Fix incautious handling of possibly-miscoded strings in client code.

An incorrectly-encoded multibyte character near the end of a string
could cause various processing loops to run past the string's
terminating NUL, with results ranging from no detectable issue to
a program crash, depending on what happens to be in the following
memory.

This isn't an issue in the server, because we take care to verify
the encoding of strings before doing any interesting processing
on them.  However, that lack of care leaked into client-side code
which shouldn't assume that anyone has validated the encoding of
its input.

Although this is certainly a bug worth fixing, the PG security team
elected not to regard it as a security issue, primarily because
any untrusted text should be sanitized by PQescapeLiteral or
the like before being incorporated into a SQL or psql command.
(If an app fails to do so, the same technique can be used to
cause SQL injection, with probably much more dire consequences
than a mere client-program crash.)  Those functions were already
made proof against this class of problem, cf CVE-2006-2313.

To fix, invent PQmblenBounded() which is like PQmblen() except it
won't return more than the number of bytes remaining in the string.
In HEAD we can make this a new libpq function, as PQmblen() is.
It seems imprudent to change libpq's API in stable branches though,
so in the back branches define PQmblenBounded as a macro in the files
that need it.  (Note that just changing PQmblen's behavior would not
be a good idea; notably, it would completely break the escaping
functions' defense against this exact problem.  So we just want a
version for those callers that don't have any better way of handling
this issue.)

Per private report from houjingyi.  Back-patch to all supported branches.

Author: tglsfdc

src/bin/psql/common.c

@@ -1846,7 +1846,7 @@ skip_white_space(const char *query)
 
 	while (*query)
 	{
-		int			mblen = PQmblen(query, pset.encoding);
+		int			mblen = PQmblenBounded(query, pset.encoding);
 
 		/*
 		 * Note: we assume the encoding is a superset of ASCII, so that for
@@ -1883,7 +1883,7 @@ skip_white_space(const char *query)
 					query++;
 					break;
 				}
-				query += PQmblen(query, pset.encoding);
+				query += PQmblenBounded(query, pset.encoding);
 			}
 		}
 		else if (cnestlevel > 0)
@@ -1918,7 +1918,7 @@ command_no_begin(const char *query)
 	 */
 	wordlen = 0;
 	while (isalpha((unsigned char) query[wordlen]))
-		wordlen += PQmblen(&query[wordlen], pset.encoding);
+		wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 	/*
 	 * Transaction control commands.  These should include every keyword that
@@ -1949,7 +1949,7 @@ command_no_begin(const char *query)
 
 		wordlen = 0;
 		while (isalpha((unsigned char) query[wordlen]))
-			wordlen += PQmblen(&query[wordlen], pset.encoding);
+			wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 		if (wordlen == 11 && pg_strncasecmp(query, "transaction", 11) == 0)
 			return true;
@@ -1983,7 +1983,7 @@ command_no_begin(const char *query)
 
 		wordlen = 0;
 		while (isalpha((unsigned char) query[wordlen]))
-			wordlen += PQmblen(&query[wordlen], pset.encoding);
+			wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 		if (wordlen == 8 && pg_strncasecmp(query, "database", 8) == 0)
 			return true;
@@ -1999,7 +1999,7 @@ command_no_begin(const char *query)
 
 			wordlen = 0;
 			while (isalpha((unsigned char) query[wordlen]))
-				wordlen += PQmblen(&query[wordlen], pset.encoding);
+				wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 		}
 
 		if (wordlen == 5 && pg_strncasecmp(query, "index", 5) == 0)
@@ -2010,7 +2010,7 @@ command_no_begin(const char *query)
 
 			wordlen = 0;
 			while (isalpha((unsigned char) query[wordlen]))
-				wordlen += PQmblen(&query[wordlen], pset.encoding);
+				wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 			if (wordlen == 12 && pg_strncasecmp(query, "concurrently", 12) == 0)
 				return true;
@@ -2027,7 +2027,7 @@ command_no_begin(const char *query)
 
