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

Re: Expand[] on the RHS of a rule

• To: mathgroup at yoda.physics.unc.edu
• Subject: Re: Expand[] on the RHS of a rule
• From: ags at seaman.cc.purdue.edu (Dave Seaman)
• Date: Thu, 28 May 92 09:30:21 EST

athos at warp.stanford.edu (Athos Kasapi) writes:

>Has anybody found a way to work around the problem that Expand does not
>work when it is on the right hand side of a rule?  For example, if I
>define
>
>p := Sqrt[ (a + b) (c + d) ]
>
>and then apply the rule
>
>p /. Sqrt[ x_ y_ ] -> Sqrt[ Expand[ x y ] ]
>
>Then all I get back is
>
>Sqrt[ (a + b)(c + d) ]
>
>What I wanted was
>
>Sqrt[ ac + ad + bc + bd ]
>
>I'd appreciate any suggestions for getting around this.

You specified immediate transformation by using "->", which means
"Expand[x y]" is immediately evaluated to "x y", and then the
resulting rule is applied.

You need to use the ":>" operator to specify delayed transformation.
The rule is

	p /. Sqrt[ x_ y_ ] :> Sqrt[ Expand[ x y ] ]

in which the delayed transformation operator ":>" causes the product
to be expanded after the substitution is made, rather than before.

Dave