Adding TabularDataReader::chunkBy method

Author: nyamsprod

CHANGELOG.md

@@ -8,6 +8,7 @@ All Notable changes to `Csv` will be documented in this file
 
 - `Statement::select`
 - `TabularDataReader::getRecordsAsObject`
+- `TabularDataReader::chunkBy`
 
 ### Deprecated
 

docs/9.0/reader/tabular-data-reader.md

@@ -472,3 +472,23 @@ $reader->matchingFirstOrFail('row=3-1;4-6'); // will throw
 
 <p class="message-info"> Wraps the functionality of <code>FragmentFinder</code> class.</p>
 <p class="message-notice">Added in version <code>9.12.0</code> for <code>Reader</code> and <code>ResultSet</code>.</p>
+
+### chunkBy
+
+<p class="message-notice">Added in version <code>9.15.0</code> for <code>Reader</code> and <code>ResultSet</code>.</p>
+
+If you are dealing with a large CSV and you want it to be split in smaller sizes for better handling you can use
+the `chunkBy` method which breaks the `TabularDataReader` into multiple, smaller instance of a given size. The
+last instance may contain fewer records because of the chunk size you have chosen.
+
+```php
+use League\Csv\Reader;
+
+$reader = Reader::createFromString($csv);
+
+foreach ($reader->chunkBy(4) as $chunk) {
+    foreach ($chunk as $record) {
+        //the actual record will be found here.
+    }
+}
+```

src/Reader.php

@@ -319,6 +319,18 @@ public function reduce(Closure $closure, mixed $initial = null): mixed
         return ResultSet::createFromTabularDataReader($this)->reduce($closure, $initial);
     }
 
+    /**
+     * @param positive-int $length
+     *
+     * @throws InvalidArgument
+     *
+     * @return Iterator<TabularDataReader>
+     */
+    public function chunkBy(int $length): Iterator
+    {
+        return ResultSet::createFromTabularDataReader($this)->chunkBy($length);
+    }
+
     /**
      * @param Closure(array<mixed>, array-key): bool $closure
      *

src/ResultSet.php

@@ -160,6 +160,36 @@ public function reduce(Closure $closure, mixed $initial = null): mixed
         return $initial;
     }
 
+    /**
+     * @param positive-int $length
+     *
+     * @throws InvalidArgument
+     *
+     * @return Iterator<TabularDataReader>
+     */
+    public function chunkBy(int $length): Iterator
+    {
+        if ($length < 1) {
+            throw InvalidArgument::dueToInvalidChunkSize($length, __METHOD__);
+        }
+
+        $records = [];
+        $i = 0;
+        foreach ($this->getRecords() as $record) {
+            $records[] = $record;
+            ++$i;
+            if ($i === $length) {
+                yield self::createFromRecords($records);
+                $i = 0;
+                $records = [];
+            }
+        }
+
+        if ([] !== $records) {
+            yield self::createFromRecords($records);
+        }
+    }
+
     public function filter(Closure $closure): TabularDataReader
     {
         return Statement::create()->where($closure)->process($this);

src/TabularDataReader.php

@@ -40,6 +40,7 @@
  * @method TabularDataReader matchingFirstOrFail(string $expression) extract the first found fragment identifier of the tabular data or fail
  * @method TabularDataReader|null matchingFirst(string $expression) extract the first found fragment identifier of the tabular data or return null if none is found
  * @method iterable<int, TabularDataReader> matching(string $expression) extract all found fragment identifiers for the tabular data
+ * @method iterable<int, TabularDataReader> chunkBy(int $length) Chunk the TabulaDataReader into smaller TabularDataReader instance of the given size or less.
  */
 interface TabularDataReader extends Countable, IteratorAggregate
 {

src/TabularDataReaderTestCase.php

@@ -449,6 +449,17 @@ public function __construct(
 
         self::assertInstanceOf($class::class, $this->tabularDataWithHeader()->firstAsObject($class::class, ['observedOn', 'temperature', 'place']));
     }
+
+    public function testChunkingTabularDataUsingTheRangeMethod(): void
+    {
+        self::assertCount(2, [...$this->tabularData()->chunkBy(4)]);
+        foreach ($this->tabularDataWithHeader()->chunkBy(4) as $offset => $item) {
+            match ($offset) {
+                0 => self::assertCount(4, $item),
+                default => self::assertCount(2, $item),
+            };
+        }
+    }
 }
 
 enum Place: string