Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.c > #399064
| From | "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> |
|---|---|
| Newsgroups | comp.lang.c |
| Subject | Re: Is reallocarray() planned to be standardized? |
| Date | 2026-05-16 12:32 -0700 |
| Organization | A noiseless patient Spider |
| Message-ID | <10uago5$16iu1$1@dont-email.me> (permalink) |
| References | (1 earlier) <10u9du6$rsgq$1@dont-email.me> <10u9ipa$k0ug$1@dont-email.me> <10u9m51$u6ms$1@dont-email.me> <10u9o9a$q3k$1@reader1.panix.com> <10u9uo3$10n3l$1@dont-email.me> |
On 5/16/2026 7:25 AM, David Brown wrote:
[...]
You don't want any bit/byte leak for say an encryption process wrt the
password. If a side channel attacker can gain any of it, its bad. For
instance, here is a crude C99 version of my experimental encryption:
**********************************************
Well, I made some alterations to my old C version of my HMAC cipher. It
uses some non-portable API's in order to try to get a TRNG. It prints
out its usage, just run the program with no arguments, look in ct_help.
Well, can anybody else get it to compile _and_ run on their end? Thanks
everybody!
The secret key is hardcoded to Password and SHA2-512:
____________________________________
/*
Chris M. Thomasson 6/4/2018
Experimental HMAC Cipher
C version with hardcoded secret key
FIXED VERSION: Now uses proper TRNG (/dev/urandom on Unix,
CryptGenRandom on Windows)
Using the following HMAC lib:
https://github.com/ogay/hmac
Here is some info on my cipher:
http://funwithfractals.atspace.cc/ct_cipher
________________________________________________________*/
#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
#include <string.h>
#ifdef _WIN32
#include <windows.h>
#include <wincrypt.h>
#else
#include <fcntl.h>
#include <unistd.h>
#endif
#include "hmac_sha2.h"
#define CT_HMAC_SZ 64
// Uncomment PYTHON_TEST_VECTOR to sync with the Python 3 test vector
// Python code: https://pastebin.com/raw/NAnsBJAZ
// plaintext 9 bytes at: "Plaintext"
// ciphertext bytes:
// 9a419a03ac79bfa74edbbdda778316f6840b1aac07910de758e03e35a0d8ff1d407d
// 757ed6b734de9f9ed339bedf73786c5130d2f1891813c179ca20b82e81375e7a64e2
// dddead403b8284b9b76d1e83eddb
//#define PYTHON_TEST_VECTOR
struct ct_secret_key
{
unsigned char* hmac_key;
size_t hmac_key_sz;
char* hmac_algo;
size_t rand_n;
};
struct ct_buf
{
unsigned char* p;
size_t sz;
};
/*
CRITICAL: Cryptographically Secure Random Number Generation
This function uses:
- /dev/urandom on Unix/Linux/macOS (non-blocking, cryptographically
secure)
- CryptGenRandom on Windows (CSPRNG)
NEVER use rand() for cryptographic purposes!
*/
int ct_get_random_bytes(
unsigned char* buf,
size_t buf_sz
) {
if (!buf || buf_sz == 0) {
fprintf(stderr, "ERROR: Invalid buffer for random bytes\n");
return 0;
}
#ifdef _WIN32
// Windows: Use CryptGenRandom
HCRYPTPROV hCryptProv;
if (!CryptAcquireContext(&hCryptProv, NULL, NULL, PROV_RSA_FULL,
CRYPT_VERIFYCONTEXT)) {
fprintf(stderr, "ERROR: CryptAcquireContext failed\n");
return 0;
}
if (!CryptGenRandom(hCryptProv, (DWORD)buf_sz, buf)) {
fprintf(stderr, "ERROR: CryptGenRandom failed\n");
CryptReleaseContext(hCryptProv, 0);
return 0;
}
CryptReleaseContext(hCryptProv, 0);
return 1;
#else
// Unix/Linux/macOS: Use /dev/urandom
int fd = open("/dev/urandom", O_RDONLY);
if (fd < 0) {
fprintf(stderr, "ERROR: Cannot open /dev/urandom\n");
perror("open");
return 0;
}
size_t bytes_read = 0;
while (bytes_read < buf_sz) {
ssize_t result = read(fd, buf + bytes_read, buf_sz - bytes_read);
if (result < 0) {
fprintf(stderr, "ERROR: Failed to read from /dev/urandom\n");
perror("read");
close(fd);
return 0;
}
bytes_read += result;
}
close(fd);
return 1;
#endif
}
void ct_hex_printf(
FILE* fout,
unsigned char* buf,
size_t buf_sz
) {
for (size_t i = 0; i < buf_sz; i++)
{
fprintf(fout, "%02x", buf[i]);
}
}
unsigned char*
ct_reverse(
unsigned char* P,
size_t P_sz
) {
for (size_t i = 0; i < P_sz / 2; ++i)
{
size_t r = P_sz - i - 1;
unsigned char t = P[i];
P[i] = P[r];
P[r] = t;
}
return P;
}
size_t
ct_file_get_size(
FILE* file
) {
size_t file_sz = 0;
for (file_sz = 0; fgetc(file) != EOF; ++file_sz);
rewind(file);
return file_sz;
}
// return value ct_buf.p needs to be freed!
