http://warbler2.cit.cornell.edu/pfw-maps/wms-without-cache.html

you can see that the tiles line up just fine (except when you zoom out from the initial zoom, i believe this is a manifold wms problem, i'm looking into that) but zooming in is ok. if you're looking at the wms requests, in firebug, these same requests generate the

An error occurred: Current y value 37.718590 is too far from tile corner y 37.968750


when i put tilecache in front of the wms.
I appreciate any advice!
chris-

On Sun, Nov 9, 2008 at 8:05 PM, Christopher Schmidt <[hidden email]> wrote:
On Sun, Nov 09, 2008 at 12:36:09PM -0500, chris marx wrote:
> Hi,
>  I just got tilecache working on xp and win 2003, and with my wms (using
> manifold gis) the open layers example works great. However, I configured
> tilecache in the hopes of using it in conjunction with google maps. However,
> it appears that tilecache is not honoring the wms requests that are being
> generated from the script in google maps (see below). The requests are
> valid, if I point the script directly at the wms, the tiles come back just
> fine. But when i put tilecache in between, I get
>
> An error occurred: Current y value 37.718590 is too far from tile
> corner y 37.968750
>
> You can see the working google map with tiles coming back directly from the
> wms here:

You must use Spherical Mercator to overlay tiles on Google Maps.

http://trac.openlayers.org/wiki/SphericalMercator


Regards,
--
Christopher Schmidt
MetaCarta



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/

_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

chris marx
anyone?
I'm so close to getting this working, I would really like to use TileCache!

On Mon, Nov 10, 2008 at 3:33 PM, chris marx <[hidden email]> wrote:
Hi,
 Sorry about the bad link everyone, the site is now upyou can see that the tiles line up just fine (except when you zoom out from the initial zoom, i believe this is a manifold wms problem, i'm looking into that) but zooming in is ok. if you're looking at the wms requests, in firebug, these same requests generate the

An error occurred: Current y value 37.718590 is too far from tile corner y 37.968750


when i put tilecache in front of the wms.
I appreciate any advice!
chris-

On Sun, Nov 9, 2008 at 8:05 PM, Christopher Schmidt <[hidden email]> wrote:
On Sun, Nov 09, 2008 at 12:36:09PM -0500, chris marx wrote:
> Hi,
>  I just got tilecache working on xp and win 2003, and with my wms (using
> manifold gis) the open layers example works great. However, I configured
> tilecache in the hopes of using it in conjunction with google maps. However,
> it appears that tilecache is not honoring the wms requests that are being
> generated from the script in google maps (see below). The requests are
> valid, if I point the script directly at the wms, the tiles come back just
> fine. But when i put tilecache in between, I get
>
> An error occurred: Current y value 37.718590 is too far from tile
> corner y 37.968750
>
> You can see the working google map with tiles coming back directly from the
> wms here:

You must use Spherical Mercator to overlay tiles on Google Maps.

http://trac.openlayers.org/wiki/SphericalMercator


Regards,
--
Christopher Schmidt
MetaCarta



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/

_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

Christopher Schmidt-2
On Wed, Nov 12, 2008 at 11:38:49AM -0500, chris marx wrote:
> anyone?
> I'm so close to getting this working, I would really like to use TileCache!

Are yu using Spherical Mercator as I suggested? If not, that's your
first step.

