ESQL: Add url_encode function

- Rename the capability to exclude the decoding portion, since this PR is only concerned with encoding.
- Mark url_encode as SNAPSHOT to allow changes in encoding logic.
- Add tests that use random strings to test a wide range of characters, that run alongside the ones that use random URLs.
- Minor change in the sample csv test string to showcase a fictional URL with path and few query params.

Author: mouhc1ne

docs/changelog/132869.yaml

@@ -2467,34 +2467,33 @@ warning:Line 2:9: java.lang.IllegalArgumentException: single-value function enco
 ;
 
 url_encode sample for docs
-required_capability: url_decode_encode
+required_capability: url_encode
 
 // tag::url_encode[]
-ROW u = "http://elastic.co" 
-| EVAL e = URL_ENCODE(u)
+ROW u = "https://www.example.com/papers?q=information+retrieval&year=2024&citations=high" | EVAL u = URL_ENCODE(u)
 // end::url_encode[]
 ;
 
 // tag::url_encode-result[]
-u:keyword           | e:keyword
-http://elastic.co   | http%3A%2F%2Felastic.co
+u:keyword
+https%3A%2F%2Fwww.example.com%2Fpapers%3Fq%3Dinformation%2Bretrieval%26year%3D2024%26citations%3Dhigh
 // end::url_encode-result[]
 ;
 
 url_encode mixed functions tests
 FROM employees 
-    | WHERE emp_no == 10001 
-    | EVAL a = TRIM(URL_ENCODE(first_name))
-    | EVAL b = URL_ENCODE(TO_LOWER(first_name))
-    | KEEP a,b;
+| WHERE emp_no == 10001 
+| EVAL a = TRIM(URL_ENCODE(first_name))
+| EVAL b = URL_ENCODE(TO_LOWER(first_name))
+| KEEP a,b;
 
 a:keyword | b:keyword
 Georgi    | georgi 
 ;
 
 url_encode mixed input tests
-ROW u = ["hello elastic!", "a+b-c%d", "", "!#$&'()+,/:;=?@[]"] | EVAL u = URL_ENCODE(u);
+ROW u = ["hello elastic!", "a+b-c%d", "", "!#$&'()*+,/:;=?@[]"] | EVAL u = URL_ENCODE(u);
 
 u:keyword
-["hello+elastic%21", "a%2Bb-c%25d", "", "%21%23%24%26%27%28%29%2B%2C%2F%3A%3B%3D%3F%40%5B%5D"]
+["hello+elastic%21", "a%2Bb-c%25d", "", "%21%23%24%26%27%28%29*%2B%2C%2F%3A%3B%3D%3F%40%5B%5D"]
 ;

docs/reference/query-languages/esql/_snippets/functions/examples/url_encode.md

@@ -1399,12 +1399,12 @@ public enum Cap {
         /**
          * Allow qualifiers in attribute names.
          */
-        NAME_QUALIFIERS(Build.current().isSnapshot());
+        NAME_QUALIFIERS(Build.current().isSnapshot()),
 
         /**
          * URL encoding function.
          */
-        URL_ENCODE;
+        URL_ENCODE(Build.current().isSnapshot());
 
         private final boolean enabled;
 

docs/reference/query-languages/esql/kibana/definition/functions/url_encode.json

@@ -79,8 +79,8 @@ public EvalOperator.ExpressionEvaluator.Factory toEvaluator(ToEvaluator toEvalua
 
     @ConvertEvaluator()
     static BytesRef process(final BytesRef val) {
-        String s = val.utf8ToString();
-        String encoded = URLEncoder.encode(s, StandardCharsets.UTF_8);
+        String input = val.utf8ToString();
+        String encoded = URLEncoder.encode(input, StandardCharsets.UTF_8);
         return new BytesRef(encoded);
     }
 

docs/reference/query-languages/esql/kibana/docs/functions/url_encode.md

@@ -11,6 +11,7 @@
 import com.carrotsearch.randomizedtesting.annotations.ParametersFactory;
 
 import org.apache.lucene.util.BytesRef;
+import org.elasticsearch.common.lucene.BytesRefs;
 import org.elasticsearch.xpack.esql.core.expression.Expression;
 import org.elasticsearch.xpack.esql.core.tree.Source;
 import org.elasticsearch.xpack.esql.core.type.DataType;
@@ -28,6 +29,9 @@
 
