2005-07-10 17:46:29 +04:00
|
|
|
/*
|
|
|
|
* pgp-pubkey.c
|
|
|
|
* Read public or secret key.
|
|
|
|
*
|
|
|
|
* Copyright (c) 2005 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.
|
|
|
|
*
|
2005-10-15 06:49:52 +04:00
|
|
|
* $PostgreSQL: pgsql/contrib/pgcrypto/pgp-pubkey.c,v 1.4 2005/10/15 02:49:06 momjian Exp $
|
2005-07-10 17:46:29 +04:00
|
|
|
*/
|
2005-07-11 19:07:59 +04:00
|
|
|
#include "postgres.h"
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
#include "px.h"
|
|
|
|
#include "mbuf.h"
|
|
|
|
#include "pgp.h"
|
|
|
|
|
2005-10-15 06:49:52 +04:00
|
|
|
int
|
|
|
|
pgp_key_alloc(PGP_PubKey ** pk_p)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
|
|
|
PGP_PubKey *pk;
|
2005-10-15 06:49:52 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
pk = px_alloc(sizeof(*pk));
|
|
|
|
memset(pk, 0, sizeof(*pk));
|
|
|
|
*pk_p = pk;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-10-15 06:49:52 +04:00
|
|
|
void
|
|
|
|
pgp_key_free(PGP_PubKey * pk)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
if (pk == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
switch (pk->algo)
|
|
|
|
{
|
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
|
|
|
pgp_mpi_free(pk->pub.elg.p);
|
|
|
|
pgp_mpi_free(pk->pub.elg.g);
|
|
|
|
pgp_mpi_free(pk->pub.elg.y);
|
|
|
|
pgp_mpi_free(pk->sec.elg.x);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_RSA_SIGN:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
pgp_mpi_free(pk->pub.rsa.n);
|
|
|
|
pgp_mpi_free(pk->pub.rsa.e);
|
|
|
|
pgp_mpi_free(pk->sec.rsa.d);
|
|
|
|
pgp_mpi_free(pk->sec.rsa.p);
|
|
|
|
pgp_mpi_free(pk->sec.rsa.q);
|
|
|
|
pgp_mpi_free(pk->sec.rsa.u);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_DSA_SIGN:
|
|
|
|
pgp_mpi_free(pk->pub.dsa.p);
|
|
|
|
pgp_mpi_free(pk->pub.dsa.q);
|
|
|
|
pgp_mpi_free(pk->pub.dsa.g);
|
|
|
|
pgp_mpi_free(pk->pub.dsa.y);
|
|
|
|
pgp_mpi_free(pk->sec.dsa.x);
|
|
|
|
break;
|
|
|
|
}
|
2005-07-10 17:46:29 +04:00
|
|
|
memset(pk, 0, sizeof(*pk));
|
|
|
|
px_free(pk);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2005-10-15 06:49:52 +04:00
|
|
|
calc_key_id(PGP_PubKey * pk)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
|
|
|
PX_MD *md;
|
|
|
|
int len;
|
|
|
|
uint8 hdr[3];
|
|
|
|
uint8 hash[20];
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
res = pgp_load_digest(PGP_DIGEST_SHA1, &md);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
|
|
|
|
len = 1 + 4 + 1;
|
|
|
|
switch (pk->algo)
|
|
|
|
{
|
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
len += 2 + pk->pub.elg.p->bytes;
|
|
|
|
len += 2 + pk->pub.elg.g->bytes;
|
|
|
|
len += 2 + pk->pub.elg.y->bytes;
|
|
|
|
break;
|
|
|
|
case PGP_PUB_RSA_SIGN:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
len += 2 + pk->pub.rsa.n->bytes;
|
|
|
|
len += 2 + pk->pub.rsa.e->bytes;
|
|
|
|
break;
|
|
|
|
case PGP_PUB_DSA_SIGN:
|
|
|
|
len += 2 + pk->pub.dsa.p->bytes;
|
|
|
|
len += 2 + pk->pub.dsa.q->bytes;
|
|
|
|
len += 2 + pk->pub.dsa.g->bytes;
|
|
|
|
len += 2 + pk->pub.dsa.y->bytes;
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
hdr[0] = 0x99;
|
|
|
|
hdr[1] = len >> 8;
|
|
|
|
hdr[2] = len & 0xFF;
|
|
|
|
px_md_update(md, hdr, 3);
|
|
|
|
|
|
|
|
px_md_update(md, &pk->ver, 1);
|
|
|
|
px_md_update(md, pk->time, 4);
|
|
|
|
px_md_update(md, &pk->algo, 1);
|
2005-10-15 06:49:52 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
switch (pk->algo)
|
|
|
|
{
|
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
pgp_mpi_hash(md, pk->pub.elg.p);
|
|
|
|
pgp_mpi_hash(md, pk->pub.elg.g);
|
|
|
|
pgp_mpi_hash(md, pk->pub.elg.y);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_RSA_SIGN:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
pgp_mpi_hash(md, pk->pub.rsa.n);
|
|
|
|
pgp_mpi_hash(md, pk->pub.rsa.e);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_DSA_SIGN:
