Replace non-portable M_PI reference with fcl::constants invocation #264
Conversation
|
Reviewed 1 of 1 files at r1. Comments from Reviewable |
|
CI failure is one test in Mac Debug and two in Mac Release - expected? Review status: all files reviewed at latest revision, all discussions resolved, some commit checks failed. Comments from Reviewable |
|
The Debug failure is the same we've seen before (box-capsule test via ccd). However, the mac release has unexpected errors. Apparently #259 introduced functionality that fails in the mac release mode; these failures were masked by the box-capsule failures). In this PR, I recognized these failures. The failure is independent of this PR. I've added an issue for the CI failures: #265. Review status: all files reviewed at latest revision, all discussions resolved, some commit checks failed. Comments from Reviewable |
|
|
||
| #include "gtest/gtest.h" | ||
| #include "fcl/fcl.h" | ||
| #include "fcl/math/constants.h" |
There was a problem hiding this comment.
Doesn't fcl/fcl.h include fcl/math/constants.h?
SeanCurtis-TRI
force-pushed
the
PR_replace_M_PI
branch
from
7b44a30 to
ebd4b5a
Compare
|
Review status: 0 of 1 files reviewed at latest revision, 1 unresolved discussion. test/test_fcl_cylinder_half_space.cpp, line 41 at r1 (raw file): Previously, jslee02 (Jeongseok Lee) wrote…
Yes indeed. Now I feel foolish. I hadn't realized that this included everything. Comments from Reviewable |
|
Reviewed 1 of 1 files at r2. Comments from Reviewable |
There is a non-portable reference to M_PI in a test. This changes it to the fcl-standard constants reference to maintain portability.
(It also shuffles up the includes in the test to be more "standard".)
This change is