Add user-callable SHA-2 functions

Add the user-callable functions sha224, sha256, sha384, sha512.  We
already had these in the C code to support SCRAM, but there was no test
coverage outside of the SCRAM tests.  Adding these as user-callable
functions allows writing some tests.  Also, we have a user-callable md5
function but no more modern alternative, which led to wide use of md5 as
a general-purpose hash function, which leads to occasional complaints
about using md5.

Also mark the existing md5 functions as leak-proof.
Reviewed-by: Michael Paquier <michael@paquier.xyz>

doc/src/sgml/func.sgml

@@ -3640,7 +3640,7 @@ SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
        returning the result in hexadecimal
       </entry>
       <entry><literal>md5(E'Th\\000omas'::bytea)</literal></entry>
-      <entry><literal>8ab2d3c9689aaf18 b4958c334c82d8b1</literal></entry>
+      <entry><literal>8ab2d3c9689aaf18&#x200B;b4958c334c82d8b1</literal></entry>
      </row>
 
       <row>
@@ -3674,6 +3674,66 @@ SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
        <entry><literal>set_byte(E'Th\\000omas'::bytea, 4, 64)</literal></entry>
        <entry><literal>Th\000o@as</literal></entry>
       </row>
+
+      <row>
+       <entry>
+        <indexterm>
+         <primary>sha224</primary>
+        </indexterm>
+        <literal><function>sha224(<type>bytea</type>)</function></literal>
+       </entry>
+       <entry><type>bytea</type></entry>
+       <entry>
+        SHA-224 hash
+       </entry>
+       <entry><literal>sha224('abc')</literal></entry>
+       <entry><literal>\x23097d223405d8228642a477bda2&#x200B;55b32aadbce4bda0b3f7e36c9da7</literal></entry>
+      </row>
+
+      <row>
+       <entry>
+        <indexterm>
+         <primary>sha256</primary>
+        </indexterm>
+        <literal><function>sha256(<type>bytea</type>)</function></literal>
+       </entry>
+       <entry><type>bytea</type></entry>
+       <entry>
+        SHA-256 hash
+       </entry>
+       <entry><literal>sha256('abc')</literal></entry>
+       <entry><literal>\xba7816bf8f01cfea414140de5dae2223&#x200B;b00361a396177a9cb410ff61f20015ad</literal></entry>
+      </row>
+
+      <row>
+       <entry>
+        <indexterm>
+         <primary>sha384</primary>
+        </indexterm>
+        <literal><function>sha384(<type>bytea</type>)</function></literal>
+       </entry>
+       <entry><type>bytea</type></entry>
+       <entry>
+        SHA-384 hash
+       </entry>
+       <entry><literal>sha384('abc')</literal></entry>
+       <entry><literal>\xcb00753f45a35e8bb5a03d699ac65007&#x200B;272c32ab0eded1631a8b605a43ff5bed&#x200B;8086072ba1e7cc2358baeca134c825a7</literal></entry>
+      </row>
+
+      <row>
+       <entry>
+        <indexterm>
+         <primary>sha512</primary>
+        </indexterm>
+        <literal><function>sha512(<type>bytea</type>)</function></literal>
+       </entry>
+       <entry><type>bytea</type></entry>
+       <entry>
+        SHA-512 hash
+       </entry>
+       <entry><literal>sha512('abc')</literal></entry>
+       <entry><literal>\xddaf35a193617abacc417349ae204131&#x200B;12e6fa4e89a97ea20a9eeee64b55d39a&#x200B;2192992a274fc1a836ba3c23a3feebbd&#x200B;454d4423643ce80e2a9ac94fa54ca49f</literal></entry>
+      </row>
     </tbody>
    </tgroup>
   </table>
@@ -3686,6 +3746,15 @@ SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
    the first byte, and bit 15 is the most significant bit of the second byte.
   </para>
 
+  <para>
+   Note that for historic reasons, the function <function>md5</function>
+   returns a hex-encoded value of type <type>text</type> whereas the SHA-2
+   functions return type <type>bytea</type>.  Use the functions
+   <function>encode</function> and <function>decode</function> to convert
+   between the two, for example <literal>encode(sha256('abc'),
+   'hex')</literal> to get a hex-encoded text representation.
+  </para>
+
   <para>
    See also the aggregate function <function>string_agg</function> in
    <xref linkend="functions-aggregate"/> and the large object functions

src/backend/utils/adt/Makefile

@@ -11,7 +11,8 @@ include $(top_builddir)/src/Makefile.global
 # keep this list arranged alphabetically or it gets to be a mess
 OBJS = acl.o amutils.o arrayfuncs.o array_expanded.o array_selfuncs.o \
 	array_typanalyze.o array_userfuncs.o arrayutils.o ascii.o \
-	bool.o cash.o char.o date.o datetime.o datum.o dbsize.o domains.o \
+	bool.o cash.o char.o cryptohashes.o \
+	date.o datetime.o datum.o dbsize.o domains.o \
 	encode.o enum.o expandeddatum.o expandedrecord.o \
 	float.o format_type.o formatting.o genfile.o \
 	geo_ops.o geo_selfuncs.o geo_spgist.o inet_cidr_ntop.o inet_net_pton.o \

src/backend/utils/adt/cryptohashes.c

+/*-------------------------------------------------------------------------
+ *
+ * cryptohashes.c
+ *	  Cryptographic hash functions
+ *
+ * Portions Copyright (c) 2018, PostgreSQL Global Development Group
+ *
+ *
+ * IDENTIFICATION
+ *	  src/backend/utils/adt/cryptohashes.c
+ *
+ *-------------------------------------------------------------------------
+ */
+#include "postgres.h"
+
+#include "common/md5.h"
+#include "common/sha2.h"
+#include "utils/builtins.h"
+
+
+/*
+ * MD5
+ */
+
+/* MD5 produces a 16 byte (128 bit) hash; double it for hex */
+#define MD5_HASH_LEN  32
+
+/*
+ * Create an MD5 hash of a text value and return it as hex string.
+ */
+Datum
+md5_text(PG_FUNCTION_ARGS)
+{
+	text	   *in_text = PG_GETARG_TEXT_PP(0);
+	size_t		len;
+	char		hexsum[MD5_HASH_LEN + 1];
+
+	/* Calculate the length of the buffer using varlena metadata */
+	len = VARSIZE_ANY_EXHDR(in_text);
+
+	/* get the hash result */
+	if (pg_md5_hash(VARDATA_ANY(in_text), len, hexsum) == false)
+		ereport(ERROR,
+				(errcode(ERRCODE_OUT_OF_MEMORY),
+				 errmsg("out of memory")));
+
+	/* convert to text and return it */
+	PG_RETURN_TEXT_P(cstring_to_text(hexsum));
+}
+
+/*
+ * Create an MD5 hash of a bytea value and return it as a hex string.
+ */
+Datum
+md5_bytea(PG_FUNCTION_ARGS)
+{
+	bytea	   *in = PG_GETARG_BYTEA_PP(0);
+	size_t		len;
+	char		hexsum[MD5_HASH_LEN + 1];
+
+	len = VARSIZE_ANY_EXHDR(in);
+	if (pg_md5_hash(VARDATA_ANY(in), len, hexsum) == false)
+		ereport(ERROR,
+				(errcode(ERRCODE_OUT_OF_MEMORY),
+				 errmsg("out of memory")));
+
+	PG_RETURN_TEXT_P(cstring_to_text(hexsum));
+}
+
+
+/*
+ * SHA-2 variants
+ */
+
+Datum
+sha224_bytea(PG_FUNCTION_ARGS)
+{
+	bytea	   *in = PG_GETARG_BYTEA_PP(0);
+	const uint8 *data;
+	size_t		len;
+	pg_sha224_ctx ctx;
+	unsigned char buf[PG_SHA224_DIGEST_LENGTH];
+	bytea	   *result;
+
+	len = VARSIZE_ANY_EXHDR(in);
+	data = (unsigned char *) VARDATA_ANY(in);
+
+	pg_sha224_init(&ctx);
+	pg_sha224_update(&ctx, data, len);
+	pg_sha224_final(&ctx, buf);
+
+	result = palloc(sizeof(buf) + VARHDRSZ);
+	SET_VARSIZE(result, sizeof(buf) + VARHDRSZ);
+	memcpy(VARDATA(result), buf, sizeof(buf));
+
+	PG_RETURN_BYTEA_P(result);
+}
+
+Datum
+sha256_bytea(PG_FUNCTION_ARGS)
+{
+	bytea	   *in = PG_GETARG_BYTEA_PP(0);
+	const uint8 *data;
+	size_t		len;
+	pg_sha256_ctx ctx;
+	unsigned char buf[PG_SHA256_DIGEST_LENGTH];
+	bytea	   *result;
+
+	len = VARSIZE_ANY_EXHDR(in);
+	data = (unsigned char *) VARDATA_ANY(in);
+
+	pg_sha256_init(&ctx);
+	pg_sha256_update(&ctx, data, len);
+	pg_sha256_final(&ctx, buf);
+
+	result = palloc(sizeof(buf) + VARHDRSZ);
+	SET_VARSIZE(result, sizeof(buf) + VARHDRSZ);
+	memcpy(VARDATA(result), buf, sizeof(buf));
+
+	PG_RETURN_BYTEA_P(result);
+}
+
+Datum
+sha384_bytea(PG_FUNCTION_ARGS)
+{
+	bytea	   *in = PG_GETARG_BYTEA_PP(0);
+	const uint8 *data;
+	size_t		len;
+	pg_sha384_ctx ctx;
+	unsigned char buf[PG_SHA384_DIGEST_LENGTH];
+	bytea	   *result;
+
+	len = VARSIZE_ANY_EXHDR(in);
+	data = (unsigned char *) VARDATA_ANY(in);
+
+	pg_sha384_init(&ctx);
+	pg_sha384_update(&ctx, data, len);
+	pg_sha384_final(&ctx, buf);
+
+	result = palloc(sizeof(buf) + VARHDRSZ);
+	SET_VARSIZE(result, sizeof(buf) + VARHDRSZ);
+	memcpy(VARDATA(result), buf, sizeof(buf));
+
+	PG_RETURN_BYTEA_P(result);
+}
+
+Datum
+sha512_bytea(PG_FUNCTION_ARGS)
+{
+	bytea	   *in = PG_GETARG_BYTEA_PP(0);
+	const uint8 *data;
+	size_t		len;
+	pg_sha512_ctx ctx;
+	unsigned char buf[PG_SHA512_DIGEST_LENGTH];
+	bytea	   *result;
+
+	len = VARSIZE_ANY_EXHDR(in);
+	data = (unsigned char *) VARDATA_ANY(in);
+
+	pg_sha512_init(&ctx);
+	pg_sha512_update(&ctx, data, len);
+	pg_sha512_final(&ctx, buf);
+
+	result = palloc(sizeof(buf) + VARHDRSZ);
+	SET_VARSIZE(result, sizeof(buf) + VARHDRSZ);
+	memcpy(VARDATA(result), buf, sizeof(buf));
+
+	PG_RETURN_BYTEA_P(result);
+}

