Thanks for the thoughts!
> not as pretty, but simpler and pretty enough?
>
> (":"0@i.@# (,~ #&'-')"1 0 ]) 0 1 2 2 1 1 2 2 3
>
> 0
>
> -1
>
> --2
>
> --3
>
> -4
>
> -5
>
> --6
>
> --7
>
> ---8
This is sort of in between the raw depth vector and the graph representation.
It's probably fine for smallish trees, but especially with larger trees, I find
it hard to visually pick out subtrees. For that, the explicit links between
sibling nodes is really helpful.
Otherwise, an even more minimal "mountain graph" is probably serviceable:
(=/ ~.) d
1 0 0 0
0 1 0 0
0 0 1 0
0 0 1 0
0 1 0 0
0 1 0 0
0 0 1 0
0 0 1 0
0 0 0 1
> J has several functions built in to process trees that are represented as
> nested box structures, and so may be useful in converting from various
> representations to and from such boxed structures. Or more interesting to
> me, but I need more thought on this.
It really depends. Nested boxes require quite a lot of overhead, no? Pilfering
the structure of an example tree from Voc, Appendix A (Foreign Conjunction),
Representation (5!:) [0]:
[ s=: (((0;1;2);3;<4;5));6;7 NB. Nested box representation
┌─────────────────┬─┬─┐
│┌───────┬─┬─────┐│6│7│
││┌─┬─┬─┐│3│┌─┬─┐││ │ │
│││0│1│2││ ││4│5│││ │ │
││└─┴─┴─┘│ │└─┴─┘││ │ │
│└───────┴─┴─────┘│ │ │
└─────────────────┴─┴─┘
7!:5 <'s'
512
[ t=: 0 1 2 3 3 3 2 2 3 3 1 1 NB. Depth vector representation
7!:5 <'t'
256
That said, it would be neat if we could trick (5!:4) into pretty-printing (s)
for us.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
