Jump to content

  • Log In with Google      Sign In   
  • Create Account

We're offering banner ads on our site from just $5!

1. Details HERE. 2. GDNet+ Subscriptions HERE. 3. Ad upload HERE.


read double from, inconsistent results with some values


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
13 replies to this topic

#1 Alessio1989   Members   -  Reputation: 2126

Like
0Likes
Like

Posted 08 November 2013 - 04:18 PM

I was just playing with visual studio 2013 rtm and compile some old demos to see if there were some changes to the standard library (like "max()" and "min()" functions moved to <algorithm>....)

 

And I had a little surprise with a sample code for write and read binary files with the standard library: using the following double values (111.111, 222.222, 333.333, 444.444), when I read back the data I have completely different data (like  -6.27744e+066 -6.27744e+066 -6.27744e+066 -6.27744e+066). It is also strange that the values are the same in debug mode in 32 bit and 64 bit configuration, but they change in release mode in 64 bit and 32 bit configurations. Changing the float precision model ( "precise" 80 bit, strict and fast) doesn't solve the problem.

 

Note also the sample works well if I change the double data to other values... Maybe that is due to some "nice" behaviour of the IEEE standard? ... I don't remember nothing about that in the floating point model, to me that seems just a bug (code or compiler... dunno XD )

 

Here is the code:



#include <fstream>
#include <iostream>

int main()
{
	double a[ 4 ] = { 111.111, 222.222, 333.333, 444.444 };
	double* i = new( double[ 4 ] );

	std::fstream file( "file", std::ios::out | std::ios::binary );

	file.write( ( char* )&a, sizeof( a ) );
	file.close();


	file.open( "file", std::ios::in | std::ios::binary );
	file.read( ( char* )&*i, sizeof ( double )* 4 );
	file.close( );

	for( int j = 0; j < 4; ++j )
	{
		std::cout << i[ j ] << " ";
	}
	std::cout << std::endl;

	return( 0 );
}

Edited by Alessio1989, 08 November 2013 - 06:31 PM.

"Software does not run in a magical fairy aether powered by the fevered dreams of CS PhDs"


Sponsor:

#2 Zaoshi Kaba   Crossbones+   -  Reputation: 4566

Like
0Likes
Like

Posted 08 November 2013 - 04:31 PM

Your std::fstream::open() calls are incorrect. You probably meant OR operator instead of comma.



#3 Alessio1989   Members   -  Reputation: 2126

Like
0Likes
Like

Posted 08 November 2013 - 05:12 PM

yes, that's a copy-typo, but that's not the problem. anyway thanky for make me note that.


Edited by Alessio1989, 08 November 2013 - 05:13 PM.

"Software does not run in a magical fairy aether powered by the fevered dreams of CS PhDs"


#4 Paradigm Shifter   Crossbones+   -  Reputation: 5432

Like
0Likes
Like

Posted 08 November 2013 - 05:26 PM

    file.write( ( char* )&a, sizeof( a ) );

 

This is incorrect. You want:

 

(char*)a

 

a is already a pointer so taking the address is bobbins.

 

EDIT: I'm surprised (char*)&a even compiles?


Edited by Paradigm Shifter, 08 November 2013 - 05:29 PM.

"Most people think, great God will come from the sky, take away everything, and make everybody feel high" - Bob Marley

#5 Alessio1989   Members   -  Reputation: 2126

Like
0Likes
Like

Posted 08 November 2013 - 05:43 PM

    file.write( ( char* )&a, sizeof( a ) );

 

This is incorrect. You want:

 

(char*)a

 

a is already a pointer so taking the address is bobbins.

 

EDIT: I'm surprised (char*)&a even compiles?

 

here the reference is redundant since they are arrays, but that's not the problem. only with that values I got incorrect read-back from the file, with any other value, the read is correct.


Edited by Alessio1989, 08 November 2013 - 05:45 PM.

"Software does not run in a magical fairy aether powered by the fevered dreams of CS PhDs"


#6 Brother Bob   Moderators   -  Reputation: 8570

Like
0Likes
Like

Posted 08 November 2013 - 05:44 PM

    file.write( ( char* )&a, sizeof( a ) );

 

This is incorrect. You want:

 

(char*)a

 

a is already a pointer so taking the address is bobbins.

 

EDIT: I'm surprised (char*)&a even compiles?

