[Tilecache] Tilecache seeding with BBOX args

Roger André randre at gmail.com
Fri Jul 24 19:34:38 EDT 2009


Any chance someone could clarify this for me?
--

On Mon, Sep 29, 2008 at 3:02 PM, Roger André<randre at gmail.com> wrote:
> Ok, so maybe there isn't a problem, but I'm still totally confused.  I reran
> the seeding command with the same extents, but widened the Zoom range to be
> from 1 - 11.  I got tiles generated, but only between Zoom levels 6 - 10.
> I'm not really sure I understand why.
> --
>
> On Mon, Sep 29, 2008 at 2:08 PM, Roger André <randre at gmail.com> wrote:
>>
>> Hi,
>>
>> I would like to test the speed that it takes to seed my cache for a single
>> tile of my raw data.  I am having trouble figuring out the correct
>> config/commandline syntax to run tilecache_seed.py so that it only runs on
>> the area occupied by my raw tile.  I *think* that since I have my Tilecache
>> layer set to use "spherical mercator", I need to specify both the layer's
>> bbox, and the commandline argument to tilecache_seed.py in spherical
>> mercator coordinates, and not in the WGS84 coords that my original data is
>> in.  I also *think* that the bbox format is "min_x, min_y, max_x, max_y",
>> based on the exaple present in the original tilecache.cfg file.  Based on
>> those assumptions, I have the following info:
>>
>> - wgs84 bbox = "-110, 30, -105, 35"
>>
>> - Based on that, I *think* the bbox in spherical mercator is:
>> sph_mercator bbox =
>> "-12245143.987260,3503549.843504,-11688546.533294,4163881.144064"
>>
>> - (That's based on using cs2cs to convert the wgs84 coords):
>> $echo "-110 30" | cs2cs + +init=epsg:4326 +to +init=epsg:900913 -f "%.6f"
>> -12245143.987260        3503549.843504 0.000000
>> $echo "-105 35" | cs2cs + +init=epsg:4326 +to +init=epsg:900913 -f "%.6f"
>> -11688546.533294        4163881.144064 0.000000
>>
>> - tilecache config:
>> [blah]
>> type=WMS
>>
>> url=http://10.10.10.143/cgi-bin/mapserv?map=/var/www/mapfiles/blah/blah.map
>> layers=imagery2
>> spherical_mercator=true
>> metabuffer=15
>> bbox="-12245143.987260,3503549.843504,-11688546.533294,4163881.144064"
>>
>> - This entry works fine when I try to view the tile in Google Maps.
>> Displays correctly with no issues.  It also works fine when I use
>> tilecache_seed.py without specifying a bbox:
>> ./tilecache_seed.py
>> "http://localhost/cgi-bin/tilecache-2.04/tilecache.cgi?" blah 1 2
>> ###### (0, 0, 1), (2, 2, 1)
>> Cache miss:
>> -20037508.34,-20037508.34,-0.000799998641014,-0.000799998641014, Tile: x: 0,
>> y: 0, z: 1, time: 0.235054969788
>> 01 (000000, 000000) = (-20037508.3400 -20037508.3400 -0.0008 -0.0008)
>> [0.2352s : 4.246/s] 1/4
>> Cache miss:
>> -0.000799998641014,-20037508.34,20037508.3384,-0.000799998641014, Tile: x:
>> 1, y: 0, z: 1, time: 0.225296020508
>> 01 (000001, 000000) = (-0.0008 -20037508.3400 20037508.3384 -0.0008)
>> [0.2254s : 4.338/s] 2/4
>> Cache miss:
>> -20037508.34,-0.000799998641014,-0.000799998641014,20037508.3384, Tile: x:
>> 0, y: 1, z: 1, time: 0.242295980453
>> 01 (000000, 000001) = (-20037508.3400 -0.0008 -0.0008 20037508.3384)
>> [0.2424s : 4.264/s] 3/4
>> Cache miss:
>> -0.000799998641014,-0.000799998641014,20037508.3384,20037508.3384, Tile: x:
>> 1, y: 1, z: 1, time: 0.221132993698
>> 01 (000001, 000001) = (-0.0008 -0.0008 20037508.3384 20037508.3384)
>> [0.2213s : 4.325/s] 4/4
>>
>> - However, when I try to specify the same bbox as the layer's in the
>> commandline, I get no tiles:
>> ./tilecache_seed.py
>> "http://localhost/cgi-bin/tilecache-2.04/tilecache.cgi?" blah 4 5
>> "-12245143.987260,3503549.843504,-11688546.533294,4163881.144064"
>>
>> I'm betting that I'm specifying incorrect coordinates somehow, but I'm not
>> sure where.
>>
>> - To recap, I'm starting with a raw tile that has these extents in
>> EPSG:4326:
>> Upper Left  (-110.0000000,  35.0000000)
>> Lower Left  (-110.0000000,  30.0000000)
>> Upper Right (-105.0000000,  35.0000000)
>> Lower Right (-105.0000000,  30.0000000)
>>
>> - I want to run Tilecache against this and generate Google Map tiles, but
>> only for the area occupied by that tile.
>>
>> Thanks,
>>
>> Roger
>>
>>
>>
>>
>>
>
>



More information about the Tilecache mailing list