Member Avatar for perlsu

I have a problem in converting char to integer. My code is as follow:

char lang [2][4] ={"en","fr"};

char ch = lang[0][0];
int i = sprintf("%d",ch); //want to get integer value of 'e' in string "en"

ch = lang[0][1];
int j = sprintf("%d",ch);

The error is "sprintf cannot accept parameter 2 from const char*to char".

pls help me.
Thanks.

Rgds,
su

  • 3 Contributors
  • 2 Replies
  • 854 Views
  • 8 Hours Discussion Span
  • Latest Post Latest Post by Narue

Dani AI

Generated

: the character already carries a numeric code; the task is to retrieve that code (or the numeric value a digit represents) and, if needed, format it into a string. is correct about "value vs representation" — a char has an integer value, and printing it with %d shows that integer. 's atoi suggestion is inappropriate here: atoi expects a NUL-terminated string and is not the right tool for a single char.

Key points and safe examples:

  • To get the byte/code of a character and avoid sign-extension on platforms where char is signed, convert through unsigned char and then to int:

    unsigned char uc = some_byte;
int code = uc;

  • To turn that integer into a textual decimal representation, write into a buffer with snprintf (not by calling sprintf incorrectly). snprintf returns the number of characters written:

    char buf[16];
int written = snprintf(buf, sizeof buf, "%d", code);
/* buf now holds the decimal text, e.g. "101" for 'e' */

  • If the goal is the numeric value of a digit character (for example '7' -> 7), subtract the character '0' after validating the range:

    if (c >= '0' && c <= '9') {
  int digit = c - '0';
}

Cautions:

  • sprintf's first argument must be the destination buffer; calling it with the format string as the first parameter is incorrect and will produce compile errors. sprintf/snprintf return the number of characters written, not the numeric value of the formatted data.
  • Do not pass a raw char to atoi; that causes undefined behavior. For multi-byte encodings (UTF-8), a single char may not represent a complete code point — use proper wide/Unicode handling if needed.

Checklist: decide whether the need is "byte code" (use unsigned cast) or "digit value" (use - '0'), then format safely with snprintf if a string is required.

Recommended Answers

All 2 Replies

Member Avatar for server_crash

I think it's:

int x = atoi(char);

Probably better ways, but that's how I do it.

Member Avatar for Narue

Getting the numeric value of a single character is simple:

char lang[2][4] = {"en","fr"};
int i = lang[0][0];

You're confused with the distinction between value and representation. The values of lang[0][0] and i are identical. The only difference is how they're printed, much like octal, decimal, and hexadecimal. The values are the same, but the representation is different. That's why you can do this:

printf ( "%c\n", i ); /* Print i as a character */
printf ( "%d\n", i ); /* Print i as an integer */

>int i = sprintf("%d",ch);
This is not how sprintf works. Until you have a stronger foundation in the idioms of C, you should refrain from guessing and use a good reference.

>int x = atoi(char);
That's broken, and you should know better. atoi takes a null terminated string. If you pass it a single character that doesn't have the value of 0, you invoke undefined behavior. Not to mention that atoi will fail and return 0 if the character is non-numeric, which in my world, 'e' most certainly is.

Here's a nice little guideline for both of you: If you don't know, go find out. In C and C++ there's no margin for error, so a guess is just asking for trouble.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.