Shallow, simple or mild flatten?

[cymplecy]

Background waffle

When manipluating pixels of costumes - I regularly convert the pixel list into a 2D matrix
e.g a simple 2x2 red green blue white costume

When I'm finished processing, I need to convert back to a standard pixel list and I use flatten to do that

But flatten does a deep flatten (ravel in APL speak) so it ends up like this

when we need this

So I reshape it using

All works but it's a two stage process and you have to work out the length of the pixel list required to use the reshape

Anyway, I decided to make my own "shallow" flatten block but I always like to shorten the block labels as much as I can ...

Which one of these would you prefer to see as a "shallow" flatten reporter compared to the standard flatten reporter that does a deep flatten

Or have you got your own alternative suggestion

[kinestheticlearning]

I vote for the 'mild' adjective because the rest of them are all 2-syllables long.

[bh]

I vote for "shallow" because that's the official term everyone understands. (Although in some systems "flatten" by itself means a shallow flatten and you say "deep flatten" if that's what you want.) We don't have a shallow-flatten primitive because you can say

to get that result.

Are you proposing a new primitive, or just asking for advice about your own projects?

[cymplecy]

I was asking about what people thought about using a name other than shallow.

Lots of other reporters are shallow ones but none of them use the name.

Shallow doesn't grab me so I was looking to come up with something more appealing.
Simple is a compromise - mild is just nice and short but less computing like

FYI I'm calling this in my custom reporter

but if using append input list works at a similar speed then I may as well use that

[bh]

@cymplecy

but if using append input list works at a similar speed then I may as well use that

APPEND is a little different because it requires that its inputs be lists.

[cymplecy]

That's not a real issue - I'm only trying to flatten lists.

It's more whether if someone saw append input list with just the one variable attached to it - would they guess that it does a shallow flatten?

I don't think its obvious that it would so I might stick to declaring a custom reporter (but that can call the append input list instead of the JS)