Casting XML to string (#1475)

* Failing example for Xml insert

* Add test on DOMDocument

* Update to use actual domdocuments

* Add example of casting to string

* Add another test for xml writing of dom document

* Lint

* Update to use loadXml

* Unified inconsistent behavior of casting xml to string

* Added test for saving also xml element entry to database through doctrine

---------

Co-authored-by: John Mortlock <jmortlock@zenglobal.au>

src/adapter/etl-adapter-doctrine/tests/Flow/ETL/Adapter/Doctrine/Tests/Integration/DbalLoaderTest.php

@@ -261,6 +261,96 @@ public function test_inserts_new_rows_or_updates_already_existed_based_on_primar
         );
     }
 
+    public function test_inserts_xml_element_entry() : void
+    {
+        $this->pgsqlDatabaseContext->createTable((new Table(
+            $table = 'flow_doctrine_bulk_test',
+            [
+                new Column('id', Type::getType(Types::INTEGER), ['notnull' => true]),
+                new Column('name', Type::getType(Types::STRING), ['notnull' => true, 'length' => 255]),
+                new Column('description', Type::getType(Types::STRING), ['notnull' => true, 'length' => 255]),
+            ],
+        ))
+            ->setPrimaryKey(['id']));
+
+        $loader = to_dbal_table_insert($this->connectionParams(), $table);
+
+        $documentA = new \DOMDocument();
+        $documentA->loadXml('<xml>Description One</xml>');
+
+        $documentB = new \DOMDocument();
+        $documentB->loadXml('<xml>Description Two</xml>');
+
+        $documentC = new \DOMDocument();
+        $documentC->loadXml('<xml>Description Three</xml>');
+
+        (data_frame())
+            ->read(
+                from_array([
+                    ['id' => 1, 'name' => 'Name One', 'description' => $documentA->getElementsByTagName('xml')[0]],
+                    ['id' => 2, 'name' => 'Name Two', 'description' => $documentB->getElementsByTagName('xml')[0]],
+                    ['id' => 3, 'name' => 'Name Three', 'description' => $documentC->getElementsByTagName('xml')[0]],
+                ]),
+            )
+            ->load($loader)
+            ->run();
+
+        self::assertEquals(3, $this->pgsqlDatabaseContext->tableCount($table));
+        self::assertEquals(
+            [
+                ['id' => 1, 'name' => 'Name One', 'description' => '<xml>Description One</xml>'],
+                ['id' => 2, 'name' => 'Name Two', 'description' => '<xml>Description Two</xml>'],
+                ['id' => 3, 'name' => 'Name Three', 'description' => '<xml>Description Three</xml>'],
+            ],
+            $this->pgsqlDatabaseContext->selectAll($table)
+        );
+    }
+
+    public function test_inserts_xml_entry() : void
+    {
+        $this->pgsqlDatabaseContext->createTable((new Table(
+            $table = 'flow_doctrine_bulk_test',
+            [
+                new Column('id', Type::getType(Types::INTEGER), ['notnull' => true]),
+                new Column('name', Type::getType(Types::STRING), ['notnull' => true, 'length' => 255]),
+                new Column('description', Type::getType(Types::STRING), ['notnull' => true, 'length' => 255]),
+            ],
+        ))
+            ->setPrimaryKey(['id']));
+
+        $loader = to_dbal_table_insert($this->connectionParams(), $table);
+
+        $documentA = new \DOMDocument();
+        $documentA->loadXml('<xml>Description One</xml>');
+
+        $documentB = new \DOMDocument();
+        $documentB->loadXml('<xml>Description Two</xml>');
+
+        $documentC = new \DOMDocument();
+        $documentC->loadXml('<b>Description Three</b>');
+
+        (data_frame())
+            ->read(
+                from_array([
+                    ['id' => 1, 'name' => 'Name One', 'description' => $documentA],
+                    ['id' => 2, 'name' => 'Name Two', 'description' => $documentB],
+                    ['id' => 3, 'name' => 'Name Three', 'description' => $documentC],
+                ]),
+            )
+            ->load($loader)
+            ->run();
+
+        self::assertEquals(3, $this->pgsqlDatabaseContext->tableCount($table));
+        self::assertEquals(
+            [
+                ['id' => 1, 'name' => 'Name One', 'description' => '<xml>Description One</xml>'],
+                ['id' => 2, 'name' => 'Name Two', 'description' => '<xml>Description Two</xml>'],
+                ['id' => 3, 'name' => 'Name Three', 'description' => '<b>Description Three</b>'],
+            ],
+            $this->pgsqlDatabaseContext->selectAll($table)
+        );
+    }
+
     public function test_update_multiple_rows_at_once() : void
     {
         $this->pgsqlDatabaseContext->createTable((new Table(

