Clean up line grouping strategies

src/Document/Object/Decorator/Page.php

@@ -13,7 +13,6 @@
 use PrinsFrank\PdfParser\Exception\ParseFailureException;
 use PrinsFrank\PdfParser\Exception\PdfParserException;
 use PrinsFrank\PdfParser\Extraction\Text\TextExtractor;
-use PrinsFrank\PdfParser\Extraction\Text\TextGrouping\LineGrouping\TextOverlapStrategy;
 
 class Page extends DecoratedObject {
     /**
@@ -27,7 +26,7 @@ public function getPositionedTextElements(): array {
 
     /** @throws PdfParserException */
     public function getText(): string {
-        return TextExtractor::extractText($this->getPositionedTextElements(), $this->document, $this, new TextOverlapStrategy());
+        return TextExtractor::extractText($this->getPositionedTextElements(), $this->document, $this);
     }
 
     /** @throws PdfParserException */

src/Extraction/Text/TextExtractor.php

@@ -6,16 +6,18 @@
 use PrinsFrank\PdfParser\Document\Document;
 use PrinsFrank\PdfParser\Document\Object\Decorator\Page;
 use PrinsFrank\PdfParser\Exception\PdfParserException;
-use PrinsFrank\PdfParser\Extraction\Text\TextGrouping\LineGrouping\LineGroupingStrategy;
+use PrinsFrank\PdfParser\Extraction\Text\TextGrouping\LineGrouping\TextOverlapStrategy;
 
 class TextExtractor {
     /**
      * @param list<PositionedTextElement> $positionedTextElements
      * @throws PdfParserException
      */
-    public static function extractText(array $positionedTextElements, Document $document, Page $page, LineGroupingStrategy $lineGroupingStrategy): string {
+    public static function extractText(array $positionedTextElements, Document $document, Page $page): string {
+        $lineGroupedElements = TextOverlapStrategy::group($positionedTextElements);
+
         $text = '';
-        foreach ($lineGroupingStrategy->group($positionedTextElements) as $i => $positionedTextElementsForLine) {
+        foreach ($lineGroupedElements as $i => $positionedTextElementsForLine) {
             if ($i !== 0) {
                 $text .= "\n";
             }

src/Extraction/Text/TextGrouping/LineGrouping/TextOverlapStrategy.php

@@ -2,7 +2,6 @@
 
 namespace PrinsFrank\PdfParser\Extraction\Text\TextGrouping\LineGrouping;
 
-use Override;
 use PrinsFrank\PdfParser\Document\ContentStream\PositionedText\PositionedTextElement;
 
 /**
@@ -17,14 +16,14 @@
  * Strategy where we sort all positioned text elements, retrieve the very first text element from the page (highest)
  * And for each text element check if there is significant overlap above a threshold. Continue until all elements are processed
  */
-class TextOverlapStrategy implements LineGroupingStrategy {
-    /** @param int<0, 100> $overlapPercentage */
-    public function __construct(
-        private readonly int $overlapPercentage = 90,
-    ) {}
+class TextOverlapStrategy {
+    private const OVERLAP_PERCENTAGE = 90;
 
-    #[Override]
-    public function group(array $positionedTextElements): iterable {
+    /**
+     * @param list<PositionedTextElement> $positionedTextElements
+     * @return iterable<list<PositionedTextElement>>
+     */
+    public static function group(array $positionedTextElements): iterable {
         usort(
             $positionedTextElements,
             fn(PositionedTextElement $a, PositionedTextElement $b): int => $b->absoluteMatrix->offsetY <=> $a->absoluteMatrix->offsetY,
@@ -66,7 +65,7 @@ public function group(array $positionedTextElements): iterable {
                 }
 
                 $overlap = min($highestElementTop, $currentElementTop) - max($highestPositionedTextElementBottom, $currentElementBottom);
-                $belongsOnLine = $overlap / $smallestElementHeight * 100 >= $this->overlapPercentage;
+                $belongsOnLine = $overlap / $smallestElementHeight * 100 >= self::OVERLAP_PERCENTAGE;
                 $isEnclosedSubscript = $overlap > 0.0
                     && $positionedTextElementHeight < $highestPositionedTextElementHeight
                     && $positionedTextElement->absoluteMatrix->offsetX >= $lineLeftX

tests/Unit/Document/ContentStream/PositionedText/LineGroupingStrategy/TextOverlapStrategyTest.php

@@ -19,7 +19,7 @@ public function testOrdersLinesTopToBottom(): void {
 
         static::assertSame(
             [[$top], [$bottom]],
-            iterator_to_array((new TextOverlapStrategy())->group([$bottom, $top]), false),
+            iterator_to_array(TextOverlapStrategy::group([$bottom, $top]), false),
         );
     }
 
@@ -33,7 +33,7 @@ public function testOrdersLinesTopToBottomOnANegativeOriginPage(): void {
 
         static::assertSame(
             [[$top], [$bottom]],
-            iterator_to_array((new TextOverlapStrategy())->group([$bottom, $top]), false),
+            iterator_to_array(TextOverlapStrategy::group([$bottom, $top]), false),
         );
     }
 }