s =: ' ';'this';'is';'a';'sentence'
(<'is') 0} s
┌──┬────┬──┬─┬────────┐
│is│this│is│a│sentence│
└──┴────┴──┴─┴────────┘
4 ({.,(<'sample'),}.) s
┌─┬────┬──┬─┬──────┬────────┐
│ │this│is│a│sample│sentence│
└─┴────┴──┴─┴──────┴────────┘
1 }. s
┌────┬──┬─┬────────┐
│this│is│a│sentence│
└────┴──┴─┴────────┘
(2 {. s), 3 }.s
┌─┬────┬─┬────────┐
│ │this│a│sentence│
└─┴────┴─┴────────┘
(<'the') (I. s = <,'a')} s
┌─┬────┬──┬─┬────────┐
│ │this│is│a│sentence│
└─┴────┴──┴─┴────────┘
(<'the') (I. s = <'a')} s
┌─┬────┬──┬───┬────────┐
│ │this│is│the│sentence│
└─┴────┴──┴───┴────────┘
s -. <'is'
┌─┬────┬─┬────────┐
│ │this│a│sentence│
└─┴────┴─┴────────┘
In other words:
1) yes
2) yes
3) yes, but be careful
In this case, your s contained five boxes. Four of those boxes where
character sequences. One of them was just a character. In the
general case, you need to be aware of the distinction between a
character (rank 0) and a sequence of characters (rank 1), and you
especially need to be aware of the distinction between a character and
a sequence of characters of length 1. They're different but seem to
be the same.
This is analogous to the distinction between
2
and
3 4 5
The first is just a number, the second is a sequence of numbers...
And you can have sequences of numbers of length 1, but that's not
quite the same thing as the number by itself.
--
Raul
On Sun, Sep 16, 2012 at 3:30 PM, Emir Ustamujic <[email protected]> wrote:
> Hello,
>
> As I'm often working with boxes of strings I have identified several things
> that I would like to do with boxed strings but haven't figured how to do yet
> as I'm still learning J.
>
> [I hope the boxes in the examples don't turn out garbled...]
>
>
> 1. If there is a list of boxed words(strings) how to replace individual items
> with another string ?
>
> For example I have :
> ] s =: ' ';'this';'is';'a';'sentence'
> ┌─┬────┬──┬─┬────────┐
> │ │ this│ is│a│ sentence│
> └─┴────┴──┴─┴────────┘
> but would like to use something like: 0 boxitemreplace s; 'is' (to replace
> the first element which can be a space with 'is)
> and get this result :
> ┌──┬────┬──┬─┬────────┐
> │is │this │is │a │sentence │
> └──┴────┴──┴─┴────────┘
> or something like 2 boxitemreplace s;' ' (to replace the third element with a
> space)
> ┌──┬────┬──┬─┬────────┐
> │ │this │ │a │sentence │
> └──┴────┴──┴─┴────────┘
>
> 2. Insert and Remove
>
> For example using the same s =: ' ';'this';'is';'a';'sentence':
> would it be possible to somehow insert a word at a specific index so: 4
> inserttobox s;'sample'
> would yield
> ┌─┬────┬──┬─┬──────┬────────┐
> │ │this │is │a │sample │sentence │
> └─┴────┴──┴─┴──────┴────────┘
> or remove nth item such as 0 removeitemfrombox s;' '
> ┌────┬──┬─┬──────┬────────┐
> │this │is │a │sample │sentence │
> └────┴──┴─┴──────┴────────┘
>
>
> 3. And consequently would it be also possible to replace all occurences of
> one string value within s with another string and remove all occurences of
> one string from a boxed list such as s ?
>
> I more or less know how to do this with regular strings such as s=: 'this is
> a sentence' but my problem is when dealing directly with boxed strings and
> when they contain spaces.
>
>
> Thanks &Regards,
> Emir
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
