reverse(zip(its...)) now checks lengths of constituent iterators for equality

[adienes]

fixes #45085, fixes #25583, as per suggested by @Moelf

Despite the fact that this might cause code to fail that previously worked, I think this is a bugfix not a breaking change.

In particular, Iterators.reverse says:

Given an iterator itr, then reverse(itr) is an iterator over the same collection but in the reverse order

And in many cases, the previous behavior of reverse(::Zip) would not return "the same collection"

[mschauer]

I think _zip_lengths_finite_equal(is) deserves a doc string, in particular it is not clear what pair of Bool and Union{Nothing, Int} (?) the function is supposed to return .

[oscardssmith]

@nanosoldier runtests()

[nanosoldier]

The package evaluation job you requested has completed - possible new issues were detected.
The full report is available.

[adienes]

looks pretty safe actually? among the packages that failed only on this branch,

GMT.jl
VCFTools.jl
Braket.jl
Test.jl

I did not see any use of reverse(zip in the source code and the failure logs did not contain the new error message in this PR, so I think it is unrelated.

I also did a code search on JuliaHub and I find use here

• InteractiveCodeSearch.jl
• KroneckerProducts.jl
• Yao.jl

The latter two it appears, as far as I can tell, to be using the pattern safely and already ensuring that the lengths being zipped match. Only InteractiveCodeSearch.jl it looks like might sometimes (presumably unwittingly) be facing this bug, but I'm not at all familiar with the package so not sure what to do about it---it seems its tests passed at least

[adienes]

bump

[adienes]

bump
does this need triage? or is it unambiguously a bugfix

[adienes]

bump

[oscardssmith]

Throwing it on the triage label just in case (and to make sure it doesn't get forgotten) IMO it's a bugfix.

[adienes]

can we merge or close?

[oscardssmith]

thanks for the reminder!