 @FunctionName("url_encode")
 public class UrlEncodeTests extends AbstractScalarFunctionTestCase {
+
+    private record RandomUrl(String plain, String encoded) {}
+
     public UrlEncodeTests(@Name("TestCase") Supplier<TestCaseSupplier.TestCase> testCaseSupplier) {
         this.testCase = testCaseSupplier.get();
     }
@@ -37,8 +41,16 @@ public static Iterable<Object[]> parameters() {
         List<TestCaseSupplier> suppliers = new ArrayList<>();
 
         for (DataType dataType : DataType.stringTypes()) {
-            TestCaseSupplier supplier = new TestCaseSupplier(List.of(dataType), () -> createTestCase(dataType));
-            suppliers.add(supplier);
+            suppliers.add(new TestCaseSupplier(List.of(dataType), () -> createTestCaseWithRandomUrl(dataType)));
+
+            for (TestCaseSupplier.TypedDataSupplier supplier : TestCaseSupplier.stringCases(dataType)) {
+                TestCaseSupplier testCaseSupplier = new TestCaseSupplier(
+                    supplier.name(),
+                    List.of(supplier.type()),
+                    () -> createTestCaseWithRandomString(dataType, supplier)
+                );
+                suppliers.add(testCaseSupplier);
+            }
         }
 
         return parameterSuppliersFromTypedDataWithDefaultChecksNoErrors(false, suppliers);
@@ -49,24 +61,45 @@ protected Expression build(Source source, List<Expression> args) {
         return new UrlEncode(source, args.get(0));
     }
 
-    private static TestCaseSupplier.TestCase createTestCase(DataType dataType) {
-        String url = generateRandomUrl();
-        BytesRef input = new BytesRef(url);
+    private static TestCaseSupplier.TestCase createTestCaseWithRandomUrl(DataType dataType) {
+        RandomUrl url = generateRandomUrl();
+        BytesRef input = new BytesRef(url.plain());
+        BytesRef output = new BytesRef(url.encoded());
+        TestCaseSupplier.TypedData fieldTypedData = new TestCaseSupplier.TypedData(input, dataType, "string");
+
+        return new TestCaseSupplier.TestCase(
+            List.of(fieldTypedData),
+            "UrlEncodeEvaluator[val=Attribute[channel=0]]",
+            dataType,
+            equalTo(output)
+        );
+    }
+
+    private static TestCaseSupplier.TestCase createTestCaseWithRandomString(
+        DataType dataType,
+        TestCaseSupplier.TypedDataSupplier supplier
+    ) {
+        TestCaseSupplier.TypedData fieldTypedData = supplier.get();
+        BytesRef input = BytesRefs.toBytesRef(fieldTypedData.data());
         BytesRef output = new BytesRef(URLEncoder.encode(input.utf8ToString(), StandardCharsets.UTF_8));
 
         return new TestCaseSupplier.TestCase(
-            List.of(new TestCaseSupplier.TypedData(input, dataType, "string")),
+            List.of(fieldTypedData),
             "UrlEncodeEvaluator[val=Attribute[channel=0]]",
             dataType,
             equalTo(output)
         );
     }
 
-    private static String generateRandomUrl() {
+    private static RandomUrl generateRandomUrl() {
         String protocol = randomFrom("http://", "https://", "");
         String domain = String.format("%s.com", randomAlphaOfLengthBetween(3, 10));
         String path = randomFrom("", "/" + randomAlphanumericOfLength(5) + "/");
         String query = randomFrom("", "?" + randomAlphaOfLength(5) + "=" + randomAlphanumericOfLength(5));
-        return String.format("%s%s%s%s", protocol, domain, path, query);
+
+        String plain = String.format("%s%s%s%s", protocol, domain, path, query);
+        String encoded = URLEncoder.encode(plain, StandardCharsets.UTF_8);
+
+        return new RandomUrl(plain, encoded);
     }
 }