Hex: Difference between revisions

From miki
Jump to navigation Jump to search
Line 397: Line 397:
import binascii
import binascii


'deadbeef' == binascii.b2a_hex('\xde\xad\xbe\xef').encode("hex")
'deadbeef' == '\xde\xad\xbe\xef'.encode("hex")
'deadbeef' == binascii.b2a_hex('\xde\xad\xbe\xef')
'deadbeef' == binascii.b2a_hex('\xde\xad\xbe\xef')
'deadbeef' == binascii.hexlify('\xde\xad\xbe\xef')
'deadbeef' == binascii.hexlify('\xde\xad\xbe\xef')

Revision as of 14:53, 29 September 2021

References

Many examples come from cplusplus.com and stackoverflow.com.

Shell

See also Linux Commands.

Convert 66 6f 6f → "foo"

hexdump -e '"%2x"' <myfile>             # Convert myfile into a long hexadecimal string - ! See DOUBLE-QUOTED parameter
hexdump -C <myfile>                     # Canonical hex + ascii display (16 bytes)
hd <myfile>                             # (idem)

echo 202122 | xxd -r -p                 # Convert hexdump to a binary string
echo -e '\x66\x6f\x6f'                  # 'foo'

or to convert hex to decimal [1]

echo $((0x15a))
printf '%d\n' 0x15a
perl -e 'printf ("%d\n", 0x15a)'
echo 'ibase=16;obase=A;15A' | bc

Convert "foo" → 66 6f 6f

xxd -g8 -c32 <file>                     # Output 32 bytes, grouped in 8-byte columns
xxd -p -c64 <file>                      # Output 64 bytes per line, plain (postscript) mode
echo -n ' !"' | xxd -p                  # Convert binary string to hexdump - DON'T FORGET -n

Or to convert decimal to hex

printf '%x\n' 346
perl -e 'printf ("%x\n", 346)'
echo 'ibase=10;obase=16;346' | bc

hex editors

hex diffs

diff -u <(hexdump -v -C file1) <(hexdump -v -C file2)

C / C ++

Convert an integer to hex string

C++

This looks simple but is in fact extremelly treacherous because of persistent iomanip flags:

#include <iostream>
#include <iomanip>

unsigned value = 10;

# 'internal' to make sure we use correct alignment (because 'left' is persistent)
cout << "0x" << setfill('0') << setw(2) << internal << hex << 10 << endl;

C char2hex

A complete C solution, char2hex:

// ############################################################################################################
static uint32_t _char2hex(char *s, const uint8_t data[], size_t data_n, size_t n)
{
    uint32_t  o = 0;

#define out(f, a ...)  {                                          \
        o += (uint32_t)snprintf(s + o, n - (uint32_t)o, f, ## a); \
        o  = o <= n ? o : n;                                      \
}
    if (n > 0) {
        s[0] = 0;
    }
    for (size_t i = 0; i < data_n; i++) {
        out("%.2x", data[i]);
    }

    return o;
#undef out
}

/**
 * Convert an array of char into a list of hexadecimal values.
 * For instance, "\xca\xfe" is converted into "cafe".
 */
char * char2hex(char *s, const uint8_t data[], size_t data_n, size_t n)
{
    uint32_t  o = _char2hex(s, data, data_n, n);

    assert(o < n);                                                   // Check if given buffer was big enough

    return s;
}

Read an integer from an hex string

(See [2], [3]).

Homebrew

const char _hex[] = "0123456789ABCDEF";

// Change a 0x0A into 'A'
char itoh(int nibble)
{
    return _hex[nibble];
}

// Change a 'A' (or 'a') into 0x0A
int htoi(char hex)
{
    return ( hex >> 6 ) + (( hex >> 6 ) << 3 ) + ( hex & 0x0F );
}

// Change "A010" into 0xA010
template<class T>
T hextoint_BE(string h)
{
    T i=0;

    for(string::const_iterator it=h.begin(); it!=h.end(); ++it)
    {
        i=(i<<4)|htoi(*it);
    }

    return i;
}

// Change "A010" into "\xA0\x10"
string hextostring_BE(const string& h)
{
    string x;
    for(unsigned int i=0; (i+1)<h.length(); i+=2) {
        x += (htoi(h[i]) << 4)+ htoi(h[i+1]);
    }
    return x;
}

// Change 0xA010 into "A010"
template<class T> 
string inttohex_BE(T i, unsigned int size)
{
    string sh;
    while( i > 0 )
    {
        sh = itoh(i&0xF) + sh;
        i>>=4;
    }

    if(!size)
        return sh;
    if(size<sh.length())
        return string(sh,sh.length()-size,size);
    else
        return string(size-sh.length(),'0')+sh;
}

