#1926 - change bechawior for entry factory

src/cli/tests/Flow/CLI/Tests/Integration/FileSchemaCommandTest.php

@@ -22,7 +22,7 @@ public function test_run_schema() : void
 
         self::assertCommandOutputIdentical(
             <<<'OUTPUT'
-[{"ref":"order_id","type":{"type":"uuid"},"nullable":false,"metadata":[]},{"ref":"created_at","type":{"type":"string"},"nullable":false,"metadata":[]},{"ref":"updated_at","type":{"type":"string"},"nullable":false,"metadata":[]},{"ref":"discount","type":{"type":"string"},"nullable":true,"metadata":[]},{"ref":"address","type":{"type":"json"},"nullable":false,"metadata":[]},{"ref":"notes","type":{"type":"json"},"nullable":false,"metadata":[]},{"ref":"items","type":{"type":"json"},"nullable":false,"metadata":[]}]
+[{"ref":"order_id","type":{"type":"string"},"nullable":false,"metadata":[]},{"ref":"created_at","type":{"type":"string"},"nullable":false,"metadata":[]},{"ref":"updated_at","type":{"type":"string"},"nullable":false,"metadata":[]},{"ref":"discount","type":{"type":"string"},"nullable":true,"metadata":[]},{"ref":"address","type":{"type":"json"},"nullable":false,"metadata":[]},{"ref":"notes","type":{"type":"json"},"nullable":false,"metadata":[]},{"ref":"items","type":{"type":"json"},"nullable":false,"metadata":[]}]
 
 OUTPUT,
             $tester->getDisplay()
@@ -40,7 +40,7 @@ public function test_run_schema_as_ascii() : void
         self::assertCommandOutputIdentical(
             <<<'OUTPUT'
 schema
-|-- order_id: uuid
+|-- order_id: string
 |-- created_at: string
 |-- updated_at: string
 |-- discount: ?string
@@ -85,7 +85,7 @@ public function test_run_schema_with_offset() : void
         self::assertCommandOutputIdentical(
             <<<'OUTPUT'
 schema
-|-- order_id: uuid
+|-- order_id: string
 |-- created_at: string
 |-- updated_at: string
 |-- discount: ?string
@@ -117,7 +117,7 @@ public function test_run_schema_with_offset_and_limit() : void
 +------------+--------+----------+----------+
 |       name |   type | nullable | metadata |
 +------------+--------+----------+----------+
-|   order_id |   uuid |    false |       [] |
+|   order_id | string |    false |       [] |
 | created_at | string |    false |       [] |
 | updated_at | string |    false |       [] |
 |   discount | string |    false |       [] |
@@ -143,7 +143,7 @@ public function test_run_schema_with_php_output() : void
         self::assertCommandOutputIdentical(
             <<<'OUTPUT'
 \Flow\ETL\DSL\schema(
-    \Flow\ETL\DSL\uuid_schema("order_id", nullable: false, metadata: \Flow\ETL\DSL\schema_metadata()),
+    \Flow\ETL\DSL\string_schema("order_id", nullable: false, metadata: \Flow\ETL\DSL\schema_metadata()),
     \Flow\ETL\DSL\string_schema("created_at", nullable: false, metadata: \Flow\ETL\DSL\schema_metadata()),
     \Flow\ETL\DSL\string_schema("updated_at", nullable: false, metadata: \Flow\ETL\DSL\schema_metadata()),
     \Flow\ETL\DSL\string_schema("discount", nullable: true, metadata: \Flow\ETL\DSL\schema_metadata()),
@@ -171,7 +171,7 @@ public function test_run_schema_with_pretty_output() : void
     {
         "ref": "order_id",
         "type": {
-            "type": "uuid"
+            "type": "string"
         },
         "nullable": false,
         "metadata": []
@@ -244,7 +244,7 @@ public function test_run_schema_with_table_output() : void
 +------------+--------+----------+----------+
 |       name |   type | nullable | metadata |
 +------------+--------+----------+----------+
-|   order_id |   uuid |    false |       [] |
+|   order_id | string |    false |       [] |
 | created_at | string |    false |       [] |
 | updated_at | string |    false |       [] |
 |   discount | string |     true |       [] |
@@ -328,7 +328,7 @@ public function test_run_schema_with_table_output_on_excel() : void
 +------------+--------+----------+----------+
 |       name |   type | nullable | metadata |
 +------------+--------+----------+----------+
-|   order_id |   uuid |    false |       [] |
+|   order_id | string |    false |       [] |
 | created_at | string |    false |       [] |
 | updated_at | string |    false |       [] |
 |   discount | string |     true |       [] |
@@ -463,7 +463,7 @@ public function test_run_schema_with_zero_offset() : void
         self::assertCommandOutputIdentical(
             <<<'OUTPUT'
 schema
-|-- order_id: uuid
+|-- order_id: string
 |-- created_at: string
 |-- updated_at: string
 |-- discount: ?string