-- Chris

> On Mon, Nov 10, 2008 at 3:33 PM, chris marx <[hidden email]> wrote:
>
> > Hi,
> >  Sorry about the bad link everyone, the site is now up
> >
> > http://warbler2.cit.cornell.edu/pfw-maps/wms-without-cache.html
> >
> > you can see that the tiles line up just fine (except when you zoom out from
> > the initial zoom, i believe this is a manifold wms problem, i'm looking into
> > that) but zooming in is ok. if you're looking at the wms requests, in
> > firebug, these same requests generate the
> >
> > An error occurred: Current y value 37.718590 is too far from tile corner y 37.968750
> >
> >
> > when i put tilecache in front of the wms.
> > I appreciate any advice!
> > chris-
> >
> > On Sun, Nov 9, 2008 at 8:05 PM, Christopher Schmidt <
> > [hidden email]> wrote:
> >
> >> On Sun, Nov 09, 2008 at 12:36:09PM -0500, chris marx wrote:
> >> > Hi,
> >> >  I just got tilecache working on xp and win 2003, and with my wms (using
> >> > manifold gis) the open layers example works great. However, I configured
> >> > tilecache in the hopes of using it in conjunction with google maps.
> >> However,
> >> > it appears that tilecache is not honoring the wms requests that are
> >> being
> >> > generated from the script in google maps (see below). The requests are
> >> > valid, if I point the script directly at the wms, the tiles come back
> >> just
> >> > fine. But when i put tilecache in between, I get
> >> >
> >> > An error occurred: Current y value 37.718590 is too far from tile
> >> > corner y 37.968750
> >> >
> >> > You can see the working google map with tiles coming back directly from
> >> the
> >> > wms here:
> >>
> >> You must use Spherical Mercator to overlay tiles on Google Maps.
> >>
> >> http://trac.openlayers.org/wiki/SphericalMercator
> >>
> >>
> >> Regards,
> >> --
> >> Christopher Schmidt
> >> MetaCarta
> >>
> >
> >
> >
> > --
> >
> > Chris Marx
> > Programmer/Analyst
> > Cornell Lab of Ornithology
> > 159 Sapsucker Woods Rd.
> > Ithaca, NY 14850
> > t. 1.607.254.1142
> > http://www.birds.cornell.edu/
> >
>
>
>
> --
>
> Chris Marx
> Programmer/Analyst
> Cornell Lab of Ornithology
> 159 Sapsucker Woods Rd.
> Ithaca, NY 14850
> t. 1.607.254.1142
> http://www.birds.cornell.edu/

--
Christopher Schmidt
MetaCarta
_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

Christopher Schmidt-2
On Wed, Nov 12, 2008 at 02:50:52PM -0500, chris marx wrote:
> Ah, ok. Yes, in fact I did, I even posted a comment there a few days ago,
> because there seemed to be something missing in the code, but perhaps you
> can elaborate. I looked at the example, and it's firing off tile requests
> that look like this
>
> http://hypercube.telascience.org/tiles/1.0.0/google-tiles/1/0/0.png?type=google
>
> this doesn't even look like a wms request to me, is there some part of the
> TileCache app that knows that to do with a request like this?

Yes.

Regards,
--
Christopher Schmidt
MetaCarta
_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

chris marx
Ok,
 I missed this sentence
"TileCache supports three tile request mechanisms: WMS requests, like this, WorldWind requests, like this one, and TMS requests, Like this one. "

The last one goes to this link
http://labs.metacarta.com/wms-c/tilecache.py/1.0.0/basic/5/32/23.png

But there is no documentation of what these parameters should be (although everything seems obvious except the first, is this the wms version?)

I tried doing this, but I'm running IIS, and the microsoft link thats posted
http://support.microsoft.com/kb/q184320/

no longer works, so I'm not sure if my path_info is configured to receive parameters this way. any pointers for getting this to work with IIS, and more about this format of TMS requests would be much appreciated-
chris-

On Wed, Nov 12, 2008 at 3:37 PM, chris marx <[hidden email]> wrote:
Ah,
 Sorry, must have missed this in the documentation, I just looked through it again, I saw some references to this configuration in the seeding tile cache, could you be a little more specific?
chris-


On Wed, Nov 12, 2008 at 3:14 PM, Christopher Schmidt <[hidden email]> wrote:
On Wed, Nov 12, 2008 at 02:50:52PM -0500, chris marx wrote:
> Ah, ok. Yes, in fact I did, I even posted a comment there a few days ago,
> because there seemed to be something missing in the code, but perhaps you
> can elaborate. I looked at the example, and it's firing off tile requests
> that look like this
>
> http://hypercube.telascience.org/tiles/1.0.0/google-tiles/1/0/0.png?type=google
>
> this doesn't even look like a wms request to me, is there some part of the
> TileCache app that knows that to do with a request like this?

