MathGroup Archive 1995

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Positive[a] = True ???

  • Subject: [mg2618] Re: [mg2518] Positive[a] = True ???
  • From: johan at kajsa.isy.liu.se (Johan Gunnarsson)
  • Date: Thu, 30 Nov 1995 20:58:27 -0500
  • Approved: usenet@wri.com
  • Distribution: local
  • Newsgroups: wri.mathgroup
  • Organization: Wolfram Research, Inc.


If the symbol a is positive then why cannot Mathematica simplify
Sqrt[a^2] to a?

In[1]:= Positive[a]^=True;

In[2]:= Simplify[Sqrt[a^2]]

              2
Out[2]= Sqrt[a ]

/Johan

In article <48ejb3$fi5 at ralph.vnet.net> Allan Hayes <hay at haystack.demon.co.uk> writes:

> Frank Domokos <domokos at chaph.usc.edu>
> in [mg2518] Positive[a] = True ???
> asks about the following kind of response
> 
> IN>> Positive[a] = True;
> OUT>> Set::write: Tag Positive in Positive[a] is Protected.
> 
> Frank,
> 
> Most system functions, like Positive, are protected (have the  
> attribute Protected) to avoid unintentional changes. You cannot make  
> a definition for them ("tagged" by them) without unprotecting them  
> (see later).
> 
> Here are some ways round this for your example
> 
> 1.Tag your definition with a instead of f:
> 
>    a. by using UpSet (^=) instead of Set (=):
> 
>       Positive[a]^= True;
> 
>    or, more specifically,
> 
>    b. by using TagSet:
> 
>       a/:Positive[a] = True;
> 
> Or
> 
> 2. Unprotect Positive to allow definitions to be tagged by it:
> 
>    Unprotect[Positive];
> 
>    Positive[a] = True;
> 
>  (eventually re-protecting Positive)
> 
> 
> Of course if a is protected or is an expression with a protected  
> head then you will need to unprotect either Positive or a or the  
> head of a.
> 
> Allan Hayes
> hay at haystack.demon.co.uk


-- 

_____________________________________________________
		Johan Gunnarsson             
     |    	Division of Automatic Control
    /|\   	Dept. of EE, Linkoping University
    \|/   	S-581 83 Linkoping, Sweden
    /|\   	Tel: 	+46 13 282913     
   / | \  	Fax: 	+46 13 282622     	
 ()  |  ()	Email: 	johan at isy.liu.se






  • Prev by Date: Peculiar behavior of Integrate
  • Next by Date: Re: Bug in symbolic inversion of matrices
  • Previous by thread: RE: Positive[a] = True ???
  • Next by thread: [Q] how fast is each machine with Mathematica ?