       Re: Problem with Solve and O[x]

• To: mathgroup at christensen.cybernetics.net
• Subject: [mg1257] Re: Problem with Solve and O[x]
• From: danl (Daniel Lichtblau)
• Date: Mon, 29 May 1995 02:38:49 -0400
• Organization: Wolfram Research, Inc.

```In article <3q3mcr\$mc5 at news0.cybernetics.net> steve at tweedledee.ucsb.edu
(Steve Trainoff) writes:
> It appears the Solve and O[x] don't mix.  Consider the following.
>
> Mathematica 2.0 for NeXT
> Copyright 1988-91 Wolfram Research, Inc.
>
> In:= Solve[x+z==1,x]
>
> Out= {{x -> 1 - z}}
>
> In:= Solve[x+z==1+O[z],x]
>
> Out= {{x -> 1}}	(* I would prefer x -> 1 + O[z] *)
>
> In:= Solve[x+z==1+O[z]^2,x]
>
> Out= {}		(* What is wrong with x -> 1 + z + O[z]^2 ??? *)
>
> --
>
> ...STeve

This is caused by the fact that Solve uses LogicalExpand, and
LogicalExpand does not do what one might wish in these cases.

In:= LogicalExpand[x == 1+O[z]]
Out= -1 + x == 0

In:= LogicalExpand[x+z==1+O[z]^2]
Out= False

If I can fix this immediate problem without breaking Solve, I'll do so.

Daniel Lichtblau, WRI

```

• Prev by Date: strings into lists
• Next by Date: Numerous Functional Forms
• Previous by thread: Problem with Solve and O[x]
• Next by thread: Re: Problem with Solve and O[x]