Yes.

Regards,
--
Christopher Schmidt
MetaCarta



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/

_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

chris marx
Ok,
 I found this command

adsutil set w3svc/AllowPathInfoForScriptMappings True


Which is apparently the command i need to run to allow uri path parameters for IIS (from the bum link to microsoft). I've done that and try navigating to this link

http://warbler2.cit.cornell.edu/tilecache/1.0.0/google-tiles/4/7/6.png?type=google

but iis says the page is not found. if i try this

http://warbler2.cit.cornell.edu/tilecache/tilecache.py/1.0.0/google-tiles/4/7/6.png?type=google

it says

An error occurred: The requested layer (tilecache.py) does not exist. Available layers are: 
* google-tiles

Thanks for any advice!
chris-

On Wed, Nov 12, 2008 at 6:38 PM, chris marx <[hidden email]> wrote:
Ok,
 I missed this sentence
"TileCache supports three tile request mechanisms: WMS requests, like this, WorldWind requests, like this one, and TMS requests, Like this one. "

The last one goes to this link
http://labs.metacarta.com/wms-c/tilecache.py/1.0.0/basic/5/32/23.png

But there is no documentation of what these parameters should be (although everything seems obvious except the first, is this the wms version?)

I tried doing this, but I'm running IIS, and the microsoft link thats posted
http://support.microsoft.com/kb/q184320/

no longer works, so I'm not sure if my path_info is configured to receive parameters this way. any pointers for getting this to work with IIS, and more about this format of TMS requests would be much appreciated-
chris-


On Wed, Nov 12, 2008 at 3:37 PM, chris marx <[hidden email]> wrote:
Ah,
 Sorry, must have missed this in the documentation, I just looked through it again, I saw some references to this configuration in the seeding tile cache, could you be a little more specific?
chris-


On Wed, Nov 12, 2008 at 3:14 PM, Christopher Schmidt <[hidden email]> wrote:
On Wed, Nov 12, 2008 at 02:50:52PM -0500, chris marx wrote:
> Ah, ok. Yes, in fact I did, I even posted a comment there a few days ago,
> because there seemed to be something missing in the code, but perhaps you
> can elaborate. I looked at the example, and it's firing off tile requests
> that look like this
>
> http://hypercube.telascience.org/tiles/1.0.0/google-tiles/1/0/0.png?type=google
>
> this doesn't even look like a wms request to me, is there some part of the
> TileCache app that knows that to do with a request like this?

Yes.

Regards,
--
Christopher Schmidt
MetaCarta



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/

_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

chris marx
All,
 I've solved the problem, it appears that the order of the parameters was coming in differently. I enabled logging, and looked at the parts variable in TMS.py

the log showed parts = ['tilecache', 'tilecache.py', '1.0.0', 'google-tiles', '4', '6', '5.png']

which meant i had to rewrite the the rest of the parse method like so

else:
            layer = self.getLayer(parts[3])
            if len(parts) < 3:
                return self.layerCapabilities(host, layer)
            else:
                parts[-1] = parts[-1].split(".")[0]
                tile = None
                if layer.tms_type == "google" or (fields.has_key('type') and fields['type'] == 'google'):
                    res = layer.resolutions[int(parts[4])]
                    maxY = int(
                      round(
                        (layer.bbox[3] - layer.bbox[1]) /
                        (res * layer.size[1])
                       )
                    ) - 1
                    tile  = Layer.Tile(layer, int(parts[5]), maxY - int(parts[6]), int(parts[4]))
                else:
                    tile  = Layer.Tile(layer, int(parts[5]), int(parts[6]), int(parts[4]))
                return tile

now, the TMS style requests are working!!

I'm still trying to get the right codes to access spherical mercator in Manifold GIS WMS. If that fails, can someone point me to the right class where the calculations are made that translate the x,y,z variables in the mercator request? I might have to convert that to lat/lng to get a proper response from the wms server-

On Thu, Nov 13, 2008 at 5:42 PM, chris marx <[hidden email]> wrote:
Ok,
 I found this command

adsutil set w3svc/AllowPathInfoForScriptMappings True


Which is apparently the command i need to run to allow uri path parameters for IIS (from the bum link to microsoft). I've done that and try navigating to this link

http://warbler2.cit.cornell.edu/tilecache/1.0.0/google-tiles/4/7/6.png?type=google

but iis says the page is not found. if i try this

http://warbler2.cit.cornell.edu/tilecache/tilecache.py/1.0.0/google-tiles/4/7/6.png?type=google

it says

An error occurred: The requested layer (tilecache.py) does not exist. Available layers are: 
* google-tiles

Thanks for any advice!
chris-


On Wed, Nov 12, 2008 at 6:38 PM, chris marx <[hidden email]> wrote:
Ok,
 I missed this sentence
"TileCache supports three tile request mechanisms: WMS requests, like this, WorldWind requests, like this one, and TMS requests, Like this one. "

The last one goes to this link
http://labs.metacarta.com/wms-c/tilecache.py/1.0.0/basic/5/32/23.png

But there is no documentation of what these parameters should be (although everything seems obvious except the first, is this the wms version?)

I tried doing this, but I'm running IIS, and the microsoft link thats posted
http://support.microsoft.com/kb/q184320/

no longer works, so I'm not sure if my path_info is configured to receive parameters this way. any pointers for getting this to work with IIS, and more about this format of TMS requests would be much appreciated-
chris-


On Wed, Nov 12, 2008 at 3:37 PM, chris marx <[hidden email]> wrote:
Ah,
 Sorry, must have missed this in the documentation, I just looked through it again, I saw some references to this configuration in the seeding tile cache, could you be a little more specific?
chris-


On Wed, Nov 12, 2008 at 3:14 PM, Christopher Schmidt <[hidden email]> wrote:
On Wed, Nov 12, 2008 at 02:50:52PM -0500, chris marx wrote:
> Ah, ok. Yes, in fact I did, I even posted a comment there a few days ago,
> because there seemed to be something missing in the code, but perhaps you
> can elaborate. I looked at the example, and it's firing off tile requests
> that look like this
>
> http://hypercube.telascience.org/tiles/1.0.0/google-tiles/1/0/0.png?type=google
>
> this doesn't even look like a wms request to me, is there some part of the
> TileCache app that knows that to do with a request like this?

Yes.

Regards,
--
Christopher Schmidt
MetaCarta



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/



--

Chris Marx
Programmer/Analyst
Cornell Lab of Ornithology
159 Sapsucker Woods Rd.
Ithaca, NY 14850
t. 1.607.254.1142
http://www.birds.cornell.edu/

_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

Christopher Schmidt-2
On Fri, Nov 14, 2008 at 04:17:49AM -0500, chris marx wrote:
> I'm still trying to get the right codes to access spherical mercator in
> Manifold GIS WMS. If that fails, can someone point me to the right class
> where the calculations are made that translate the x,y,z variables in the
> mercator request? I might have to convert that to lat/lng to get a proper
> response from the wms server-

This will not help you. If you ask for unprojected data, this will *not*
overlay correctly on Google maps -- Google Maps data is projected.
Laying data down in two different projections will not work.

TileCache does no calculation -- it works entirely in planes. Google
Maps is projected in Mercator, so the tiles it cares about are mercator
tiles, so there is no conversion to lat/long that matters in this
process.

(Is there a reason you're not using OpenLayers?)

Regards,
--
Christopher Schmidt
MetaCarta
_______________________________________________
Tilecache mailing list
[hidden email]
http://openlayers.org/mailman/listinfo/tilecache
Reply | Threaded
Open this post in threaded view
|

Re: Why is the bounding box/resolution from the original request not honored?

longred
In reply to this post by chris marx
Awesome, the TMS.py modification works