Gateway mexFunction - Matrix equal to zero

Goodmorning. I have some problem on writing and using mexFunction. I have a fortran code where I read some input file. I write the mexFunction for this code and so I can read all this file. The fortran code is right and It can read the files and in matlab I can call the fortran function (by building the mexw32 file). My problem is that the matlab matrix generated by this function have the values all equal to zero. If you have any suggestion...thanks a lot. Stefano

Here It is my code:

#include <fintrf.h>

        SUBROUTINE MEXFUNCTION(NLHS, PLHS, NRHS, PRHS)
  C-----------------------------------------------------------------------
  C (pointer) Replace integer by integer*8 on 64-bit platforms
  C
        MWPOINTER PLHS(*), PRHS(*)

C-----------------------------------------------------------------------
C

        INTEGER NLHS, NRHS
  C
  C-----------------------------------------------------------------------
  C (pointer) Replace integer by integer*8 on 64-bit platforms
  C
        MWPOINTER MXCREATEDOUBLEMATRIX, MXGETPR

C-----------------------------------------------------------------------
C

        MWSIZE MXGETM, MXGETN
  C
  C KEEP THE ABOVE SUBROUTINE, ARGUMENT, AND FUNCTION DECLARATIONS FOR USE
  C IN ALL YOUR FORTRAN MEX FILES.
  C---------------------------------------------------------------------
  C
  C-----------------------------------------------------------------------
  C (pointer) Replace integer by integer*8 on 64-bit platforms
  C

MWPOINTER ncolture1P, VECthetaFCP, VECthetaWPP, VECREWP,
     $ phiP, zetaP, frazP, nrsbP

C-----------------------------------------------------------------------
C

        MWSIZE M, N, NEL
  REAL*8 Rncolture1P, RVECthetaFCP(625), RVECthetaWPP(625),
       $ RVECREWP(625), RphiP(625), RzetaP(625), RfrazP(625,8),
       $ RnrsbP
  #if defined MSWIND
  C For Windows only!
  C This resets the floating point exception to allow divide by zero,
  C overflow and invalid numbers.
  C
  INTEGER(2) CONTROL
  CALL GETCONTROLFPQQ(CONTROL)
  CONTROL = CONTROL .OR. FPCW$ZERODIVIDE
        CONTROL = CONTROL .OR. FPCW$INVALID
        CONTROL = CONTROL .OR. FPCW$OVERFLOW
  CALL SETCONTROLFPQQ(CONTROL)
  #endif
  C
  C CHECK FOR PROPER NUMBER OF ARGUMENTS
  C
        IF (NRHS .NE. 1) THEN
          CALL MEXERRMSGTXT('YPRIME requires one input arguments')
        ELSEIF (NLHS .GT. 7) THEN
          CALL MEXERRMSGTXT('YPRIME requires seven output argument')
        ENDIF

C
C CREATE A MATRIX FOR RETURN ARGUMENT
C
      PLHS(1) = MXCREATEDOUBLEMATRIX(1,1,0)
PLHS(2) = MXCREATEDOUBLEMATRIX(625,1,0)
PLHS(3) = MXCREATEDOUBLEMATRIX(625,1,0)
PLHS(4) = MXCREATEDOUBLEMATRIX(625,1,0)
PLHS(5) = MXCREATEDOUBLEMATRIX(625,1,0)
PLHS(6) = MXCREATEDOUBLEMATRIX(625,1,0)
PLHS(7) = MXCREATEDOUBLEMATRIX(625,8,0)

C
C ASSIGN POINTERS TO THE VARIOUS PARAMETERS
C
      ncolture1P = MXGETPR(PLHS(1)) ! LEFT HAND -> PARAMETRI DI OUTPUT
VECthetaFCP = MXGETPR(PLHS(2))
VECthetaWPP = MXGETPR(PLHS(3))
VECREWP = MXGETPR(PLHS(4))
phiP = MXGETPR(PLHS(5))
zetaP = MXGETPR(PLHS(6))
frazP = MXGETPR(PLHS(7))

nrsbP = MXGETPR(PRHS(1)) ! RIGHT HAND -> PARAMETRO DI INPUT

C
C COPY RIGHT HAND ARGUMENTS TO LOCAL ARRAYS OR VARIABLES
      NEL = 1
      CALL MXCOPYPTRTOREAL8(nrsbP, RnrsbP, NEL)

C
C DO THE ACTUAL COMPUTATIONS IN A SUBROUTINE
C CREATED ARRAYS.
C
CALL leggiETparam(Rncolture1P, RVECthetaFCP, RVECthetaWPP,
     $ RVECREWP, RphiP, RzetaP, RfrazP, RnrsbP)

C
C COPY OUTPUT WHICH IS STORED IN LOCAL ARRAY TO MATRIX OUTPUT
      NEL = 1
      CALL MXCOPYREAL8TOPTR(Rncolture1P, ncolture1P, NEL)
      NEL = 625
      CALL MXCOPYREAL8TOPTR(RVECthetaFCP, VECthetaFCP, NEL)
      NEL = 625
      CALL MXCOPYREAL8TOPTR(RVECthetaWPP, VECthetaWPP, NEL)
      NEL = 625
      CALL MXCOPYREAL8TOPTR(RVECREWP, VECREWP, NEL)
      NEL = 625
      CALL MXCOPYREAL8TOPTR(RphiP, phiP, NEL)
      NEL = 625
      CALL MXCOPYREAL8TOPTR(RzetaP, zetaP, NEL)
      NEL = 625
      CALL MXCOPYREAL8TOPTR(RfrazP, frazP, NEL)

C
      RETURN
      END

[Stefano Basso]

The line of the code is right in my code. I don't know why here they aren't new line when they must be

[Laurens Bakker]

Hi Stefano,

Could you please use the 'code' button at the top of the edit screen to re-format your code as actual code? That will help us understand your question.

Cheers,

Laurens

[James Tursa]

How do you know the file is being read properly? Could you put in some mexPrintf statements right after the leggiETparam call and print out the results directly to make sure they are not all zero?

[James Tursa]

P.S. Why isn't your last NEL = 625 * 8 for RfrazP?

[webdesign]

Hi Stefano,

It looks like you're encountering an issue where the matrices generated by your Fortran function in MATLAB are all zeros. Here are a few suggestions that might help you diagnose and solve the problem:

• Check Data Transfer: Make sure that the data is being correctly transferred from your Fortran code to MATLAB. Specifically, check the calls to MXCOPYREAL8TOPTR and the number of elements (NEL) being transferred.
• Variable Initialization: Ensure all variables in your Fortran code are properly initialized before attempting to pass them to MATLAB. Uninitialized variables could lead to undefined behavior.
• Error Handling: Check if there are error handling mechanisms in your Fortran code that might catch any errors that could occur. It's possible that an error is occurring before the data is successfully passed to MATLAB, but without proper error handling, this might go unnoticed.

If nothing helps, feel free to write me again :).