Artem Pavlenko
242c9eba3c
Revert "opps, adding lost change (re: boost>=1.67)"
...
This reverts commit 22f913d892
.
2018-08-07 14:26:34 +01:00
Artem Pavlenko
22f913d892
opps, adding lost change (re: boost>=1.67)
2018-08-07 14:09:30 +01:00
Jiri Drbalek
8b27aba0c4
Fix seams of mesh faces when warping transparent raster
2018-08-07 11:57:38 +00:00
talaj
1a87e920da
Merge pull request #3962 from lightmare/bootstrap-manners
...
bootstrap.sh improvements
2018-08-07 09:28:33 +02:00
Jiri Drbalek
cfbc044fd2
Force full qualification of ICU types
...
https://github.com/mapnik/mapnik/issues/3961
2018-08-03 16:02:20 +00:00
Mickey Rose
f400f17c75
travis: some tweaks
...
- remove unused MASON_PUBLISH variables
- defer downloading test-data until after successful compilation;
they are big and don't update frequently, so I added depth limit;
if that doesn't work well, feel free to increase or remove it
2018-08-01 17:59:09 +02:00
Mickey Rose
7f0daee8b3
configure: posix shell can handle this
...
It's such a trivial script, no need to bash.
2018-08-01 17:37:15 +02:00
Mickey Rose
33a42303d1
bootstrap: behave when sourced, minor improvements
...
- don't exit sourcing shell on error
- don't git pull mason, checkout --detach instead
- only git fetch mason when the desired version is unknown
- use arguments of the form NAME=VALUE as variables to export;
this is to unify syntax with ./configure, for example:
./bootstrap.sh CC='cc' CXX='ccache c++'
2018-08-01 15:24:34 +02:00
Jiri Drbalek
8a0dca38a3
Fix buffer-size with scale-factor
2018-08-01 08:06:37 +00:00
Artem Pavlenko
3683c30557
Merge branch 'lightmare-pi-is-wrong'
2018-07-30 10:41:13 +01:00
Artem Pavlenko
af842a8bf9
Merge branch 'pi-is-wrong' of https://github.com/lightmare/mapnik into lightmare-pi-is-wrong
2018-07-30 10:40:51 +01:00
lightmare
686355cbbe
Merge pull request #3506 from mapycz/fix-building-symbolizer-transform
...
Building symbolizer improvements
2018-07-27 12:55:53 +02:00
Jiri Drbalek
acbf80677a
update visual tests data
2018-07-27 10:23:34 +00:00
Jiri Drbalek
4a93209423
render building symbolizer rewrite
...
- make projection and view transforms first
- removed dependency on segment_t
- no need for sorting faces
2018-07-27 10:23:33 +00:00
Mickey Rose
6e71bad70c
Revert "update test data"
...
This reverts commit 81183d6fe6
.
It removed recent tests, didn't update.
2018-07-27 10:58:31 +02:00
Artem Pavlenko
b796cc9912
[skip ci]
2018-07-25 17:04:37 +01:00
Artem Pavlenko
ace2baafbd
Update README.md
2018-07-25 16:01:41 +01:00
Artem Pavlenko
50caf755c3
Update README.md
2018-07-25 15:55:18 +01:00
Artem Pavlenko
7cf93eb70f
Update README.md
2018-07-25 15:45:40 +01:00
Artem Pavlenko
f930bea045
Update README.md
2018-07-25 15:45:09 +01:00
Artem Pavlenko
174c82662d
Merge pull request #3952 from mapnik/circleci-20-test
...
Circleci 20 test
2018-07-25 15:26:28 +01:00
Artem Pavlenko
fed8effb67
remove circleci v1.0 config
2018-07-25 14:17:46 +01:00
Artem Pavlenko
64462b8dca
cleanup
2018-07-25 14:16:04 +01:00
Artem Pavlenko
39b6dca059
Adding auto-generated CircleCI 2.0 config file
2018-07-25 13:53:56 +01:00
Artem Pavlenko
70fbdd21c7
Merge pull request #3948 from mapnik/boost_1_67
...
