Newsgroups: comp.lang.apl
Path: watmath!watserv2.uwaterloo.ca!torn!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!jato!csi!sam
From: sam@csi.jpl.nasa.gov (Sam Sirlin)
Subject: Re:      Questions using J version 5
Message-ID: <1992Aug31.185438.29879@csi.jpl.nasa.gov>
Originator: sam@kalessin
Sender: usenet@csi.jpl.nasa.gov (Network Noise Transfer Service)
Nntp-Posting-Host: kalessin
Organization: Jet Propulsion Laboratory, Pasadena, CA
References:  <920831.093911.ITA.AGRTQB@IRMFAO01>
Date: Mon, 31 Aug 1992 18:54:38 GMT
Lines: 66


In article <920831.093911.ITA.AGRTQB@IRMFAO01>, agrtqb%irmfao01.bitnet@utcs.utoronto.ca writes:
|> 
|> Using J version 5 (from APL92) we encountered some problems or bugs
(on an IBM?)
Using J 5.1 on a sun sparc, I get
|> a) external function save workspace

These seemed to work for me.

|> b) with the Member of Interval
|> 
|>       NB. from `Dictionnary of J` page 34
|>       'co' E. 'cocoa'

this worked.

|>       (0 1 +/ i. 3) E. 4 | +/~ i. 5

E. with an array right argument gives a nonce error. I'd guess E.
isn't quite implemented yet (the status file says NYA).

|> 
|> c) with the fit function
|> 
|>       x=. 20.5 20.6
|>       y=. 20.52 20.61
|>       x=y
|> 
|>       x =!.0.2 y

domain error
As with normal APL systems, there seems to be a maximum you can set
the fuzz to. I'd expect this to be machine dependant. Try something
like: 
   y=. x*1+1e_12
   y
20.5 20.6
   y-x
2.05027e_11 2.06022e_11
   x=y
0 0
   x=!.1e_12 y
0 0
   x=!.1e_11 y
1 1


|> d) How do you, in a explicit function, create a prompt for input, like:

This is very machine dependant. But it might not be to hard to do if
you know what your screen driver does, using components from a. Trying
this using an Xterm, I can get the prompt on any line, but not in any
but the first column... 

|> PS moving file from the PC to the mainframe I get incorrect characters
|>    as well as when I receive files or messages from the network; this
|>    most of the time concerns the square brackets and caret, sometimes the
|>    stile. Solutions?

I think you need a translator ASCII/EBCDIC, or use xxencode at both
ends.  
-- 
Sam Sirlin 
Jet Propulsion Laboratory        sam@kalessin.jpl.nasa.gov

