Support hex-string input and output for type BYTEA.

Both hex format and the traditional "escape" format are automatically
handled on input.  The output format is selected by the new GUC variable
bytea_output.

As committed, bytea_output defaults to HEX, which is an *incompatible
change*.  We will keep it this way for awhile for testing purposes, but
should consider whether to switch to the more backwards-compatible
default of ESCAPE before 8.5 is released.

Peter Eisentraut

src/backend/utils/adt/encode.c

@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/encode.c,v 1.23 2009/01/01 17:23:49 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/encode.c,v 1.24 2009/08/04 16:08:36 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -109,7 +109,7 @@ binary_decode(PG_FUNCTION_ARGS)
  * HEX
  */
 
-static const char *hextbl = "0123456789abcdef";
+static const char hextbl[] = "0123456789abcdef";
 
 static const int8 hexlookup[128] = {
 	-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
@@ -122,7 +122,7 @@ static const int8 hexlookup[128] = {
 	-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
 };
 
-static unsigned
+unsigned
 hex_encode(const char *src, unsigned len, char *dst)
 {
 	const char *end = src + len;
@@ -136,7 +136,7 @@ hex_encode(const char *src, unsigned len, char *dst)
 	return len * 2;
 }
 
-static char
+static inline char
 get_hex(char c)
 {
 	int			res = -1;
@@ -152,7 +152,7 @@ get_hex(char c)
 	return (char) res;
 }
 
-static unsigned
+unsigned
 hex_decode(const char *src, unsigned len, char *dst)
 {
 	const char *s,

src/backend/utils/adt/selfuncs.c

@@ -15,7 +15,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/selfuncs.c,v 1.261 2009/06/11 14:49:04 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/selfuncs.c,v 1.262 2009/08/04 16:08:36 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -109,6 +109,7 @@
 #include "parser/parse_coerce.h"
 #include "parser/parsetree.h"
 #include "utils/builtins.h"
+#include "utils/bytea.h"
 #include "utils/date.h"
 #include "utils/datum.h"
 #include "utils/fmgroids.h"

src/backend/utils/adt/varlena.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/varlena.c,v 1.171 2009/06/11 14:49:04 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/varlena.c,v 1.172 2009/08/04 16:08:36 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -24,10 +24,14 @@
 #include "parser/scansup.h"
 #include "regex/regex.h"
 #include "utils/builtins.h"
+#include "utils/bytea.h"
 #include "utils/lsyscache.h"
 #include "utils/pg_locale.h"
 
 
+/* GUC variable */
+int		bytea_output = BYTEA_OUTPUT_HEX;
+
 typedef struct varlena unknown;
 
 typedef struct
@@ -186,10 +190,24 @@ byteain(PG_FUNCTION_ARGS)
 	char	   *inputText = PG_GETARG_CSTRING(0);
 	char	   *tp;
 	char	   *rp;
-	int byte;
+	int			bc;
 	bytea	   *result;
 
-	for (byte = 0, tp = inputText; *tp != '\0'; byte ++)
+	/* Recognize hex input */
+	if (inputText[0] == '\\' && inputText[1] == 'x')
+	{
+		size_t len = strlen(inputText);
+
+		bc = (len - 2)/2 + VARHDRSZ;		/* maximum possible length */
+		result = palloc(bc);
+		bc = hex_decode(inputText + 2, len - 2, VARDATA(result));
+		SET_VARSIZE(result, bc + VARHDRSZ);	/* actual length */
+
+		PG_RETURN_BYTEA_P(result);
+	}
+
+	/* Else, it's the traditional escaped style */
+	for (bc = 0, tp = inputText; *tp != '\0'; bc++)
 	{
 		if (tp[0] != '\\')
 			tp++;
@@ -204,7 +222,7 @@ byteain(PG_FUNCTION_ARGS)
 		else
 		{
 			/*
-			 * one backslash, not followed by 0 or ### valid octal
+			 * one backslash, not followed by another or ### valid octal
 			 */
 			ereport(ERROR,
 					(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
@@ -212,10 +230,10 @@ byteain(PG_FUNCTION_ARGS)
 		}
 	}
 
-	byte	  +=VARHDRSZ;
+	bc += VARHDRSZ;
 
