CODEWITHSUNDEEP
c
Lyle Ayden Fetterly
Lyle Ayden Fetterly

Reputation: 19

Syntax causing Segmentation Fault?

I downloaded a text file, I want to read a line, chop the lead and trail whitespace, and write the updated (new) line into another file. I'm sure there are more eloquent ways of doing, but I'm trying to do it this way:

char *make_dst( char *src, int beg, int end ) {
    char *dst = ( char * )malloc( MAXLEN * sizeof( char ));
    int i = 0;

    if( dst != NULL ) {
        while( beg <= end ) {
            /* this works fine */
            dst[i] = src[beg];
            i++;
            beg++;
            /* this causes the segmentation fault */
            dst[i] = src[i + beg];
            i++;
        }
    dst[i] = '\0';

    return dst;
}

I don't understand why the second manner of doing it causing the error? Can anybody explain this for me? I'm using lubuntu 14.04 - is it an operating system thing? I thought it was fine to use "math" in that manner to reference different indexes in an array?

Upvotes: 1

Views: 91

Answers (2)

user3629249
user3629249

Reputation: 16540

the following code will go the job

char *make_dst( char *src, int beg, int end )
{
    int i = 0;
    char *dst = malloc( MAXLEN);
    if( NULL == dst )
    { // then malloc failed
        perror( "malloc failed" );
        exit( EXIT_FAILURE );
    }

    // implied else, malloc successful

    memset( dst, 0x00, MAXLEN );
    for( i=0, i<(end-beg+1); i++ )
    {
        dst[i] = src[beg++];
    } // end for

    return dst;
} // end function: make_dst

Upvotes: 0

user123
user123

Reputation: 9071

while( beg <= end ) {
    dst[i] = src[beg];
    i++;
    beg++;
}

This is correct because you're advancing both i and beg and assuring that beg <= end.

while( beg <= end ) {
    dst[i] = src[i + beg];
    i++;
}

In this case, you have an infinite loop, because if beg <= end was true initially, it will always be true after N iterations since the value of beg is never modified.

To correct it, the condition must ensure that i + beg <= end (This is assuming you want [begin, end] ranges instead of [begin, end), which is not wrong, per se).

[begin, end) ranges have the advantage of allowing you to specify cardinality conveniently.
[x, x + 10) means you have a range of size 10. The equivalent is [x, x + 10 - 1].

You want the size of the range to be a simple end - begin, and you want to include the lower bound. [begin, end] means you need special handling for empty ranges, and it would incur a lot of noise (-1, +1, you know the kind).

Upvotes: 3

Related Questions