// Change "\xA0\x10" into "A010"
string stringtohex_BE(const string& s, unsigned int size)
{
    string h;
    for(string::const_iterator it=s.begin();it!=s.end();++it)
    {
        h += itoh((*it & 0xF0) >> 4);
        h += itoh((*it & 0x0F) >> 0);
    }

    if(!size)
        return h;
    if(size<h.length())
        return string(h,h.length()-size,size);
    else
        return string(size-h.length(),'0')+h;
}

Using atol

/* atol example */
#include <stdio.h>
#include <stdlib.h>

int main ()
{
  long int li;
  char szInput [256];
  printf ("Enter a long number: ");
  gets ( szInput );
  li = atol (szInput);
  printf ("The value entered is %d. The double is %d.\n",li,li*2);
  return 0;
}

Using strtol, strtoll

strtol (resp. strtoll) returns LONG_MIN, LONG_MAX (resp. LLONG_MIN, LLONGMAX) in case of under/overflow.

/* strtol example */
#include <stdio.h>
#include <stdlib.h>

int main ()
{
  char szNumbers[] = "2001 60c0c0 -1101110100110100100000 0x6fffff";
  char * pEnd;
  long int li1, li2, li3, li4;
  li1 = strtol (szNumbers,&pEnd,10);
  li2 = strtol (pEnd,&pEnd,16);
  li3 = strtol (pEnd,&pEnd,2);
  li4 = strtol (pEnd,NULL,0);
  printf ("The decimal equivalents are: %ld, %ld, %ld and %ld.\n", 
          li1, li2, li3, li4);
  return 0;
}
#include <cstdlib>
#include <iostream>
using namespace std;

int main() {
    string s = "abcd";
    char * p;
    long n = strtol( s.c_str(), & p, 16 );
    if ( * p != 0 ) {
        cout << "not a number" << endl;
    }
    else {
        cout << n << endl;
    }
}

Using strtoul, strtoull

Use strtoul (resp. strtoull) when value to read is greater than numeric_limits<long>::max() (resp. numeric_limits<long long>::max()).

/* strtoul example */
#include <stdio.h>
#include <stdlib.h>

int main ()
{
  char szInput [256];
  unsigned long ul;
  printf ("Enter an unsigned number: ");
  fgets (szInput,256,stdin);
  ul = strtoul (szInput,NULL,0);
  printf ("Value entered: %lu. Its double: %lu\n",ul,ul*2);
  return 0;
}
#include <cstdlib>
#include <iostream>
using namespace std;

int main() { 
    string s = "fffefffe";
    char * p;
    long n = strtoul( s.c_str(), & p, 16 ); 
    if ( * p != 0 ) {  
        cout << "not a number" << endl;
    }    else {  
        cout << n << endl;
    }
}

Using sscanf

#include <stdio.h>
main()
{
    char s[] = "fffffffe";
    int x;
    sscanf(s, "%x", &x);
    printf("%u\n", x);
}

Using stringstream

#include <sstream>
#include <iostream>

int main() {
    unsigned int x;   
    std::stringstream ss;
    ss << std::hex << "fffefffe";
    ss >> x;
    // output it as a signed type
    std::cout << static_cast<int>(x) << std::endl;
}

C - hex2char

// ############################################################################################################
static uint8_t _htoi(char hex)
{
    return ((uint8_t)hex >> 6) + (((uint8_t)hex >> 6) << 3) + ((uint8_t)hex & 0x0F);
}

// ############################################################################################################

/** Convert an hex null-terminated string ("cafe") into a byte array ({0xca, 0xfe}).
 *
 * Convert at most `n` bytes from an hexadecimal null-terminated string into a byte array.
 * For instance, it converts "cafe" to {0xca, 0xfe}.
 *
 * @param[out]  dst pointer to destination buffer (at least #n-byte long).
 * @param[in]   src pointer to source string.
 * @param[in]   n   maximum number of bytes to write in #dst.
 *
 * @return          Number of bytes converted.
 */
uint32_t hex2char(void *dst, const char *src, uint32_t n)
{
    uint8_t *   p;
    uint8_t *   _dst = dst;
    const char *_src = src;

    for (p = _dst; (p < _dst + n) && _src[0] && _src[1]; p++, _src += 2) {
        *p = (_htoi(_src[0]) << 4) + _htoi(_src[1]);
    }

    return p - _dst;
}

C - hex2char - fastest and lowercase/uppercase

Inspired from SO:

// ############################################################################################################
// 74ns on i686 (32-bit), vs 150ns for version above
uint32_t hex2char(void *dst, const char *src, uint32_t n)
{
    uint8_t *   p;
    uint8_t *   _dst = dst;
    const char *_src = src;