No, a is an array, not a pointer, so &a is a pointer to the array. This is the same as just a in this context when the array decays into a pointer to its first element, but &a is not wrong.


Edited by Brother Bob, 08 November 2013 - 05:44 PM.


#7 Brother Bob   Moderators   -  Reputation: 8570

Like
2Likes
Like

Posted 08 November 2013 - 05:45 PM

Alessio, you still have the third incorrect parameter in your call to open(). You need to OR the last two parameters there as well.



#8 Paradigm Shifter   Crossbones+   -  Reputation: 5432

Like
1Likes
Like

Posted 08 November 2013 - 05:55 PM

Well that's silly. a, &a and &a[0] are all the same? &a is not like the others and should really be a compiler error, if C was a sensible language (which it isn't, of course). a and &a[0] are of course the same. &a should be of type char** (since it is the address of a decayed array).

 

I guess I don't know the correct answer since I'd never use the address of an array. Taking the address of an array suggests to me it could be changed!?


"Most people think, great God will come from the sky, take away everything, and make everybody feel high" - Bob Marley

#9 Brother Bob   Moderators   -  Reputation: 8570

Like
3Likes
Like

Posted 08 November 2013 - 05:59 PM

They all have the same value, but they all have different types.

  1. a is of type double[4].
  2. &a is of type double (*)[4].
  3. &a[0] is of type double *.

And no, &a should not have the type double ** since a is not a pointer. See point 2, a is of type double[4], so &a is a pointer to a double[4], thus a double (*)[4].



#10 swiftcoder   Senior Moderators   -  Reputation: 10361

Like
1Likes
Like

Posted 08 November 2013 - 06:18 PM


-6.27744e+066

That looks a lot like accidental/uninitialised memory reinterpreted as a double - the upper bytes are 0xCDCDCDCE...

 

Are you sure you are running the exact same program you have reproduced here? Apart from the missing bitwise-OR already mentioned, there isn't anything particularly wrong with your code snippet, and a compiler bug is beyond unlikely.


Tristam MacDonald - Software Engineer @Amazon - [swiftcoding]


#11 Alessio1989   Members   -  Reputation: 2126

Like
0Likes
Like

Posted 08 November 2013 - 06:34 PM

Alessio, you still have the third incorrect parameter in your call to open(). You need to OR the last two parameters there as well.

 

thank you. now it works with that values too.

 

 

 

 


-6.27744e+066

That looks a lot like accidental/uninitialised memory reinterpreted as a double - the upper bytes are 0xCDCDCDCE...

 

Are you sure you are running the exact same program you have reproduced here? Apart from the missing bitwise-OR already mentioned, there isn't anything particularly wrong with your code snippet, and a compiler bug is beyond unlikely.

it was the comma in the input-mode of open()... anyway it is so strange, the comma don't brake me the read-back if I use other floating values.. Probably since ios openmodes are macros, some "funny" things happened...


Edited by Alessio1989, 08 November 2013 - 06:43 PM.

"Software does not run in a magical fairy aether powered by the fevered dreams of CS PhDs"


#12 swiftcoder   Senior Moderators   -  Reputation: 10361

Like
0Likes
Like

Posted 08 November 2013 - 06:51 PM


it was the comma in the input-mode of open()... anyway it is so strange, the comma don't brake me the read-back if I use other floating values.. Probably since ios openmodes are macros, some "funny" things happened...

Weird. That just results in a compile error on Clang.


Tristam MacDonald - Software Engineer @Amazon - [swiftcoding]


#13 Alessio1989   Members   -  Reputation: 2126

Like
0Likes
Like

Posted 08 November 2013 - 06:53 PM

 


it was the comma in the input-mode of open()... anyway it is so strange, the comma don't brake me the read-back if I use other floating values.. Probably since ios openmodes are macros, some "funny" things happened...

Weird. That just results in a compile error on Clang.

 

 

Pure Microsoft magic laugh.png


"Software does not run in a magical fairy aether powered by the fevered dreams of CS PhDs"


#14 Brother Bob   Moderators   -  Reputation: 8570

Like
2Likes
Like

Posted 08 November 2013 - 06:59 PM

The library that ships with MSVC apparently has an extra implementation defined parameter for (I guess) internal use. Since it has a default value, the public interface remains unchanged, but unexpected things happen when you call it the wrong way.






Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS