From 5b64368742b5cdf73395bd5ddd66b5edcbc20823 Mon Sep 17 00:00:00 2001
From: Tom Lane <tgl@sss.pgh.pa.us>
Date: Mon, 7 Jun 2021 14:15:25 -0400
Subject: [PATCH] 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.
---
 src/bin/psql/common.c               | 28 +++++++++++++++-------------
 src/bin/psql/psqlscanslash.l        |  4 +++-
 src/bin/psql/stringutils.c          |  8 +++++---
 src/bin/psql/tab-complete.c         |  4 +++-
 src/bin/scripts/common.c            |  4 +++-
 src/common/jsonapi.c                |  6 +++---
 src/common/wchar.c                  | 15 +++++++++++++++
 src/fe_utils/print.c                |  3 +++
 src/include/mb/pg_wchar.h           |  1 +
 src/interfaces/libpq/fe-print.c     |  3 ++-
 src/interfaces/libpq/fe-protocol3.c |  6 ++++--
 11 files changed, 57 insertions(+), 25 deletions(-)

diff --git a/src/bin/psql/common.c b/src/bin/psql/common.c
index 10314288499..94d5d60f879 100644
--- a/src/bin/psql/common.c
+++ b/src/bin/psql/common.c
@@ -29,6 +29,8 @@
 #include "portability/instr_time.h"
 #include "settings.h"
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
+
 static bool DescribeQuery(const char *query, double *elapsed_msec);
 static bool ExecQueryUsingCursor(const char *query, double *elapsed_msec);
 static bool command_no_begin(const char *query);
@@ -1842,7 +1844,7 @@ 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
@@ -1879,7 +1881,7 @@ skip_white_space(const char *query)
 					query++;
 					break;
 				}
-				query += PQmblen(query, pset.encoding);
+				query += PQmblenBounded(query, pset.encoding);
 			}
 		}
 		else if (cnestlevel > 0)
@@ -1914,7 +1916,7 @@ 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
@@ -1945,7 +1947,7 @@ 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;
@@ -1979,7 +1981,7 @@ 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;
@@ -1995,7 +1997,7 @@ 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)
@@ -2006,7 +2008,7 @@ 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;
@@ -2023,7 +2025,7 @@ 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)
@@ -2046,7 +2048,7 @@ 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;
@@ -2061,7 +2063,7 @@ 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
@@ -2080,7 +2082,7 @@ 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;
@@ -2100,7 +2102,7 @@ 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;
@@ -2136,7 +2138,7 @@ 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;
diff --git a/src/bin/psql/psqlscanslash.l b/src/bin/psql/psqlscanslash.l
index 4dff84d6271..390dc6c5c90 100644
--- a/src/bin/psql/psqlscanslash.l
+++ b/src/bin/psql/psqlscanslash.l
@@ -28,6 +28,8 @@
 %{
 #include "fe_utils/psqlscan_int.h"
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
+
 /*
  * We must have a typedef YYSTYPE for yylex's first argument, but this lexer
  * doesn't presently make use of that argument, so just declare it as int.
@@ -753,7 +755,7 @@ 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);
 		}
 	}
 }
diff --git a/src/bin/psql/stringutils.c b/src/bin/psql/stringutils.c
index c521749661c..c96b2fe1965 100644
--- a/src/bin/psql/stringutils.c
+++ b/src/bin/psql/stringutils.c
@@ -12,6 +12,8 @@
 #include "common.h"
 #include "stringutils.h"
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
+
 
 /*
  * Replacement for strtok() (a.k.a. poor man's flex)
@@ -143,7 +145,7 @@ 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 +264,7 @@ 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 +326,7 @@ 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++;
 	}
diff --git a/src/bin/psql/tab-complete.c b/src/bin/psql/tab-complete.c
index eb018854a5c..4f46cd949d9 100644
--- a/src/bin/psql/tab-complete.c
+++ b/src/bin/psql/tab-complete.c
@@ -73,6 +73,8 @@
 #define USE_FILENAME_QUOTING_FUNCTIONS 1
 #endif
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
+
 /* word break characters */
 #define WORD_BREAKS		"\t\n@$><=;|&{() "
 
