Set default _numOutputs to 1 in splitEngine

[brianhou]

Resolves #187.

@mkoval is this what you meant? My initial thought was to have the two methods:

• splitEngine(RNG& _engine, size_t _numOutputs, size_t _numSeeds = NUM_DEFAULT_SEEDS) (already exists)
• splitEngine(RNG& _engine, size_t _numSeeds = NUM_DEFAULT_SEEDS) (call the former with _numOutputs = 1)

but this makes splitEngine(RNG&, size_t) ambiguous.

[codecov]

Codecov Report

Merging #214 into master will decrease coverage by 7.14%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #214      +/-   ##
==========================================
- Coverage   72.03%   64.88%   -7.15%     
==========================================
  Files         157      156       -1     
  Lines        4487     4443      -44     
  Branches      707      703       -4     
==========================================
- Hits         3232     2883     -349     
- Misses        793     1164     +371     
+ Partials      462      396      -66

Impacted Files	Coverage Δ
include/aikido/util/RNG.hpp	100% <ø> (ø)	⬆️
src/util/RNG.cpp	92.3% <100%> (ø)	⬆️
src/constraint/JointStateSpaceHelpers.cpp	70.83% <100%> (ø)	⬆️
include/aikido/planner/ompl/CRRT.hpp	0% <0%> (-100%)	⬇️
...lude/aikido/planner/ompl/BackwardCompatibility.hpp	25% <0%> (-75%)	⬇️
src/planner/ompl/CRRTConnect.cpp	0% <0%> (-64.03%)	⬇️
src/planner/ompl/CRRT.cpp	0% <0%> (-63.96%)	⬇️
src/planner/ompl/Planner.cpp	7.3% <0%> (-50.57%)	⬇️
src/planner/ompl/MotionValidator.cpp	5.4% <0%> (-24.33%)	⬇️
src/planner/ompl/StateValidityChecker.cpp	15.38% <0%> (-23.08%)	⬇️
... and 4 more

[jslee02]

According to the description of #187, it seems @brianhou 's initial thought is what @mkoval means. To avoid the function name's ambiguity, we could use different name. If the intention is to create one object with the same type of another existing object, I would like to call the function clone (or cloneFrom).

Also, we could use cloneRNGs for multiple cloning version instead of splitEngine.

I'm open to using different name from what I proposed, but my point is to use different function name to avoid the ambiguity.

Or, we could remove all the default parameters of the original function.

[brianhou]

Different function names sound reasonable, although I think it'd make more sense to keep the names similar. How about cloneRngFrom (for the new one-output version) and cloneRngsFrom (for the current splitEngine)?

I'd prefer to keep the default parameters; I don't want to arbitrarily choose a value for _numSeeds each time I use it 😅

[jslee02]

cloneRngFrom and cloneRngsFrom overall sound good to me, but have a minor concern that they use different capitalization (RNG vs Rng). I personally prefer Rng for any other acronyms, but it seems the current convention is RNG. So we may want to stay with it like cloneRNGFrom and cloneRNGsFrom unless we change the convention and apply it to all other names.

Also, out of curiosity, do we have a specific reason why not we have this functions as member functions?