It is neither.

   $1 2 +/ i. 4
2 4

Which, perhaps would explain your dilemma?

Thanks,

-- 
Raul


On Thu, Sep 25, 2014 at 2:09 AM, Chris Wright <[email protected]> wrote:
> thanks Robert!
>
> OK. boxes are atoms withb an empty shape.
>
> I've looked at the docs and can't see where I could read to understand why
> shape of
>
> 1 2 +/ i. 4
>
> is 2 1 and not 1 2
>
> thanks for your help!
>
> Chris
> On 25/09/2014 4:01 PM, "robert therriault" <[email protected]> wrote:
>
>> Hi Chris,
>>
>> Well there is information on the vocabulary page for Table
>> http://www.jsoftware.com/help/dictionary/d420.htm
>>
>> also Nuvoc has good information on this verb
>> http://www.jsoftware.com/jwiki/Vocabulary/slash#dyadic
>>
>> The actual shape of a box is empty since the box is an atom. In the case
>> of 1 2 +/ i. 4 the shape is actually 2 4 i.e. 2 rows of 4 not 2 1
>>
>>    < 1 2 +/ i. 4
>> ┌───────┐
>> │1 2 3 4│
>> │2 3 4 5│
>> └───────┘
>>    $ < 1 2 +/ i. 4
>>    $ each < 1 2 +/ i. 4
>> ┌───┐
>> │2 4│
>> └───┘
>>
>> If you wanted the two rows to be in their own boxes then you could do
>>
>>    ;/  1 2 +/ i. 4
>> ┌───────┬───────┐
>> │1 2 3 4│2 3 4 5│
>> └───────┴───────┘
>>    or
>>
>>    <"1 [ 1 2 +/ i. 4
>> ┌───────┬───────┐
>> │1 2 3 4│2 3 4 5│
>> └───────┴───────┘
>>
>>    $ <"1 [ 1 2 +/ i. 4
>> 2
>>
>> The above shows that the shape of the two boxes is not actually 1 2
>> because each box is an atom the shape is actually just 2.
>>
>> You are correct that an understanding of rank and shape is key to getting
>> J to work for you. If you are looking for information I have always found
>> Henry Rich's J for C Programmers much more accessible than the J
>> dictionary.  Chapters 5 and 6 are really good at explaining rank and shape
>> with regard to verbs. http://www.jsoftware.com/help/jforc/contents.htm
>>
>> Also, look at Nuvoc on the wiki. This is a crowd sourced resource
>> organized by Ian Clark and Henry Rich which many have found useful.
>> http://www.jsoftware.com/jwiki/NuVoc
>>
>> Enjoy the ride as you learn J. Welcome aboard.
>>
>> Cheers, bob
>>
>>
>> On Sep 24, 2014, at 10:35 PM, Chris Wright <[email protected]> wrote:
>>
>> > Hi
>> >
>> > Could I work out from the docs that
>> >
>> >  < 1 2 +/ i. 4
>> >
>> > ┌───────┐
>> >
>> > │1 2 3 4│
>> >
>> > │2 3 4 5│
>> >
>> > └───────┘
>> >
>> > will give the above (2 1) shape result
>> >
>> >
>> > and not a (1 2) result like below?
>> >
>> >
>> > ┌───────┬───────┐
>> >
>> > │1 2 3 4│2 3 4 5│
>> >
>> > └───────┴───────┘
>> >
>> >
>> >
>> >
>> > With the (2,1) it's obvious that two 1-cells are returned,  and with
>> (1,2)
>> > it's not obvious (where is the boundary b/n them), but I wonder if I
>> could
>> > have worked out the shape of the output from the docs alone?
>> >
>> >
>> > with thanks
>> >
>> >
>> > Chris
>> > ----------------------------------------------------------------------
>> > For information about J forums see http://www.jsoftware.com/forums.htm
>>
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to