From b8f40ced2f7daebb13a3c9ab9aece5a969e2c28a Mon Sep 17 00:00:00 2001
From: Joe Conway <mail@joeconway.com>
Date: Sun, 30 Nov 2003 20:55:09 +0000
Subject: [PATCH] Make PQescapeBytea and byteaout consistent with each other,
and octal escape all octets outside the range 0x20 to 0x7e. This fixes the
problem pointed out by Sergey Yatskevich here:
http://archives.postgresql.org/pgsql-bugs/2003-11/msg00140.php
---
doc/src/sgml/datatype.sgml | 28 ++++++++++++++++++++--------
src/backend/utils/adt/varlena.c | 14 +++++++-------
src/interfaces/libpq/fe-exec.c | 9 +++++----
3 files changed, 32 insertions(+), 19 deletions(-)
diff --git a/doc/src/sgml/datatype.sgml b/doc/src/sgml/datatype.sgml
index d47cee96338..7707aed2a47 100644
--- a/doc/src/sgml/datatype.sgml
+++ b/doc/src/sgml/datatype.sgml
@@ -1,5 +1,5 @@
<!--
-$PostgreSQL: pgsql/doc/src/sgml/datatype.sgml,v 1.132 2003/11/29 19:51:36 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/datatype.sgml,v 1.133 2003/11/30 20:55:09 joe Exp $
-->
<chapter id="datatype">
@@ -1076,9 +1076,10 @@ SELECT b, char_length(b) FROM test2;
strings are distinguished from characters strings by two
characteristics: First, binary strings specifically allow storing
octets of value zero and other <quote>non-printable</quote>
- octets. Second, operations on binary strings process the actual
- bytes, whereas the encoding and processing of character strings
- depends on locale settings.
+ octets (defined as octets outside the range 32 to 126).
+ Second, operations on binary strings process the actual bytes,
+ whereas the encoding and processing of character strings depends
+ on locale settings.
</para>
<para>
@@ -1131,14 +1132,25 @@ SELECT b, char_length(b) FROM test2;
<entry><literal>\\</literal></entry>
</row>
+ <row>
+ <entry>0 to 31 and 127 to 255</entry>
+ <entry><quote>non-printable</quote> octets</entry>
+ <entry><literal>'\\<replaceable>xxx'</></literal> (octal value)</entry>
+ <entry><literal>SELECT '\\001'::bytea;</literal></entry>
+ <entry><literal>\001</literal></entry>
+ </row>
+
</tbody>
</tgroup>
</table>
<para>
- Note that the result in each of the examples in <xref linkend="datatype-binary-sqlesc"> was exactly one
- octet in length, even though the output representation of the zero
- octet and backslash are more than one character.
+ The requirement to escape <quote>non-printable</quote> octets actually
+ varies depending on locale settings. In some instances you can get away
+ with leaving them unescaped. Note that the result in each of the examples
+ in <xref linkend="datatype-binary-sqlesc"> was exactly one octet in
+ length, even though the output representation of the zero octet and
+ backslash are more than one character.
</para>
<para>
@@ -1206,7 +1218,7 @@ SELECT b, char_length(b) FROM test2;
<row>
<entry>32 to 126</entry>
<entry><quote>printable</quote> octets</entry>
- <entry>ASCII representation</entry>
+ <entry>client character set representation</entry>
<entry><literal>SELECT '\\176'::bytea;</literal></entry>
<entry><literal>~</literal></entry>
</row>
diff --git a/src/backend/utils/adt/varlena.c b/src/backend/utils/adt/varlena.c
index 244e2fbc698..8e01f9f539e 100644
--- a/src/backend/utils/adt/varlena.c
+++ b/src/backend/utils/adt/varlena.c
@@ -8,7 +8,7 @@
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/utils/adt/varlena.c,v 1.107 2003/11/29 19:51:59 pgsql Exp $
+ * $PostgreSQL: pgsql/src/backend/utils/adt/varlena.c,v 1.108 2003/11/30 20:55:09 joe Exp $
*
*-------------------------------------------------------------------------
*/
@@ -186,10 +186,10 @@ byteaout(PG_FUNCTION_ARGS)
{
if (*vp == '\\')
len += 2;
- else if (isprint((unsigned char) *vp))
- len++;
- else
+ else if ((unsigned char) *vp < 0x20 || (unsigned char) *vp > 0x7e)
len += 4;
+ else
+ len++;
}
rp = result = (char *) palloc(len);
vp = VARDATA(vlena);
@@ -200,9 +200,7 @@ byteaout(PG_FUNCTION_ARGS)
*rp++ = '\\';
*rp++ = '\\';
}
- else if (isprint((unsigned char) *vp))
- *rp++ = *vp;
- else
+ else if ((unsigned char) *vp < 0x20 || (unsigned char) *vp > 0x7e)
{
val = *vp;
rp[0] = '\\';
@@ -213,6 +211,8 @@ byteaout(PG_FUNCTION_ARGS)
rp[1] = DIG(val & 03);
rp += 4;
}
+ else
+ *rp++ = *vp;
}
*rp = '\0';
PG_RETURN_CSTRING(result);
diff --git a/src/interfaces/libpq/fe-exec.c b/src/interfaces/libpq/fe-exec.c
index 19cb840af71..84b7d2e1c19 100644
--- a/src/interfaces/libpq/fe-exec.c
+++ b/src/interfaces/libpq/fe-exec.c
@@ -8,7 +8,7 @@
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/interfaces/libpq/fe-exec.c,v 1.154 2003/11/29 19:52:11 pgsql Exp $
+ * $PostgreSQL: pgsql/src/interfaces/libpq/fe-exec.c,v 1.155 2003/11/30 20:55:09 joe Exp $
*
*-------------------------------------------------------------------------
*/
@@ -2261,7 +2261,8 @@ PQescapeString(char *to, const char *from, size_t length)
* '\0' == ASCII 0 == \\000
* '\'' == ASCII 39 == \'
* '\\' == ASCII 92 == \\\\
- * anything >= 0x80 ---> \\ooo (where ooo is an octal expression)
+ * anything < 0x20, or > 0x7e ---> \\ooo
+ * (where ooo is an octal expression)
*/
unsigned char *
PQescapeBytea(const unsigned char *bintext, size_t binlen, size_t *bytealen)
@@ -2280,7 +2281,7 @@ PQescapeBytea(const unsigned char *bintext, size_t binlen, size_t *bytealen)
vp = bintext;
for (i = binlen; i > 0; i--, vp++)
{
- if (*vp == 0 || *vp >= 0x80)
+ if (*vp < 0x20 || *vp > 0x7e)
len += 5; /* '5' is for '\\ooo' */
else if (*vp == '\'')
len += 2;
@@ -2299,7 +2300,7 @@ PQescapeBytea(const unsigned char *bintext, size_t binlen, size_t *bytealen)
for (i = binlen; i > 0; i--, vp++)
{
- if (*vp == 0 || *vp >= 0x80)
+ if (*vp < 0x20 || *vp > 0x7e)
{
(void) sprintf(rp, "\\\\%03o", *vp);
rp += 5;
--
GitLab