Add double metaphone code from Andrew Dunstan. Also change metaphone so that

an empty input string causes an empty output string to be returned, instead of
throwing an ERROR -- per complaint from Aaron Hillegass, and consistent with
double metaphone. Fix examples in README.soundex pointed out by James Robinson.

contrib/fuzzystrmatch/Makefile

-# $PostgreSQL: pgsql/contrib/fuzzystrmatch/Makefile,v 1.3 2003/11/29 19:51:35 pgsql Exp $
+# $PostgreSQL: pgsql/contrib/fuzzystrmatch/Makefile,v 1.4 2004/07/01 03:25:48 joe Exp $
 
 subdir = contrib/fuzzystrmatch
 top_builddir = ../..
 include $(top_builddir)/src/Makefile.global
 
-MODULES = fuzzystrmatch
+MODULE_big = fuzzystrmatch
+SRCS += fuzzystrmatch.c dmetaphone.c
+OBJS = $(SRCS:.c=.o)
 DATA_built = fuzzystrmatch.sql
 DOCS = README.fuzzystrmatch README.soundex
 

contrib/fuzzystrmatch/README.fuzzystrmatch

@@ -23,6 +23,11 @@
  * Metaphone was originally created by Lawrence Philips and presented in article
  * in "Computer Language" December 1990 issue.
  *
+ * dmetaphone() and dmetaphone_alt()
+ * ---------------------------------
+ * A port of the DoubleMetaphone perl module by Andrew Dunstan. See dmetaphone.c
+ * for more detail.
+ *
  * soundex()
  * -----------
  * Folded existing soundex contrib into this one. Renamed text_soundex() (C function)
@@ -48,11 +53,14 @@
  */
 
 
-Version 0.2 (7 August, 2001):
-  Functions to calculate the degree to which two strings match in a "fuzzy" way
-  Tested under Linux (Red Hat 6.2 and 7.0) and PostgreSQL 7.2devel
+Version 0.3 (30 June, 2004):
 
 Release Notes:
+  Version 0.3
+   - added double metaphone code from Andrew Dunstan
+   - change metaphone so that an empty input string causes an empty
+     output string to be returned, instead of throwing an ERROR
+   - fixed examples in README.soundex
 
   Version 0.2
     - folded soundex contrib into this one

contrib/fuzzystrmatch/README.soundex

@@ -20,7 +20,7 @@ insert into s values ('wobbly')\g
 select * from s
 where soundex(nm) = soundex('john')\g
 
-select nm from s a, s b
+select a.nm, b.nm from s a, s b
 where soundex(a.nm) = soundex(b.nm)
 and a.oid <> b.oid\g
 
@@ -51,7 +51,7 @@ LANGUAGE 'sql';
 DROP OPERATOR #= (text,text)\g
 
 CREATE OPERATOR #= (leftarg=text, rightarg=text, procedure=text_sx_eq,
-commutator=text_sx_eq)\g
+commutator = #=)\g
 
 SELECT *
 FROM s

contrib/fuzzystrmatch/fuzzystrmatch.c

@@ -202,6 +202,8 @@ levenshtein(PG_FUNCTION_ARGS)
  * Returns number of characters requested
  * (suggested value is 4)
  */
+#define GET_TEXT(cstrp) DatumGetTextP(DirectFunctionCall1(textin, CStringGetDatum(cstrp)))
+
 PG_FUNCTION_INFO_V1(metaphone);
 Datum
 metaphone(PG_FUNCTION_ARGS)
@@ -216,6 +218,10 @@ metaphone(PG_FUNCTION_ARGS)
 	str_i = DatumGetCString(DirectFunctionCall1(textout, PointerGetDatum(PG_GETARG_TEXT_P(0))));
 	str_i_len = strlen(str_i);
 
+	/* return an empty string if we receive one */
+	if (!(str_i_len > 0))
+		PG_RETURN_TEXT_P(GET_TEXT(""));
+
 	if (str_i_len > MAX_METAPHONE_STRLEN)
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_PARAMETER_VALUE),

contrib/fuzzystrmatch/fuzzystrmatch.sql.in

@@ -18,3 +18,11 @@ LANGUAGE 'C' WITH (iscachable, isstrict);
 CREATE FUNCTION text_soundex(text) RETURNS text
 AS 'MODULE_PATHNAME', 'soundex'
 LANGUAGE 'C';
+
+CREATE FUNCTION dmetaphone (text) RETURNS text 
+LANGUAGE C IMMUTABLE STRICT
+AS 'MODULE_PATHNAME', 'dmetaphone';
+
+CREATE FUNCTION dmetaphone_alt (text) RETURNS text 
+LANGUAGE C IMMUTABLE STRICT
+AS 'MODULE_PATHNAME', 'dmetaphone_alt';