src/core/etl/src/Flow/ETL/Row/EntryFactory.php

@@ -21,26 +21,16 @@ enum_entry,
     uuid_entry,
     xml_element_entry,
     xml_entry};
-use function Flow\Types\DSL\{type_date,
-    type_datetime,
-    type_html,
-    type_json,
-    type_optional,
-    type_string,
-    type_time,
-    type_uuid,
-    type_xml,
-    type_xml_element};
-use Flow\ETL\Exception\{InvalidArgumentException,
-    RuntimeException,
-    SchemaDefinitionNotFoundException};
+use function Flow\Types\DSL\{type_optional, type_string};
+use Flow\ETL\Exception\{InvalidArgumentException, SchemaDefinitionNotFoundException};
 use Flow\ETL\Row\Entry\{ListEntry, MapEntry, StringEntry, StructureEntry};
 use Flow\ETL\Schema;
 use Flow\ETL\Schema\{Definition, Metadata};
 use Flow\Types\Exception\CastingException;
 use Flow\Types\Type;
 use Flow\Types\Type\Logical\{DateTimeType,
     DateType,
+    HTMLType,
     InstanceOfType,
     JsonType,
     ListType,
@@ -52,7 +42,6 @@ enum_entry,
     UuidType,
     XMLElementType,
     XMLType};
-use Flow\Types\Type\Logical\HTMLType;
 use Flow\Types\Type\Native\{
     ArrayType,
     BooleanType,
@@ -63,17 +52,14 @@ enum_entry,
     StringType,
     UnionType
 };
-use Flow\Types\Type\Native\String\StringTypeChecker;
 use Flow\Types\Type\TypeDetector;
-use Flow\Types\Value\Uuid;
 
 final readonly class EntryFactory
 {
     /**
      * @param null|Definition<mixed>|Schema $schema
      *
      * @throws InvalidArgumentException
-     * @throws RuntimeException
      * @throws SchemaDefinitionNotFoundException
      *
      * @return Entry<mixed>
@@ -96,45 +82,6 @@ public function create(string $entryName, mixed $value, Schema|Definition|null $
 
         $valueType = (new TypeDetector())->detectType($value);
 
-        if ($valueType instanceof StringType) {
-            $value = type_string()->assert($value);
-            $stringChecker = new StringTypeChecker($value);
-
-            if ($stringChecker->isJson()) {
-                $valueType = type_json();
-            } elseif ($stringChecker->isUuid()) {
-                $valueType = type_uuid();
-            } elseif ($stringChecker->isHTML()) {
-                $valueType = type_html();
-            } elseif ($stringChecker->isXML()) {
-                $valueType = type_xml();
-            }
-        }
-
-        if ($valueType instanceof InstanceOfType) {
-            if ($valueType->class === \DOMDocument::class) {
-                $valueType = type_xml();
-            } elseif ($valueType->class === \DOMElement::class) {
-                $valueType = type_xml_element();
-            } elseif ($valueType->class === \DateInterval::class) {
-                $valueType = type_time();
-            } elseif (\in_array($valueType->class, [\DateTimeImmutable::class, \DateTimeInterface::class, \DateTime::class], true)) {
-                if ($value instanceof \DateTimeInterface && $value->format('H:i:s') === '00:00:00') {
-                    $valueType = type_date();
-                } else {
-                    $valueType = type_datetime();
-                }
-            } else {
-                foreach (['Ramsey\Uuid\UuidInterface', Uuid::class, 'Symfony\Component\Uid\Uuid'] as $uuidClass) {
-                    if (\is_a($valueType->class, $uuidClass, true)) {
-                        $valueType = type_uuid();
-
-                        break;
-                    }
-                }
-            }
-        }
-
         return $this->createAs($entryName, $value, $valueType);
     }
 
