postgresql/contrib/pgcrypto/encode.c

347 lines
7.1 KiB
C

/*
* encode.c
* Various data encoding/decoding things.
*
* Copyright (c) 2001 Marko Kreen
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $Id: encode.c,v 1.3 2001/02/10 02:31:25 tgl Exp $
*/
#include "postgres.h"
#include "fmgr.h"
#include "encode.h"
/*
* NAMEDATALEN is used for hash names
*/
#if NAMEDATALEN < 16
#error "NAMEDATALEN < 16: too small"
#endif
static pg_coding *
find_coding(pg_coding *hbuf, text *name, int silent);
static pg_coding *
pg_find_coding(pg_coding *res, char *name);
/* SQL function: encode(bytea, text) returns text */
PG_FUNCTION_INFO_V1(encode);
Datum
encode(PG_FUNCTION_ARGS)
{
text *arg;
text *name;
uint len, rlen, rlen0;
pg_coding *c, cbuf;
text *res;
if (PG_ARGISNULL(0) || PG_ARGISNULL(1))
PG_RETURN_NULL();
name = PG_GETARG_TEXT_P(1);
c = find_coding(&cbuf, name, 0); /* will give error if fails */
arg = PG_GETARG_TEXT_P(0);
len = VARSIZE(arg) - VARHDRSZ;
rlen0 = c->encode_len(len);
res = (text *)palloc(rlen0 + VARHDRSZ);
rlen = c->encode(VARDATA(arg), len, VARDATA(res));
VARATT_SIZEP(res) = rlen + VARHDRSZ;
if (rlen > rlen0)
elog(FATAL, "pg_encode: overflow, encode estimate too small");
PG_FREE_IF_COPY(arg, 0);
PG_FREE_IF_COPY(name, 1);
PG_RETURN_TEXT_P(res);
}
/* SQL function: decode(text, text) returns bytea */
PG_FUNCTION_INFO_V1(decode);
Datum
decode(PG_FUNCTION_ARGS)
{
text *arg;
text *name;
uint len, rlen, rlen0;
pg_coding *c, cbuf;
text *res;
if (PG_ARGISNULL(0) || PG_ARGISNULL(1))
PG_RETURN_NULL();
name = PG_GETARG_TEXT_P(1);
c = find_coding(&cbuf, name, 0); /* will give error if fails */
arg = PG_GETARG_TEXT_P(0);
len = VARSIZE(arg) - VARHDRSZ;
rlen0 = c->decode_len(len);
res = (text *)palloc(rlen0 + VARHDRSZ);
rlen = c->decode(VARDATA(arg), len, VARDATA(res));
VARATT_SIZEP(res) = rlen + VARHDRSZ;
if (rlen > rlen0)
elog(FATAL, "pg_decode: overflow, decode estimate too small");
PG_FREE_IF_COPY(arg, 0);
PG_FREE_IF_COPY(name, 1);
PG_RETURN_TEXT_P(res);
}
static pg_coding *
find_coding(pg_coding *dst, text *name, int silent)
{
pg_coding *p;
char buf[NAMEDATALEN];
uint len;
len = VARSIZE(name) - VARHDRSZ;
if (len >= NAMEDATALEN) {
if (silent)
return NULL;
elog(ERROR, "Encoding type does not exist (name too long)");
}
memcpy(buf, VARDATA(name), len);
buf[len] = 0;
p = pg_find_coding(dst, buf);
if (p == NULL && !silent)
elog(ERROR, "Encoding type does not exist: '%s'", buf);
return p;
}
static char *hextbl = "0123456789abcdef";
uint
hex_encode(uint8 *src, uint len, uint8 *dst)
{
uint8 *end = src + len;
while (src < end) {
*dst++ = hextbl[(*src >> 4) & 0xF];
*dst++ = hextbl[*src & 0xF];
src++;
}
return len*2;
}
/* probably should use lookup table */
static uint8
get_hex(char c)
{
uint8 res = 0;
if (c >= '0' && c <= '9')
res = c - '0';
else if (c >= 'a' && c <= 'f')
