cleanup code

src/Knowledge/ChunkManager.php

@@ -17,11 +17,6 @@ final class ChunkManager
         $this->indexPath = rtrim($projectDir, '/') . $relativeIndexPath;
     }
 
-    public function getIndexPath(): string
-    {
-        return $this->indexPath;
-    }
-
     // ============================================================
     // COUNT (Streaming, robust)
     // ============================================================

src/Knowledge/Ingest/SimpleChunker.php

@@ -6,15 +6,16 @@ declare(strict_types=1);
 namespace App\Knowledge\Ingest;
 
 use App\Index\IndexConfigurationProvider;
+use App\Knowledge\Text\TextNormalizer;
 
-final class SimpleChunker
+final readonly class SimpleChunker
 {
-    private IndexConfigurationProvider $configurationProvider;
 
     public function __construct(
-        IndexConfigurationProvider $configurationProvider
-    ) {
-        $this->configurationProvider = $configurationProvider;
+        private IndexConfigurationProvider $configurationProvider,
+        private TextNormalizer             $textNormalizer
+    )
+    {
     }
 
     /** @return string[] */
@@ -25,7 +26,7 @@ final class SimpleChunker
         $maxWords = $config->getChunkSize();
         $overlapWords = $config->getChunkOverlap();
 
-        $text = $this->normalize($text);
+        $text = $this->textNormalizer->normalize($text);
         if ($text === '') {
             return [];
         }
@@ -60,7 +61,7 @@ final class SimpleChunker
             $wordEnd = min($wordPos + $maxWords, $totalWords);
 
             $tokenStart = $wordTokenIndexes[$wordPos];
-            $tokenEnd   = $wordTokenIndexes[$wordEnd - 1] + 1;
+            $tokenEnd = $wordTokenIndexes[$wordEnd - 1] + 1;
 
             $tokenEnd = $this->adjustCutToBoundary($tokens, $tokenStart, $tokenEnd);
 
@@ -84,15 +85,6 @@ final class SimpleChunker
         return $this->dedupe($chunks);
     }
 
-    private function normalize(string $text): string
-    {
-        $text = str_replace(["\r\n", "\r"], "\n", $text);
-        $text = preg_replace("/[ \t]+/u", " ", $text);
-        $text = preg_replace("/\n{3,}/u", "\n\n", $text);
-
-        return trim((string) $text);
-    }
-
     private function adjustCutToBoundary(array $tokens, int $start, int $end): int
     {
         $startToken = $tokens[$start] ?? '';
@@ -122,7 +114,7 @@ final class SimpleChunker
     private function dedupe(array $chunks): array
     {
         $seen = [];
-        $out  = [];
+        $out = [];
 
         foreach ($chunks as $chunk) {
             $key = mb_strtolower(