Add overload for foldl and aggregate that don't require a seed value. about shittylinq HOT 5 OPEN

@jreina would it be acceptable to add other extension methods for IEnumerable<T> to the repo?

Thanks!

from shittylinq.

@Programm3r
Absolutely! Just opened #4 specifically for this.

from shittylinq.

@jreina many thanks!

from shittylinq.

Hope that this helps :)

One sidenote: I opted to go for the parameterless constructor constraint, since just using default will result in null for objects/classes.
Using the constructor constraint solves this and it also works for structs and types.

I am really bad at unit testing so I was hoping someone else has idea's for good cases.

from shittylinq.

@tosti007 thanks for your PR!

Looking at these specifications for the Aggregate function, it looks like the signature for this overload should be different and the behavior is slightly different as well.

Instead of constructing a new instance of T and creating a constraint on what types can be used here, the function should instead use the first element in the sequence as the seed value and throw an InvalidOperationException if the sequence is empty. That allows us to get rid of the U type parameter altogether.

Regarding unit tests, I think the tests should ensure this new function throws an InvalidOperationException when the sequence is empty, and make sure the first item in the sequence is used as the seed.

from shittylinq.