src/adapter/etl-adapter-json/tests/Flow/ETL/Adapter/JSON/Tests/Integration/JsonTest.php

@@ -9,12 +9,34 @@
 use function Flow\ETL\DSL\{average, df, from_array, overwrite, ref};
 use function Flow\ETL\DSL\{config, flow_context, rows};
 use function Flow\Filesystem\DSL\path;
+
 use Flow\ETL\Adapter\JSON\JsonLoader;
 use Flow\ETL\Tests\Double\FakeExtractor;
 use Flow\ETL\{Tests\FlowTestCase};
 
 final class JsonTest extends FlowTestCase
 {
+    public function test_domdocument_json_file() : void
+    {
+        $domDocument = new \DOMDocument();
+        $domDocument->loadXml('<b>red</b>');
+
+        df()
+            ->read(from_array([
+                ['id' => 1, 'descriptionHtml' => $domDocument, 'size' => 'small'],
+            ]))
+            ->saveMode(overwrite())
+            ->write(to_json($path = __DIR__ . '/var/test_domdocument.json'))
+            ->run();
+
+        self::assertStringContainsString(
+            <<<'JSON'
+[{"id":1,"descriptionHtml":"<b>red<\/b>","size":"small"}]
+JSON,
+            \file_get_contents($path)
+        );
+    }
+
     public function test_json_loader() : void
     {
         $path = __DIR__ . '/var/test_json_loader.json';

src/core/etl/src/Flow/ETL/PHP/Type/Caster/StringCastingHandler.php

@@ -41,7 +41,7 @@ public function value(mixed $value, Type $type, Caster $caster, Options $options
         }
 
         if ($value instanceof \DOMDocument) {
-            return $value->saveXML() ?: '';
+            return $value->saveXML($value->documentElement) ?: '';
         }
 
         if ($value instanceof \DOMElement) {

src/core/etl/tests/Flow/ETL/Tests/Integration/Function/DOMElementValueTest.php

@@ -4,11 +4,37 @@
 
 namespace Flow\ETL\Tests\Integration\Function;
 
-use function Flow\ETL\DSL\{df, from_rows, ref, row, rows, xml_element_entry, xml_entry};
+use function Flow\ETL\DSL\{df, from_rows, ref, row, rows, type_string, xml_element_entry, xml_entry};
+
 use Flow\ETL\Tests\FlowTestCase;
 
 final class DOMElementValueTest extends FlowTestCase
 {
+    public function test_dom_element_cast_as_string() : void
+    {
+        $document = new \DOMDocument();
+        $document->loadXml('<b>User Name 01</b>');
+
+        $rows = df()
+            ->read(from_rows(
+                rows(
+                    row(
+                        xml_entry('html_raw', $document)
+                    )
+                )
+            ))
+            ->withEntry('html', ref('html_raw')->cast(type_string()))
+            ->drop('html_raw')
+            ->fetch();
+
+        self::assertSame(
+            [
+                ['html' => '<b>User Name 01</b>'],
+            ],
+            $rows->toArray()
+        );
+    }
+
     public function test_dom_element_value() : void
     {
         $rows = df()
@@ -53,6 +79,31 @@ public function test_dom_element_value_from_dom_document() : void
         );
     }
 
