Garbbage in the variable

Hello everyone... I am having a problem with a couple of variables.

I declare the variable as char jar[100]=" ", tmp[6]=" ", cup[6]=" ";

In the watch window the value for tmp & cup looks like this: " \000\000\000", but not jar.
As a consequence the program is not running properly. Any thoughts about this problem? Let me know them...

I am learning how to program in c++ win32 application on Code::Blocks 10.05 running on win xp... Thanx

The problem is probably somewhere else. Since the three variables are the same in every respect except size, jar may have been displayed differently just to look nice, but it should still say exactly what the value is. What about the rest of the code?

Originally Posted by whiteflags

What about the rest of the code?

Hi Whiteflags,
Well, the garbbage in cup and tmp are transferred in to the bucket, and that messes up the flow of the code.
The goal of the code is to sort the largest string two char at the time by extracting and inserting them with the strncpy() like this...

Code:

.
.
.
if( strcmp( jar, cup ) > 0 ) {
   strncpy( tmp, &bucket[m], 2 ); /* I am aware tmp is not NULL terminated */
   strncpy( &bucket[m], cup, 2 ); /* The gabbage moves in and destroys the string */
   strncpy( &bucket[n], tmp, 2 ); /* and more garbbage is added */
}
.
.
.

Here is the line where the variables are declared locally to the WindProc()...
char bucket[100]="", track1[]="654321", jar[100]=" ", cup[6]={""}, tmp[6]="";

OK, one thing we need to note here.

char *strncpy(char *dest, const char *src, size_t n);

The strncpy() function is similar, except that not more than n bytes of
src are copied. Thus, if(3,n) there is no null byte among the first n bytes
of src, the result will not be null-terminated.

http://swoolley.org/man.cgi/3/strcpy

So for the first strncpy, if there is not a '\0' in bucket[m + 1] or bucket[m + 2] tmp is not terminated.

For the third strncpy, since tmp is not terminated within the first two characters, neither is the string at bucket + n.

So bucket is probably terminated somewhere in the middle of bucket[m] and bucket[n].

I can only say you should use strncpy right, or if you want to attach a (zero terminated!) string to an empty string, you should really be using strncat().

Originally Posted by whiteflags

So bucket is probably terminated somewhere in the middle of bucket[m] and bucket[n].

The bucket is holding the content of track1 three times. So there is no terminating '\0' NULL string in between.
The problem is with cup and tmp having the '\0' in them.

Originally Posted by whiteflags

I can only say you should use strncpy right, or if you want to attach a (zero terminated!) string to
an empty string, you should really be using strncat().

Cool! strncat(), but I need strncpy() at the moment to place characters between an array's content.

I think I may have found a solution... declaring cup and tmp as pointers and using malloc(). I no longer see the garbage
I was seeing in before. Will run it few more times and then on to correct a problem with the sorting loop.

But I am still at a loss as to what was going on in the codes to cause this terrible side effect. Will keep an eye on google
for anything similar to my problem...

Thanx Whiteflags...

Cool! strncat(), but I need strncpy() at the moment to place characters between an array's content.

So that is what you are doing! To do that you're actually supposed to shift elements over to make room. If you don't then what happens is that the string you want to insert will actually overwrite the part you should have moved before.

memmove is just the function for this. Make sure that your destination string is big enough to accommodate this move:

Code:

#include <string.h>
#include <stdio.h>
int main(void)
{
    char bucket[100] = "hello world";
    size_t x = 6;
    char cup[] = "sick, sad ";
    size_t cuplen = strlen(cup);
    size_t afterlen = strlen(bucket + x) + 1; /* x is where you want the cup string to be. */
    memmove(bucket + x + cuplen, bucket + x, afterlen);
    /* move afterlen bytes of stuff from bucket + x to bucket + x + cuplen */
    strncpy(bucket + x, cup, cuplen);
    /* fill in the hole */ 
    printf("'%s'\n", bucket);
    return 0;
}

Just an example, but I think you can apply it in reality. In any case, all of the strings are zero terminated. It's never OK if they aren't really.

Are you using C or C++?

Originally Posted by whiteflags

So that is what you are doing!

I am using two loops to move around the characters. The inner loop does the moving
with this

Code:

if( strcmp( jar, cup ) > 0 ) {
   strncpy( &bucket[m], cup, 2 );
   strncpy( &bucket[n], jar, 2 );
   strcpy( jar, cup );
}

Anyway, will dissect your code to figure out how it works, and increase my knowledge, thanx.

Originally Posted by Elysia

Are you using C or C++?

Learning C++ win32 GUI with Code::Blocks

Then why are you using C?

Originally Posted by Elysia

Then why are you using C?

Um! It looks like I need more help than I thought.
Please point me in the right direction so that I can learn the
correct way of C++ programming for a GUI or Windows... Will appreciate it - a lot!

He probably means you should use the std::string class to do whatever you are doing. It makes things easier.

Code:

std::string bucket("hello world");
bucket.insert(6, "sick, sad ");
std::cout <<'\'' << bucket << '\'' << std::endl;

See? There is a starting tutorial over here: Cprogramming.com - C++ Standard Library - String Class. That said, it wasn't really wrong of you to learn character arrays first. People have different approaches to learning C++. But std::string is basically the preferred type.

Basically there is no point in using char arrays and str(n)cpy et all in C++, because we have better tools that usually work all the time.
There is nothing wrong in knowing or learning them--especially for educational purposes--but a C++ programmer really should be using the C++ standard library and all its contents, or we'll get what is called a "C+" programmer. A programmer which is basically stuck in C with using some C++ features.

What is the point of a language if you don't use it to its fullest extent, as fully possible?

Well, using a language "to its fullest extent" will change meaning depending on for what you use it. For example, if you use C++ to learn about character arrays, but don't use character arrays in order to learn them, then you aren't using the language to its fullest extent.

Originally Posted by whiteflags

There is a starting tutorial over here: Cprogramming.com - C++ Standard Library - String Class.

To delete an entire string, you could use
str.erase(0, str.length());

I like this part... wiping a string...

Code:

using namespace std::string cup;
.
.
.
cup.erase( 0, cup.length() );
.
.
.

Is that correct?
With std::string may help avoid getting garbage into the cup variable... amazing! Awhile ago I could not grasp
the concept behind std::xxxxxxx ... I take it that I am ready now... Do you have any links to sample code of
simple programs designed for windows?

Thank you Whiteflags, and Elysia... looking forward not being a C+ programmer.... hehehe