@@ -223,6 +170,10 @@ public function createAs(string $entryName, mixed $value, Definition|Type $defin
                 return string_entry($entryName, type_optional(type_string())->cast($value), $metadata);
             }
 
+            if ($type instanceof NullType) {
+                return StringEntry::fromNull($entryName, $metadata);
+            }
+
             if ($type instanceof EnumType) {
                 $castValue = type_optional($type)->cast($value);
 

src/core/etl/tests/Flow/ETL/Tests/Unit/Function/CastTest.php

@@ -44,7 +44,7 @@ public static function cast_provider() : array
             'xml_to_array' => [$xml, 'array', ['root' => ['foo' => ['@attributes' => ['baz' => 'buz'], '@value' => 'bar']]]],
             'string_to_xml' => [$xmlString, 'xml', $xml],
             'xml_to_string' => [$xml, 'string', '<root><foo baz="buz">bar</foo></root>'],
-            'full_xml_to_string' => [$fullXMLString, 'string', '<root><foo baz="buz">bar</foo></root>'],
+            'full_xml_to_string' => [$fullXMLString, 'string', $fullXMLString],
             'datetime' => [new \DateTimeImmutable('2023-01-01 00:00:00 UTC'), 'string', '2023-01-01T00:00:00+00:00'],
             'datetime_to_date' => [new \DateTimeImmutable('2023-01-01 00:01:00 UTC'), 'date', new \DateTimeImmutable('2023-01-01T00:00:00+00:00')],
             'string_to_timezone' => ['UTC', 'timezone', new \DateTimeZone('UTC')],

src/core/etl/tests/Flow/ETL/Tests/Unit/Row/EntryFactoryTest.php

@@ -14,6 +14,7 @@ enum_entry,
     json_object_entry,
     list_entry,
     str_entry,
+    string_entry,
     time_entry,
     uuid_entry,
     xml_entry};
@@ -35,9 +36,9 @@ enum_schema,
     time_schema,
     uuid_schema,
     xml_schema};
-use function Flow\Types\DSL\{type_datetime, type_float, type_integer, type_list, type_map, type_string, type_structure, type_time_zone};
+use function Flow\Types\DSL\{type_datetime, type_float, type_integer, type_list, type_map, type_null, type_string, type_structure, type_time_zone};
 use Flow\ETL\Exception\{InvalidArgumentException, SchemaDefinitionNotFoundException};
-use Flow\ETL\Row\Entry\TimeEntry;
+use Flow\ETL\Row\Entry\{StringEntry, TimeEntry};
 use Flow\ETL\Row\{Entry, EntryFactory};
 use Flow\ETL\Schema\Metadata;
 use Flow\ETL\Tests\Fixtures\Enum\BackedIntEnum;
@@ -59,17 +60,17 @@ public static function provide_recognized_data() : \Generator
 
         yield 'xml' => [
             $xml = '<root><foo>1</foo><bar>2</bar><baz>3</baz></root>',
-            xml_entry('e', $xml),
+            string_entry('e', $xml),
         ];
 
         yield 'html' => [
             $html = '<!DOCTYPE html><html><head></head><body><div id="id">2</div><p>3</p></body></html>',
-            html_entry('e', $html),
+            string_entry('e', $html),
         ];
 
         yield 'uuid' => [
             $uuid = '00000000-0000-0000-0000-000000000000',
-            uuid_entry('e', $uuid),
+            string_entry('e', $uuid),
         ];
     }
 
@@ -263,7 +264,7 @@ public function test_html_from_dom_document() : void
     public function test_html_from_string() : void
     {
         self::assertEquals(
-            html_entry('e', $html = '<!DOCTYPE html><html><head></head><body><div>foo</div><p>3</p></body></html>'),
+            string_entry('e', $html = '<!DOCTYPE html><html><head></head><body><div>foo</div><p>3</p></body></html>'),
             $this->entryFactory->create('e', $html)
         );
     }
@@ -418,6 +419,14 @@ public function test_nested_structure() : void
         );
     }
 
