cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - If community subscription notifications are filling up your inbox you can set up a daily digest and get all your notifications in a single email. X

To Tom Gutman

ValeryOchkov
24-Ruby IV

To Tom Gutman

Dear Tom!
Some yeas ago you have sent me a Expresion Compiler (Text2Func function) for Mathcad 11.
If You have same for Mathcad 12, 13 or 13.5 - send me please!
Val
http://twt.mpei.ac.ru/ochkov/v_ochkov.htm
7 REPLIES 7

Sorry -- the changes made in MC12 (and perpetrated in MC13) have killed that compiler. I just use MC11 for any uses of that (and many other things).

� � � � Tom Gutman
SteenGroðe
6-Contributor
(To:TomGutman)

But I would like to see the original version.

Steen Gro�e

The original compiler was posted in http://collab.mathsoft.com/read?50345,63e#50345 . There is another version (one compatible with the symbolic processor) in the NIST worksheet.

� � � � Tom Gutman

On 8/9/2006 3:55:39 PM, sgrode wrote:
>But I would like to see the
>original version.
>
>Steen Gro�e

Please - see attach

Val
http://twt.mpei.ac.ru/ochkov/v_ochkov.htm

On 8/10/2006 3:29:37 AM, VFO wrote:
>Please - see attach

I had a brief look last night at the expression compiler Tom referenced earlier in the thread.

Unfortunately, as Tom alluded, M13 static type checking severely breaks the underlying structure of the compiler.

M13 and M12 do not allow arrays to mix functions and scalars, nor do they allow a function to accept a function as a parameter unless all such functions have the same number of arguments.

For example:

x:=[1 2 sin]

and

f(g,v):=
g(v) if IsScalar(v)
g(v0,v1) if rows(v)=2
g(v0,v1,v2) otherwise

are both legal and useful constructs in M11 but are illegal in M12 and M13.

The compiler makes use of these constructs and hence fails under M13.

It may be possible to rewrite the compiler by splitting up the arrays and having separate handling for monadic and dyadic functions, but the result will be quite messy and lose the elegance of Tom's implementation.

I believe that the following thread partially addresses some of the issues (I'm Mathcadless so can't check)

http://collab.mathsoft.com/read?75228,85e#75228

Like Tom, I have several *very* useful M11 worksheets that are conceptually simple but which would require a total rewrite to get working in M13, and introduce a maintenance nightmare in the process ... provided I could find a way round all of the difficulties.

Stuart

(M13 is a lot better than M12, where I've never found a workaround for most of my sheets).

Solution for Mathcad 12 & 13 but not for future version of Mathcad and without symbolic Math - see attach
Val
http://twt.mpei.ac.ru/ochkov/v_ochkov.htm

There is a little bug in "converter.xmcd" if one of contained functions is used with ORIGIN not equal to zero.
I have corrected the problem and added two examples. See attach.

Massimo Priano
Top Tags