+    public function test_dom_element_value_on_dom_document() : void
+    {
+        $document = new \DOMDocument();
+        $document->loadXml('<b>User Name 01</b>');
+
+        $rows = df()
+            ->read(from_rows(
+                rows(
+                    row(
+                        xml_entry('html_raw', $document)
+                    )
+                )
+            ))
+            ->withEntry('html', ref('html_raw')->domElementValue())
+            ->drop('html_raw')
+            ->fetch();
+
+        self::assertSame(
+            [
+                ['html' => 'User Name 01'],
+            ],
+            $rows->toArray()
+        );
+    }
+
     public function test_dom_element_value_on_xpath_result() : void
     {
         $rows = df()

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

@@ -41,7 +41,7 @@ public static function cast_provider() : array
             'json_pretty' => [[1], 'json_pretty', "[\n    1\n]"],
             'xml_to_array' => [$xml, 'array', ['root' => ['foo' => ['@attributes' => ['baz' => 'buz'], '@value' => 'bar']]]],
             'string_to_xml' => [$xmlString, 'xml', $xml],
-            'xml_to_string' => [$xml, 'string', $fullXMLString],
+            'xml_to_string' => [$xml, 'string', '<root><foo baz="buz">bar</foo></root>'],
             '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')],
             'uuid' => [Uuid::fromString('a0eebc99-9c0b-4ef8-bb6d-6bb9bd380a11'), 'string', 'a0eebc99-9c0b-4ef8-bb6d-6bb9bd380a11'],

src/core/etl/tests/Flow/ETL/Tests/Unit/PHP/Type/Caster/StringCastingHandlerTest.php

@@ -27,6 +27,11 @@ public function __toString() : string
             }
         }, 'stringable'];
         yield 'DOMDocument' => [new \DOMDocument(), '<?xml version="1.0"?>'];
+
+        $xml = (new \DOMDocument());
+        $xml->loadXML('<xml>Some Happy XML</xml>');
+
+        yield 'Not Empty DOMDocument' => [$xml, '<xml>Some Happy XML</xml>'];
         yield 'DOMElement' => [new \DOMElement('element'), '<element/>'];
     }
 

src/lib/doctrine-dbal-bulk/src/Flow/Doctrine/Bulk/BulkData.php

@@ -4,6 +4,7 @@
 
 namespace Flow\Doctrine\Bulk;
 
+use function Flow\ETL\DSL\dom_element_to_string;
 use Doctrine\DBAL\Types\{Type, Types};
 use Flow\Doctrine\Bulk\Exception\RuntimeException;
 
@@ -131,7 +132,7 @@ public function toSqlParameters(TableDefinition $table) : array
                         default => $entry,
                     },
                     'array' => match (Type::getTypeRegistry()->lookupName($table->dbalColumn($column)->getType())) {
-                        Types::TEXT => \json_encode($entry, JSON_THROW_ON_ERROR),
+                        Types::TEXT, Types::STRING => \json_encode($entry, JSON_THROW_ON_ERROR),
                         default => $entry,
                     },
                     'object' => match ($entry::class) {
@@ -143,6 +144,16 @@ public function toSqlParameters(TableDefinition $table) : array
                             Types::DATETIME_IMMUTABLE => \DateTimeImmutable::createFromMutable($entry),
                             default => $entry,
                         },
+                        \DOMDocument::class => match (Type::getTypeRegistry()->lookupName($table->dbalColumn($column)->getType())) {
+                            Types::TEXT,
+                            Types::STRING => $entry->saveXML($entry->documentElement),
+                            default => $entry,
+                        },
+                        \DOMElement::class => match (Type::getTypeRegistry()->lookupName($table->dbalColumn($column)->getType())) {
+                            Types::TEXT,
+                            Types::STRING => (string) dom_element_to_string($entry),
+                            default => $entry,
+                        },
                         default => $entry,
                     },
                     default => $entry,