struct ct_buf
ct_file_copy(
FILE* file
) {
size_t file_sz = ct_file_get_size(file);
struct ct_buf buf = { calloc(1, file_sz), file_sz };
assert(buf.p);
if (buf.p)
{
for (size_t i = 0; i < file_sz; ++i)
{
int byte = fgetc(file);
assert(byte != EOF);
buf.p[i] = byte;
}
}
return buf;
}
// return value ct_buf.p needs to be freed!
struct ct_buf
ct_prepend_from_file(
struct ct_secret_key const* const SK,
const char* fname
) {
FILE* file = fopen(fname, "rb");
assert(file);
size_t file_sz = ct_file_get_size(file) + SK->rand_n;
struct ct_buf buf = { calloc(1, file_sz), file_sz };
if (buf.p)
{
// Prepend the random bytes.
// CRITICAL: These are drawn from a TRNG (cryptographically secure)
#if defined (PYTHON_TEST_VECTOR)
// Test vector mode: deterministic for testing only
printf("WARNING: Using test vector mode - NOT SECURE!\n");
for (size_t i = 0; i < SK->rand_n; ++i)
{
buf.p[i] = (unsigned char)i;
}
#else
// Production mode: Use proper TRNG
printf("Generating %zu bytes of cryptographically secure random
data...\n", SK->rand_n);
if (!ct_get_random_bytes(buf.p, SK->rand_n)) {
fprintf(stderr, "FATAL: Failed to generate secure random
bytes!\n");
free(buf.p);
fclose(file);
exit(EXIT_FAILURE);
}
printf("Random prefix generated successfully\n");
#endif
// Append the original plaintext
for (size_t i = SK->rand_n; i < file_sz; ++i)
{
int byte = fgetc(file);
assert(byte != EOF);
buf.p[i] = byte;
}
}
fclose(file);
return buf;
}
struct ct_buf
ct_load_from_file(
const char* fname
) {
FILE* file = fopen(fname, "rb");
assert(file);
size_t file_sz = ct_file_get_size(file);
struct ct_buf buf = { calloc(1, file_sz), file_sz };
if (buf.p)
{
// Append the original plaintext
for (size_t i = 0; i < file_sz; ++i)
{
int byte = fgetc(file);
assert(byte != EOF);
buf.p[i] = byte;
}
}
fclose(file);
return buf;
}
void ct_hmac_sha512_digest(
hmac_sha512_ctx* ctx,
unsigned char* digest
) {
hmac_sha512_ctx ctx_copy = *ctx;
hmac_sha512_final(&ctx_copy, digest, CT_HMAC_SZ);
}
unsigned char*
ct_crypt_round(
struct ct_secret_key* SK,
unsigned char* P,
size_t P_sz,
int M
) {
hmac_sha512_ctx H;
hmac_sha512_init(&H, SK->hmac_key, SK->hmac_key_sz);
ct_reverse(SK->hmac_key, SK->hmac_key_sz);
hmac_sha512_update(&H, SK->hmac_key, SK->hmac_key_sz);
ct_reverse(SK->hmac_key, SK->hmac_key_sz);
unsigned char D[256] = { 0 };
size_t P_I = 0;
unsigned long di = 0;
while (P_I < P_sz)
{
ct_hmac_sha512_digest(&H, D);
// Progress indicator
if (!(di % 128))
{
printf("P_I = %zu of %zu\r", P_I, P_sz);
}
size_t D_I = 0;
++di;
unsigned char update[CT_HMAC_SZ * 2];
size_t bytes_written = 0;
while (P_I < P_sz && D_I < CT_HMAC_SZ)
{
unsigned char P_byte = P[P_I];
unsigned char C_byte = P_byte ^ D[D_I];
P[P_I] = C_byte;
if (M == 0)
{
update[D_I * 2] = P_byte;
update[D_I * 2 + 1] = C_byte;
}
else
{
update[D_I * 2] = C_byte;
update[D_I * 2 + 1] = P_byte;
}
++P_I;
++D_I;
bytes_written += 2;
}
// Update with ACTUAL bytes, not full buffer!
