> [EMAIL PROTECTED] On Behalf Of Miller, Raul D
>
> Jose Mario Quintana wrote:
> > The change in behavior that I suggested would only
> > apply to expressions that currently generate an error;
>
> Ok, let's take as given that you specified
> box-if-open only for cases which would otherwise
> trigger domain errors
>
> Note that this means
> 0`1`2`#`4`5`6
> would be the same as
> 0 1 2`#`4`5`6
> but not the same as
> 0`1`2`#`4 5 6
>
> I think this is asymmetrical enough that it warrants
> mention.
Thanks for the feedback. That behavior would be an enhancement as far as I
am concerned. Yet, the way I would really want tie to operate (to simplify
the typing and reading of mixed gerunds) is as follows:
0`1`2`#`4`5`6
+-+-+-+-+-+-+-+
|0|1|2|#|4|5|6|
+-+-+-+-+-+-+-+
0 1 2`#`4`5`6
+-----+-+-+-+-+
|0 1 2|#|4|5|6|
+-----+-+-+-+-+
0`1`2`#`4 5 6
+-+-+-+-+-----+
|0|1|2|#|4 5 6|
+-+-+-+-+-----+
and even,
0`1 2`#`(^&*)`(i.2 3)
+-+---+-+---------+-----+
|0|1 2|#|+-+-----+|0 1 2|
| | | ||&|+-+-+||3 4 5|
| | | || ||^|*||| |
| | | || |+-+-+|| |
| | | |+-+-----+| |
+-+---+-+---------+-----+
This change would break existent code and I guess there are not many J users
out there using mixed gerunds; thus, I do not expect this change to be
implemented. So, I just wrote a conjunction (l) that performs that job
instead.
>
> --
> Raul
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
