Add InfExtendedTime for TimeType support

[fchorney]

Alright here's a first pass at attempting to add TimeType support.
Tried to take #4 and #5 into account.

Closes #3

[fchorney]

I don't know that I really need that isfinite(x.finitevalue) to be there.

[omus]

I suspect most of this code was copied from InfExtended. It would probably be best to do some code generation rather than copy/paste. For example:

for InfExt in (InfExtended, InfExtendedTime)
    @eval begin 
        Base.promote_rule(::Type{$InfExt{T}}, ::Type{$InfExt{S}}) where {T, S} = $InfExt(promote_type(T, S))
        Base.promote_rule(::Type{$InfExt{T}}, ::Type{S}) where {T, S} = $InfExt(promote_type(T, S))
        Base.promote_rule(::Type{$InfExt{T}}, ::Type{Infinite}) where T = $InfExt{T}
    end
end

[omus]

Currently wouldn't support hashing different infinities (e.g. InfExtended vs. InfExtendedTime) to the same result. I'm not sure what the desired behaviour is.

[fchorney]

Good point.

[cjdoris]

I'd like the hash to simply defer to whatever the true value is, i.e. the finitevalue or PosInf or NegInf. It's fine that the infinite date has the same hash as the infinite number.

[fchorney]

I suspect most of this code was copied from InfExtended. It would probably be best to do some code generation rather than copy/paste. For example:

for InfExt in (InfExtended, InfExtendedTime)
    @eval begin 
        Base.promote_rule(::Type{$InfExt{T}}, ::Type{$InfExt{S}}) where {T, S} = $InfExt(promote_type(T, S))
        Base.promote_rule(::Type{$InfExt{T}}, ::Type{S}) where {T, S} = $InfExt(promote_type(T, S))
        Base.promote_rule(::Type{$InfExt{T}}, ::Type{Infinite}) where T = $InfExt{T}
    end
end

I guess this kind of breaks down the separated structure that #5 was looking for. Though I guess a few things could be moved into a more generic section. I'm thinking maybe this could be part of #5?

[omus]

I'm thinking maybe this could be part of #5?

Very true, maybe then it would good to add comments as to parts that were copied from InfExtended but had the type changed to InfExtendedTime? Would make it easier to catch mistakes

[cjdoris]

Thanks for this, it looks really good. Please merge in the master branch --- I renamed the (badly thought through) InfFlag constants to be uppercase (e.g. POSINF), because they conflicted with things from Infinity.Utils, which will break your code somewhere.

[cjdoris]

I'm thinking maybe this could be part of #5?

Very true, maybe then it would good to add comments as to parts that were copied from InfExtended but had the type changed to InfExtendedTime? Would make it easier to catch mistakes

If there is code that is sensible to generate together for both InfExtended (which I'm considering renaming to InfExtendedReal) and InfExtendedTime then please do. Maybe there should be directories infextendedcommon, infextendedreal and infextendedtime.

[codecov-commenter]

Codecov Report

Merging #10 into master will increase coverage by 13.16%.
The diff coverage is 90.62%.

@@             Coverage Diff             @@
##           master      #10       +/-   ##
===========================================
+ Coverage   54.23%   67.40%   +13.16%     
===========================================
  Files          13       18        +5     
  Lines         118      181       +63     
===========================================
+ Hits           64      122       +58     
- Misses         54       59        +5     

Impacted Files	Coverage Δ
src/Infinity.jl	100.00% <ø> (ø)
src/utils.jl	100.00% <ø> (ø)
src/infextendedtime/conversion.jl	57.14% <57.14%> (ø)
src/infextendedtime/arithmetic.jl	91.30% <91.30%> (ø)
src/infextendedtime/comparison.jl	93.75% <93.75%> (ø)
src/infextendedtime/base.jl	100.00% <100.00%> (ø)
src/infextendedtime/io.jl	100.00% <100.00%> (ø)
src/infinite/base.jl	75.00% <100.00%> (+8.33%)	⬆️
src/infinite/conversion.jl	53.84% <0.00%> (-3.30%)	⬇️
... and 5 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 b50a1a5...29b02e0. Read the comment docs.

[cjdoris]

Thanks :)

[cjdoris]

It helps if I click the merge button!