Joey,
Thank you finding the bug, your fix has been committed to the
base library.
There had been some work done to profile.ijs and boot.ijs, but
the 9!:37 was not touched. May be it was fixed indirectly, so
that it may also be broken again indirectly in the future.
Вт, 21 янв 2014, Joey K Tuttle писал(а):
> Bill,
>
> Sooo, my confusion was caused by the fact that in file
> J64-801/system/mail/stdlib.ijs on line 194 expand is defined:
>
> expand=: # inverse
>
> but, unfortunately, due to "cleanup" it isn't until line 225 that
> inv is defined:
>
> inv=: inverse=: ^:_1
>
> I believe you need to be sure inverse is defined before expand. This
> is an example of making things "readable" isn't as functional as just
> defining them directly ... i.e. expand=: #^:_1 NB. would "fix the
> problem" without reordering the lines.
>
> Sometimes making things read better (or redundantly defined) causes
> problems (thinking back to my complaints about my use of 9!:37 in my
> profile being stomped on by boot) - HOWEVER, I want to thank you for
> fixing that in util/boot.ijs so that my profile works OK in j8 -
> THANK YOU!
>
> - joey
>
> On 2014/01/21, 15:38 , Tom Allen wrote:
> >Sub-topic about plot problem -
> >
> >plot with " pd 'pdf' " results in
> >
> > |domain error: inverse
> >
> >| ext=.msk expand(;len{.each 1)+/;.1 ext
> >
> >
> >as described by
> >
> >
> >Joey
> >
> >when using SUSE Linux 12.3.
> >
> >eg.
> >
> > load 'plot'
> >
> >
> >pd 1 2 3 4 5;1 4 9 16 25
> >
> >
> >pd 'pdf'
> >
> >
> >pd 'show'
> >
> >
> >
> >On Wed, Jan 22, 2014 at 5:08 AM, Joey K Tuttle <[email protected]> wrote:
> >
> >>Sub-topic about plot problem -
> >>
> >>Adding some echos (e.g. echo 'len';len) in afm.ijs
> >>
> >>Tue Jan 21 09:49:47
> >>iMi7:~ jkt$ j8
> >> load 'plot'
> >> plot i.6
> >>┌───┬─────────────────────┐
> >>│len│1 3 1 3 1 3 1 3 1 3 1│
> >>└───┴─────────────────────┘
> >>┌───┬───────────────────────────────────────────────────────
> >>────────────────────────────┐
> >>│ext│556 556 278 556 556 556 278 556 556 556 278 556 556 556 278 556 556
> >>556 278 556 556│
> >>└───┴───────────────────────────────────────────────────────
> >>────────────────────────────┘
> >>┌───┬─────────────────────┐
> >>│msk│1 1 1 1 1 1 1 1 1 1 1│
> >>└───┴─────────────────────┘
> >>
> >>|domain error: inverse
> >>| ext=.msk expand(;len{.each 1)+/;.1 ext
> >>
> >>But trying to track down the problem, I get confused (not an unusual state
> >>for me...)
> >>
> >> expand
> >># inverse
> >> inverse
> >>^:_1
> >> 1 0 1 expand 'ac'
> >>|domain error: inverse
> >>| 1 0 1 expand'ac'
> >> 1 0 1 #^:_1 'ac'
> >>a c
> >> 1 0 1 (#^:_1) 'ac'
> >>a c
> >> 1 0 1 (# inverse) 'ac'
> >>a c
> >> 1 0 1 # inverse 'ac'
> >>a c
> >> 1 0 1 expand 'ac'
> >>|domain error: inverse
> >>| 1 0 1 expand'ac'
> >> expand
> >># inverse
> >>
> >>NB. sooo --- Replacing the line in afm.ijs with
> >>
> >>ext=.msk #inverse (;len{.each 1)+/;.1 ext
> >>
> >>iMi7:afm jkt$ j8
> >> load 'plot'
> >> plot i. 6
> >>
> >>It works fine (except that I am confused ... ;-)
> >>
> >>- joey
> >>
> >>
> >>
> >>On 2014/01/20, 22:49 , bill lam wrote:
> >>
> >>>the plot domain error actually occurs inside afm. Can you help
> >>>investigate the problem by adding smoutout msk len ext just before
> >>>that line in afm.ijs or advise what will be the fix to it?
> >>>
> ----------------------------------------------------------------------
> 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