|
|
|
|
pgp_mpi_hash(md, pk->pub.dsa.p);
|
|
|
|
pgp_mpi_hash(md, pk->pub.dsa.q);
|
|
|
|
pgp_mpi_hash(md, pk->pub.dsa.g);
|
|
|
|
pgp_mpi_hash(md, pk->pub.dsa.y);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
px_md_finish(md, hash);
|
|
|
|
px_md_free(md);
|
|
|
|
|
|
|
|
memcpy(pk->key_id, hash + 12, 8);
|
|
|
|
memset(hash, 0, 20);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-10-15 06:49:52 +04:00
|
|
|
int
|
|
|
|
_pgp_read_public_key(PullFilter * pkt, PGP_PubKey ** pk_p)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
PGP_PubKey *pk;
|
|
|
|
|
|
|
|
res = pgp_key_alloc(&pk);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
/* get version */
|
|
|
|
GETBYTE(pkt, pk->ver);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (pk->ver != 4)
|
|
|
|
{
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = PXE_PGP_NOT_V4_KEYPKT;
|
|
|
|
goto out;
|
2005-07-10 17:46:29 +04:00
|
|
|
}
|
2005-10-15 06:49:52 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
/* read time */
|
|
|
|
res = pullf_read_fixed(pkt, 4, pk->time);
|
|
|
|
if (res < 0)
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
goto out;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
/* pubkey algorithm */
|
|
|
|
GETBYTE(pkt, pk->algo);
|
|
|
|
|
2005-10-15 06:49:52 +04:00
|
|
|
switch (pk->algo)
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
case PGP_PUB_DSA_SIGN:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.dsa.p);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.dsa.q);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.dsa.g);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.dsa.y);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
|
|
|
res = calc_key_id(pk);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
|
|
|
case PGP_PUB_RSA_SIGN:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.rsa.n);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.rsa.e);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
|
|
|
res = calc_key_id(pk);
|
|
|
|
|
|
|
|
if (pk->algo != PGP_PUB_RSA_SIGN)
|
|
|
|
pk->can_encrypt = 1;
|
|
|
|
break;
|
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.elg.p);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.elg.g);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->pub.elg.y);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
res = calc_key_id(pk);
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
|
|
|
pk->can_encrypt = 1;
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
default:
|
|
|
|
px_debug("unknown public algo: %d", pk->algo);
|
|
|
|
res = PXE_PGP_UNKNOWN_PUBALGO;
|
|
|
|
}
|
|
|
|
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
out:
|
|
|
|
if (res < 0)
|
|
|
|
pgp_key_free(pk);
|
|
|
|
else
|
|
|
|
*pk_p = pk;
|
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define HIDE_CLEAR 0
|
|
|
|
#define HIDE_CKSUM 255
|
|
|
|
#define HIDE_SHA1 254
|
|
|
|
|
|
|
|
static int
|
2005-10-15 06:49:52 +04:00
|
|
|
check_key_sha1(PullFilter * src, PGP_PubKey * pk)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
|
|
|
uint8 got_sha1[20];
|
|
|
|
uint8 my_sha1[20];
|
|
|
|
PX_MD *md;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
res = pullf_read_fixed(src, 20, got_sha1);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
|
|
|
|
res = pgp_load_digest(PGP_DIGEST_SHA1, &md);
|
|
|
|
if (res < 0)
|
|
|
|
goto err;
|
|
|
|
switch (pk->algo)
|
|
|
|
{
|
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
pgp_mpi_hash(md, pk->sec.elg.x);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_RSA_SIGN:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
pgp_mpi_hash(md, pk->sec.rsa.d);
|
|
|
|
pgp_mpi_hash(md, pk->sec.rsa.p);
|
|
|
|
pgp_mpi_hash(md, pk->sec.rsa.q);
|
|
|
|
pgp_mpi_hash(md, pk->sec.rsa.u);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_DSA_SIGN:
|
|
|
|
pgp_mpi_hash(md, pk->sec.dsa.x);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
px_md_finish(md, my_sha1);
|
|
|
|
px_md_free(md);
|
|
|
|
|
|
|
|
if (memcmp(my_sha1, got_sha1, 20) != 0)
|
|
|
|
{
|
|
|
|
px_debug("key sha1 check failed");
|
|
|
|
res = PXE_PGP_KEYPKT_CORRUPT;
|
|
|
|
}
|
|
|
|
err:
|
|
|
|
memset(got_sha1, 0, 20);
|
|
|
|
memset(my_sha1, 0, 20);
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2005-10-15 06:49:52 +04:00
|
|
|
check_key_cksum(PullFilter * src, PGP_PubKey * pk)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
|
|
|
unsigned got_cksum,
|
|
|
|
my_cksum = 0;
|
|
|
|
uint8 buf[2];
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
res = pullf_read_fixed(src, 2, buf);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
|
2005-10-15 06:49:52 +04:00
|
|
|
got_cksum = ((unsigned) buf[0] << 8) + buf[1];
|
2005-07-10 17:46:29 +04:00
|
|
|
switch (pk->algo)
|
|
|
|
{
|
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
my_cksum = pgp_mpi_cksum(0, pk->sec.elg.x);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_RSA_SIGN:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
my_cksum = pgp_mpi_cksum(0, pk->sec.rsa.d);
|
|
|
|
my_cksum = pgp_mpi_cksum(my_cksum, pk->sec.rsa.p);
|
|
|
|
my_cksum = pgp_mpi_cksum(my_cksum, pk->sec.rsa.q);
|
|
|
|
my_cksum = pgp_mpi_cksum(my_cksum, pk->sec.rsa.u);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_DSA_SIGN:
|
|
|
|
my_cksum = pgp_mpi_cksum(0, pk->sec.dsa.x);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (my_cksum != got_cksum)
|
|
|
|
{
|
|
|
|
px_debug("key cksum check failed");
|
|
|
|
return PXE_PGP_KEYPKT_CORRUPT;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-10-15 06:49:52 +04:00
|
|
|
static int
|
|
|
|
process_secret_key(PullFilter * pkt, PGP_PubKey ** pk_p,
|
|
|
|
const uint8 *key, int key_len)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
|
|
|
int hide_type;
|
|
|
|
int cipher_algo;
|
|
|
|
int bs;
|
|
|
|
uint8 iv[512];
|
|
|
|
PullFilter *pf_decrypt = NULL,
|
|
|
|
*pf_key;
|
|
|
|
PGP_CFB *cfb = NULL;
|
|
|
|
PGP_S2K s2k;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
PGP_PubKey *pk;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
/* first read public key part */
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = _pgp_read_public_key(pkt, &pk);
|
2005-07-10 17:46:29 +04:00
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* is secret key encrypted?
|
|
|
|
*/
|
|
|
|
GETBYTE(pkt, hide_type);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (hide_type == HIDE_SHA1 || hide_type == HIDE_CKSUM)
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
if (key == NULL)
|
|
|
|
return PXE_PGP_NEED_SECRET_PSW;
|
|
|
|
GETBYTE(pkt, cipher_algo);
|
|
|
|
res = pgp_s2k_read(pkt, &s2k);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
|
|
|
|
res = pgp_s2k_process(&s2k, cipher_algo, key, key_len);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
2005-10-15 06:49:52 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
bs = pgp_get_cipher_block_size(cipher_algo);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (bs == 0)
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
px_debug("unknown cipher algo=%d", cipher_algo);
|
|
|
|
return PXE_PGP_UNSUPPORTED_CIPHER;
|
|
|
|
}
|
|
|
|
res = pullf_read_fixed(pkt, bs, iv);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
2005-10-15 06:49:52 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
/*
|
|
|
|
* create decrypt filter
|
|
|
|
*/
|
|
|
|
res = pgp_cfb_create(&cfb, cipher_algo, s2k.key, s2k.key_len, 0, iv);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
res = pullf_create(&pf_decrypt, &pgp_decrypt_filter, cfb, pkt);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
pf_key = pf_decrypt;
|
2005-10-15 06:49:52 +04:00
|
|
|
}
|
|
|
