Hi everybody, one thing is not still clear to me about assume: for example  I have to find the solution of this simple equation:

e1 := q=ks*h^(5/3)*sqrt(If);

If i write solve(e1,h);   I get also all the complex solutions. I want just the real solution so i wrote:

assume(h::real);
solve(e1,h);

but I still have all the complex solutions. Why? I tried also  with assume(h::real,h>=0); but it's the same.

thanks
 

alberto

Hi to everybody. I have to derive in "x" a few times an expression that contains the term signum(Q), where Q =Q(x). The problem is that maple derive also signum and gives as results signum(1,Q) so in my long expression there are a lot of signum(1,Q) . The problem is that when I create a fortran subroutine that compute this expression i get the warning

"the function "signum"  is not recognized in the target language".

So the problem is duplex:

Hi, I have a code that generate a fortran subroutine that calculate the high order derivatives of the unknowns of a partial differential system. it  works well but when the list of local variables is too long (it happen for some kind of complex(but not too much) differential equations) i get the following error from the "codegeneration" command:

"assigning to a long list, please use Arrays"

Hi I used the "search" option of this forum. I have seen old messages about  the continuation of line "#" of fortran77 that is different from the ones in fortran90. I have thousands lines of fortran code created by codegeneration and I can't use "replace all # with &", because # are at the beginning and "&" at the end of the line. It takes a while to substitute all of them by hand every time I generate the fortran subroutine. Do you know if they are going to introduce the continuation of line with & instead #?

thanks

Hi. I have some doubts about how to optimize a fortran very complicated output. I have an array called "stampa" that contain high order derivatives(very long expression). I create a fortran soubroutine for "stampa" with CodeGeneration with the following commands:

codegen[makeproc](stampa,[CS]);
codegen[optimize](%);
codegen[prep2trans](%);
codegen[split](%):
Fortran(%, optimize,defaulttype=numeric,deducetypes=false);

I have the following problems or doubts,: