Fix parquet fixed array length (#2257)

* refactor: parquet binary encoding to standalone functions

- Convert ByteConverter class to standalone functions in
Binary/functions.php (encode_i8, decode_i8, encode_i32, decode_i32,
encode_f64, decode_f64, etc.)
- Remove ByteOrder parameter from i8/u8 functions (single bytes have no
endianness)
- Add precision validation to encode_decimal and decode_decimal
functions
- Fix UuidConverter to handle Stringable objects (e.g.,
Flow\Types\Value\Uuid)
- Fix static analysis errors:
  - Add Bytes type to RLEDictionaryChunkBuilder::$pageValues annotation
  - Handle Bytes objects as array keys in ScalarDictionaryBuilder
  - Remove unnecessary @phpstan-ignore annotations in RLEBitPackedHybrid
  - Regenerate CLI test fixture orders.parquet with correct binary UUID
  - Update CLI test expectations to match regenerated fixture

* refactor: parquet performance optimizations

* refactor: optimize dremel shredder algorithm

* refactor: reduce computation of parquet flat path

* refactor: ColumnDataalidator optimizations

* fix: missing dsl definitions

Author: norberttech

composer.json

@@ -201,6 +201,7 @@
             "src/lib/azure-sdk/src/Flow/Azure/SDK/DSL/functions.php",
             "src/lib/filesystem/src/Flow/Filesystem/DSL/functions.php",
             "src/lib/parquet/src/Flow/Parquet/functions.php",
+            "src/lib/parquet/src/Flow/Parquet/Binary/functions.php",
             "src/lib/parquet/src/stubs.php",
             "src/lib/postgresql/src/Flow/PostgreSql/DSL/functions.php",
             "src/lib/postgresql/src/stubs.php",

src/adapter/etl-adapter-parquet/src/Flow/ETL/Adapter/Parquet/ParquetExtractor.php

@@ -8,7 +8,8 @@
 use Flow\ETL\{Exception\InvalidArgumentException, Extractor, FlowContext};
 use Flow\ETL\Extractor\{FileExtractor, Limitable, LimitableExtractor, PathFiltering, Signal};
 use Flow\Filesystem\{Path, SourceStream};
-use Flow\Parquet\{ByteOrder, Options, ParquetFile, Reader};
+use Flow\Parquet\Binary\ByteOrder;
+use Flow\Parquet\{Options, ParquetFile, Reader};
 
 final class ParquetExtractor implements Extractor, FileExtractor, LimitableExtractor
 {

src/adapter/etl-adapter-parquet/src/Flow/ETL/Adapter/Parquet/functions.php

@@ -12,7 +12,8 @@
 };
 use Flow\ETL\Schema;
 use Flow\Filesystem\Path;
-use Flow\Parquet\{ByteOrder, Options};
+use Flow\Parquet\Binary\ByteOrder;
+use Flow\Parquet\Options;
 use Flow\Parquet\ParquetFile\Compressions;
 
 /**

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

@@ -98,15 +98,15 @@ public function test_read_rows_parquet() : void
 
         self::assertCommandOutputIdentical(
             <<<'OUTPUT'
-+----------------------+----------------------+----------------------+-----------+----------------------+-------------------+----------------------+----------------------+----------------------+
-|             order_id |           created_at |           updated_at |  discount |                email |          customer |              address |                notes |                items |
-+----------------------+----------------------+----------------------+-----------+----------------------+-------------------+----------------------+----------------------+----------------------+
-| 1e4544ab-7c94-3d39-b | 2024-10-02T23:01:19+ | 2024-11-18T12:21:16+ | 26.209999 | shany14@baumbach.org |  Rafaela Hartmann | {"street":"64610 Kat | ["Deleniti vitae dol | [{"sku":"SKU_0005"," |
-| f1aba27a-3387-3e10-b | 2024-09-04T01:26:42+ | 2024-11-15T05:27:03+ |           | okon.haley@yahoo.com | Marjolaine Kohler | {"street":"14054 Ker | ["Nulla exercitation | [{"sku":"SKU_0003"," |
-| c8d23f0b-c157-323f-8 | 2024-02-14T14:07:07+ | 2024-11-06T13:38:25+ |           | rolfson.noble@hotmai |  Loyce McLaughlin | {"street":"9058 Kess | ["Laborum molestiae  | [{"sku":"SKU_0005"," |
-| d6215090-cea0-3fd9-a | 2024-10-12T09:18:12+ | 2024-11-21T09:38:15+ |           | lruecker@hotmail.com |  Estelle Schinner | {"street":"68058 Dav | ["In dolore nam et s | [{"sku":"SKU_0003"," |
-| ac622a00-7de2-3eb0-b | 2024-06-07T10:27:53+ | 2024-11-23T21:03:36+ |           | morar.beth@mayer.net |  Ethan Hodkiewicz | {"street":"84594 Vla | ["Vel ipsam id quos  | [{"sku":"SKU_0004"," |
-+----------------------+----------------------+----------------------+-----------+----------------------+-------------------+----------------------+----------------------+----------------------+
++----------------------+----------------------+----------------------+-----------+----------------------+------------------+----------------------+----------------------+----------------------+
+|             order_id |           created_at |           updated_at |  discount |                email |         customer |              address |                notes |                items |
++----------------------+----------------------+----------------------+-----------+----------------------+------------------+----------------------+----------------------+----------------------+
+| 4e93f175-0c89-30df-a | 2026-01-29T13:48:47+ | 2026-02-15T13:48:47+ | 27.969999 | fahey.aurelie@denesi |     Jerod Abbott | {"street":"5567 Grim | ["Maxime et sed impe | [{"sku":"SKU_0004"," |
+| db560d65-3a26-359e-8 | 2026-02-15T18:39:41+ |                      | 35.990002 |   dusty70@howell.com |    Omari McGlynn | {"street":"86184 Mck | ["Mollitia eos optio | [{"sku":"SKU_0001"," |
+| f24d7f7d-1615-3b5b-a | 2026-01-26T05:23:13+ | 2026-02-01T05:23:13+ | 42.669998 |   linnea91@gmail.com | Kendall Weissnat | {"street":"65268 Apr | ["Ut nesciunt volupt | [{"sku":"SKU_0003"," |
+| 78fe7069-f06f-3081-b | 2026-03-12T13:08:56+ |                      |           | glubowitz@morissette | Roman Balistreri | {"street":"1722 Hall | ["Velit vero invento | [{"sku":"SKU_0005"," |
+| e3fb781c-34ff-380f-8 | 2026-01-10T18:34:34+ | 2026-02-06T18:34:34+ |           |    walton60@hand.com | Sedrick Ondricka | {"street":"201 Mosci | ["Doloremque culpa i | [{"sku":"SKU_0004"," |
++----------------------+----------------------+----------------------+-----------+----------------------+------------------+----------------------+----------------------+----------------------+
 5 rows
 
 OUTPUT,

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

@@ -388,7 +388,7 @@ public function test_run_schema_with_table_output_on_parquet() : void
 +------------+----------+----------+----------+
 |   order_id |     uuid |    false |       [] |
 | created_at | datetime |    false |       [] |
-| updated_at | datetime |    false |       [] |
+| updated_at | datetime |     true |       [] |
 |   discount |    float |     true |       [] |
 |      email |   string |    false |       [] |
 |   customer |   string |    false |       [] |

src/cli/tests/Flow/CLI/Tests/Integration/Fixtures/orders.parquet

@@ -33,6 +33,7 @@
         },
         "files": [
             "src/Flow/Parquet/functions.php",
+            "src/Flow/Parquet/Binary/functions.php",
             "src/stubs.php"
         ]
     },

src/lib/parquet/composer.json

@@ -2,7 +2,7 @@
 
 declare(strict_types=1);
 
-namespace Flow\Parquet;
+namespace Flow\Parquet\Binary;
 
 enum ByteOrder : string
 {

…b/parquet/src/Flow/Parquet/ByteOrder.php …et/src/Flow/Parquet/Binary/ByteOrder.phpsrc/lib/parquet/src/Flow/Parquet/ByteOrder.php renamed to src/lib/parquet/src/Flow/Parquet/Binary/ByteOrder.php src/lib/parquet/src/Flow/Parquet/ByteOrder.php renamed to src/lib/parquet/src/Flow/Parquet/Binary/ByteOrder.php

@@ -2,16 +2,24 @@
 
 declare(strict_types=1);
 
-namespace Flow\Parquet\BinaryReader;
+namespace Flow\Parquet\Binary;
 
-use Flow\Parquet\{ByteOrder, DataSize};
+use Flow\Parquet\DataSize;
 
+/**
+ * @implements \ArrayAccess<int, int>
+ * @implements \IteratorAggregate<int, int>
+ */
 final class Bytes implements \ArrayAccess, \Countable, \IteratorAggregate
 {
+    /** @var ?\ArrayIterator<int, int> */
     private ?\ArrayIterator $iterator = null;
 
     private readonly DataSize $size;
 
+    /**
+     * @param array<int, int> $bytes
+     */
     public function __construct(
         private array $bytes,
         private readonly ByteOrder $byteOrder = ByteOrder::LITTLE_ENDIAN,
@@ -31,7 +39,9 @@ public function count() : int
         return \count($this->bytes);
     }
 
-    // IteratorAggregate methods
+    /**
+     * @return \ArrayIterator<int, int>
+     */
     public function getIterator() : \ArrayIterator
     {
         if ($this->iterator === null) {
@@ -41,18 +51,17 @@ public function getIterator() : \ArrayIterator
         return $this->iterator;
     }
 
-    // ArrayAccess methods
-    public function offsetExists($offset) : bool
+    public function offsetExists(mixed $offset) : bool
     {
         return isset($this->bytes[$offset]);
     }
 
-    public function offsetGet($offset) : mixed
+    public function offsetGet(mixed $offset) : int
     {
         return $this->bytes[$offset];
     }
 
-    public function offsetSet($offset, $value) : void
+    public function offsetSet(mixed $offset, mixed $value) : void
     {
         if ($offset === null) {
             $this->bytes[] = $value;
@@ -61,7 +70,7 @@ public function offsetSet($offset, $value) : void
         }
     }
 
-    public function offsetUnset($offset) : void
+    public function offsetUnset(mixed $offset) : void
     {
         unset($this->bytes[$offset]);
     }