ivy: unexpected indentation in matrix of vectors

[rsc]

mix (1 2 3) (4 5 6)
1 2 3
4 5 6

mix ('abc' ('d' 'efg')) ('hij' ('k' 'lmn'))
    (abc) (d (efg))
    (hij) (k (lmn))

mix ('abc' ('d' ('efg' 'EFG'))) ('hij' ('k' ('lmn' 'LMN')))
            (abc) (d ((efg) (EFG)))
            (hij) (k ((lmn) (LMN)))

I'm not sure what the leading indentation is, but note how it increases with the vector nesting depth.

[rsc]

Oh, I see. It is making all columns the same width and right-aligning the data.
I wonder whether right-alignment should be reserved for numbers or even integers.

I'm also unsure about making all columns the same width when the widths are very different.
I'm playing with matrices as dictionaries, with each row a key-value pair, and they typically have short keys and long values, but the key column ends up with many many spaces at the front.

[robpike]

There is a lot of judgement involved here. I tried to make the output generally readable, but with large elements of course that's very hard. It's had several iterations.

For simple things the alignment is how I remember APL\360 working, but of course that was half a century ago. (Really.) My memory may be faulty, or things may have changed, or other changing things may argue for other tradeoffs. The trick is doing the right thing most of the time, perhaps at the cost of some ugliness in oddball cases.

[rsc]

Another example to consider in any future tweak of the heuristics. Playing with the game of life from #194, displaying a matrix of boxed matrices could possibly use blank lines between rows of the outer matrix.

R = -1 flip -2 rot 5 7 take (3 3 rho iota 9) in 2 3 4 5 8

1 0 -1 @rot R
(0 0 0 0 0 0 0| (0 0 0 0 0 0 0| (0 0 0 0 0 0 0|
|0 0 1 1 0 0 0| |0 0 0 1 1 0 0| |0 0 0 0 1 1 0|
|0 1 1 0 0 0 0| |0 0 1 1 0 0 0| |0 0 0 1 1 0 0|
|0 0 1 0 0 0 0| |0 0 0 1 0 0 0| |0 0 0 0 1 0 0|
|0 0 0 0 0 0 0) |0 0 0 0 0 0 0) |0 0 0 0 0 0 0)

1 0 -1 o.flip 1 0 -1 @rot R
(0 0 1 1 0 0 0| (0 0 0 1 1 0 0| (0 0 0 0 1 1 0|
|0 1 1 0 0 0 0| |0 0 1 1 0 0 0| |0 0 0 1 1 0 0|
|0 0 1 0 0 0 0| |0 0 0 1 0 0 0| |0 0 0 0 1 0 0|
|0 0 0 0 0 0 0| |0 0 0 0 0 0 0| |0 0 0 0 0 0 0|
|0 0 0 0 0 0 0) |0 0 0 0 0 0 0) |0 0 0 0 0 0 0)
(0 0 0 0 0 0 0| (0 0 0 0 0 0 0| (0 0 0 0 0 0 0|
|0 0 1 1 0 0 0| |0 0 0 1 1 0 0| |0 0 0 0 1 1 0|
|0 1 1 0 0 0 0| |0 0 1 1 0 0 0| |0 0 0 1 1 0 0|
|0 0 1 0 0 0 0| |0 0 0 1 0 0 0| |0 0 0 0 1 0 0|
|0 0 0 0 0 0 0) |0 0 0 0 0 0 0) |0 0 0 0 0 0 0)
(0 0 0 0 0 0 0| (0 0 0 0 0 0 0| (0 0 0 0 0 0 0|
|0 0 0 0 0 0 0| |0 0 0 0 0 0 0| |0 0 0 0 0 0 0|
|0 0 1 1 0 0 0| |0 0 0 1 1 0 0| |0 0 0 0 1 1 0|
|0 1 1 0 0 0 0| |0 0 1 1 0 0 0| |0 0 0 1 1 0 0|
|0 0 1 0 0 0 0) |0 0 0 1 0 0 0) |0 0 0 0 1 0 0)

[robpike]

Reopening because the fix only addresses the last comment.