C ++时间库和Octave .oct文件

Question

我正在尝试编写一个使用linasm-1.13 library的Octave C ++ .oct函数，但我似乎无法从/ usr / share / zoneinfo /中获得tzdata的基本加载工作。到目前为止我的简单测试功能是

#include <octave/oct.h>
#include <Time.h> // the linasm-1.13 library

DEFUN_DLD ( tz, args, nargout,
"-*- texinfo -*-\n\
@deftypefn {Function File} {} tz (@var{YYYYMMDDHHMMSS})\n\
\n\
@end deftypefn" )

{
octave_value_list retval_list ;
unsigned int tz ;

const char *ny_time = "/usr/share/zoneinfo/America/New_York" ; 

tz = Time::LoadTimeZone( ny_time ) ;

return retval_list ;

在使用mkoctfile进行编译时，会出现此错误

>> mkoctfile tz.cc
tz.cc: In function ‘octave_value_list Ftz(const octave_value_list&, int)’:
tz.cc:24:34: error: cannot call member function ‘unsigned int  Time::LoadTimeZone(const char*)’ without object
tz = Time::LoadTimeZone( ny_time ) ;
                              ^
warning: mkoctfile: building exited with failure status

我对此的理解是ny_time不是一个被识别的对象，但我已经尝试将ny_time作为字符串文字投射，如this accepted SO answer中详述的那样。

我这样做是因为根据linasm page输入的LoadTimeZone应该是“tzfile的路径，它描述了所需的时区”。我哪里错了？

Answer 1

我认为你还必须使用#include "source.cc"文件，而不仅仅是#include "header.h"文件。在你的情况下，我想你应该添加：#include "Time.cc"或类似的东西。我不知道为什么但是这对我来说在使用Rafat's Hussain wavemin library时起作用，但我只有4个文件，它必须是非常繁琐的大量文件。

这就是我所做的（它是Rafat和他的库提供的测试代码的修改版本）。

#include "wavemin.h"
#include "waveaux.h"
#include "wavemin.cc"
#include "waveaux.cc"
#include <octave/oct.h>

double ensayo();
double absmax(double *array, int N);

DEFUN_DLD(helloctave2, argv, , "Usage: hello()"){   

    wave_object obj;
    wt_object wt;
    double *inp, *out, *diff;
    int N, i, J;
    char *name = "db4";
    obj = wave_init(name);// Initialize the wavelet
    N = 14;  //Length of Signal
    inp = (double*)malloc(sizeof(double)* N); //Input signal
    out = (double*)malloc(sizeof(double)* N);
    diff = (double*)malloc(sizeof(double)* N);
    //wmean = mean(temp, N);
    for (i = 0; i < N; ++i) {
        inp[i] = i;
    }
    J = 1; //Decomposition Levels
    wt = wt_init(obj, "dwt", N, J);// Initialize the wavelet transform object
    setDWTExtension(wt, "sym");// Options are "per" and "sym". Symmetric is the default option
    setWTConv(wt, "direct");
    dwt(wt, inp);// Perform DWT
    //DWT output can be accessed using wt->output vector. Use wt_summary to find out how to extract appx and detail coefficients
    for (i = 0; i < wt->outlength; ++i) {
    octave_stdout << wt->output[i];
    octave_stdout << "\n";
    }

    idwt(wt, out);// Perform IDWT (if needed)
    // Test Reconstruction
    for (i = 0; i < wt->siglength; ++i) {
        diff[i] = out[i] - inp[i];
    }

  octave_stdout << absmax(diff, wt->siglength);
  octave_stdout << "\n"; 

  octave_value_list retval;             

  return retval;                            
}       

double
absmax(double *array, int N) {
    double max;
    int i;
    max = 0.0;
    for (i = 0; i < N; ++i) {
        if (fabs(array[i]) >= max) {
            max = fabs(array[i]);
        }
    }
    return max;
}