[Date Index]
[Thread Index]
[Author Index]
Re: Releasing several Holds simultaneously
*To*: mathgroup at smc.vnet.net
*Subject*: [mg82271] Re: Releasing several Holds simultaneously
*From*: "Szabolcs HorvÃt" <szhorvat at gmail.com>
*Date*: Wed, 17 Oct 2007 03:46:17 -0400 (EDT)
*References*: <ff1po5$90n$1@smc.vnet.net> <47147224.60309@gmail.com>
On 10/16/07, Andrew Moylan <andrew.j.moylan at gmail.com> wrote:
> Hi Szabolcs,
>
> My general expression is of the form Hold[SetDelayed[a, b]] where both a and
> b may contain one or more Holds. The expression is the result of using
> Mathematica to generate Mathematica code.
>
> I start with Hold[SetDelayed[a, b]] and then insert (via /. and
> placeholders) code into b. The reason b contains Holds is that some of the
> pieces of code that I insert into it are wrapped in Hold because they can't
> be evaluated:
>
> pieceofcode = Hold[z[[1]]];
>
> So I could generalise my question like this:
>
> Suppose I have Hold[a := b] and I have pieceofcode = Hold[z[[1]]]. How can I
> arrange to evaluate a := z[[1]]?
>
> Here's my present solution:
>
> SetAttributes[SuperHold, HoldAll];
> ReleaseSuperHold[expr_] := expr /. SuperHold[x_] :> x;
>
> pieceofcode = Hold[z[[1]]];
> SuperHold[a := b] /. b -> pieceofcode //
> ReleaseHold // ReleaseSuperHold;
>
> Andrew
>
Here's an alternative:
In[1]:= z={1,2,3}
Out[1]= {1,2,3}
In[2]:= pieceofcode:=Hold[z[[1]]]
In[3]:= toBeAssginedTo=a
Out[3]= a
In[4]:= Function[{val},Evaluate[toBeAssginedTo]:=val,{HoldAll}]@@pieceofcode
In[5]:= ?a
Global`a
a:=z[[1]]
You could also "Block out" (Block[{z}, ...]) 'z' and other variables
instead of Holding them, but this may not work for all types of code;
or wrap expressions containing 'z' in Unevaluated while you pass them
on. It all depends one exactly what kind of Mathematica code are you
building (z[[1]] was of course a simplified example):
In[1]:= SetAttributes[fun,HoldRest]
In[2]:= fun[a_,z_]:=a:=Unevaluated[z[[1]]]
In[3]:= z={1,2,3}
Out[3]= {1,2,3}
In[4]:= fun[a,z]
In[5]:= ?a
Global`a
a:=z[[1]]
Prev by Date:
**Re: Efficient creation of regression design matrix**
Next by Date:
**Re: Recursion limit: a good idea to ignore?**
Previous by thread:
**Re: Releasing several Holds simultaneously**
Next by thread:
**RE: Releasing several Holds simultaneously**
| |