Damiano,
Thanks!
I tested on my own just now to make sure nik2img was not adding any
substantial overhead.
I can confirm what you describe and commented on the ticket:
http://trac.mapnik.org/ticket/156#comment:1
Dane
On Dec 2, 2008, at 9:43 AM, Damiano Morosi wrote:
Ticket opened:
http://trac.mapnik.org/ticket/156
it's my first ticket, so please check it to see if it's good!
Thank you!
2008/12/2 Dane Springmeyer <[EMAIL PROTECTED]>
Damiano,
No, Mapnik does not read from the .prj file AFAIK.
Dane
On Dec 2, 2008, at 9:17 AM, Damiano Morosi wrote:
One more question: did Mapnik use the .prj file when reading shape
files? And how?
2008/12/2 Damiano Morosi <[EMAIL PROTECTED]>
Hi,
another hint: if I remove all the srs from the mapfile, the tiles
are generated CORRECTLY while Dane said that the result should be
something screwy... The only thing that doesn't work here are the
scale based rules. I think that Mapnik ignore what the measure unit
is, so it cannot calculate the scale correctly. Maybe introducing a
unit="..." param for the map should be a good thing?
2008/12/1 Robert Coup <[EMAIL PROTECTED]>
I would assume Proj4 does a no-op if the source and destination
SRSs were the same, but maybe it doesn't. If it's reprojecting, it
does some reasonably complex calcuations on every vertex of every
geometry, which would explain the added time.
In https://trac.mapnik.org/ticket/138 I added a no-op around not
calculating extents for same-SRS map/layers, but maybe
proj_transform should do a similar no-op rather than pass it
through to Proj4?
Rob :)
_______________________________________________
Mapnik-users mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/mapnik-users
_______________________________________________
Mapnik-users mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/mapnik-users
