Ensure Soundex difference() function handles empty input sanely.

tglsfdc · tglsfdc · commit 1d369c9e90f3 · 2023-05-16T10:53:42.000-04:00
fuzzystrmatch's difference() function assumes that _soundex() always initializes its output buffer fully. This was not so for the case of a string containing no alphabetic characters, resulting in unstable output and Valgrind complaints. Fix by using memset() to fill the whole buffer in the early-exit case. Also make some cosmetic improvements (I didn't care for the random switches between "instr[0]" and "*instr" notation). Report and diagnosis by Alexander Lakhin (bug #17935). Back-patch to all supported branches. Discussion: https://postgr.es/m/17935-b99316aa79c18513@postgresql.org
diff --git a/contrib/fuzzystrmatch/expected/fuzzystrmatch.out b/contrib/fuzzystrmatch/expected/fuzzystrmatch.out
@@ -23,6 +23,12 @@ SELECT soundex('Anne'), soundex('Margaret'), difference('Anne', 'Margaret');
  A500    | M626    |          0
 (1 row)
 
+SELECT soundex(''), difference('', '');
+ soundex | difference 
+---------+------------
+         |          4
+(1 row)
+
 SELECT levenshtein('GUMBO', 'GAMBOL');
  levenshtein 
 -------------
diff --git a/contrib/fuzzystrmatch/fuzzystrmatch.c b/contrib/fuzzystrmatch/fuzzystrmatch.c
@@ -728,16 +728,14 @@ _soundex(const char *instr, char *outstr)
 	Assert(instr);
 	Assert(outstr);
 
-	outstr[SOUNDEX_LEN] = '\0';
-
 	/* Skip leading non-alphabetic characters */
-	while (!isalpha((unsigned char) instr[0]) && instr[0])
+	while (*instr && !isalpha((unsigned char) *instr))
 		++instr;
 
-	/* No string left */
-	if (!instr[0])
+	/* If no string left, return all-zeroes buffer */
+	if (!*instr)
 	{
-		outstr[0] = (char) 0;
+		memset(outstr, '\0', SOUNDEX_LEN + 1);
 		return;
 	}
 
@@ -750,7 +748,7 @@ _soundex(const char *instr, char *outstr)
 		if (isalpha((unsigned char) *instr) &&
 			soundex_code(*instr) != soundex_code(*(instr - 1)))
 		{
-			*outstr = soundex_code(instr[0]);
+			*outstr = soundex_code(*instr);
 			if (*outstr != '0')
 			{
 				++outstr;
@@ -767,6 +765,9 @@ _soundex(const char *instr, char *outstr)
 		++outstr;
 		++count;
 	}
+
+	/* And null-terminate */
+	*outstr = '\0';
 }
 
 PG_FUNCTION_INFO_V1(difference);
diff --git a/contrib/fuzzystrmatch/sql/fuzzystrmatch.sql b/contrib/fuzzystrmatch/sql/fuzzystrmatch.sql
@@ -6,6 +6,7 @@ SELECT soundex('hello world!');
 SELECT soundex('Anne'), soundex('Ann'), difference('Anne', 'Ann');
 SELECT soundex('Anne'), soundex('Andrew'), difference('Anne', 'Andrew');
 SELECT soundex('Anne'), soundex('Margaret'), difference('Anne', 'Margaret');
+SELECT soundex(''), difference('', '');
 
 
 SELECT levenshtein('GUMBO', 'GAMBOL');

Original file line number	Diff line number	Diff line change
`@@ -728,16 +728,14 @@ _soundex(const char instr, char outstr)`
`728`	`728`	`Assert(instr);`
`729`	`729`	`Assert(outstr);`
`730`	`730`
`731`		`- outstr[SOUNDEX_LEN] = '\0';`
`732`		`-`
`733`	`731`	`/* Skip leading non-alphabetic characters */`
`734`		`- while (!isalpha((unsigned char) instr[0]) && instr[0])`
	`732`	`+ while (instr && !isalpha((unsigned char) instr))`
`735`	`733`	`++instr;`
`736`	`734`
`737`		`- /* No string left */`
`738`		`- if (!instr[0])`
	`735`	`+ /* If no string left, return all-zeroes buffer */`
	`736`	`+ if (!*instr)`
`739`	`737`	`{`
`740`		`- outstr[0] = (char) 0;`
	`738`	`+ memset(outstr, '\0', SOUNDEX_LEN + 1);`
`741`	`739`	`return;`
`742`	`740`	`}`
`743`	`741`
`@@ -750,7 +748,7 @@ _soundex(const char instr, char outstr)`
`750`	`748`	`if (isalpha((unsigned char) *instr) &&`
`751`	`749`	`soundex_code(instr) != soundex_code((instr - 1)))`
`752`	`750`	`{`
`753`		`- *outstr = soundex_code(instr[0]);`
	`751`	`+ outstr = soundex_code(instr);`
`754`	`752`	`if (*outstr != '0')`
`755`	`753`	`{`
`756`	`754`	`++outstr;`
`@@ -767,6 +765,9 @@ _soundex(const char instr, char outstr)`
`767`	`765`	`++outstr;`
`768`	`766`	`++count;`
`769`	`767`	`}`
	`768`	`+`
	`769`	`+ /* And null-terminate */`
	`770`	`+ *outstr = '\0';`
`770`	`771`	`}`
`771`	`772`
`772`	`773`	`PG_FUNCTION_INFO_V1(difference);`