Any alternative to `List.zip(List).toMap(mutableMa...

# stdlib

[Sam Stone]

Any alternative to List.zip(List).toMap(mutableMapOf())?

[nkiesel]

I guess I would use List.zip(List).toMap().toMutableMap() but that is mostly personal preference

[nkiesel]

btw: this would better be asked in #getting-started, this channel is for discussions about stdlib extensions/bugs/missing features

[gildor]

toMap(mutableMapOf()) is more efficient

[gildor]

I think it’s a good candidate for stdlib function, or with own function if performance is a concern, it’s pretty simple to write one and use case maybe not so common, but still I can imagine real cases for it

[gildor]

Something like this:
https://pl.kotl.in/XNP4yGL9D

[Sam Stone]

@gildor why inline?

[Sam Stone]

Maybe zipTo?

[gildor]

I follow the same convention as all existing associateWith/associateBy functions from stdlib which create map from keys/values

[gildor]

inline as all similar stdlib functions (see for example List.zip)

[gildor]

probably zipTo also reasonable, but it would be confusing to have zipTo which returns map, and no zipTo with list as destination

[gildor]

but for own project of course you can use any naming, no need to worry too much about consistency with stdlib

[nkiesel]

I still fail to see the common use cases which would support adding this method. In all the cases I came across so far, I only needed an immutable result and thus was happy enough with zip+toMap.

[gildor]

well, the reason why we have toMap(destination) includes support of mutability too
I agree, it’s not so common, but there are cases when you need mutability anyway

[gildor]

happy enough with zip+toMap

To be honest it’s not super efficient way, if you use it often, I would just write own function as one which I posted, just without destination

[nkiesel]

Understood. But the barrier of entry to stdlib is quite high per my understanding (for good reasons). So why is List.zip(List).toMap(mutableMapOf()) not good enough? And yes: if that method would become a hotspot, I would try to optimize it along your code.

[gildor]

I’m not advocating for adding it to stdlib, just said that it may be reasonable to have something like this

[gildor]

if that method would become a hotspot

Not only because it’s hotspot, just a bit less GC pressure and clearly shows intent of the code