builtins: add fuzzystrmatch metaphone()

Release note (sql change): metaphone() builtin function was added

docs/generated/sql/functions.md

@@ -1050,6 +1050,8 @@ available replica will error.</p>
 </span></td><td>Immutable</td></tr>
 <tr><td><a name="levenshtein"></a><code>levenshtein(source: <a href="string.html">string</a>, target: <a href="string.html">string</a>, ins_cost: <a href="int.html">int</a>, del_cost: <a href="int.html">int</a>, sub_cost: <a href="int.html">int</a>) &rarr; <a href="int.html">int</a></code></td><td><span class="funcdesc"><p>Calculates the Levenshtein distance between two strings. The cost parameters specify how much to charge for each edit operation. Maximum input length is 255 characters.</p>
 </span></td><td>Immutable</td></tr>
+<tr><td><a name="metaphone"></a><code>metaphone(source: <a href="string.html">string</a>, max_output_length: <a href="int.html">int</a>) &rarr; <a href="string.html">string</a></code></td><td><span class="funcdesc"><p>Convert a string to its Metaphone code. Maximum input length is 255 characters</p>
+</span></td><td>Immutable</td></tr>
 <tr><td><a name="soundex"></a><code>soundex(source: <a href="string.html">string</a>) &rarr; <a href="string.html">string</a></code></td><td><span class="funcdesc"><p>Convert a string to its Soundex code.</p>
 </span></td><td>Immutable</td></tr></tbody>
 </table>
@@ -2870,7 +2872,7 @@ Can be used to define the tile bounds required by ST_AsMVTGeom to convert geomet
 </span></td><td>Immutable</td></tr>
 <tr><td><a name="decompress"></a><code>decompress(data: <a href="bytes.html">bytes</a>, codec: <a href="string.html">string</a>) &rarr; <a href="bytes.html">bytes</a></code></td><td><span class="funcdesc"><p>Decompress <code>data</code> with the specified <code>codec</code> (<code>gzip</code>, ‘lz4’, ‘snappy’, 'zstd).</p>
 </span></td><td>Immutable</td></tr>
-<tr><td><a name="difference"></a><code>difference(source: <a href="string.html">string</a>, target: <a href="string.html">string</a>) &rarr; <a href="int.html">int</a></code></td><td><span class="funcdesc"><p>Convert two strings to their Soundex codes and then reports the number of matching code positions.</p>
+<tr><td><a name="difference"></a><code>difference(source: <a href="string.html">string</a>, target: <a href="string.html">string</a>) &rarr; <a href="int.html">int</a></code></td><td><span class="funcdesc"><p>Convert two strings to their Soundex codes and report the number of matching code positions.</p>
 </span></td><td>Immutable</td></tr>
 <tr><td><a name="encode"></a><code>encode(data: <a href="bytes.html">bytes</a>, format: <a href="string.html">string</a>) &rarr; <a href="string.html">string</a></code></td><td><span class="funcdesc"><p>Encodes <code>data</code> using <code>format</code> (<code>hex</code> / <code>escape</code> / <code>base64</code>).</p>
 </span></td><td>Immutable</td></tr>

pkg/sql/logictest/testdata/logic_test/fuzzystrmatch

@@ -54,3 +54,13 @@ query TTTI
 SELECT soundex('Anne'), soundex(NULL), difference('Anne', NULL), difference(NULL, 'Bob');
 ----
 A500  NULL  NULL  NULL
+
+query TT
+SELECT metaphone('GUMBO', 4), metaphone(NULL, 4);
+----
+KM  NULL
+
+query TTT
+SELECT metaphone('Night', 4), metaphone('Knight', 4), metaphone('Knives', 4);
+----
+NFT NFT NFS

