Open seidlmic opened 9 years ago
a lillte history,
the change from round to cast a long time ago broke the server. I changed it back in 2011, I am hesatant to mess with that
this is also why i asked for clarification and a bug report on that pull request.
is this related to your other bug? https://github.com/OSGeo/tilecache/issues/14
the change from cast to round would cause that number to change perhaps seed should be using its own logic here?
it does no good to break the server to fix the client
link to that pull request on seed https://github.com/OSGeo/tilecache/pull/8
Acually I have no knowledge about TileCache. I just start to dig for reason why tilecache_seed.py creates strange protocols that does not reflect what is actually created.
Strange:
It look that all of this is also connected with my "crazy" settings.
metatile=yes metasize=1,1 metabuffer=100
that I use as workaround of label problem at edges of tiles.
The #14 an #13 issues are just side efects.
Despite this i guess the x,y,z should be integers not floats.
metabuffer should be 2 ints, other than that you set it however big you need it, nothing crazy about that
Thanks for hint. Maybe will help.
from tilecache.cfg
# metaBuffer=10 *** size of metatile buffer in px ***
This BUG is already mentioned in Pull Requests.
File "/var/www/dev/3data/minemap/map-layer/cgi-bin/tilecache-github/tilecache/TileCache/Client.py", line 117, in seed for y in range(startY, endY, stepY): TypeError: range() integer end argument expected, got float.
It is caused by commit from 2011 where a change in Layer.py was made
` I think better solution then change line #117 in Client.py is to fix Layer.py by