 		wordlen = 0;
 		while (isalpha((unsigned char) query[wordlen]))
-			wordlen += PQmblen(&query[wordlen], pset.encoding);
+			wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 		/* ALTER SYSTEM isn't allowed in xacts */
 		if (wordlen == 6 && pg_strncasecmp(query, "system", 6) == 0)
@@ -2050,7 +2050,7 @@ command_no_begin(const char *query)
 
 		wordlen = 0;
 		while (isalpha((unsigned char) query[wordlen]))
-			wordlen += PQmblen(&query[wordlen], pset.encoding);
+			wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 		if (wordlen == 8 && pg_strncasecmp(query, "database", 8) == 0)
 			return true;
@@ -2065,7 +2065,7 @@ command_no_begin(const char *query)
 			query = skip_white_space(query);
 			wordlen = 0;
 			while (isalpha((unsigned char) query[wordlen]))
-				wordlen += PQmblen(&query[wordlen], pset.encoding);
+				wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 			/*
 			 * REINDEX [ TABLE | INDEX ] CONCURRENTLY are not allowed in
@@ -2084,7 +2084,7 @@ command_no_begin(const char *query)
 
 			wordlen = 0;
 			while (isalpha((unsigned char) query[wordlen]))
-				wordlen += PQmblen(&query[wordlen], pset.encoding);
+				wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 			if (wordlen == 12 && pg_strncasecmp(query, "concurrently", 12) == 0)
 				return true;
@@ -2104,7 +2104,7 @@ command_no_begin(const char *query)
 
 		wordlen = 0;
 		while (isalpha((unsigned char) query[wordlen]))
-			wordlen += PQmblen(&query[wordlen], pset.encoding);
+			wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 		if (wordlen == 3 && pg_strncasecmp(query, "all", 3) == 0)
 			return true;
@@ -2140,7 +2140,7 @@ is_select_command(const char *query)
 	 */
 	wordlen = 0;
 	while (isalpha((unsigned char) query[wordlen]))
-		wordlen += PQmblen(&query[wordlen], pset.encoding);
+		wordlen += PQmblenBounded(&query[wordlen], pset.encoding);
 
 	if (wordlen == 6 && pg_strncasecmp(query, "select", 6) == 0)
 		return true;

src/bin/psql/psqlscanslash.l

@@ -753,7 +753,7 @@ dequote_downcase_identifier(char *str, bool downcase, int encoding)
 		{
 			if (downcase && !inquotes)
 				*cp = pg_tolower((unsigned char) *cp);
-			cp += PQmblen(cp, encoding);
+			cp += PQmblenBounded(cp, encoding);
 		}
 	}
 }

src/bin/psql/stringutils.c

@@ -143,7 +143,7 @@ strtokx(const char *s,
 		/* okay, we have a quoted token, now scan for the closer */
 		char		thisquote = *p++;
 
-		for (; *p; p += PQmblen(p, encoding))
+		for (; *p; p += PQmblenBounded(p, encoding))
 		{
 			if (*p == escape && p[1] != '\0')
 				p++;			/* process escaped anything */
@@ -262,7 +262,7 @@ strip_quotes(char *source, char quote, char escape, int encoding)
 		else if (c == escape && src[1] != '\0')
 			src++;				/* process escaped character */
 
-		i = PQmblen(src, encoding);
+		i = PQmblenBounded(src, encoding);
 		while (i--)
 			*dst++ = *src++;
 	}
@@ -324,7 +324,7 @@ quote_if_needed(const char *source, const char *entails_quote,
 		else if (strchr(entails_quote, c))
 			need_quotes = true;
 
-		i = PQmblen(src, encoding);
+		i = PQmblenBounded(src, encoding);
 		while (i--)
 			*dst++ = *src++;
 	}

src/bin/psql/tab-complete.c

@@ -4397,7 +4397,7 @@ _complete_from_query(const char *simple_query,
 		while (*pstr)
 		{
 			char_length++;
-			pstr += PQmblen(pstr, pset.encoding);
+			pstr += PQmblenBounded(pstr, pset.encoding);
 		}
 
 		/* Free any prior result */

src/bin/scripts/common.c

@@ -52,7 +52,7 @@ splitTableColumnsSpec(const char *spec, int encoding,
 			cp++;
 		}
 		else
-			cp += PQmblen(cp, encoding);
+			cp += PQmblenBounded(cp, encoding);
 	}
 	*table = pnstrdup(spec, cp - spec);
 	*columns = cp;

src/common/jsonapi.c

@@ -740,7 +740,7 @@ json_lex_string(JsonLexContext *lex)
 						ch = (ch * 16) + (*s - 'A') + 10;
 					else
 					{
-						lex->token_terminator = s + pg_encoding_mblen(lex->input_encoding, s);
+						lex->token_terminator = s + pg_encoding_mblen_bounded(lex->input_encoding, s);
 						return JSON_UNICODE_ESCAPE_FORMAT;
 					}
 				}
@@ -846,7 +846,7 @@ json_lex_string(JsonLexContext *lex)
 					default:
 						/* Not a valid string escape, so signal error. */
 						lex->token_start = s;
-						lex->token_terminator = s + pg_encoding_mblen(lex->input_encoding, s);
+						lex->token_terminator = s + pg_encoding_mblen_bounded(lex->input_encoding, s);
 						return JSON_ESCAPING_INVALID;
 				}
 			}
@@ -860,7 +860,7 @@ json_lex_string(JsonLexContext *lex)
 				 * shown it's not a performance win.
 				 */
 				lex->token_start = s;
-				lex->token_terminator = s + pg_encoding_mblen(lex->input_encoding, s);
+				lex->token_terminator = s + pg_encoding_mblen_bounded(lex->input_encoding, s);
 				return JSON_ESCAPING_INVALID;
 			}
 

src/common/wchar.c

@@ -1911,6 +1911,11 @@ const pg_wchar_tbl pg_wchar_table[] = {
 
 /*
  * Returns the byte length of a multibyte character.
+ *
+ * Caution: when dealing with text that is not certainly valid in the
+ * specified encoding, the result may exceed the actual remaining
+ * string length.  Callers that are not prepared to deal with that
+ * should use pg_encoding_mblen_bounded() instead.
  */
 int
 pg_encoding_mblen(int encoding, const char *mbstr)
@@ -1920,6 +1925,16 @@ pg_encoding_mblen(int encoding, const char *mbstr)
 			pg_wchar_table[PG_SQL_ASCII].mblen((const unsigned char *) mbstr));
 }
 
+/*
+ * Returns the byte length of a multibyte character; but not more than
+ * the distance to end of string.
+ */
+int
+pg_encoding_mblen_bounded(int encoding, const char *mbstr)
+{
+	return strnlen(mbstr, pg_encoding_mblen(encoding, mbstr));
+}
+
 /*
  * Returns the display length of a multibyte character.
  */

src/fe_utils/print.c

@@ -3636,6 +3636,9 @@ strlen_max_width(unsigned char *str, int *target_width, int encoding)
 		curr_width += char_width;
 
 		str += PQmblen((char *) str, encoding);
+
+		if (str > end)			/* Don't overrun invalid string */
+			str = end;
 	}
 
 	*target_width = curr_width;

src/fe_utils/string_utils.c

@@ -1072,12 +1072,9 @@ patternToSQLRegex(int encoding, PQExpBuffer dbnamebuf, PQExpBuffer schemabuf,
 				appendPQExpBufferChar(curbuf, '\\');
 			else if (ch == '[' && cp[1] == ']')
 				appendPQExpBufferChar(curbuf, '\\');
-			i = PQmblen(cp, encoding);
-			while (i-- && *cp)
-			{
-				appendPQExpBufferChar(curbuf, *cp);
-				cp++;
-			}
+			i = PQmblenBounded(cp, encoding);
+			while (i--)
+				appendPQExpBufferChar(curbuf, *cp++);
 		}
 	}
 	appendPQExpBufferStr(curbuf, ")$");

src/include/mb/pg_wchar.h

@@ -574,6 +574,7 @@ extern int	pg_valid_server_encoding_id(int encoding);
  * earlier in this file are also available from libpgcommon.
  */
 extern int	pg_encoding_mblen(int encoding, const char *mbstr);
+extern int	pg_encoding_mblen_bounded(int encoding, const char *mbstr);
 extern int	pg_encoding_dsplen(int encoding, const char *mbstr);
 extern int	pg_encoding_verifymbchar(int encoding, const char *mbstr, int len);
 extern int	pg_encoding_verifymbstr(int encoding, const char *mbstr, int len);

src/interfaces/libpq/exports.txt

@@ -184,3 +184,4 @@ PQexitPipelineMode        181
 PQpipelineSync            182
 PQpipelineStatus          183
 PQtraceSetFlags           184
+PQmblenBounded              185

src/interfaces/libpq/fe-misc.c

@@ -1180,8 +1180,13 @@ pqSocketPoll(int sock, int forRead, int forWrite, time_t end_time)
  */
 
 /*
- * returns the byte length of the character beginning at s, using the
+ * Returns the byte length of the character beginning at s, using the
  * specified encoding.
+ *
+ * Caution: when dealing with text that is not certainly valid in the
+ * specified encoding, the result may exceed the actual remaining
+ * string length.  Callers that are not prepared to deal with that
+ * should use PQmblenBounded() instead.
  */
 int
 PQmblen(const char *s, int encoding)
@@ -1190,7 +1195,17 @@ PQmblen(const char *s, int encoding)
 }
 
 /*
- * returns the display length of the character beginning at s, using the
+ * Returns the byte length of the character beginning at s, using the
+ * specified encoding; but not more than the distance to end of string.
+ */
+int
+PQmblenBounded(const char *s, int encoding)
+{
+	return strnlen(s, pg_encoding_mblen(encoding, s));
+}
+
+/*
+ * Returns the display length of the character beginning at s, using the
  * specified encoding.
  */
 int

src/interfaces/libpq/fe-print.c

@@ -365,7 +365,7 @@ do_field(const PQprintOpt *po, const PGresult *res,
 			/* Detect whether field contains non-numeric data */
 			char		ch = '0';
 
-			for (p = pval; *p; p += PQmblen(p, res->client_encoding))
+			for (p = pval; *p; p += PQmblenBounded(p, res->client_encoding))
 			{
 				ch = *p;
 				if (!((ch >= '0' && ch <= '9') ||

src/interfaces/libpq/fe-protocol3.c

@@ -1296,7 +1296,7 @@ reportErrorPosition(PQExpBuffer msg, const char *query, int loc, int encoding)
 			if (w <= 0)
 				w = 1;
 			scroffset += w;
-			qoffset += pg_encoding_mblen(encoding, &wquery[qoffset]);
+			qoffset += PQmblenBounded(&wquery[qoffset], encoding);
 		}
 		else
 		{
@@ -1364,7 +1364,7 @@ reportErrorPosition(PQExpBuffer msg, const char *query, int loc, int encoding)
 		 * width.
 		 */
 		scroffset = 0;
-		for (; i < msg->len; i += pg_encoding_mblen(encoding, &msg->data[i]))
+		for (; i < msg->len; i += PQmblenBounded(&msg->data[i], encoding))
 		{
 			int			w = pg_encoding_dsplen(encoding, &msg->data[i]);
 

src/interfaces/libpq/libpq-fe.h

@@ -625,6 +625,9 @@ extern int	PQlibVersion(void);
 /* Determine length of multibyte encoded char at *s */
 extern int	PQmblen(const char *s, int encoding);
 
+/* Same, but not more than the distance to the end of string s */
+extern int	PQmblenBounded(const char *s, int encoding);
+
 /* Determine display length of multibyte encoded char at *s */
 extern int	PQdsplen(const char *s, int encoding);