it seems a bug in floating point numbers in-place concatenation as illustrated
below
a=. i.4 4
a=. a,("1) 1 2 3
a
0 1 2 3 1 2 3
4 5 6 7 1 2 3
8 9 10 11 1 2 3
12 13 14 15 1 2 3
NB. so far correct, but for floating point numbers
a=. (1.1-1.1)+ i.4 4
a
0 1 2 3
4 5 6 7
8 9 10 11
12 13 14 15
a=. a,("1) 1 2 3
a
0 1 2 3 1 2 3 0 0 0 0 0 0
4 5 6 7 1 2 3 0 0 0 0 0 0
8 9 10 11 1 2 3 1 2 3 1 2 3
12 13 14 15 1 2 3 1 2 3 1 2 3
NB. rubbish columns appended
a=. (1.1-1.1)+ i.4 4
a=. a,("1) 1 2 3
a
0 1 2 3 1 2 3 4.58553e_72 1.5549e_259 0 0 0 0
4 5 6 7 1 2 3 0 0 0 0 0 0
8 9 10 11 1 2 3 1 2 3 1 2 3
12 13 14 15 1 2 3 1 2 3 1 2 3
NB. rubbish columns with strange numbers
a=. (1.1-1.1)+ i.4 4
]a=. a,("1) 1 2 3
0 1 2 3 1 2 3
4 5 6 7 1 2 3
8 9 10 11 1 2 3
12 13 14 15 1 2 3
a
0 1 2 3 1 2 3
4 5 6 7 1 2 3
8 9 10 11 1 2 3
12 13 14 15 1 2 3
NB. strange, if the value is used for further evaluation, the result seems
correct.
It runs fine if not in-place , eg
b=. a,("1) 1 2 3
Tested on j6, j7 , 32-bit windows, linux and iphone. not all combinations.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