hmac_sha512_update(&H, update, bytes_written);
}
printf("P_I = %zu of %zu\n", P_I, P_sz);
return P;
}
unsigned char*
ct_crypt(
struct ct_secret_key* SK,
unsigned char* P,
size_t P_sz,
int M
) {
printf("Crypt Round 0:\n________________________\n");
unsigned char* C = ct_crypt_round(SK, P, P_sz, M);
unsigned char* C_1 = ct_reverse(C, P_sz);
printf("\n\nCrypt Round 1:\n________________________\n");
C = ct_crypt_round(SK, C_1, P_sz, M);
return C;
}
int
ct_ciphertext_to_file(
FILE* fout,
struct ct_buf const* buf
) {
for (size_t i = 0; i < buf->sz; ++i)
{
int status = fputc((int)buf->p[i], fout);
if (status == EOF)
{
assert(status != EOF);
return 0;
}
}
return 1;
}
int
ct_plaintext_to_file(
FILE* fout,
struct ct_secret_key* SK,
struct ct_buf const* buf
) {
assert(SK->rand_n <= buf->sz);
for (size_t i = SK->rand_n; i < buf->sz; ++i)
{
int status = fputc((int)buf->p[i], fout);
if (status == EOF)
{
assert(status != EOF);
return 0;
}
}
return 1;
}
int
ct_encrypt(
struct ct_secret_key* SK,
char const* fname_in,
char const* fname_out
) {
int status = 0;
// Prepend the random bytes to the file...
struct ct_buf buf = ct_prepend_from_file(SK, fname_in);
if (buf.p)
{
unsigned char* C = ct_crypt(SK, buf.p, buf.sz, 0);
//printf("\n\n\nCiphertext:");
//ct_hex_printf(stdout, C, buf.sz);
//printf("\n\n\n");
// Write encrypted buffer to out file
{
FILE* fout = fopen(fname_out, "wb");
assert(fout);
status = ct_ciphertext_to_file(fout, &buf);
fclose(fout);
}
free(buf.p);
}
return status;
}
int
ct_decrypt(
struct ct_secret_key* SK,
char const* fname_in,
char const* fname_out
) {
int status = 0;
// Load the file...
struct ct_buf buf = ct_load_from_file(fname_in);
if (buf.p)
{
unsigned char* C = ct_crypt(SK, buf.p, buf.sz, 1);
//printf("\n\n\nPlaintext:");
//ct_hex_printf(stdout, C, buf.sz);
//printf("\n\n\n");
// Write decrypted buffer to out file
{
FILE* fout = fopen(fname_out, "wb");
assert(fout);
status = ct_plaintext_to_file(fout, SK, &buf);
fclose(fout);
}
free(buf.p);
}
return status;
}
void ct_help(void)
{
printf(
"\n\n\n"
"DrMoron Cipher - HMAC-based Stream Cipher\n"
"==========================================\n\n"
"Usage: program in_file out_file mode_flag\n\n"
"mode_flag -e is encrypt where the in_file gets encrypted as
out_file\n\n"
"mode_flag -d is decrypt where the in_file gets decrypted as
out_file\n\n"
"Example:\n\n"
"program plaintext.txt ciphertext.bin -e\n"
"program ciphertext.bin plaintext_decrypt.txt -d\n\n"
"SECURITY NOTES:\n"
"- Uses cryptographically secure RNG (/dev/urandom on Unix,
CryptGenRandom on Windows)\n"
"- Hardcoded key in this version - REPLACE with proper key
management for real use!\n"
"- This is an EXPERIMENTAL cipher - not recommended for
production use\n\n"
);
}
int main(int argc, char* argv[])
{
printf("\n=== DrMoron Cipher (Fixed TRNG Version) ===\n\n");
if (argc != 4)
{
printf("ERROR: Incorrect argument count!\n");
ct_help();
return EXIT_FAILURE;
}
{
int mode = 0;
if (strcmp(argv[3], "-e") == 0)
{
mode = 0;
printf("Mode: ENCRYPT\n");
}
else if (strcmp(argv[3], "-d") == 0)
{
mode = 1;
printf("Mode: DECRYPT\n");
}
else
{
printf("ERROR: Invalid encrypt/decrypt flag!\n");
ct_help();
return EXIT_FAILURE;
}
// WARNING: This is a hardcoded key for demonstration only!