src/backend/utils/adt/varlena.c

@@ -22,7 +22,6 @@
 #include "catalog/pg_collation.h"
 #include "catalog/pg_type.h"
 #include "common/int.h"
-#include "common/md5.h"
 #include "lib/hyperloglog.h"
 #include "libpq/pqformat.h"
 #include "miscadmin.h"
@@ -4564,53 +4563,6 @@ to_hex64(PG_FUNCTION_ARGS)
 	PG_RETURN_TEXT_P(cstring_to_text(ptr));
 }
 
-/*
- * Create an md5 hash of a text string and return it as hex
- *
- * md5 produces a 16 byte (128 bit) hash; double it for hex
- */
-#define MD5_HASH_LEN  32
-
-Datum
-md5_text(PG_FUNCTION_ARGS)
-{
-	text	   *in_text = PG_GETARG_TEXT_PP(0);
-	size_t		len;
-	char		hexsum[MD5_HASH_LEN + 1];
-
-	/* Calculate the length of the buffer using varlena metadata */
-	len = VARSIZE_ANY_EXHDR(in_text);
-
-	/* get the hash result */
-	if (pg_md5_hash(VARDATA_ANY(in_text), len, hexsum) == false)
-		ereport(ERROR,
-				(errcode(ERRCODE_OUT_OF_MEMORY),
-				 errmsg("out of memory")));
-
-	/* convert to text and return it */
-	PG_RETURN_TEXT_P(cstring_to_text(hexsum));
-}
-
-/*
- * Create an md5 hash of a bytea field and return it as a hex string:
- * 16-byte md5 digest is represented in 32 hex characters.
- */
-Datum
-md5_bytea(PG_FUNCTION_ARGS)
-{
-	bytea	   *in = PG_GETARG_BYTEA_PP(0);
-	size_t		len;
-	char		hexsum[MD5_HASH_LEN + 1];
-
-	len = VARSIZE_ANY_EXHDR(in);
-	if (pg_md5_hash(VARDATA_ANY(in), len, hexsum) == false)
-		ereport(ERROR,
-				(errcode(ERRCODE_OUT_OF_MEMORY),
-				 errmsg("out of memory")));
-
-	PG_RETURN_TEXT_P(cstring_to_text(hexsum));
-}
-
 /*
  * Return the size of a datum, possibly compressed
  *

src/include/catalog/pg_proc.h

@@ -3917,10 +3917,18 @@ DATA(insert OID = 3314 (  system			PGNSP PGUID 12 1 0 0 0 f f f f t f v s 1 0 33
 DESCR("SYSTEM tablesample method handler");
 
 /* cryptographic */
-DATA(insert OID =  2311 (  md5	   PGNSP PGUID 12 1 0 0 0 f f f f t f i s 1 0 25 "25" _null_ _null_ _null_ _null_ _null_ md5_text _null_ _null_ _null_ ));
+DATA(insert OID =  2311 (  md5	   PGNSP PGUID 12 1 0 0 0 f f f t t f i s 1 0 25 "25" _null_ _null_ _null_ _null_ _null_ md5_text _null_ _null_ _null_ ));
 DESCR("MD5 hash");
-DATA(insert OID =  2321 (  md5	   PGNSP PGUID 12 1 0 0 0 f f f f t f i s 1 0 25 "17" _null_ _null_ _null_ _null_ _null_ md5_bytea _null_ _null_ _null_ ));
+DATA(insert OID =  2321 (  md5	   PGNSP PGUID 12 1 0 0 0 f f f t t f i s 1 0 25 "17" _null_ _null_ _null_ _null_ _null_ md5_bytea _null_ _null_ _null_ ));
 DESCR("MD5 hash");
+DATA(insert OID =  3419 (  sha224  PGNSP PGUID 12 1 0 0 0 f f f t t f i s 1 0 17 "17" _null_ _null_ _null_ _null_ _null_ sha224_bytea _null_ _null_ _null_ ));
+DESCR("SHA-224 hash");
+DATA(insert OID =  3420 (  sha256  PGNSP PGUID 12 1 0 0 0 f f f t t f i s 1 0 17 "17" _null_ _null_ _null_ _null_ _null_ sha256_bytea _null_ _null_ _null_ ));
+DESCR("SHA-256 hash");
+DATA(insert OID =  3421 (  sha384  PGNSP PGUID 12 1 0 0 0 f f f t t f i s 1 0 17 "17" _null_ _null_ _null_ _null_ _null_ sha384_bytea _null_ _null_ _null_ ));
+DESCR("SHA-384 hash");
+DATA(insert OID =  3422 (  sha512  PGNSP PGUID 12 1 0 0 0 f f f t t f i s 1 0 17 "17" _null_ _null_ _null_ _null_ _null_ sha512_bytea _null_ _null_ _null_ ));
+DESCR("SHA-512 hash");
 
 /* crosstype operations for date vs. timestamp and timestamptz */
 DATA(insert OID = 2338 (  date_lt_timestamp		   PGNSP PGUID 12 1 0 0 0 f f f f t f i s 2 0 16 "1082 1114" _null_ _null_ _null_ _null_ _null_ date_lt_timestamp _null_ _null_ _null_ ));

src/test/regress/expected/opr_sanity.out

@@ -699,6 +699,8 @@ timestamp_lt(timestamp without time zone,timestamp without time zone)
 timestamp_le(timestamp without time zone,timestamp without time zone)
 timestamp_ge(timestamp without time zone,timestamp without time zone)
 timestamp_gt(timestamp without time zone,timestamp without time zone)
+md5(text)
+md5(bytea)
 tidgt(tid,tid)
 tidlt(tid,tid)
 tidge(tid,tid)
@@ -711,6 +713,10 @@ uuid_gt(uuid,uuid)
 uuid_ne(uuid,uuid)
 xidneq(xid,xid)
 xidneqint4(xid,integer)
+sha224(bytea)
+sha256(bytea)
+sha384(bytea)
+sha512(bytea)
 macaddr8_eq(macaddr8,macaddr8)
 macaddr8_lt(macaddr8,macaddr8)
 macaddr8_le(macaddr8,macaddr8)

src/test/regress/expected/strings.out

@@ -1439,6 +1439,58 @@ select md5('12345678901234567890123456789012345678901234567890123456789012345678
  t
 (1 row)
 
