C String strerror() Function

What You’ll Learn

The strerror() function turns a numeric error code into a readable English message. After a failed call like fopen, you read errno and pass it to strerror to explain what went wrong to users or logs.

Definition and Usage

When a C library function fails, it often sets the global variable errno to a positive integer code. strerror(errnum) maps that integer to a short descriptive string supplied by the implementation.

The returned pointer usually points to static memory inside the C library. Treat it as read-only and copy the text with strdup or snprintf if you need to keep it across later library calls.

📝 Syntax

Standard C declaration:

char *strerror(int errnum);

Headers

• #include <string.h> — for strerror
• #include <errno.h> — for errno and codes like ENOENT

Parameter

• errnum — error number (typically the value of errno).

Return Value

• Pointer to an implementation-defined error message string.
• For unknown codes, message may be "Unknown error" or similar.

Examples Gallery

Compile with gcc strerror.c -std=c11 -o strerror. Error text may vary slightly by operating system, but common codes like ENOENT are portable.

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

int main(void) {
    FILE *file = fopen("nonexistent.txt", "r");

    if (file == NULL) {
        int err = errno;
        printf("Error opening file: %s\n", strerror(err));
        return 1;
    }

    fclose(file);
    return 0;
}

Error opening file: No such file or directory

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

int main(void) {
    FILE *f = fopen("missing.dat", "rb");
    if (f == NULL) {
        int saved = errno;

        /* Other work here will not lose the original errno */
        printf("Failed (code %d): %s\n", saved, strerror(saved));
        return 1;
    }

    fclose(f);
    return 0;
}

Failed (code 2): No such file or directory

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

int main(void) {
    FILE *f = fopen("nope.txt", "r");
    if (f == NULL) {
        int err = errno;

        fprintf(stderr, "Custom: %s\n", strerror(err));
        errno = err;
        perror("perror prefix");
    }

    return f == NULL ? 1 : 0;
}

Custom: No such file or directory
perror prefix: No such file or directory

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

int main(void) {
    printf("ENOENT: %s\n", strerror(ENOENT));
    printf("EINVAL: %s\n", strerror(EINVAL));

    return 0;
}

ENOENT: No such file or directory
EINVAL: Invalid argument

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

void log_errno(const char *action) {
    int err = errno;
    fprintf(stderr, "[ERROR] %s failed: %s (code %d)\n",
            action, strerror(err), err);
}

int main(void) {
    if (fopen("ghost.txt", "r") == NULL) {
        log_errno("fopen");
        return 1;
    }
    return 0;
}

[ERROR] fopen failed: No such file or directory (code 2)

🚀 Common Use Cases

• File I/O failures — explain missing paths or permission errors.
• Network/socket code — combine with platform APIs that set errno.
• Debugging — print numeric code plus readable text in logs.
• User interfaces — show friendly messages instead of raw integers.
• Unit tests — assert expected error codes and look up messages.

📝 Notes

• Declared in <string.h>; errno lives in <errno.h>.
• Return pointer may alias static storage—do not modify or free it.
• Read errno immediately after detecting failure.
• Success paths may leave errno unchanged—only trust it after errors.
• Multithreaded programs may prefer POSIX strerror_r.

⚡ Optimization

Error handling should run on exceptional paths only, so strerror cost rarely matters. Avoid calling it in tight success loops. Cache messages only if you copy them—never cache the raw pointer across later strerror calls.

💡 Best Practices