After studying APL2 language reference for a while, it seems like
APL's rho(reshape) and J's shape($) has different behaviour.
R=:$@,
spd=: 4 : 0
X=.y
col=. x
((1{f),*/ 0 2{f ) R 1 0 2|:(f=.col,c%col,1)R(c=.d+(col|col-0{(d=.
$X)),0){. X
)
mat=.3 4 $'can you see '
2 f mat
can see
you
3 f mat
can you see
$2 f mat
2 8
$3 f mat
1 12
This works. Didn't I make any mistakes?
2007/2/18, June Kim <[EMAIL PROTECTED]>:
Hello
I've recently read an interesting article by Michael Berry and Roland
Pesch. The title is Style and Literacy in APL.
The whole point seems an APL code that's written with the style of
each line being short(broken into pieces) isn't really helpful in
improving the readability of the code, and actually it decreases the
readability. Moreover, reading from left to right, the longer line
style enables skimming and getting the whole picture easily.
The author quotes some example codes from the literatures. One of them
is from Standards for APL Applications Development and Enhancement, by
Eugene R. Mannacio.
SPD is a long line function and SPREAD is many lines composed of short
ones, but both do the same thing.
I don't know much about APL so it is hard for me to understand either
of the functions. (the author's intention was showing SPD being easier
to comprehend, though)
spd=: 4 : 0
X=.y
col=. x
((1{f),*/ 0 2{f )$1 0 2|:(f=.col,c%col,1)$(c=.d+(col|col-0{(d=. $X)),0){. X
)
This is my translation but obviously there was a few mis-translations,
I suspect.
There's a comment that describes what SPD does here: spreads a right
argument matrix X of shape <M,N> to a result of shape
<(T=.>.M%COLS),COLS*N>. Places first group of T rows in the first N
columns and the next group of T rows in next N columns, etc.
SPREAD also has its comment describing what it does(same thing with
SPD): spreads a right argument matrix to a multiple of its current
width. The multiple is specified by its left argument. Left arugment:
the multiple desired (an integer). Right argument: the martix to be
spread(usually character data). Result: the spread matrix.
Now, could anyone help me translating the function spd into J(then I
might easily translate SPREAD, too)?
(I temporarily put the pdf file for the paper on my server. I might
remove it in a couple of days. Get it from
http://agcns.com/p1-berry.pdf )
Thanks
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
