I think the issue is tolerance, it works for 0-tolerance.
eg.
(#,{.)/.!.0~ tb
1 1
1 1
1 1
1 1
1 1
1 1
1 1
Сб, 10 окт 2015, Henry Rich написал(а):
> B Jonas reported that ~. is inconsistent with the definition:
>
> tb=. 1+2e_14*i:3
>
> ~: tb
>
> 1 0 0 0 0 0 0
>
> =tb
>
> 1 1 1 0 0 0 0
>
>
> You could fix that in the docs. But the problem is much more serious: the
> partitioning induced by ~. does not match that induced by u/. . That these
> be identical is a canon of J, relied on by code whose name is legion.
>
>
> ~. tb
>
> 1
>
> tb </. tb
>
> +-----+-+-+-+-+
>
> |1 1 1|1|1|1|1|
>
> +-----+-+-+-+-+
>
>
> What!? Only 1 unique item, but 4 partitions???
>
>
>
>
> The special code for #/. shares the problem:
>
>
> #/.~ tb
>
> 3
>
>
> only 3 items from a list of 7!
>
>
>
> The interpreter itself relies on the equivalence, and is let down:
>
> (#,{.)/.~ tb
>
> |length error
>
> | (#,{.)/.~tb
>
>
> there shouldn't be any error.
>
>
>
> IMO this will be the first bug for us to fix, if Jsoftware ever creates a
> proper open-source setup that allows user contributions to get into the
> official release.
>
> Henry Rich
>
> ----------------------------------------------------------------------
> 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
gpg --keyserver subkeys.pgp.net --armor --export 4434BAB3
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