+    public function test_null_type_handled() : void
+    {
+        self::assertEquals(
+            StringEntry::fromNull('e'),
+            $this->entryFactory->createAs('e', null, type_null())
+        );
+    }
+
     public function test_object() : void
     {
         $this->expectExceptionMessage("e: object<ArrayIterator> can't be converted to any known Entry, please normalize that object first");
@@ -521,16 +530,17 @@ public function test_uuid_from_ramsey_uuid_library() : void
             self::markTestSkipped("Package 'ramsey/uuid' is required for this test.");
         }
 
+        $uuidObject = Uuid::uuid4();
         self::assertEquals(
-            uuid_entry('e', $uuid = Uuid::uuid4()->toString()),
-            $this->entryFactory->create('e', $uuid)
+            uuid_entry('e', $uuidObject->toString()),
+            $this->entryFactory->create('e', $uuidObject)
         );
     }
 
     public function test_uuid_from_string() : void
     {
         self::assertEquals(
-            uuid_entry('e', $uuid = '00000000-0000-0000-0000-000000000000'),
+            string_entry('e', $uuid = '00000000-0000-0000-0000-000000000000'),
             $this->entryFactory->create('e', $uuid)
         );
     }
@@ -546,7 +556,7 @@ public function test_uuid_string_with_uuid_definition_provided() : void
     public function test_uuid_type() : void
     {
         self::assertEquals(
-            uuid_entry('e', '00000000-0000-0000-0000-000000000000'),
+            string_entry('e', '00000000-0000-0000-0000-000000000000'),
             $this->entryFactory->create('e', '00000000-0000-0000-0000-000000000000')
         );
     }
@@ -580,7 +590,7 @@ public function test_xml_from_dom_document() : void
     public function test_xml_from_string() : void
     {
         self::assertEquals(
-            xml_entry('e', $xml = '<root><foo>1</foo><bar>2</bar><baz>3</baz></root>'),
+            string_entry('e', $xml = '<root><foo>1</foo><bar>2</bar><baz>3</baz></root>'),
             $this->entryFactory->create('e', $xml)
         );
     }

src/lib/types/src/Flow/Types/Type/AutoCaster.php

@@ -4,16 +4,9 @@
 
 namespace Flow\Types\Type;
 
-use function Flow\Types\DSL\{get_type,
-    type_boolean,
-    type_date,
-    type_datetime,
-    type_float,
-    type_integer,
-    type_json,
-    type_time_zone,
-    type_uuid};
-use Flow\Types\Type\Native\String\StringTypeChecker;
+use function Flow\Types\DSL\{get_type, type_float};
+use Flow\Types\Type\Native\NullType;
+use Flow\Types\Type\Native\String\StringTypeNarrower;
 
 final readonly class AutoCaster
 {
@@ -62,44 +55,12 @@ private function castArray(array $value) : array
 
     private function castToString(string $value) : mixed
     {
-        $typeChecker = new StringTypeChecker($value);
+        $narrowedType = (new StringTypeNarrower())->narrow($value);
 
-        if ($typeChecker->isNull()) {
+        if ($narrowedType instanceof NullType) {
             return null;
         }
 
-        if ($typeChecker->isInteger()) {
-            return type_integer()->cast($value);
-        }
-
-        if ($typeChecker->isFloat()) {
-            return type_float()->cast($value);
-        }
-
-        if ($typeChecker->isBoolean()) {
-            return type_boolean()->cast($value);
-        }
-
-        if ($typeChecker->isJson()) {
-            return type_json()->cast($value);
-        }
-
-        if ($typeChecker->isUuid()) {
-            return type_uuid()->cast($value);
-        }
-
-        if ($typeChecker->isTimeZone()) {
-            return type_time_zone()->cast($value);
-        }
-
-        if ($typeChecker->isDate()) {
-            return type_date()->cast($value);
-        }
-
-        if ($typeChecker->isDateTime()) {
-            return type_datetime()->cast($value);
-        }
-
-        return $value;
+        return $narrowedType->cast($value);
     }
 }