In J gtk, the session expects data in utf8 format. Obviously a. won't work,
but could be converted to utf8 for display.
The utility that formats data for display is octal_j_. This does a simple
check on the argument string and either returns it unchanged, or formats to
octal. A simple improvement is to run the formatter on each line, rather
than the string as a whole, as below:
octal_j_=: 3 : 0
t=. ,y
x=. a. i. t
n=. x e. 9 10 13
m=. n < 32 > x
if. (isutf8 t) > 1 e. m do. t return. end.
if. LF e. t do.
t=. octal each <;._2 t,LF
}: ; t ,each LF return.
end.
r=. t ,"0 1 [ 3 # EAV
if. #m=. I. m +. x>126 do.
s=. '\',.}.1 ": 8 (#.^:_1) 255,m{x
r=. s m} r
end.
EAV -.~ ,r
)
For example, this works reasonably well with the new definition:
load 'unicode'
toutf8 &> _16<\a.
On my machine, the first two rows (first 32 chars) are still in octal, but
the rest get displayed using the proper characters (where available in the
font).
I would appreciate suggestions for futher improvements to the octal utility.
On Sun, Feb 5, 2012 at 1:14 PM, bill lam <bbill....@gmail.com> wrote:
> behaviour of < is different if the string is unicode. u: is required in
> some places and it is not related to simplicity or complexity if you
> wants correct results.
>
> compare the outputs of _16{.a. with u: _16{.a.
>
> the first contains illegal utf8 characters which can be checked by
> isutf8 _16{.a.
>
> What would you expect from gtkide to display those illegal characters?
>
> Вск, 05 Фев 2012, Fraser Jackson писал(а):
> > Why should I have to use u: to box elements of a. ? Seems an
> unnecessary
> > complication - and a bigger problem for beginners.
> >
> > In gtkide <"0 [2}. u: i. 16 16 still has all the problems. That
> suggests
> > some problems are associated with the unicode treatment of control
> > characters.
> >
> > Even in Jconsole the first and second line are also irregular.When the
> > configure specifies a mono spaced font in gtkide we should surely expect
> > characters within the ascii set will be treated as atoms and assigned a
> > single space.
> >
> > I am really seeking that members of the alphabet ( and hence any literal
> > nouns defined using that set ) have each member treated as an atom and
> that
> > that treatment is clearly separated from their role in controlling the
> > nature of the display. That seems to have been achieved in J602.
> >
> > The treatment of type 131072 - unicode obviously involves a host of other
> > issues with respect to display and treatment of aoms. However there are
> a
> > large number of problems within the narrower framework of atoms within
> a. or
> > represented by those characters. Precise, simple and consistent display
> of
> > those atoms is extremely valuable and important and has not yet been
> > achieved with the gtkide.
> >
> > Fraser
> > ----- Original Message -----
> > From: "bill lam" <bbill....@gmail.com>
> > To: <programming@jsoftware.com>
> > Sent: Sunday, February 05, 2012 3:04 PM
> > Subject: Re: [Jprogramming] Problems displaying the alphabet in J701gtk
> and
> > J701jhs
> >
> >
> > > It seemes that the characters in the first two rows caused the trouble,
> > > the following display ok in gtkide,
> > >
> > > <"0 [ u: 2}. i.16 16
> > >
> > > jconsole have no problem in displaying them.
> > >
> > > <"0[u: i.16 16
> > >
> >
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
>
> --
> regards,
> ====================================================
> GPG key 1024D/4434BAB3 2008-08-24
> gpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