res = c - 'a' + 10;
else if (c >= 'A' && c <= 'F')
res = c - 'A' + 10;
else
elog(ERROR, "Bad hex code: '%c'", c);
return res;
}
uint
hex_decode(uint8 *src, uint len, uint8 *dst)
{
uint8 *s, *srcend, v1, v2, *p = dst;
srcend = src + len;
s = src; p = dst;
while (s < srcend) {
if (*s == ' ' || *s == '\n' || *s == '\t' || *s == '\r') {
s++;
continue;
}
v1 = get_hex(*s++) << 4;
if (s >= srcend)
elog(ERROR, "hex_decode: invalid data");
v2 = get_hex(*s++);
*p++ = v1 | v2;
}
return p - dst;
}
static unsigned char _base64[] =
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
uint
b64_encode(uint8 *src, uint len, uint8 *dst)
{
uint8 *s, *p, *end = src + len, *lend = dst + 76;
int pos = 2;
unsigned long buf = 0;
s = src; p = dst;
while (s < end) {
buf |= *s << (pos << 3);
pos--;
s++;
/* write it out */
if (pos < 0) {
*p++ = _base64[(buf >> 18) & 0x3f];
*p++ = _base64[(buf >> 12) & 0x3f];
*p++ = _base64[(buf >> 6) & 0x3f];
*p++ = _base64[buf & 0x3f];
pos = 2;
buf = 0;
}
if (p >= lend) {
*p++ = '\n';
lend = p + 76;
}
}
if (pos != 2) {
*p++ = _base64[(buf >> 18) & 0x3f];
*p++ = _base64[(buf >> 12) & 0x3f];
*p++ = (pos == 0) ? _base64[(buf >> 6) & 0x3f] : '=';
*p++ = '=';
}
return p - dst;
}
/* probably should use lookup table */
uint
b64_decode(uint8 *src, uint len, uint8 *dst)
{
char *srcend = src + len, *s = src;
uint8 *p = dst;
char c;
uint b = 0;
unsigned long buf = 0;
int pos = 0, end = 0;
while (s < srcend) {
c = *s++;
if (c >= 'A' && c <= 'Z')
b = c - 'A';
else if (c >= 'a' && c <= 'z')
b = c - 'a' + 26;
else if (c >= '0' && c <= '9')
b = c - '0' + 52;
else if (c == '+')
b = 62;
else if (c == '/')
b = 63;
else if (c == '=') {
/* end sequence */
if (!end) {
if (pos == 2) end = 1;
else if (pos == 3) end = 2;
else
elog(ERROR, "base64: unexpected '='");
}
b = 0;
} else if (c == ' ' || c == '\t' || c == '\n' || c == '\r')
continue;
else
elog(ERROR, "base64: Invalid symbol");
/* add it to buffer */
buf = (buf << 6) + b;
pos++;
if (pos == 4) {
*p++ = (buf >> 16) & 255;
if (end == 0 || end > 1)
*p++ = (buf >> 8) & 255;
if (end == 0 || end > 2)
*p++ = buf & 255;
buf = 0;
pos = 0;
}
}
if (pos != 0)
elog(ERROR, "base64: invalid end sequence");
return p - dst;
}
uint
hex_enc_len(uint srclen)
{
return srclen << 1;
}
uint
hex_dec_len(uint srclen)
{
return srclen >> 1;
}
uint
b64_enc_len(uint srclen)
{
return srclen + (srclen / 3) + (srclen / (76 / 2));
}
uint
b64_dec_len(uint srclen)
{
return (srclen * 3) >> 2;
}
static pg_coding
encoding_list [] = {
{ "hex", hex_enc_len, hex_dec_len, hex_encode, hex_decode},
{ "base64", b64_enc_len, b64_dec_len, b64_encode, b64_decode},
{ NULL, NULL, NULL, NULL, NULL}
};
static pg_coding *
pg_find_coding(pg_coding *res, char *name)
{
pg_coding *p;
for (p = encoding_list; p->name; p++) {
if (!strcasecmp(p->name, name))
return p;
}
return NULL;
}