Closed johnnychen94 closed 5 years ago
Unfortunately, ambiguities still exist (moves from evaluate
to result_type
). I plan to fix it in the next PR.
Merging #32 into master will increase coverage by
18.5%
. The diff coverage is66.66%
.
@@ Coverage Diff @@
## master #32 +/- ##
==========================================
+ Coverage 69.1% 87.61% +18.5%
==========================================
Files 6 6
Lines 123 113 -10
==========================================
+ Hits 85 99 +14
+ Misses 38 14 -24
Impacted Files | Coverage Δ | |
---|---|---|
src/ImageDistances.jl | 100% <ø> (+25%) |
:arrow_up: |
src/metrics_distances.jl | 100% <100%> (+37.5%) |
:arrow_up: |
src/hausdorff.jl | 86.95% <33.33%> (+49.74%) |
:arrow_up: |
src/metrics.jl | 66.66% <50%> (-22.23%) |
:arrow_down: |
src/ciede2000.jl | 87.5% <83.33%> (-2.5%) |
:arrow_down: |
src/generic.jl | 94.87% <84.61%> (+4.87%) |
: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 7d01ad6...d9a2777. Read the comment docs.
The purpose of this rewrite is to close #22, close #28 and close #25 . Also, it's much easier to understand and maintain the source codes.
This rework consists of three updates:
result_type
on::Type
rather than::AbstractArray
( ~blocked by https://github.com/JuliaStats/Distances.jl/pull/140~)eval_op
instead ofevaluate
. The latter introduces a lot of method ambiguities (e.g., https://github.com/JuliaImages/ImageDistances.jl/issues/28#issuecomment-495010897 reported by @tlnagy)ImageCore >= 0.8.3
--> I'd like to make this a seperate PR but it's required to simplify the dispatching.Todo:
cc: @juliohm