@@ -4140,7 +4142,7 @@ _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 */
diff --git a/src/bin/scripts/common.c b/src/bin/scripts/common.c
index afa62e4da73..d446d7af9fa 100644
--- a/src/bin/scripts/common.c
+++ b/src/bin/scripts/common.c
@@ -25,6 +25,8 @@
 
 #define ERRCODE_UNDEFINED_TABLE  "42P01"
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
+
 /*
  * Provide strictly harmonized handling of --help and --version
  * options.
@@ -368,7 +370,7 @@ splitTableColumnsSpec(const char *spec, int encoding,
 			cp++;
 		}
 		else
-			cp += PQmblen(cp, encoding);
+			cp += PQmblenBounded(cp, encoding);
 	}
 	*table = pnstrdup(spec, cp - spec);
 	*columns = cp;
diff --git a/src/common/jsonapi.c b/src/common/jsonapi.c
index 9326f805366..6fe17a3d378 100644
--- a/src/common/jsonapi.c
+++ b/src/common/jsonapi.c
@@ -738,7 +738,7 @@ 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;
 					}
 				}
@@ -844,7 +844,7 @@ 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;
 				}
 			}
@@ -858,7 +858,7 @@ 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;
 			}
 
diff --git a/src/common/wchar.c b/src/common/wchar.c
index efaf1c155bb..08734802236 100644
--- a/src/common/wchar.c
+++ b/src/common/wchar.c
@@ -1549,6 +1549,11 @@ 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)
@@ -1558,6 +1563,16 @@ 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.
  */
diff --git a/src/fe_utils/print.c b/src/fe_utils/print.c
index f3c176aa555..966a7721801 100644
--- a/src/fe_utils/print.c
+++ b/src/fe_utils/print.c
@@ -3652,6 +3652,9 @@ 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;
diff --git a/src/include/mb/pg_wchar.h b/src/include/mb/pg_wchar.h
index 494aefc7fab..b43454b64ce 100644
--- a/src/include/mb/pg_wchar.h
+++ b/src/include/mb/pg_wchar.h
@@ -553,6 +553,7 @@ 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_verifymb(int encoding, const char *mbstr, int len);
 extern int	pg_encoding_max_length(int encoding);
diff --git a/src/interfaces/libpq/fe-print.c b/src/interfaces/libpq/fe-print.c
index 784eaae48e5..61ba4e4ce7b 100644
--- a/src/interfaces/libpq/fe-print.c
+++ b/src/interfaces/libpq/fe-print.c
@@ -36,6 +36,7 @@
 #include "libpq-fe.h"
 #include "libpq-int.h"
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
 
 static void do_field(const PQprintOpt *po, const PGresult *res,
 					 const int i, const int j, const int fs_len,
@@ -365,7 +366,7 @@ 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') ||
diff --git a/src/interfaces/libpq/fe-protocol3.c b/src/interfaces/libpq/fe-protocol3.c
index 686299dce91..04fd4ba874b 100644
--- a/src/interfaces/libpq/fe-protocol3.c
+++ b/src/interfaces/libpq/fe-protocol3.c
@@ -39,6 +39,8 @@
 	((id) == 'T' || (id) == 'D' || (id) == 'd' || (id) == 'V' || \
 	 (id) == 'E' || (id) == 'N' || (id) == 'A')
 
+#define PQmblenBounded(s, e)  strnlen(s, PQmblen(s, e))
+
 
 static void handleSyncLoss(PGconn *conn, char id, int msgLength);
 static int	getRowDescriptions(PGconn *conn, int msgLength);
@@ -1241,7 +1243,7 @@ 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
 		{
@@ -1309,7 +1311,7 @@ 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]);
 
-- 
2.39.5