Gpt4o: Write me a hexdump for the amiga

[alkis]

Gpt4o wrote a decent hexdump. I was testing it with SAS C on Amiga 500 Os 1.3.

I did 2 corrections on follow-up prompts.

1. "I think the compiler is C89"
2. "It doesn't have getopt"

Not bad.

#include <dos/dos.h>
#include <exec/types.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stddef.h> // for size_t

#ifndef SIZE_MAX
#define SIZE_MAX ((size_t)-1)
#endif

void hexdump(const char *filename, size_t offset, size_t length) {
    FILE *file;
    unsigned char buffer[16];
    size_t bytesRead;
    size_t totalRead;
    size_t i;

    file = fopen(filename, "rb"); // Open file in binary read mode
    if (!file) {
        printf("Failed to open file: %s\n", filename);
        return;
    }

    fseek(file, offset, SEEK_SET); // Set the file position to the offset

    totalRead = 0; // Total bytes read

    // Read the file and display its contents in hexadecimal format
    while ((bytesRead = fread(buffer, 1, sizeof(buffer), file)) > 0 && totalRead < length) {
        printf("%08lx ", (unsigned long)(offset + totalRead)); // Print the offset

        // Print the hexadecimal values of the bytes
        for (i = 0; i < bytesRead; ++i) {
            printf("%02x ", buffer[i]);
            if (i == 7) printf(" ");
        }

        // If the line is not full, pad with spaces
        if (bytesRead < 16) {
            for (i = bytesRead; i < 16; ++i) {
                printf("   ");
                if (i == 7) printf(" ");
            }
        }

        printf(" |");

        // Print the ASCII representation of the bytes
        for (i = 0; i < bytesRead; ++i) {
            if (buffer[i] >= 32 && buffer[i] <= 126) {
                printf("%c", buffer[i]);
            } else {
                printf(".");
            }
        }

        printf("|\n");
        totalRead += bytesRead; // Update total bytes read
    }

    fclose(file); // Close the file
}

int main(int argc, char **argv) {
    size_t offset = 0; // Default offset is 0
    size_t length = SIZE_MAX; // Default length is maximum size
    int i;

    // Manually parse command-line options
    for (i = 1; i < argc; i++) {
        if (strcmp(argv[i], "-o") == 0) {
            if (i + 1 < argc) {
                offset = strtoul(argv[++i], NULL, 0); // Set offset
            } else {
                printf("Option -o requires an argument.\n");
                return 1;
            }
        } else if (strcmp(argv[i], "-n") == 0) {
            if (i + 1 < argc) {
                length = strtoul(argv[++i], NULL, 0); // Set length
            } else {
                printf("Option -n requires an argument.\n");
                return 1;
            }
        } else {
            // Assume the remaining argument is the filename
            break;
        }
    }

    // Ensure a filename is provided
    if (i >= argc) {
        printf("Usage: hexdump [-o offset] [-n length] <filename>\n");
        return 1;
    }

    // Perform hexdump on the provided file
    hexdump(argv[i], offset, length);
    return 0;
}

[phx]

You shouldn't need any AmigaOS headers in a completely portable source text.

[Karlos]

My comment is the opposite. If the application is for the Amiga, why is it not using dos.library IO/ReadArgs and why is it not using exec.library RawDoFmt (or a simple custom online nybble to char conversion) and types?

[Dunny]

Originally Posted by Karlos

My comment is the opposite. If the application is for the Amiga, why is it not using dos.library IO/ReadArgs and why is it not using exec.library RawDoFmt (or a simple custom online nybble to char conversion) and types?

I'd guess that ChatGPT hasn't had a lot of Amiga sources trained into it?

[Karlos]

Originally Posted by Dunny

I'd guess that ChatGPT hasn't had a lot of Amiga sources trained into it?

Sure. To be honest, if I was writing a hex dump tool I'd probably just use C standard library too.