Blake Thompson
b48bf5f48e
Fix for benchmark
2015-04-30 17:29:35 -05:00
Dane Springmeyer
3baf6f3450
clipping test: expect 30 verts [skip ci]
2015-04-30 15:07:02 -07:00
Dane Springmeyer
a9bda0a5df
remove agg_conv clipper - we are only interested in PolyTree
2015-04-30 14:27:53 -07:00
Blake Thompson
1bb8136390
Fix for bench
2015-04-21 10:29:22 -05:00
artemp
33661c818d
Merge branch 'mapnik-geometry' into mapnik-geometry-template
...
Conflicts:
benchmark/test_polygon_clipping.cpp
include/mapnik/geometry.hpp
include/mapnik/geometry_adapters.hpp
include/mapnik/marker_helpers.hpp
2015-04-10 14:52:40 +02:00
artemp
d5f9a14b58
more fixes (dependent names etc)
2015-04-10 14:17:36 +02:00
artemp
2ebc729cc8
Clipper - fix PolyTree traversal
2015-04-10 14:13:08 +02:00
Blake Thompson
87fb41907a
Added template to geometries
2015-04-09 15:22:51 -05:00
artemp
3f1aa7b4b4
geometry : allow boost::geoemetry algorithms to work directly with box2d<double>, removing bounding_box
2015-04-09 12:33:31 +02:00
artemp
d1b971f389
fix benchmark/test_polygon_clipping.cpp compilation
2015-04-07 12:14:17 +02:00
artemp
03a118a8ba
add ClipperLib::PolyTree based clipping test (4)
2015-04-02 18:24:25 +02:00
Dane Springmeyer
a2185bb9d8
add vertex_adapter.cpp tests
...
- focus is on behavior of polygons (including with holes)
- reverts 6a827f501
because it seems incorrect to me and without it the vertex_adapter.cpp tests pass
- @artemp if you see something I don't please restore 6a827f501
, but updated the vertex_adapter.cpp tests
to show what the expected behavior should be
2015-03-27 23:16:49 -07:00
artemp
f06443910f
update polygon clipping test
2015-03-26 17:31:31 +01:00
Dane Springmeyer
dfddd53c84
renable all polygon clippers
2015-03-24 18:47:44 -07:00
Dane Springmeyer
c964b090ed
first pass updating the clipping test
...
- actual results are still slightly broken for agg and clipper
- todo: try clipper directly rather than through agg interface
- todo: color polygon parts differently in test renderer
2015-03-24 18:43:29 -07:00
artemp
4389c80d84
and finaly, rename 'new_geometry' namespace to 'geometry'
2015-03-24 14:21:28 +01:00
artemp
ac1ffa2bfb
move geometry.hpp into path.hpp and update code base to reflect this change
2015-03-24 12:42:09 +01:00
artemp
71c9c04e92
Merge branch 'master' into mapnik-geometry
2015-03-16 16:20:46 +01:00
Blake Thompson
46ea6a5b1a
Added impl for image_view_any and fix to benchmark that was missing include for mapnik/color.hpp
2015-03-13 17:28:22 -05:00
artemp
aa5ae6b843
update markers_symbolizer to work with mapnik-geometry (work-in-progress)
2015-03-05 15:07:23 +01:00
Blake Thompson
5180168491
Merge branch 'master' into release/image_data_any
2015-02-18 23:03:51 -06:00
Blake Thompson
b2c1c86d99
Merge branch 'master' into release/image_data_Any
...
Conflicts:
benchmark/test_polygon_clipping.cpp
2015-02-16 14:37:01 -06:00
Dane Springmeyer
435f329cef
fix agg::conv_clip_polygon validation
2015-02-15 13:01:57 -08:00
Dane Springmeyer
e1941f87f0
clipping bench: validate per geom
2015-02-15 12:59:55 -08:00
Dane Springmeyer
81ef45b06c
fixup polygon clipping test build
2015-02-11 23:47:26 -08:00
Dane Springmeyer
f562f6e5d4
further validate results of clipping tests
2015-02-11 22:42:31 -08:00
Dane Springmeyer
216d1e9324
adapt clipping benchmark to vertex_adapter interface - refs #2688
2015-02-11 15:53:09 -08:00
Blake Thompson
bc28c12572
Found more situations where getRowSize should be used over multiplying the width by the pixel size.
2015-01-27 15:29:43 -06:00
Blake Thompson
22a384ef33
Moved image_data_rgba8 to image_rgba8. Ref #2633
2015-01-21 20:31:02 -06:00
Blake Thompson
93f835177b
Removed all the code for the previous type of image and image_32. Ref #2633
2015-01-21 17:57:16 -06:00
Blake Thompson
badb0c9a97
This is a complete removal of code that utilizes image_32 in the library. It is a sweeping change that does some of the following:
...
* Changes all agg_renderers to use a image_data_any variant (only image_data_rgba8 is implemented currently)
* Changes the marker and marker_cache to use image_data_any images
* Changes the symbolizers so that they must be aware of the source data type they are attempting to render and the render type that is expected to be rendered into.
* Moves many utilities into image_utils, that were previously in image_32.
The kicker is that this still isn't working perfectly yet, but I am commiting so I don't have tears in case everything is lost on my computer.
Ref #2633
2015-01-20 18:30:10 -06:00
Dane Springmeyer
fbecd5ff33
always pass image_data to save_to_string/file
2015-01-08 13:54:43 -08:00
Dane Springmeyer
cdd817d772
benchmarks: refactor to allow early exit if tests do not validate
2014-12-12 11:59:45 -05:00
artemp
1ac0ebdd83
rename coord_transform to transform_path_adapter and move implementation into separate header
2014-08-28 10:59:08 +01:00
artemp
ecfaec1027
rename ctrans.hpp to view_transform.hpp
2014-08-28 10:29:04 +01:00
artemp
de22d5900c
rename CoordTransform to view_transform to better reflect its purpose and be consistent
2014-08-28 10:17:15 +01:00
Dane Springmeyer
c5be70ff65
refactor spirit grammars
2014-07-24 14:31:59 -07:00
Dane Springmeyer
64e70fb374
iwyu
2014-07-22 22:40:39 -07:00
artemp
fe215a684e
c++11 style : replace (almost) all typedef with type alias's
2014-07-07 18:23:15 +01:00
Dane Springmeyer
e313b5ba62
benchmark: use halo-rasterizer='fast' (doubles perf)
2014-05-16 19:30:35 -07:00
Dane Springmeyer
7cfb1e417a
revised benchmarking system
2013-11-19 12:23:37 -08:00