Number: 980

Date:  4-May-84 12':03':39

Submitter: Sannella.PA


Subject: Want general support for rational numbers

Lisp Version: 

Description: '
Date':  1 May 84 17':51 PDT'
Subject': Lisp': Rational numbers not yet quite "rational"'
To': Kaplan '
Lisp System Date': 27-Apr-84 12':25':44'
Machine': Dorado (PASTEUR)'
Microcode version': 24,4'
Memory size': 10000'
Frequency': Always'
Impact':  Minor '
Rational numbers aren''t quite usable in general because they are merely a cons of two fixps (rather than being a datatype).  '
You are, I believe, the source of interest in rationals.   Do you know of anyone else using them?  Is all of your code "clean" -- i.e., do you use the record forms "create" and "fetch" rather than CONS and CAR ?'
A side comment': (MAKERATIONAL <rational>) drops into a break.'
-- JonL --'
Date':  2 May 84 12':18 PDT'
Subject': Re': Lisp': Rational numbers not yet quite "rational"'
In-reply-to':''s message of 1 May 84 17':51 PDT'
I use rationals in INTERPRESS, which requires them.  There are some records there and a couple of operations defined for them (RPLUS, RTIMES--just the ones I needed).  The function MAKERATIONAL was provide in the floating point package by Bill, to help me do the conversions that I needed.  Interpress''s use of these things is entirely thru field accesses, even though they are in fact represented by  a CONS pair.'
There has been no attempt to do a general implementation of rational arithmetic.  If someone is tempted, feel free to provide a more general interface and to remove the code from INTERPRESS to somemore desireable file.'


Test Case: 


Edit-Date:  4-May-84 12':02':03

Attn: kaplan, jonl

Assigned To: 



System: Language Support

Subsystem: Arithmetic



Microcode Version: 

Memory Size: 

File Server: 

Server Software Version: 

Difficulty: Hard

Frequency: Everytime

Impact: Minor

Priority: Perhaps

Status: Open

Problem Type: Design - Impl

Source Files: