Embedding text file in c++ code

Question

I have the following text file: g.dat

16
0.0950125098376374401853193354250
0.281603550779258913230460501460
0.458016777657227386342419442983
0.617876244402643748446671764049
0.755404408355003033895101194847
0.865631202387831743880467897713
0.944575023073232576077988415535
0.989400934991649932596154173450

0.189450610455068496285396723209
0.182603415044923588866763667977
0.169156519395002538189312079058
0.149595988816576732081501730116
0.124628971255533872052476277863
0.0951585116824927848099251053810
0.0622535239386478928628438391746
0.0271524594117540948517805723700

And the following C++ code to ready it:

ifstream In;
  In.open(("g.dat").c_str());
  In>>gaussdim;
  gt = new double[gaussdim];
  gc = new double[gaussdim];
  for(int i=0;i<gaussdim/2;i++)
  {
    In>>gt[i];
    gt[gaussdim-i-1]=-gt[i];
  }
  for(int i=0;i<gaussdim/2;i++)
  {
    In>>gc[i];
    gc[gaussdim-i-1]=gc[i];
  }
  In.close();

I would like to embed this data file into my program so that I can easily redistribute it and not be dependent on always taking care of many different files. I am targeting linux and mac so xxd would be a possibility for me to convert the data to a big char. I would need help with the next step, i.e. how to convert this char into a stream or whatever better solution there is. What do you recommend?

Edit: a lot of good answers, the chosen one has for me the advantage that the change to the code is minimal and is general enough that I can use it in other similar parts of the code where several input files with different structure are parsed.

Sorry If I couldn't got your question, but from what I understand you need this data accessible through out the life of your application, without having to read the data file.Right?In that case, what you are doing is perfectly fine. You read data one time and use it endlessly. Or you can go by the way below answer suggest. — mawia
– mawia, Commented May 24, 2012 at 0:31
Actually I want to get rid of the data file, insert it somehow as text in the code, and then change the code so that 'gc' and 'gt' finally stay with the same values as in the original code — elyase
– elyase, Commented May 24, 2012 at 0:41
But I didn't got 'then change the code', is it that your code is overwriting its own behavior that is to say it is spitting code at run time which is replacing the orignal binary.As you may have guessed, I'm puzzled at the moment by your need.Apology if you mean something simpler than this, but plz do explain.. — mawia
– mawia, Commented May 24, 2012 at 1:02

abarnert · Accepted Answer · 2012-05-24 01:33:19Z

4

I'm guessing that you don't actually want to embed a text file in your code, so the answers by Blastfurnace and Kyle C for much more reasonable solutions to your problem.

But if you really do actually want to embed the text file in your code and then read it in as a stream, the easiest way to do that is something like this.

First, embed the text file as a string:

static const char *gdat="16\n"
"0.0950125098376374401853193354250\n"
"0.281603550779258913230460501460\n"
"0.458016777657227386342419442983\n"
"0.617876244402643748446671764049\n"
"0.755404408355003033895101194847\n"
"0.865631202387831743880467897713\n"
"0.944575023073232576077988415535\n"
"0.989400934991649932596154173450\n"
"\n"
"0.189450610455068496285396723209\n"
"0.182603415044923588866763667977\n"
"0.169156519395002538189312079058\n"
"0.149595988816576732081501730116\n"
"0.124628971255533872052476277863\n"
"0.0951585116824927848099251053810\n"
"0.0622535239386478928628438391746\n"
"0.0271524594117540948517805723700\n";

Then, instead of using ifstream("g.dat") you can use stringstream(gdat), and get a stream which is, for your purposes, effectively identical.

answered May 24, 2012 at 1:33

abarnert

368k54 gold badges626 silver badges691 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

abarnert Over a year ago

If you're going to do this in multiple places, possibly with larger files, you really want to look into writing a simple code generator to do the grunt work, as described in Basile Starynkevitch's answer.

Art Swri · Accepted Answer · 2012-05-23 23:51:29Z

2

Put it into an array of double that you initialize. The general form is

double my_values[16] = { 1.234, 2.345, ... etc .. };

answered May 23, 2012 at 23:51

Art Swri

2,8243 gold badges29 silver badges37 bronze badges

Comments

Kyle C · Accepted Answer · 2012-05-23 23:55:31Z

2

const double data[] = {
    0.0950125098376374401853193354250,
    0.281603550779258913230460501460 ,
    0.458016777657227386342419442983 ,
    0.617876244402643748446671764049 ,
    0.755404408355003033895101194847 ,
    0.865631202387831743880467897713 ,
    0.944575023073232576077988415535 ,
    0.989400934991649932596154173450 ,                                       
    0.189450610455068496285396723209 ,
    0.182603415044923588866763667977 ,
    0.169156519395002538189312079058 ,
    0.149595988816576732081501730116 ,
    0.124628971255533872052476277863 ,
    0.0951585116824927848099251053810,
    0.0622535239386478928628438391746,
    0.0271524594117540948517805723700
};

answered May 23, 2012 at 23:55

Kyle C

1,63712 silver badges25 bronze badges

Comments

Basile Starynkevitch · Accepted Answer · 2012-05-24 05:22:58Z

2

Use (or write) some program, perhaps a small awk or python or ocaml script, to transform your g.dat file into a C file.

Modify your build procedure (e.g. your Makefile) to add a dependency for that generated C file from your original g.dat file.

Compile that generated C file, and link it into your binary.

answered May 24, 2012 at 5:22

Basile Starynkevitch

231k18 gold badges323 silver badges578 bronze badges

1 Comment

abarnert Over a year ago

If I had to do something like this, and g.dat was more than a few lines long, this is exactly what I'd do. The question is which of the above answers to have your code generator output—but honestly, it doesn't really matter that much. You could generate a single big array of numbers (like Kyle C's answer) with a trivial sed one-liner, while pre-processing it into the final results (like Blastfurnace's answer) would be more like a few lines of Python, but they'd all be pretty easy.

Blastfurnace · Accepted Answer · 2012-05-24 00:51:46Z

This will define and initialize the gt and gc arrays similar to your current code. Note that these aren't dynamically allocated so you don't need to delete[] them when you're done.

double gt[] =
{
     0.0950125098376374401853193354250, 0.281603550779258913230460501460,
     0.458016777657227386342419442983,  0.617876244402643748446671764049,
     0.755404408355003033895101194847,  0.865631202387831743880467897713,
     0.944575023073232576077988415535,  0.989400934991649932596154173450,
    -0.989400934991649932596154173450, -0.944575023073232576077988415535,
    -0.865631202387831743880467897713, -0.755404408355003033895101194847,
    -0.617876244402643748446671764049, -0.458016777657227386342419442983,
    -0.281603550779258913230460501460, -0.0950125098376374401853193354250
};

double gc[] =
{
    0.189450610455068496285396723209,  0.182603415044923588866763667977,
    0.169156519395002538189312079058,  0.149595988816576732081501730116,
    0.124628971255533872052476277863,  0.0951585116824927848099251053810,
    0.0622535239386478928628438391746, 0.0271524594117540948517805723700,
    0.0271524594117540948517805723700, 0.0622535239386478928628438391746,
    0.0951585116824927848099251053810, 0.124628971255533872052476277863,
    0.149595988816576732081501730116,  0.169156519395002538189312079058,
    0.182603415044923588866763667977,  0.189450610455068496285396723209
};

Collectives™ on Stack Overflow

Embedding text file in c++ code

5 Answers 5

1 Comment

Comments

Comments

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

5 Answers 5

1 Comment

Comments

Comments

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related