String comparison function in C

Question

#include <stdio.h>
#include <string.h>

int in_mot(char *str) //check whether input from text file in mot table
{
    FILE *fp_mot;
    int i = 0, j = 0, r;
    char check[300], first_field[7];

    fp_mot = fopen("mot.txt", "r");

    while (1)
    {
        if (fgets(check, sizeof(check), fp_mot) == NULL)
        {
            printf("EOF");
            return 0;
        }
        else
        {
            i = 0;
            for (i = 0; i < 7; i++)
                first_field[i] = 0;

            i = 0;
            while (check[i] != ' ')
            {
                first_field[i] = check[i];
                i++;
            }
            first_field[i] = '\0';

            if (strcmp((char*)str, (char*)first_field) == 0)
            {
                puts(first_field);
                return 1;
            }
        }
    }
    if (j == 18)
        return 0;
}

int main()
{
    char temp[30], wc[10], str[4][10];
    int i = 0, j = 0, k = 0, z;
    FILE *fp;

    int LC = 0, RC = 0, STP = 1;

    fp = fopen("ip.txt", "r");

    while (1)
    {   //1-break a line and store in array of strings(str)
        if (fgets(temp, 30, fp) == NULL)
            break;
        else
        {
            j = 0; i = 0;
            while (j < 3)
            {
                k = 0;
                for (z = 0; z < 10; z++)
                    wc[z] = 0;
                while (temp[i] != ' ' && temp[i] != ',')
                {
                        wc[k] = temp[i];
                        i++;
                        k++;
                }
                i++;
                wc[k] = '\0';
                strcpy(str[j], wc);
                j++;
            }
        }

        //str[0][3] = '\0';
        if (in_mot(str[0]))
        {
            //printf("\n yesssssssss");
        }
    }

    return 0;    
}

str and first_field are my two strings. By puts(), when both the strings are equal they print same output but strcmp() returns non zero value? Aren't str and first_field being considered as strings? When I tried to print their lengths by strlen, it didn't show any output.

Try printing their strlen, it will likely be different due to non-printable characters — StoryTeller - Unslander Monica
– StoryTeller - Unslander Monica, Commented Feb 21, 2017 at 6:24
@ash Please, provide a complete test program with/and your test data. Otherwise, your observation is not reproducable. — Scheff's Cat
– Scheff's Cat, Commented Feb 21, 2017 at 6:25
@ash Just a guess: strcmp() was intended to be used with ASCII. If applied to UTF-8, it will work for identical encoded strings. However, in Unicode/UTF-8 are in certain cases multiple way to express the same glyph e.g. 'ü' can be code point 252 or a 'u' and a diaresis (the upper double points). In this case, it looks equal but strcmp()will consider this as different. — Scheff's Cat
– Scheff's Cat, Commented Feb 21, 2017 at 6:33
@ash end yet you cast to char*, which is pointless since arrays decay to pointers without any cast. It's just not idiomatic to stick a cast there. — StoryTeller - Unslander Monica
– StoryTeller - Unslander Monica, Commented Feb 21, 2017 at 6:39

Chanhee Jo · Accepted Answer · 2017-02-21 06:38:18Z

2

Try to print each character code, not the characters only visible(printable). Then you would find out what is different between two strings.

Like,

char* p = str;
while (*p)
{
    printf("%02x ", p++);
}

answered Feb 21, 2017 at 6:38

Chanhee Jo

3091 silver badge6 bronze badges

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

Comments

VolAnd · Accepted Answer · 2017-02-21 07:31:10Z

1

Perhaps, strings are different in some non-printable characters, e.g. \n or \r or (space) that you cannot see at the screen.

E.g.:

char first_field[] = "Test\r";
char str[] = "Test\n";
puts(first_field);
puts(str);
if ( !strcmp(str, first_field) )
{
    puts(first_field);
    return 1;
}

will give the same output but strings are really different.

