Barend Gehrels
|
cad4707a7d
|
[geometry] Fix ticket 7462, degenerate union for float. This was caused by two distances being wrong (by floating point precision). We now use earlier the alternative sort method (sorting on left/right), using relaxed_epsilon. This fixes the cause.
[SVN r83481]
|
2013-03-17 21:26:01 +00:00 |
|
Barend Gehrels
|
bc35eb29b6
|
[geometry] added case for ticket 8254, which had the same problem as buffer_mp1
[SVN r83478]
|
2013-03-17 17:20:47 +00:00 |
|
Barend Gehrels
|
0445cf26d7
|
[geometry] limited case #buffer_mp2 for traverse only for double, where the problem was
[SVN r83477]
|
2013-03-17 16:57:43 +00:00 |
|
Barend Gehrels
|
72f3b3f18e
|
[geometry] test cases for #buffer_mp1 and #buffer_mp2, see previous two commits
[SVN r83476]
|
2013-03-17 16:34:55 +00:00 |
|
Barend Gehrels
|
8fbc112550
|
[geometry] fixed issue (found myself with testcase #buffer_mp2) where, in double precision, a false intersection point was inserted. This was a robustness issue, an inconsistent side-value in colinear-opposite handler of get_turn_info
[SVN r83475]
|
2013-03-17 16:28:15 +00:00 |
|
Barend Gehrels
|
c992eb61e8
|
[geometry] fixed issue (found myself with testcase #buffer_mp1) where two specific circle-like polygons were not overlayed correctly. This was caused by sections not containing all points. We now make the section boxes a little smaller, such that they are not disjoint (10 * epsilon). This margin is a little fuzzy, but that is not harmful, they only influence might be some more comparisons. But that is intentional, we missed now one.
[SVN r83474]
|
2013-03-17 16:26:34 +00:00 |
|
Barend Gehrels
|
6ffb28d6ae
|
[geometry] reorganized buffer strategies (unit tests)
[SVN r83273]
|
2013-03-03 12:29:29 +00:00 |
|
Barend Gehrels
|
ce3e4e858e
|
[geometry] reorganized buffer strategies
[SVN r83272]
|
2013-03-03 12:28:16 +00:00 |
|
Barend Gehrels
|
e8422b1be3
|
[geometry] fixed/commented some buffer testcases
[SVN r83244]
|
2013-03-02 15:19:00 +00:00 |
|
Barend Gehrels
|
59ab90e79e
|
[geometry] Buffer update, added buffer support for multi_linestring, and the number of steps for round buffers can be specified (both requests of John Lilley)
[SVN r83243]
|
2013-03-02 14:30:50 +00:00 |
|
Barend Gehrels
|
9808bf8b45
|
[geometry] updated offset/buffer w.r.t. changes in library
[SVN r83189]
|
2013-02-27 17:25:07 +00:00 |
|
Barend Gehrels
|
7e949875aa
|
[geometry] applied patch of Vladimir Petrovic and changed std::string (which causes string-construction) to const char* for non-debug mode
[SVN r82690]
|
2013-02-02 18:47:00 +00:00 |
|
Bruno Lalande
|
6090b106af
|
Made the length algorithm variant-aware.
[SVN r82484]
|
2013-01-13 23:09:27 +00:00 |
|
Barend Gehrels
|
f3735f6475
|
[geometry] Made concept-checks variant aware. For now the underlying types are skipped. So a specialization for variants is made, being empty. This also solves two omissions: checks for Ring and Segment. Besides that, it now derives from not-inherited. This was earlier not the case (i.e. concepts were not checked, if type was not found). The commit makes it also consistent with the new system (template parameters auto-derived if possible)
[SVN r82482]
|
2013-01-13 17:34:50 +00:00 |
|
Bruno Lalande
|
2da1ca09d1
|
Made wkt write variant-aware.
[SVN r82481]
|
2013-01-13 17:25:15 +00:00 |
|
Barend Gehrels
|
d575690df3
|
[geometry] added include files for concept checking
[SVN r82476]
|
2013-01-13 15:05:24 +00:00 |
|
Bruno Lalande
|
e8b2693d74
|
Integrated wkt into support_status.
[SVN r82475]
|
2013-01-13 12:13:34 +00:00 |
|
Bruno Lalande
|
8ae4743abb
|
Adapted multi/io/wkt to latest change.
[SVN r82474]
|
2013-01-13 12:04:01 +00:00 |
|
Bruno Lalande
|
acba0901a7
|
Made dispatch::wkt able to retrieve the tag by itself + used not_implemented.
[SVN r82473]
|
2013-01-13 11:57:45 +00:00 |
|
Bruno Lalande
|
dfb6a611db
|
Made the convert algorithm variant-aware.
[SVN r82472]
|
2013-01-12 23:06:18 +00:00 |
|
Bruno Lalande
|
4be6f96160
|
Reimplemented variants handling in num_points algorithm.
[SVN r82470]
|
2013-01-12 21:28:59 +00:00 |
|
Bruno Lalande
|
1be15177cb
|
Reimplemented variants handling in clear algorithm.
[SVN r82469]
|
2013-01-12 21:06:44 +00:00 |
|
Bruno Lalande
|
0394222c2b
|
Small fix - missing header.
