Re: Evaluate/Module

*To*: mathgroup at smc.vnet.net*Subject*: [mg16106] Re: Evaluate/Module*From*: Mark Fisher <mefisher at bellsouth.net>*Date*: Thu, 25 Feb 1999 08:25:01 -0500*References*: <7atq9s$7kr@smc.vnet.net>*Sender*: owner-wri-mathgroup at wolfram.com

Bernd, I don't think Module is playing a central role here. Rather I think CompoundExpression is. The FullForm of (a; b) is CompoundExpression[a,b]. It has the attribute HoldAll. When you write (a; Evaluate[b]) you have told Mathematica to find out what b is *before* starting to execute the progam (a; b). Thus if you write tmp = x^2+1; (tmp = tmp-1; Evaluate[tmp]) --> (tmp = tmp-1; x^2+1) Of course, when you check tmp afterwards, you find that it is x^2 now. When you change your code to tmp = x^2+1; (tmp = tmp-1; l = Evaluate[tmp]) the second argument in the CompoundExpression has the head Set, not Evaluate and hence it does not get evaluated "out of order". You see that both tmp and l are now set to x^2. If you had instead written tmp = x^2+1; (tmp = tmp-1; Evaluate[l = tmp]) you *would* have forced the evaluation of the Set statement before the reassignment of tmp. Afterwards you will find that tmp is x^2 but l is x^2+1. Thinking about your example helped me sharpen my understanding about this stuff: Thanks. --Mark. Bernd Brandt wrote: > Hi, > > I'm having some trouble with Evaluate and i cannot find this in the > documentation. > Does anybody know what and esp. why the following is happening? > > In1:= tmp = x^2 +1; > > In2:= Module[{x=4},tmp=tmp-1;Print[tmp]; Evaluate[tmp]] > > x^2 > x^2 + 1 > > In the above Evaluate[tmp] somehow takes the latest value of tmp and not > the current one after the "tmp=tmp-1". Why? > > Changing "Evaluate[tmp] to "l=Evaluate[tmp]" does give the right answer > (in this case x^2) immediately. Does this assignment force Evaluate to > Evaluate tmp before ending the Module. Does someone know more about > Evaluate and Modelule? > > lM[x0_]:= Module[{lonediff,lMdiff,ldead,lone,search}, > lonediff= 1 + lone/2 - x0/lone - 2 / (1 + Exp[2*ldead - 2*lone]); > lMdiff = (2*(x0/(1+x0) )*E^(-ldead+ lone))/(E^(2*lone-2*ldead)+1)- > 0.1; > search=FindRoot[{lMdiff==0, lonediff==0 },{ldead,x0+1} ,{lone,x0}]; > Evaluate[ldead /. search] > ]; > > lM[6] > > ReplaceAll::"reps": > {search$77} is neither a list of replacement rules nor a valid > dispatch table, and so cannot be used for replacing." > > 5.44865 > > In this module Evaluate somehow is looking for an global "search" > variable. > Why does it not just evaluate this local "search"? > > Thanks, > > Bernd