builtins: Implement unaccent function

docs/generated/sql/functions.md

@@ -2432,6 +2432,8 @@ The swap_ordinate_string parameter is a 2-character string naming the ordinates
 <tr><td><a name="translate"></a><code>translate(input: <a href="string.html">string</a>, find: <a href="string.html">string</a>, replace: <a href="string.html">string</a>) &rarr; <a href="string.html">string</a></code></td><td><span class="funcdesc"><p>In <code>input</code>, replaces the first character from <code>find</code> with the first character in <code>replace</code>; repeat for each character in <code>find</code>.</p>
 <p>For example, <code>translate('doggie', 'dog', '123');</code> returns <code>1233ie</code>.</p>
 </span></td></tr>
+<tr><td><a name="unaccent"></a><code>unaccent(val: <a href="string.html">string</a>) &rarr; <a href="string.html">string</a></code></td><td><span class="funcdesc"><p>Removes accents (diacritic signs) from the text provided in <code>val</code>.</p>
+</span></td></tr>
 <tr><td><a name="upper"></a><code>upper(val: <a href="string.html">string</a>) &rarr; <a href="string.html">string</a></code></td><td><span class="funcdesc"><p>Converts all characters in <code>val</code> to their to their upper-case equivalents.</p>
 </span></td></tr></tbody>
 </table>

pkg/sql/logictest/testdata/logic_test/builtin_function

@@ -102,6 +102,21 @@ SELECT upper('roacH7')
 upper
 ROACH7
 
+query T
+SELECT unaccent(str) FROM ( VALUES 
+   ('no_special_CHARACTERS1!'),
+   ('Żółć'),
+   ('⃞h̀ＥＬＬＯ`̀́⃞'),
+   ('Softhyphen­separator'),
+   ('some ̂thing')
+) tbl(str)
+----
+no_special_CHARACTERS1!
+Zolc
+hELLO`
+Softhyphen-separator
+something
+
 statement error unknown signature: upper\(decimal\)
 SELECT upper(2.2)
 

pkg/sql/sem/builtins/builtins.go

@@ -61,6 +61,7 @@ import (
 	"github.com/cockroachdb/cockroach/pkg/util/timeofday"
 	"github.com/cockroachdb/cockroach/pkg/util/timetz"
 	"github.com/cockroachdb/cockroach/pkg/util/timeutil"
+	"github.com/cockroachdb/cockroach/pkg/util/unaccent"
 	"github.com/cockroachdb/cockroach/pkg/util/uuid"
 	"github.com/cockroachdb/errors"
 	"github.com/knz/strtime"
@@ -260,6 +261,26 @@ var builtins = map[string]builtinDefinition{
 		),
 	),
 
+	"unaccent": makeBuiltin(tree.FunctionProperties{Category: categoryString},
+		stringOverload1(
+			func(evalCtx *tree.EvalContext, s string) (tree.Datum, error) {
+				var b strings.Builder
+				for _, ch := range s {
+					v, ok := unaccent.Dictionary[ch]
+					if ok {
+						b.WriteString(v)
+					} else {
+						b.WriteRune(ch)
+					}
+				}
+				return tree.NewDString(b.String()), nil
+			},
+			types.String,
+			"Removes accents (diacritic signs) from the text provided in `val`.",
+			tree.VolatilityImmutable,
+		),
+	),
+
 	"upper": makeBuiltin(tree.FunctionProperties{Category: categoryString},
 		stringOverload1(
 			func(evalCtx *tree.EvalContext, s string) (tree.Datum, error) {