pkg/sql/sem/builtins/builtins.go

@@ -3972,7 +3972,7 @@ value if you rely on the HLC for accuracy.`,
 				diff := fuzzystrmatch.Difference(s, t)
 				return tree.NewDInt(tree.DInt(diff)), nil
 			},
-			Info:       "Convert two strings to their Soundex codes and then reports the number of matching code positions.",
+			Info:       "Convert two strings to their Soundex codes and report the number of matching code positions.",
 			Volatility: volatility.Immutable,
 		},
 	),
@@ -4012,10 +4012,39 @@ value if you rely on the HLC for accuracy.`,
 			Info: "Calculates the Levenshtein distance between two strings. The cost parameters specify how much to " +
 				"charge for each edit operation. Maximum input length is 255 characters.",
 			Volatility: volatility.Immutable,
-		}),
+		},
+	),
 	"levenshtein_less_equal": makeBuiltin(tree.FunctionProperties{UnsupportedWithIssue: 56820, Category: builtinconstants.CategoryFuzzyStringMatching}),
-	"metaphone":              makeBuiltin(tree.FunctionProperties{UnsupportedWithIssue: 56820, Category: builtinconstants.CategoryFuzzyStringMatching}),
-	"dmetaphone_alt":         makeBuiltin(tree.FunctionProperties{UnsupportedWithIssue: 56820, Category: builtinconstants.CategoryFuzzyStringMatching}),
+	"metaphone": makeBuiltin(
+		tree.FunctionProperties{Category: builtinconstants.CategoryFuzzyStringMatching},
+		tree.Overload{
+			Types:      tree.ParamTypes{{Name: "source", Typ: types.String}, {Name: "max_output_length", Typ: types.Int}},
+			ReturnType: tree.FixedReturnType(types.String),
+			Fn: func(_ context.Context, _ *eval.Context, args tree.Datums) (tree.Datum, error) {
+				const maxDefaultLen = 255
+				s := string(tree.MustBeDString(args[0]))
+				maxOutputLen := int(tree.MustBeDInt(args[1]))
+				if len(s) > maxDefaultLen {
+					return nil, pgerror.Newf(pgcode.InvalidParameterValue,
+						"argument exceeds maximum length of %d characters", maxDefaultLen)
+				}
+				if maxOutputLen > maxDefaultLen {
+					return nil, pgerror.Newf(pgcode.InvalidParameterValue,
+						"output exceeds maximum length of %d characters", maxDefaultLen)
+				}
+				if maxOutputLen <= 0 {
+					return nil, pgerror.Newf(pgcode.InvalidParameterValue,
+						"output length must be > 0")
+				}
+				m := fuzzystrmatch.Metaphone(s, maxDefaultLen)
+				return tree.NewDString(m), nil
+			},
+			Info:       "Convert a string to its Metaphone code. Maximum input length is 255 characters",
+			Volatility: volatility.Immutable,
+		},
+	),
+	"dmetaphone":     makeBuiltin(tree.FunctionProperties{UnsupportedWithIssue: 56820, Category: builtinconstants.CategoryFuzzyStringMatching}),
+	"dmetaphone_alt": makeBuiltin(tree.FunctionProperties{UnsupportedWithIssue: 56820, Category: builtinconstants.CategoryFuzzyStringMatching}),
 
 	// JSON functions.
 	// The behavior of both the JSON and JSONB data types in CockroachDB is

pkg/sql/sem/builtins/fixed_oids.go

@@ -2538,6 +2538,7 @@ var builtinOidsArray = []string{
 	2570: `array_position(array: refcursor[], elem: refcursor, start: int) -> int`,
 	2571: `bit_count(val: bytes) -> int`,
 	2572: `bit_count(val: varbit) -> int`,
+	2573: `metaphone(source: string, max_output_length: int) -> string`,
 }
 
 var builtinOidsBySignature map[string]oid.Oid

pkg/util/fuzzystrmatch/BUILD.bazel

@@ -4,6 +4,7 @@ go_library(
     name = "fuzzystrmatch",
     srcs = [
         "leven.go",
+        "metaphone.go",
         "soundex.go",
     ],
     importpath = "github.com/cockroachdb/cockroach/pkg/util/fuzzystrmatch",
@@ -15,6 +16,7 @@ go_test(
     size = "small",
     srcs = [
         "leven_test.go",
+        "metaphone_test.go",
         "soundex_test.go",
     ],
     embed = [":fuzzystrmatch"],