+--
+-- SHA-2
+--
+SET bytea_output TO hex;
+SELECT sha224('');
+                           sha224                           
+------------------------------------------------------------
+ \xd14a028c2a3a2bc9476102bb288234c415a2b01f828ea62ac5b3e42f
+(1 row)
+
+SELECT sha224('The quick brown fox jumps over the lazy dog.');
+                           sha224                           
+------------------------------------------------------------
+ \x619cba8e8e05826e9b8c519c0a5c68f4fb653e8a3d8aa04bb2c8cd4c
+(1 row)
+
+SELECT sha256('');
+                               sha256                               
+--------------------------------------------------------------------
+ \xe3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
+(1 row)
+
+SELECT sha256('The quick brown fox jumps over the lazy dog.');
+                               sha256                               
+--------------------------------------------------------------------
+ \xef537f25c895bfa782526529a9b63d97aa631564d5d789c2b765448c8635fb6c
+(1 row)
+
+SELECT sha384('');
+                                               sha384                                               
+----------------------------------------------------------------------------------------------------
+ \x38b060a751ac96384cd9327eb1b1e36a21fdb71114be07434c0cc7bf63f6e1da274edebfe76f65fbd51ad2f14898b95b
+(1 row)
+
+SELECT sha384('The quick brown fox jumps over the lazy dog.');
+                                               sha384                                               
+----------------------------------------------------------------------------------------------------
+ \xed892481d8272ca6df370bf706e4d7bc1b5739fa2177aae6c50e946678718fc67a7af2819a021c2fc34e91bdb63409d7
+(1 row)
+
+SELECT sha512('');
+                                                               sha512                                                               
+------------------------------------------------------------------------------------------------------------------------------------
+ \xcf83e1357eefb8bdf1542850d66d8007d620e4050b5715dc83f4a921d36ce9ce47d0d13c5d85f2b0ff8318d2877eec2f63b931bd47417a81a538327af927da3e
+(1 row)
+
+SELECT sha512('The quick brown fox jumps over the lazy dog.');
+                                                               sha512                                                               
+------------------------------------------------------------------------------------------------------------------------------------
+ \x91ea1245f20d46ae9a037a989f54f1f790f0a47607eeb8a14d12890cea77a1bbc6c7ed9cf205e67b7f2b8fd4c7dfd3a7a8617e45f3c463d481c7e586c39ac1ed
+(1 row)
+
 --
 -- test behavior of escape_string_warning and standard_conforming_strings options
 --
@@ -1525,6 +1577,7 @@ select 'a\\bcd' as f1, 'a\\b\'cd' as f2, 'a\\b\'''cd' as f3, 'abcd\\'   as f4, '
 --
 -- Additional string functions
 --
+SET bytea_output TO escape;
 SELECT initcap('hi THOMAS');
   initcap  
 -----------

src/test/regress/sql/strings.sql

@@ -506,6 +506,23 @@ select md5('ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789'::byt
 
 select md5('12345678901234567890123456789012345678901234567890123456789012345678901234567890'::bytea) = '57edf4a22be3c955ac49da2e2107b67a' AS "TRUE";
 
+--
+-- SHA-2
+--
+SET bytea_output TO hex;
+
+SELECT sha224('');
+SELECT sha224('The quick brown fox jumps over the lazy dog.');
+
+SELECT sha256('');
+SELECT sha256('The quick brown fox jumps over the lazy dog.');
+
+SELECT sha384('');
+SELECT sha384('The quick brown fox jumps over the lazy dog.');
+
+SELECT sha512('');
+SELECT sha512('The quick brown fox jumps over the lazy dog.');
+
 --
 -- test behavior of escape_string_warning and standard_conforming_strings options
 --
@@ -540,6 +557,7 @@ select 'a\\bcd' as f1, 'a\\b\'cd' as f2, 'a\\b\'''cd' as f3, 'abcd\\'   as f4, '
 --
 -- Additional string functions
 --
+SET bytea_output TO escape;
 
 SELECT initcap('hi THOMAS');