|
else if (hide_type == HIDE_CLEAR)
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
pf_key = pkt;
|
2005-10-15 06:49:52 +04:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
px_debug("unknown hide type");
|
|
|
|
return PXE_PGP_KEYPKT_CORRUPT;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* read secret key */
|
2005-10-15 06:49:52 +04:00
|
|
|
switch (pk->algo)
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
case PGP_PUB_RSA_SIGN:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
case PGP_PUB_RSA_ENCRYPT:
|
|
|
|
case PGP_PUB_RSA_ENCRYPT_SIGN:
|
|
|
|
res = pgp_mpi_read(pkt, &pk->sec.rsa.d);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->sec.rsa.p);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->sec.rsa.q);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pkt, &pk->sec.rsa.u);
|
2005-10-15 06:49:52 +04:00
|
|
|
if (res < 0)
|
|
|
|
break;
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
case PGP_PUB_ELG_ENCRYPT:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = pgp_mpi_read(pf_key, &pk->sec.elg.x);
|
|
|
|
break;
|
|
|
|
case PGP_PUB_DSA_SIGN:
|
|
|
|
res = pgp_mpi_read(pf_key, &pk->sec.dsa.x);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
px_debug("unknown public algo: %d", pk->algo);
|
|
|
|
res = PXE_PGP_KEYPKT_CORRUPT;
|
|
|
|
}
|
|
|
|
/* read checksum / sha1 */
|
|
|
|
if (res >= 0)
|
|
|
|
{
|
|
|
|
if (hide_type == HIDE_SHA1)
|
|
|
|
res = check_key_sha1(pf_key, pk);
|
|
|
|
else
|
|
|
|
res = check_key_cksum(pf_key, pk);
|
|
|
|
}
|
|
|
|
if (res >= 0)
|
|
|
|
res = pgp_expect_packet_end(pf_key);
|
|
|
|
|
|
|
|
if (pf_decrypt)
|
|
|
|
pullf_free(pf_decrypt);
|
|
|
|
if (cfb)
|
|
|
|
pgp_cfb_free(cfb);
|
|
|
|
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
if (res < 0)
|
|
|
|
pgp_key_free(pk);
|
|
|
|
else
|
|
|
|
*pk_p = pk;
|
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2005-10-15 06:49:52 +04:00
|
|
|
internal_read_key(PullFilter * src, PGP_PubKey ** pk_p,
|
|
|
|
const uint8 *psw, int psw_len, int pubtype)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
|
|
|
PullFilter *pkt = NULL;
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
|
|
|
uint8 tag;
|
|
|
|
int len;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
PGP_PubKey *enc_key = NULL;
|
2005-07-10 17:46:29 +04:00
|
|
|
PGP_PubKey *pk = NULL;
|
2005-10-15 06:49:52 +04:00
|
|
|
int got_main_key = 0;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
|
|
|
/*
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
* Search for encryption key.
|
2005-07-10 17:46:29 +04:00
|
|
|
*
|
|
|
|
* Error out on anything fancy.
|
|
|
|
*/
|
2005-10-15 06:49:52 +04:00
|
|
|
while (1)
|
|
|
|
{
|
2005-07-10 17:46:29 +04:00
|
|
|
res = pgp_parse_pkt_hdr(src, &tag, &len, 0);
|
|
|
|
if (res <= 0)
|
|
|
|
break;
|
|
|
|
res = pgp_create_pkt_reader(&pkt, src, len, res, NULL);
|
|
|
|
if (res < 0)
|
|
|
|
break;
|
2005-10-15 06:49:52 +04:00
|
|
|
|
|
|
|
switch (tag)
|
|
|
|
{
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
case PGP_PKT_PUBLIC_KEY:
|
2005-07-10 17:46:29 +04:00
|
|
|
case PGP_PKT_SECRET_KEY:
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
if (got_main_key)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
|
|
|
res = PXE_PGP_MULTIPLE_KEYS;
|
|
|
|
break;
|
|
|
|
}
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
got_main_key = 1;
|
|
|
|
res = pgp_skip_packet(pkt);
|
|
|
|
break;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
case PGP_PKT_PUBLIC_SUBKEY:
|
|
|
|
if (pubtype != 0)
|
|
|
|
res = PXE_PGP_EXPECT_SECRET_KEY;
|
2005-07-10 17:46:29 +04:00
|
|
|
else
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = _pgp_read_public_key(pkt, &pk);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
case PGP_PKT_SECRET_SUBKEY:
|
|
|
|
if (pubtype != 1)
|
|
|
|
res = PXE_PGP_EXPECT_PUBLIC_KEY;