UPDATE:

Possible ways to check:

// check pointers
if (str != NULL && first_field != NULL)
{
    // check length
    if (strlen(str) == strlen(first_field))
    {
        // check content
        if (!strcmp(str, first_field))
        {
            //do something for equal strings
            return 1;
        }
    }
}

As an option, you can write your own comparison, that skips some characters when compare strings, or delete all non-printable characters at the end of strings before using strcmp, e.g.:

void removeEmptyEnd(char * str)
// removes insignificant endings (including space with code 32)
{
    if (!str)
    {
        return;
    }
    // Start from the end
    int i = strlen(str);
    // and skip space and other non-printable chars
    while (i >= 0 && str[i] <= 32)
    {
        i--; // move to next char
    }
    // put new null-termitator to the end of string
    str[i+1] = '\0';
}

UPDATE 2:

Also check file-pointers after opening:

fp=fopen("ip.txt","r");
if( fp )
{
     // use file - read data
}

edited Feb 21, 2017 at 7:31

answered Feb 21, 2017 at 6:35

VolAnd

6,4173 gold badges29 silver badges44 bronze badges

4 Comments

ash Over a year ago

any way to check this?

Scheff's Cat Over a year ago

@ash As StoryTeller wrote: you could check the length additionally. strlen() will count \r and \nas well as any character != 0.

ash Over a year ago

yes they are showing different length due to escape characters maybe,but they contain same characters(as in alphabets) .How do i find out the extra escape sequence character?

VolAnd Over a year ago

@ash I have added my solution - you can call removeEmptyEnd(first_field) to ensure that here is no ' ', '\n' and other in the end of string

chqrlie · Accepted Answer · 2017-02-21 07:54:11Z

Your parser does not verify the lengths of the destination arrays. If ip.txt contains words (separated by a single space or comma) longer than 9 characters or if the first word on any line in file mot.txt is longer than 6 characters, you would have a buffer overflow, causing undefined behavior.

You also have undefined behavior if the lines in ip.txt do not contain at least 3 separators and if the lines in mot.txt do not all contain a space.

Here is a version that checks and reports buffer overflows:

#include <stdio.h>
#include <string.h>

int copy_word(char *dest, int size, const char *src, int len,
              const char *filename, int lineno) {
    *dest = '\0';
    if (len >= size) {
        fprintf(stderr, "%s:%d: word longer than %d characters: %.*s\n",
                filename, lineno, size - 1, len, src);
        return -1;
    }
    strncat(dest, src, len);
    return len;
}

int in_mot(const char *str) { //check whether input from text file in mot table
    FILE *fp_mot;
    char check[300], first_field[7];
    int i, lineno = 0;

    fp_mot = fopen("mot.txt", "r");
    if (fp_mot == NULL) {
        perror("cannot open file mot.txt");
        return 0;
    }

    while (fgets(check, sizeof(check), fp_mot)) {
        lineno++;
        i = strcspn(check, " \n");
        copy_word(first_field, sizeof first_field, check, i, "mot.txt", lineno);
        if (strcmp(str, first_field) == 0) {
            puts(first_field);
            return 1;
        }
    }
    return 0;
}

int main(void) {
    char temp[30], wc[10], str[4][10];
    int i, j, k;
    int lineno = 0;
    FILE *fp;

    //int LC = 0, RC = 0, STP = 1;

    fp = fopen("ip.txt", "r");
    if (fp == NULL) {
        perror("cannot open file ip.txt");
        return 1;
    }

    while (fgets(temp, sizeof(temp), fp)) {
        lineno++;
        for (j = i = 0; j < 3; j++) {
            k = strcspn(temp + i, " ,\n");
            copy_word(wc, sizeof(wc), temp + i, k, "ip.txt", lineno);
            strcpy(str[j], wc);
            i += k;
            if (temp[i] != '\0')
                i++;
        }
        if (in_mot(str[0])) {
            printf("yes!\n");
        }
    }
    return 0;
}

Collectives™ on Stack Overflow

String comparison function in C

3 Answers 3

Comments

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related