Closed peterbygrave closed 5 years ago
Merging #37 into master will increase coverage by
0.02%
. The diff coverage is100%
.
@@ Coverage Diff @@
## master #37 +/- ##
==========================================
+ Coverage 98.26% 98.28% +0.02%
==========================================
Files 20 20
Lines 2012 2036 +24
==========================================
+ Hits 1977 2001 +24
Misses 35 35
Impacted Files | Coverage Δ | |
---|---|---|
include/libirc/connectivity.h | 98.35% <100%> (ø) |
:arrow_up: |
include/libirc/wilson.h | 95.2% <100%> (ø) |
:arrow_up: |
src/test/connectivity_test.cpp | 100% <100%> (ø) |
:arrow_up: |
src/test/wilson_test.cpp | 100% <100%> (ø) |
:arrow_up: |
include/libirc/irc.h | 80.55% <0%> (ø) |
:arrow_up: |
include/libirc/atom.h | 100% <0%> (ø) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 22726d8...cfacd6e. Read the comment docs.
I updated the documentation for the linear_angle_gradient
function. @peterbygrave do you have some references I can add?
Merged master
after the dihedral bugfix #35 and corrected the failing tests accordingly (already identified by @peterbygrave).
I'll merge now since it's a critical bug. Documentation will be added later.
This PR corrects the linear angle and its gradient
The angle was being calculated defining the orthogonal axis as the additional position rather than using it as a displacement from the central atom. Previous tests worked because its not a "bad" position when molecule is close to the origin.
The second fix is that the angle gradient was wrong. Angle bends has
st
vector for the terminal atoms, and the central atom is the negative of the sum of these twost
vector.