Adjust datetime parsing to be more robust. We now pass the length of the

working buffer into ParseDateTime() and reject too-long input there,
rather than checking the length of the input string before calling
ParseDateTime(). The old method was bogus because ParseDateTime() can use
a variable amount of working space, depending on the content of the
input string (e.g. how many fields need to be NUL terminated). This fixes
a minor stack overrun -- I don't _think_ it's exploitable, although I
won't claim to be an expert.

Along the way, fix a bug reported by Mark Dilger: the working buffer
allocated by interval_in() was too short, which resulted in rejecting
some perfectly valid interval input values. I added a regression test for
this fix.

4 files changed, 74 insertions, 73 deletions

diff --git a/src/backend/utils/adt/date.c b/src/backend/utils/adt/date.c
index 4e344eb091e..0e2eb954656 100644
--- a/src/backend/utils/adt/date.c
+++ b/src/backend/utils/adt/date.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/date.c,v 1.104.4.1 2005/04/23 22:53:26 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/date.c,v 1.104.4.2 2005/05/26 02:10:02 neilc Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -65,12 +65,10 @@ date_in(PG_FUNCTION_ARGS)
 	int			dterr;
 	char	   *field[MAXDATEFIELDS];
 	int			ftype[MAXDATEFIELDS];
-	char		lowstr[MAXDATELEN + 1];
+	char		workbuf[MAXDATELEN + 1];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeDateTime(field, ftype, nf, &dtype, tm, &fsec, &tzp);
 	if (dterr != 0)
@@ -894,15 +892,13 @@ time_in(PG_FUNCTION_ARGS)
 	int			tz;
 	int			nf;
 	int			dterr;
-	char		lowstr[MAXDATELEN + 1];
+	char		workbuf[MAXDATELEN + 1];
 	char	   *field[MAXDATEFIELDS];
 	int			dtype;
 	int			ftype[MAXDATEFIELDS];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeTimeOnly(field, ftype, nf, &dtype, tm, &fsec, &tz);
 	if (dterr != 0)
@@ -1734,15 +1730,13 @@ timetz_in(PG_FUNCTION_ARGS)
 	int			tz;
 	int			nf;
 	int			dterr;
-	char		lowstr[MAXDATELEN + 1];
+	char		workbuf[MAXDATELEN + 1];
 	char	   *field[MAXDATEFIELDS];
 	int			dtype;
 	int			ftype[MAXDATEFIELDS];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeTimeOnly(field, ftype, nf, &dtype, tm, &fsec, &tz);
 	if (dterr != 0)
diff --git a/src/backend/utils/adt/datetime.c b/src/backend/utils/adt/datetime.c
index ae85fcaf302..996583061a4 100644
--- a/src/backend/utils/adt/datetime.c
+++ b/src/backend/utils/adt/datetime.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/datetime.c,v 1.137.4.1 2005/04/20 17:14:58 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/datetime.c,v 1.137.4.2 2005/05/26 02:10:02 neilc Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -700,21 +700,23 @@ TrimTrailingZeros(char *str)
 	}
 }
 
-
 /* ParseDateTime()
  *	Break string into tokens based on a date/time context.
  *	Returns 0 if successful, DTERR code if bogus input detected.
  *
  * timestr - the input string
- * lowstr - workspace for field string storage (must be large enough for
- *		a copy of the input string, including trailing null)
+ * workbuf - workspace for field string storage. This must be
+ *   larger than the largest legal input for this datetime type --
+ *   some additional space will be needed to NUL terminate fields.
+ * buflen - the size of workbuf
  * field[] - pointers to field strings are returned in this array
  * ftype[] - field type indicators are returned in this array
  * maxfields - dimensions of the above two arrays
  * *numfields - set to the actual number of fields detected
  *
- * The fields extracted from the input are stored as separate, null-terminated
- * strings in the workspace at lowstr.	Any text is converted to lower case.
+ * The fields extracted from the input are stored as separate,
+ * null-terminated strings in the workspace at workbuf. Any text is
+ * converted to lower case.
  *
  * Several field types are assigned:
  *	DTK_NUMBER - digits and (possibly) a decimal point
@@ -730,12 +732,27 @@ TrimTrailingZeros(char *str)
  *	DTK_DATE can hold Posix time zones (GMT-8)
  */
 int
-ParseDateTime(const char *timestr, char *lowstr,
+ParseDateTime(const char *timestr, char *workbuf, size_t buflen,
 			  char **field, int *ftype, int maxfields, int *numfields)
 {
 	int			nf = 0;
 	const char *cp = timestr;
-	char	   *lp = lowstr;
+	char	   *bufp = workbuf;
+	const char *bufend = workbuf + buflen;
+
+	/*
+	 * Set the character pointed-to by "bufptr" to "newchar", and
+	 * increment "bufptr". "end" gives the end of the buffer -- we
+	 * return an error if there is no space left to append a character
+	 * to the buffer. Note that "bufptr" is evaluated twice.
+	 */
+#define APPEND_CHAR(bufptr, end, newchar)		\
+	do											\
+	{											\
+		if (((bufptr) + 1) >= (end))			\
+			return DTERR_BAD_FORMAT;			\
+		*(bufptr)++ = newchar;					\
+	} while (0)
 