// In real use, generate a proper 64-byte key from a TRNG
unsigned char hmac_key[] = "Password";
printf("WARNING: Using hardcoded demo key - NOT SECURE for
production!\n");
printf("Input file: %s\n", argv[1]);
printf("Output file: %s\n\n", argv[2]);
struct ct_secret_key SK = {
hmac_key,
sizeof(hmac_key) - 1,
"sha512",
73 // >64 bytes for SHA-512 digest size requirement
};
if (mode == 0)
{
ct_encrypt(&SK, argv[1], argv[2]);
printf("\n\nEncryption complete!\n");
}
else
{
ct_decrypt(&SK, argv[1], argv[2]);
printf("\n\nDecryption complete!\n");
}
}
return EXIT_SUCCESS;
}
____________________________________
Back to comp.lang.c | Previous | Next — Previous in thread | Next in thread | Find similar
Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-14 20:09 +0000
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-14 21:10 +0000
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-14 23:35 +0000
Re: Is reallocarray() planned to be standardized? Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2026-05-14 17:46 -0700
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 02:00 +0000
Re: Is reallocarray() planned to be standardized? Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2026-05-14 19:22 -0700
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 11:02 +0000
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-15 14:44 -0700
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 21:50 +0000
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-15 14:55 -0700
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-15 15:02 -0700
Re: Is reallocarray() planned to be standardized? scott@slp53.sl.home (Scott Lurndal) - 2026-05-15 22:12 +0000
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-16 11:38 +0200
Re: Is reallocarray() planned to be standardized? Janis Papanagnou <janis_papanagnou+ng@hotmail.com> - 2026-05-16 13:00 +0200
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-16 13:58 +0200
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-16 12:34 +0000
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-16 16:25 +0200
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-16 12:32 -0700
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-17 11:10 +0200
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-19 14:03 -0700
Re: Is reallocarray() planned to be standardized? Janis Papanagnou <janis_papanagnou+ng@hotmail.com> - 2026-05-17 06:36 +0200
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-16 12:25 -0700
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-17 11:12 +0200
Re: Is reallocarray() planned to be standardized? Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2026-05-17 02:26 -0700
Leader Keith has spoken! (Was: Is reallocarray() planned to be standardized?) gazelle@shell.xmission.com (Kenny McCormack) - 2026-05-17 12:54 +0000
Re: Is reallocarray() planned to be standardized? scott@slp53.sl.home (Scott Lurndal) - 2026-05-17 19:19 +0000
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-17 21:53 +0200
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-17 20:30 +0000
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-18 08:14 +0200
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-17 20:01 +0000
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-18 09:13 +0200
Re: Is reallocarray() planned to be standardized? Janis Papanagnou <janis_papanagnou+ng@hotmail.com> - 2026-05-18 09:51 +0200
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-18 14:34 +0000
Re: Is reallocarray() planned to be standardized? David Brown <david.brown@hesbynett.no> - 2026-05-18 16:49 +0200
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-18 16:19 -0700
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 22:32 +0000
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 15:23 +0000
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 17:28 +0000
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 17:40 +0000
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 17:50 +0000
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 18:28 +0000
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 21:35 +0000
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-15 15:05 -0700
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 22:35 +0000
Re: Is reallocarray() planned to be standardized? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2026-05-15 17:18 -0700
Re: Is reallocarray() planned to be standardized? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-05-16 08:07 +0000
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-16 12:44 +0000
Re: Is reallocarray() planned to be standardized? Janis Papanagnou <janis_papanagnou+ng@hotmail.com> - 2026-05-15 23:44 +0200
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 21:47 +0000
Re: Is reallocarray() planned to be standardized? scott@slp53.sl.home (Scott Lurndal) - 2026-05-15 22:10 +0000
Re: Is reallocarray() planned to be standardized? Janis Papanagnou <janis_papanagnou+ng@hotmail.com> - 2026-05-16 01:13 +0200
AIX (was Re: Is reallocarray() planned to be standardized?) cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 22:30 +0000
Re: Is reallocarray() planned to be standardized? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-05-16 08:04 +0000
Re: Is reallocarray() planned to be standardized? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-05-15 07:07 +0000
Re: Is reallocarray() planned to be standardized? kalevi@kolttonen.fi (Kalevi Kolttonen) - 2026-05-15 13:58 +0000
Re: Is reallocarray() planned to be standardized? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-05-16 00:25 +0000
Re: Is reallocarray() planned to be standardized? Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2026-05-14 17:26 -0700
Re: Is reallocarray() planned to be standardized? cross@spitfire.i.gajendra.net (Dan Cross) - 2026-05-15 02:08 +0000
Re: Is reallocarray() planned to be standardized? Bonita Montero <Bonita.Montero@gmail.com> - 2026-05-16 11:28 +0200
csiph-web