Boost 1 67
2018-07-23 14:05:52 +01:00
Artem Pavlenko
5eca7731e9
rely on implicit std::reference_wrapper<T> -> T conversions (thanks @lightmare)
2018-07-23 13:26:22 +01:00
Artem Pavlenko
93d0eea8af
Merge branch 'master' into boost_1_67
2018-07-23 11:35:14 +01:00
Artem Pavlenko
65080e2a95
add support for BOOST_VERSION < 106700
2018-07-23 10:58:27 +01:00
talaj
c819a03bb3
Merge pull request #3944 from EmilGedda/gedda-patch-1
...
remove old render_box function declaration from cairo_renderer.hpp
2018-07-20 14:43:48 +02:00
Emil Gedda
93c566638f
remove old render_box function declaration from cairo_renderer.hpp
2018-07-20 14:02:51 +02:00
Artem Pavlenko
5fe0dda890
Merge branch 'master' into boost_1_67
2018-07-20 11:53:47 +01:00
Mickey Rose
1ddd3afc51
postgis+pgraster: substitute numbers always with decimal point
2018-07-19 22:52:55 +02:00
Mickey Rose
774a618d13
postgis: add test for pg_typeof substituted tokens
2018-07-19 22:51:01 +02:00
Artem Pavlenko
bdf7876652
const-ness
2018-07-19 15:36:00 +01:00
Artem Pavlenko
67dbb96580
use uint8_t to parse hex2 values to avoid overflowing (boost >= 1.67)
2018-07-19 15:34:40 +01:00
Mickey Rose
f1d07d6718
proj_transform test: also check that transform succeeded
2018-07-18 19:31:12 +02:00
Artem Pavlenko
a98d1b8071
use std::uint8_t as an attribute for hex2 parser
2018-07-18 10:40:41 +01:00
Artem Pavlenko
4c31bd16d2
Merge branch 'master' into boost_1_67
2018-07-18 09:27:56 +01:00
Sandro Santilli
69c317cd1f
Fix some comments and error messages referencing raster parameters
2018-07-16 19:27:12 +02:00
Mickey Rose
68a7f5f63c
proj_transform: fix strided coordinate array transform
2018-07-16 14:18:49 +02:00
Mickey Rose
1e6feecb91
proj_transform: add test for strided coordinate array
2018-07-16 14:18:49 +02:00
Mickey Rose
f84191204e
eradicate M_PI
2018-07-16 14:18:49 +02:00
Mickey Rose
32e7202223
well_known_srs: move implementation to .cpp
...
- add lonlat2merc, merc2lonlat overloads for single point
- add stride argument to overloads for coordinate arrays
- make all functions exported
- change static string constants to exported
- change static numeric constants to constexpr
- remove superfluous numeric constants
2018-07-16 14:18:32 +02:00
Mickey Rose
e11f3a660a
offset_converter: minor refactoring
...
DRY: extract method for `joint_angle` calculation.
2018-07-14 12:46:53 +02:00
Mickey Rose
a0a2c78950
util/math: clamp, degrees, radians
...
Use common functions for degrees <-> radians conversions, except when
dealing with agg, where it makes sense to use available deg2rad/rad2deg
functions instead of including another header.
2018-07-13 12:34:25 +02:00
lightmare
187c1df332
Merge pull request #3337 from lightmare/normalize-angle
...
speed up util::normalize_angle for large values
2018-07-12 19:30:37 +02:00
Mickey Rose
0a6e82e555
move src/math.cpp under src/util/
2018-07-11 23:55:21 +02:00
Mickey Rose
7c4ccf0209
make util::normalize_angle reasonably fast on large values
...
Because it simply calculates the remainder after division by full turn,
it shouldn't take time proportional to the magnitude of its operand.
2018-07-11 23:14:23 +02:00
Mickey Rose
354be41811
benchmarks: clarify output
...
Make it clear in benchmark output which mode it ran in:
- "main thread" only
- "1 worker" thread
- "N workers"
2018-07-11 23:14:04 +02:00
Mickey Rose
65f3bb75dd
Revert "benchmark - slightly better stderr"
...
This partially reverts commit 5ab6db2607
.
The change of argument --threads N doesn't just change output, it
changes the mode of operation.
With --threads 0, the benchmark runs wholly in the main thread.
With --threads 1, it starts 1 worker thread, which does the work, and
the main thread measures how long it takes to complete.
2018-07-11 23:14:03 +02:00