 	/* outer loop through fields */
 	while (*cp != '\0')
@@ -750,23 +767,23 @@ ParseDateTime(const char *timestr, char *lowstr,
 		/* Record start of current field */
 		if (nf >= maxfields)
 			return DTERR_BAD_FORMAT;
-		field[nf] = lp;
+		field[nf] = bufp;
 
 		/* leading digit? then date or time */
 		if (isdigit((unsigned char) *cp))
 		{
-			*lp++ = *cp++;
+			APPEND_CHAR(bufp, bufend, *cp++);
 			while (isdigit((unsigned char) *cp))
-				*lp++ = *cp++;
+				APPEND_CHAR(bufp, bufend, *cp++);
 
 			/* time field? */
 			if (*cp == ':')
 			{
 				ftype[nf] = DTK_TIME;
-				*lp++ = *cp++;
+				APPEND_CHAR(bufp, bufend, *cp++);
 				while (isdigit((unsigned char) *cp) ||
 					   (*cp == ':') || (*cp == '.'))
-					*lp++ = *cp++;
+					APPEND_CHAR(bufp, bufend, *cp++);
 			}
 			/* date field? allow embedded text month */
 			else if ((*cp == '-') || (*cp == '/') || (*cp == '.'))
@@ -774,13 +791,13 @@ ParseDateTime(const char *timestr, char *lowstr,
 				/* save delimiting character to use later */
 				char		delim = *cp;
 
-				*lp++ = *cp++;
+				APPEND_CHAR(bufp, bufend, *cp++);
 				/* second field is all digits? then no embedded text month */
 				if (isdigit((unsigned char) *cp))
 				{
 					ftype[nf] = ((delim == '.') ? DTK_NUMBER : DTK_DATE);
 					while (isdigit((unsigned char) *cp))
-						*lp++ = *cp++;
+						APPEND_CHAR(bufp, bufend, *cp++);
 
 					/*
 					 * insist that the delimiters match to get a
@@ -789,16 +806,16 @@ ParseDateTime(const char *timestr, char *lowstr,
 					if (*cp == delim)
 					{
 						ftype[nf] = DTK_DATE;
-						*lp++ = *cp++;
+						APPEND_CHAR(bufp, bufend, *cp++);
 						while (isdigit((unsigned char) *cp) || (*cp == delim))
-							*lp++ = *cp++;
+							APPEND_CHAR(bufp, bufend, *cp++);
 					}
 				}
 				else
 				{
 					ftype[nf] = DTK_DATE;
 					while (isalnum((unsigned char) *cp) || (*cp == delim))
-						*lp++ = pg_tolower((unsigned char) *cp++);
+						APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
 				}
 			}
 
@@ -812,9 +829,9 @@ ParseDateTime(const char *timestr, char *lowstr,
 		/* Leading decimal point? Then fractional seconds... */
 		else if (*cp == '.')
 		{
-			*lp++ = *cp++;
+			APPEND_CHAR(bufp, bufend, *cp++);
 			while (isdigit((unsigned char) *cp))
-				*lp++ = *cp++;
+				APPEND_CHAR(bufp, bufend, *cp++);
 
 			ftype[nf] = DTK_NUMBER;
 		}
@@ -826,9 +843,9 @@ ParseDateTime(const char *timestr, char *lowstr,
 		else if (isalpha((unsigned char) *cp))
 		{
 			ftype[nf] = DTK_STRING;
-			*lp++ = pg_tolower((unsigned char) *cp++);
+			APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
 			while (isalpha((unsigned char) *cp))
-				*lp++ = pg_tolower((unsigned char) *cp++);
+				APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
 
 			/*
 			 * Full date string with leading text month? Could also be a
@@ -839,15 +856,15 @@ ParseDateTime(const char *timestr, char *lowstr,
 				char		delim = *cp;
 
 				ftype[nf] = DTK_DATE;
-				*lp++ = *cp++;
+				APPEND_CHAR(bufp, bufend, *cp++);
 				while (isdigit((unsigned char) *cp) || (*cp == delim))
-					*lp++ = *cp++;
+					APPEND_CHAR(bufp, bufend, *cp++);
 			}
 		}
 		/* sign? then special or numeric timezone */
 		else if ((*cp == '+') || (*cp == '-'))
 		{
-			*lp++ = *cp++;
+			APPEND_CHAR(bufp, bufend, *cp++);
 			/* soak up leading whitespace */
 			while (isspace((unsigned char) *cp))
 				cp++;
@@ -855,18 +872,18 @@ ParseDateTime(const char *timestr, char *lowstr,
 			if (isdigit((unsigned char) *cp))
 			{
 				ftype[nf] = DTK_TZ;
-				*lp++ = *cp++;
+				APPEND_CHAR(bufp, bufend, *cp++);
 				while (isdigit((unsigned char) *cp) ||
 					   (*cp == ':') || (*cp == '.'))
-					*lp++ = *cp++;
+					APPEND_CHAR(bufp, bufend, *cp++);
 			}
 			/* special? */
 			else if (isalpha((unsigned char) *cp))
 			{
 				ftype[nf] = DTK_SPECIAL;
-				*lp++ = pg_tolower((unsigned char) *cp++);
+				APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
 				while (isalpha((unsigned char) *cp))
-					*lp++ = pg_tolower((unsigned char) *cp++);
+					APPEND_CHAR(bufp, bufend, pg_tolower((unsigned char) *cp++));
 			}
 			/* otherwise something wrong... */
 			else
@@ -883,7 +900,7 @@ ParseDateTime(const char *timestr, char *lowstr,
 			return DTERR_BAD_FORMAT;
 
 		/* force in a delimiter after each field */
-		*lp++ = '\0';
+		*bufp++ = '\0';
 		nf++;
 	}
 
diff --git a/src/backend/utils/adt/nabstime.c b/src/backend/utils/adt/nabstime.c
index 4e2069f5fb6..dd2f6fe95f6 100644
--- a/src/backend/utils/adt/nabstime.c
+++ b/src/backend/utils/adt/nabstime.c
@@ -10,7 +10,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/nabstime.c,v 1.127 2004/12/31 22:01:22 pgsql Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/nabstime.c,v 1.127.4.1 2005/05/26 02:10:02 neilc Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -306,15 +306,13 @@ abstimein(PG_FUNCTION_ARGS)
 			   *tm = &date;
 	int			dterr;
 	char	   *field[MAXDATEFIELDS];
-	char		lowstr[MAXDATELEN + 1];
+	char		workbuf[MAXDATELEN + 1];
 	int			dtype;
 	int			nf,
 				ftype[MAXDATEFIELDS];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeDateTime(field, ftype, nf, &dtype, tm, &fsec, &tz);
 	if (dterr != 0)
@@ -711,12 +709,10 @@ reltimein(PG_FUNCTION_ARGS)
 	char	   *field[MAXDATEFIELDS];
 	int			nf,
 				ftype[MAXDATEFIELDS];
-	char		lowstr[MAXDATELEN + 1];
+	char		workbuf[MAXDATELEN + 1];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeInterval(field, ftype, nf, &dtype, tm, &fsec);
 	if (dterr != 0)
diff --git a/src/backend/utils/adt/timestamp.c b/src/backend/utils/adt/timestamp.c
index fe9cfbcf946..6ade3fb71fd 100644
--- a/src/backend/utils/adt/timestamp.c
+++ b/src/backend/utils/adt/timestamp.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/timestamp.c,v 1.117.4.1 2005/04/01 14:25:39 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/timestamp.c,v 1.117.4.2 2005/05/26 02:10:02 neilc Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -77,12 +77,10 @@ timestamp_in(PG_FUNCTION_ARGS)
 	int			dterr;
 	char	   *field[MAXDATEFIELDS];
 	int			ftype[MAXDATEFIELDS];
-	char		lowstr[MAXDATELEN + MAXDATEFIELDS];
+	char		workbuf[MAXDATELEN + MAXDATEFIELDS];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeDateTime(field, ftype, nf, &dtype, tm, &fsec, &tz);
 	if (dterr != 0)
@@ -318,12 +316,10 @@ timestamptz_in(PG_FUNCTION_ARGS)
 	int			dterr;
 	char	   *field[MAXDATEFIELDS];
 	int			ftype[MAXDATEFIELDS];
-	char		lowstr[MAXDATELEN + MAXDATEFIELDS];
+	char		workbuf[MAXDATELEN + MAXDATEFIELDS];
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf),
+						  field, ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeDateTime(field, ftype, nf, &dtype, tm, &fsec, &tz);
 	if (dterr != 0)
@@ -494,7 +490,7 @@ interval_in(PG_FUNCTION_ARGS)
 	int			dterr;
 	char	   *field[MAXDATEFIELDS];
 	int			ftype[MAXDATEFIELDS];
-	char		lowstr[MAXDATELEN + MAXDATEFIELDS];
+	char		workbuf[256];
 
 	tm->tm_year = 0;
 	tm->tm_mon = 0;
@@ -504,10 +500,8 @@ interval_in(PG_FUNCTION_ARGS)
 	tm->tm_sec = 0;
 	fsec = 0;
 
-	if (strlen(str) >= sizeof(lowstr))
-		dterr = DTERR_BAD_FORMAT;
-	else
-		dterr = ParseDateTime(str, lowstr, field, ftype, MAXDATEFIELDS, &nf);
+	dterr = ParseDateTime(str, workbuf, sizeof(workbuf), field,
+						  ftype, MAXDATEFIELDS, &nf);
 	if (dterr == 0)
 		dterr = DecodeInterval(field, ftype, nf, &dtype, tm, &fsec);
 	if (dterr != 0)