#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Winitializer-overrides"
static const long  hextable[] = { // Reduce to signed char for space saving
    [0 ... 255] = -1,
    ['0']       =  0,  1,  2,  3,  4,  5,  6,  7,  8, 9,
    ['A']       = 10, 11, 12, 13, 14, 15,
    ['a']       = 10, 11, 12, 13, 14, 15,
};
#pragma clang diagnostic pop

    for (p = _dst; (p < _dst + n) && _src[0] && _src[1]; p++) {
        *p = hextable[*_src++];
        *p = (*p << 4) | hextable[*_src++];
    }

    return p - _dst;
}

Python

Integer
'4d2'      == '%x' % 1234             # to hex
'0x4d2'    == hex(1234)               # to hex (with '0x' prefix)
'0x4d2'    == '0x%x' % 1234
'000004d2' == '%08x' % 1234           # to hex (padded)
'000004D2' == '%08X' % 1234           # to hex (padded,caps)
'\x41'     == chr(0x41)               # to hex char
1234       == int('04d2',16)          # from hex
1234       == int('0x04d2',0)         # from hex - base 0, autodetect
1234       == 0x04d2                  # from hex
'1234'     == '%d' % 0x04d2           # from hex
'0x1234'   == '0x%d' % 0x04d2         # from hex (with '0x' prefix)
Strings, arrays or lists - Python 2
import binascii

'deadbeef' == '\xde\xad\xbe\xef'.encode("hex")
'deadbeef' == binascii.b2a_hex('\xde\xad\xbe\xef')
'deadbeef' == binascii.hexlify('\xde\xad\xbe\xef')
'deadbeef' == str(bytearray([0xde,0xad,0xbe,0xef])).encode('hex')
'deadbeef' == ''.join(map(chr,[0xde,0xad,0xbe,0xef])).encode('hex')  # Likely slow, but need benchmarks on 2.x / 3.x
'deadbeef' == str(bytearray([222, 173, 190, 239])).encode('hex')

'\xde\xad\xbe\xef' == 'deadbeef'.decode("hex")
'\xde\xad\xbe\xef' == binascii.a2b_hex('deadbeef')
'\xde\xad\xbe\xef' == binascii.unhexlify('deadbeef')
'\xde\xad\xbe\xef' == bytearray.fromhex('deadbeef')          # IGNORE SPACES, IF ANY
'\xde\xad\xbe\xef' == bytearray('\xde\xad\xbe\xef')
'\xde\xad\xbe\xef' == bytearray([0xde,0xad,0xbe,0xef])
'\xde\xad\xbe\xef' == str(bytearray([0xde,0xad,0xbe,0xef]))

[0xde,0xad,0xbe,0xef] == map(ord,'\xde\xad\xbe\xef')
[0xde,0xad,0xbe,0xef] == list(bytearray.fromhex('deadbeef')) # IGNORE SPACES, IF ANY
Strings, arrays or lists - Python 3
import binascii

'deadbeef' == binascii.b2a_hex(b'\xde\xad\xbe\xef').decode()
'deadbeef' == binascii.hexlify(b'\xde\xad\xbe\xef').decode()
'deadbeef' == binascii.b2a_hex(bytes([0xde,0xad,0xbe,0xef])).decode()
'deadbeef' == binascii.hexlify(bytes([0xde,0xad,0xbe,0xef])).decode()

b'deadbeef' == binascii.b2a_hex(b'\xde\xad\xbe\xef')
b'deadbeef' == binascii.hexlify(b'\xde\xad\xbe\xef')
b'deadbeef' == binascii.b2a_hex(bytes([0xde,0xad,0xbe,0xef]))
b'deadbeef' == binascii.hexlify(bytes([0xde,0xad,0xbe,0xef]))

b'\xde\xad\xbe\xef' == binascii.a2b_hex('deadbeef')
b'\xde\xad\xbe\xef' == binascii.unhexlify('deadbeef')
b'\xde\xad\xbe\xef' == bytes.fromhex('deadbeef')             # IGNORE SPACES, IF ANY
b'\xde\xad\xbe\xef' == binascii.a2b_hex(b'deadbeef')
b'\xde\xad\xbe\xef' == binascii.unhexlify(b'deadbeef')
b'\xde\xad\xbe\xef' == bytes([0xde,0xad,0xbe,0xef])

[0xde,0xad,0xbe,0xef] == list(b'\xde\xad\xbe\xef')
[0xde,0xad,0xbe,0xef] == list(bytes.fromhex('deadbeef'))     # IGNORE SPACES, IF ANY
[0xde,0xad,0xbe,0xef] == list(bytearray.fromhex('deadbeef')) # IGNORE SPACES, IF ANY

# Note that in Python3, bytearray is a different from bytes, but both are equivalent
bytearray(b'\xde\xad\xbe\xef') == bytearray([0xde,0xad,0xbe,0xef])