Closed neil-lindquist closed 3 years ago
Thanks for the fix. You might add a test case for that.
Merging #46 into master will increase coverage by
0.36%
. The diff coverage isn/a
.
@@ Coverage Diff @@
## master #46 +/- ##
==========================================
+ Coverage 91.08% 91.45% +0.36%
==========================================
Files 12 12
Lines 1873 1942 +69
==========================================
+ Hits 1706 1776 +70
+ Misses 167 166 -1
Impacted Files | Coverage Δ | |
---|---|---|
src/higham.jl | 94.63% <ø> (+0.16%) |
:arrow_up: |
src/MatrixDepot.jl | 95.45% <0%> (-4.55%) |
:arrow_down: |
src/matrixmarket.jl | 91.4% <0%> (-1.53%) |
:arrow_down: |
src/download.jl | 93.46% <0%> (-0.93%) |
:arrow_down: |
src/common.jl | 96.81% <0%> (+0.02%) |
:arrow_up: |
src/datareader.jl | 88.31% <0%> (+0.15%) |
:arrow_up: |
src/regu.jl | 92.23% <0%> (+0.36%) |
:arrow_up: |
src/markdown.jl | 98.7% <0%> (+1.51%) |
:arrow_up: |
src/types.jl | 71.92% <0%> (+1.75%) |
:arrow_up: |
... and 1 more |
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 5b7010a...c6e1aee. Read the comment docs.
I added a test case for the case that errors. As a side note, I realized that my initial comment wasn't clear that the error only arises when using the reflected form of the matrix.
bump @andreasnoack
If the frank matrix is generated for a non-integer type, it errors. The error comes from trying to use
p
as an index whenp
is an array of non-integers.