-	result = (bytea *) palloc(byte);
-	SET_VARSIZE(result, byte);
+	result = (bytea *) palloc(bc);
+	SET_VARSIZE(result, bc);
 
 	tp = inputText;
 	rp = VARDATA(result);
@@ -228,11 +246,11 @@ byteain(PG_FUNCTION_ARGS)
 				 (tp[2] >= '0' && tp[2] <= '7') &&
 				 (tp[3] >= '0' && tp[3] <= '7'))
 		{
-			byte = VAL(tp[1]);
-			byte	 <<=3;
-			byte	  +=VAL(tp[2]);
-			byte	 <<=3;
-			*rp++ = byte +VAL(tp[3]);
+			bc = VAL(tp[1]);
+			bc <<= 3;
+			bc += VAL(tp[2]);
+			bc <<= 3;
+			*rp++ = bc + VAL(tp[3]);
 
 			tp += 4;
 		}
@@ -259,21 +277,30 @@ byteain(PG_FUNCTION_ARGS)
 /*
  *		byteaout		- converts to printable representation of byte array
  *
- *		Non-printable characters are inserted as '\nnn' (octal) and '\' as
- *		'\\'.
- *
- *		NULL vlena should be an error--returning string with NULL for now.
+ *		In the traditional escaped format, non-printable characters are
+ *		printed as '\nnn' (octal) and '\' as '\\'.
  */
 Datum
 byteaout(PG_FUNCTION_ARGS)
 {
 	bytea	   *vlena = PG_GETARG_BYTEA_PP(0);
 	char	   *result;
-	char	   *vp;
 	char	   *rp;
-	int			val;			/* holds unprintable chars */
-	int			i;
+
+	if (bytea_output == BYTEA_OUTPUT_HEX)
+	{
+		/* Print hex format */
+		rp = result = palloc(VARSIZE_ANY_EXHDR(vlena) * 2 + 2 + 1);
+		*rp++ = '\\';
+		*rp++ = 'x';
+		rp += hex_encode(VARDATA_ANY(vlena), VARSIZE_ANY_EXHDR(vlena), rp);
+	}
+	else if (bytea_output == BYTEA_OUTPUT_ESCAPE)
+	{
+	/* Print traditional escaped format */
+	char	   *vp;
 	int			len;
+	int			i;
 
 	len = 1;					/* empty string has 1 char */
 	vp = VARDATA_ANY(vlena);
@@ -297,6 +324,8 @@ byteaout(PG_FUNCTION_ARGS)
 		}
 		else if ((unsigned char) *vp < 0x20 || (unsigned char) *vp > 0x7e)
 		{
+			int			val;			/* holds unprintable chars */
+
 			val = *vp;
 			rp[0] = '\\';
 			rp[3] = DIG(val & 07);
@@ -309,6 +338,13 @@ byteaout(PG_FUNCTION_ARGS)
 		else
 			*rp++ = *vp;
 	}
+	}
+	else
+	{
+		elog(ERROR, "unrecognized bytea_output setting: %d",
+			 bytea_output);
+		rp = result = NULL;		/* keep compiler quiet */
+	}
 	*rp = '\0';
 	PG_RETURN_CSTRING(result);
 }

src/backend/utils/misc/guc.c

@@ -10,7 +10,7 @@
  * Written by Peter Eisentraut <peter_e@gmx.net>.
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/misc/guc.c,v 1.509 2009/07/22 17:00:23 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/misc/guc.c,v 1.510 2009/08/04 16:08:36 tgl Exp $
  *
  *--------------------------------------------------------------------
  */
@@ -61,6 +61,7 @@
 #include "tcop/tcopprot.h"
 #include "tsearch/ts_cache.h"
 #include "utils/builtins.h"
+#include "utils/bytea.h"
 #include "utils/guc_tables.h"
 #include "utils/memutils.h"
 #include "utils/pg_locale.h"
@@ -180,6 +181,12 @@ static char *config_enum_get_options(struct config_enum * record,
  * NOTE! Option values may not contain double quotes!
  */
 
+static const struct config_enum_entry bytea_output_options[] = {
+	{"escape", BYTEA_OUTPUT_ESCAPE, false},
+	{"hex", BYTEA_OUTPUT_HEX, false},
+	{NULL, 0, false}
+};
+
 /*
  * We have different sets for client and server message level options because
  * they sort slightly different (see "log" level)
@@ -2540,6 +2547,15 @@ static struct config_enum ConfigureNamesEnum[] =
 		BACKSLASH_QUOTE_SAFE_ENCODING, backslash_quote_options, NULL, NULL
 	},
 
+	{
+		{"bytea_output", PGC_USERSET, CLIENT_CONN_STATEMENT,
+			gettext_noop("Sets the output format for bytea."),
+			NULL
+		},
+		&bytea_output,
+		BYTEA_OUTPUT_HEX, bytea_output_options, NULL, NULL
+	},
+
 	{
 		{"client_min_messages", PGC_USERSET, LOGGING_WHEN,
 			gettext_noop("Sets the message levels that are sent to the client."),

src/backend/utils/misc/postgresql.conf.sample

@@ -424,6 +424,7 @@
 #statement_timeout = 0			# in milliseconds, 0 is disabled
 #vacuum_freeze_min_age = 50000000
 #vacuum_freeze_table_age = 150000000
+#bytea_output = 'hex'			# hex, escape
 #xmlbinary = 'base64'
 #xmloption = 'content'