|
2005-07-10 17:46:29 +04:00
|
|
|
else
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
res = process_secret_key(pkt, &pk, psw, psw_len);
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
2005-07-10 17:46:29 +04:00
|
|
|
case PGP_PKT_SIGNATURE:
|
|
|
|
case PGP_PKT_MARKER:
|
|
|
|
case PGP_PKT_TRUST:
|
|
|
|
case PGP_PKT_USER_ID:
|
|
|
|
case PGP_PKT_USER_ATTR:
|
|
|
|
case PGP_PKT_PRIV_61:
|
|
|
|
res = pgp_skip_packet(pkt);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
px_debug("unknown/unexpected packet: %d", tag);
|
|
|
|
res = PXE_PGP_UNEXPECTED_PKT;
|
|
|
|
}
|
|
|
|
pullf_free(pkt);
|
2005-10-15 06:49:52 +04:00
|
|
|
pkt = NULL;
|
2005-07-10 17:46:29 +04:00
|
|
|
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
if (pk != NULL)
|
|
|
|
{
|
|
|
|
if (res >= 0 && pk->can_encrypt)
|
|
|
|
{
|
|
|
|
if (enc_key == NULL)
|
|
|
|
{
|
|
|
|
enc_key = pk;
|
|
|
|
pk = NULL;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
res = PXE_PGP_MULTIPLE_SUBKEYS;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (pk)
|
|
|
|
pgp_key_free(pk);
|
|
|
|
pk = NULL;
|
|
|
|
}
|
2005-07-10 17:46:29 +04:00
|
|
|
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
if (res < 0)
|
2005-07-10 17:46:29 +04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (pkt)
|
|
|
|
pullf_free(pkt);
|
|
|
|
|
|
|
|
if (res < 0)
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
{
|
|
|
|
if (enc_key)
|
|
|
|
pgp_key_free(enc_key);
|
|
|
|
return res;
|
2005-07-10 17:46:29 +04:00
|
|
|
}
|
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
---------------------------------------------------------------------------
The SHLIB section was copy-pasted from somewhere and contains
several unnecessary libs. This cleans it up a bit.
-lcrypt
we don't use system crypt()
-lssl, -lssleay32
no SSL here
-lz in win32 section
already added on previous line
-ldes
The chance anybody has it is pretty low.
And the chance pgcrypto works with it is even lower.
Also trim the win32 section.
---------------------------------------------------------------------------
It is already disabled in Makefile, remove code too.
---------------------------------------------------------------------------
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
This brings major speedup to Elgamal encryption.
---------------------------------------------------------------------------
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
---------------------------------------------------------------------------
o Support for RSA encryption
o Big reorg to better separate generic and algorithm-specific code.
o Regression tests for RSA.
---------------------------------------------------------------------------
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
---------------------------------------------------------------------------
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
---------------------------------------------------------------------------
Marko Kreen
2005-08-13 06:06:21 +04:00
|
|
|
|
|
|
|
if (!enc_key)
|
|
|
|
res = PXE_PGP_NO_USABLE_KEY;
|
|
|
|
else
|
|
|
|
*pk_p = enc_key;
|
|
|
|
return res;
|
2005-07-10 17:46:29 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2005-10-15 06:49:52 +04:00
|
|
|
pgp_set_pubkey(PGP_Context * ctx, MBuf * keypkt,
|
|
|
|
const uint8 *key, int key_len, int pubtype)
|
2005-07-10 17:46:29 +04:00
|
|
|
{
|
2005-10-15 06:49:52 +04:00
|
|
|
int res;
|
2005-07-10 17:46:29 +04:00
|
|
|
PullFilter *src;
|
|
|
|
PGP_PubKey *pk = NULL;
|
|
|
|
|
|
|
|
res = pullf_create_mbuf_reader(&src, keypkt);
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
|
|
|
|
res = internal_read_key(src, &pk, key, key_len, pubtype);
|
|
|
|
pullf_free(src);
|
|
|
|
|
|
|
|
if (res >= 0)
|
|
|
|
ctx->pub_key = pk;
|
|
|
|
|
|
|
|
return res < 0 ? res : 0;
|
|
|
|
}
|