Thank you very much. That works as I wanted.

 

Please:

 

What is the difference between

... assuming n::integer;

AND

... assuming n,integer;

 

The second form works as expected, but only sometimes, and I can never figure out when it is going to work.

What is that form supposed to do?

Thank you very much. That works as I wanted.

 

Please:

 

What is the difference between

... assuming n::integer;

AND

... assuming n,integer;

 

The second form works as expected, but only sometimes, and I can never figure out when it is going to work.

What is that form supposed to do?

Thanks.

You (or whoever worries about these things) might want to consider that the installation instructions are kind of misleading - the wording states that MSVC++ is "required" for installation.

 

Anyway, its installed now and working OK.

Thanks again.

Thanks.

You (or whoever worries about these things) might want to consider that the installation instructions are kind of misleading - the wording states that MSVC++ is "required" for installation.

 

Anyway, its installed now and working OK.

Thanks again.

Whoops - you are correct - I forgot that statements within a procedure aren't printed, without a print command.

So, that explains that.

Thank you.

Whoops - you are correct - I forgot that statements within a procedure aren't printed, without a print command.

So, that explains that.

Thank you.

Thank you.

I tried that and it works well, except in a procedure (which of course is exactly where I need to use it).

The help file talks about levels of recursion inside procedures, but there is no way to set a level when the second parameter of "eval" is already in use otherwise. See the example below.

Is there any way to do what I am trying to do,which is to substitute a function anme and transform the parameters all at the same time?

> restart;
> M1:=proc(X1,X2);
> f(X1,X2);
> end proc;

                M1 := proc(X1, X2) f(X1, X2) end proc

> try1:=M1(A,B);
> try2:=G(A,B);
> eval(try2,G=M1);

                           try1 := f(A, B)

                           try2 := G(A, B)

                               f(A, B)

> test:=proc(U,V) global M1; local try1, try2;
> try1:=M1(U,V);
> try2:=G(U,V);
> eval(try2,G=M1);
> print ("try1=",try1);
> print ("try2=",try2);
> end proc;

 ...

 

> test(a,b);

                           "try1=", f(a, b)

                           "try2=", G(a, b)

Thank you.

I tried that and it works well, except in a procedure (which of course is exactly where I need to use it).

The help file talks about levels of recursion inside procedures, but there is no way to set a level when the second parameter of "eval" is already in use otherwise. See the example below.

Is there any way to do what I am trying to do,which is to substitute a function anme and transform the parameters all at the same time?

> restart;
> M1:=proc(X1,X2);
> f(X1,X2);
> end proc;

                M1 := proc(X1, X2) f(X1, X2) end proc

> try1:=M1(A,B);
> try2:=G(A,B);
> eval(try2,G=M1);

                           try1 := f(A, B)

                           try2 := G(A, B)

                               f(A, B)

> test:=proc(U,V) global M1; local try1, try2;
> try1:=M1(U,V);
> try2:=G(U,V);
> eval(try2,G=M1);
> print ("try1=",try1);
> print ("try2=",try2);
> end proc;

 ...

 

> test(a,b);

                           "try1=", f(a, b)

                           "try2=", G(a, b)

I just figured out that "apply" is used incorrectly in the above example, and that the statement "eval(try2)" will do what I want, but I still don't understand why.

Can anyone explain?

Thank you

I may be naive, but wouldn't you think that evaluating the following:

 sin(Pi (j + 1))/j/(j-1) 

for j=0 and/or j=1 should trigger some sort of exception, since it leads to 0/0 in both cases.

From what I can see, the "try...catch" commands were created precisely to handle this 
kind of situation.

I'd rather receive an error message than get the wrong answer and spend hours 
trying to figure out why, or, worse yet, never notice.
 
(The example I gave was buried deep in a long complicated expression I was trying to 
reduce to something useful).

So, I call it a bug. 

Could you please clarify - what is the "specialization problem" you refer to and 
why is it so hard to fix. 

The example shows that it's not hard to detect.

Thank you.

I may be naive, but wouldn't you think that evaluating the following:

 sin(Pi (j + 1))/j/(j-1) 

for j=0 and/or j=1 should trigger some sort of exception, since it leads to 0/0 in both cases.

From what I can see, the "try...catch" commands were created precisely to handle this 
kind of situation.

I'd rather receive an error message than get the wrong answer and spend hours 
trying to figure out why, or, worse yet, never notice.
 
(The example I gave was buried deep in a long complicated expression I was trying to 
reduce to something useful).

So, I call it a bug. 

Could you please clarify - what is the "specialization problem" you refer to and 
why is it so hard to fix. 

The example shows that it's not hard to detect.

Thank you.

@Thomas Richard 

 

Yes, that is the one. The virtual machine is not limited to XP - you can load all the other old Windows operating systems if you are so inclined - I haven't tried.

Your recommendation makes no sense - you state that Maple is slower on virtual XP. That may be true but "slower" is still faster than "not at all".

By the way - I always use the classic worksheet anyway (I have troglodytic tendancies), and have noticed no speed issues between what runs on XP and what runs on Win 7 (when it does run).

Its probably the fancy front end that runs slowly on XP and notatall on Win 7 (just a guess).

The alternative of course is a yearly upgrade to the newest Maple version.

@Thomas Richard 

 

Yes, that is the one. The virtual machine is not limited to XP - you can load all the other old Windows operating systems if you are so inclined - I haven't tried.

Your recommendation makes no sense - you state that Maple is slower on virtual XP. That may be true but "slower" is still faster than "not at all".

By the way - I always use the classic worksheet anyway (I have troglodytic tendancies), and have noticed no speed issues between what runs on XP and what runs on Win 7 (when it does run).

Its probably the fancy front end that runs slowly on XP and notatall on Win 7 (just a guess).

The alternative of course is a yearly upgrade to the newest Maple version.

Well, that is a very nice bit of semantics. I think it would be helpful if some authority from MapleSoft could intercede at this point and present the developer's definition - when is a bug not a bug? Personally, I find it very frustrating (and I suspect that Alex does too) to continually run into "ordering problems"? Whose "problems"? And why do we never see such characteristics discussed in the Maple advertising or reviews? Why do reviewers not discuss "weaknesses" in this sense? Do "ordering problems" exist in other symbolic codes? Is there a literature on the subject?

Well, that is a very nice bit of semantics. I think it would be helpful if some authority from MapleSoft could intercede at this point and present the developer's definition - when is a bug not a bug? Personally, I find it very frustrating (and I suspect that Alex does too) to continually run into "ordering problems"? Whose "problems"? And why do we never see such characteristics discussed in the Maple advertising or reviews? Why do reviewers not discuss "weaknesses" in this sense? Do "ordering problems" exist in other symbolic codes? Is there a literature on the subject?