[SVN r82468]
|
2013-01-12 20:59:42 +00:00 |
|
Bruno Lalande
|
ab7220987e
|
Removed a useless template parameter + used devariant_area in the no-strategy overload.
[SVN r82467]
|
2013-01-12 20:56:00 +00:00 |
|
Bruno Lalande
|
36c9694321
|
Reimplemented variant handling in area algorithm.
[SVN r82466]
|
2013-01-12 20:46:13 +00:00 |
|
Bruno Lalande
|
4d59f27f94
|
Better aligned the recent new code.
[SVN r82465]
|
2013-01-12 20:34:40 +00:00 |
|
Bruno Lalande
|
136d84356f
|
Used 'devarianted_' term in append algorithm, like for equals.
[SVN r82464]
|
2013-01-12 18:33:20 +00:00 |
|
Bruno Lalande
|
c3c0e57304
|
Made the equals algorithm variant-aware.
[SVN r82463]
|
2013-01-12 18:28:00 +00:00 |
|
Bruno Lalande
|
ecf429362e
|
Adding back mistakenly removed polygon specialization
[SVN r82452]
|
2013-01-12 00:24:44 +00:00 |
|
Bruno Lalande
|
684945f1c5
|
Reimplemented variants handling in clear() to avoid function overloading.
[SVN r82450]
|
2013-01-11 21:54:41 +00:00 |
|
Bruno Lalande
|
13e2dc0984
|
Reimplemented variants handling in append to avoid function overloading.
[SVN r82449]
|
2013-01-11 21:16:55 +00:00 |
|
Bruno Lalande
|
2cc83e50a1
|
Made the append algorithm variant-aware.
[SVN r82412]
|
2013-01-08 23:35:28 +00:00 |
|
Bruno Lalande
|
79ce0f7bb5
|
Gave variant dispatcher a more algorithm-specific name.
[SVN r82409]
|
2013-01-08 21:58:04 +00:00 |
|
Bruno Lalande
|
22fff3f43b
|
Made clear() variant-aware.
[SVN r82395]
|
2013-01-08 00:05:36 +00:00 |
|
Bruno Lalande
|
f78cee7c30
|
Integrated not_implemented into clear.
[SVN r82390]
|
2013-01-07 22:39:17 +00:00 |
|
Bruno Lalande
|
2a5b55ffe9
|
Made num_points variant-aware.
[SVN r82378]
|
2013-01-06 22:27:38 +00:00 |
|
Bruno Lalande
|
da4868a421
|
Made clearer that variant's tag is irrelevant.
[SVN r82377]
|
2013-01-06 21:35:07 +00:00 |
|
Barend Gehrels
|
3fc118f30e
|
[geometry] updated release notes
[SVN r82021]
|
2012-12-16 13:40:23 +00:00 |
|
Barend Gehrels
|
202f8db2b2
|
[geometry] updated traverse to avoid output of nearly empty rings, like in convert_rings. This also affects unit tests
[SVN r82020]
|
2012-12-16 13:26:21 +00:00 |
|
Barend Gehrels
|
129f6cf7dd
|
[geometry] updated testsuite for difference, added test which checks convert_ring on inner-ring point-count
[SVN r82019]
|
2012-12-16 12:18:44 +00:00 |
|
Barend Gehrels
|
281bbfb52c
|
[geometry] pending commit, unit test for problems with integer coordinates reported by Volker (again) at 2012/07/17 and fixed in r79505
[SVN r81967]
|
2012-12-15 12:31:06 +00:00 |
|
Barend Gehrels
|
2f61bd3bda
|
[geometry] pending for commit, cosmetic change, done for MeetingC++ 9/10 nov 2012
[SVN r81959]
|
2012-12-14 22:36:42 +00:00 |
|
Barend Gehrels
|
b1d571076a
|
[geometry] added combinations point/ring, point/polygon, point/multi_polygon for disjoint and intersects (unit tests, multi)
[SVN r81957]
|
2012-12-14 22:31:19 +00:00 |
|
Barend Gehrels
|
86dd6e20e9
|
[geometry] added combinations point/ring, point/polygon, point/multi_polygon for disjoint and intersects (unit tests)
[SVN r81956]
|
2012-12-14 22:28:48 +00:00 |
|
Barend Gehrels
|
f7040ca437
|
[geometry] added combinations point/ring, point/polygon, point/multi_polygon for disjoint and intersects
[SVN r81955]
|
2012-12-14 22:12:12 +00:00 |
|
Barend Gehrels
|
30cd790db3
|
[geometry] fix in variable names
[SVN r81951]
|
2012-12-14 21:19:49 +00:00 |
|
Barend Gehrels
|
54a64a7e35
|
[geometry] fixed comparable_distance point-linestring (and -range, -polygon)
[SVN r81580]
|
2012-11-26 22:12:35 +00:00 |
|
Barend Gehrels
|
3ec9d48e43
|
[geometry] fixed for_each using C++11 lambda's, including adding unit test for them
[SVN r81418]
|
2012-11-18 19:51:00 +00:00 |
|
Bruno Lalande
|
df13f5a861
|
Integrated reverse into support_status.
[SVN r81078]
|
2012-10-28 20:54:42 +00:00 |
|
Bruno Lalande
|
cd44f9090e
|
Made dispatch::reverse able to find the tag by itself.
[SVN r81077]
|
2012-10-28 19:25:13 +00:00 |
|