Don't prioritize simple encoding when font has differences (#395)

Author: PrinsFrank

src/Document/ContentStream/PositionedText/PositionedTextElement.php

@@ -42,7 +42,7 @@ public function getText(Document $document, Page $page): string {
                     $chars = $glyph->getChar();
                 } elseif (strlen($unescapedChars) === 1 && ($glyph = $font->getDifferences()?->getGlyph(ord($unescapedChars))) !== null) {
                     $chars = $glyph->getChar();
-                } elseif (in_array($encoding = $font->getEncoding(), [EncodingNameValue::MacExpertEncoding, EncodingNameValue::WinAnsiEncoding], true)) {
+                } elseif (in_array($encoding = $font->getEncoding(), [EncodingNameValue::MacExpertEncoding, EncodingNameValue::WinAnsiEncoding], true) && $font->getDifferences() === null) {
                     $chars = $encoding->decodeString($unescapedChars);
                 } elseif (($toUnicodeCMap = $font->getToUnicodeCMap() ?? $font->getToUnicodeCMapDescendantFont()) !== null) {
                     $chars = $toUnicodeCMap->textToUnicode(bin2hex($unescapedChars));

tests/Samples/files/issue-393/contents.yml

@@ -0,0 +1,23 @@
+# yaml-language-server: $schema=../../schema.json
+version: '1.6'
+userPassword: null
+ownerPassword: null
+fileEncryptionKey: null
+title: 'Custom encoding extraction reproducer (control-char range)'
+producer: 'GPL Ghostscript 10.07.1'
+author: null
+creator: 'synthetic reproducer for PDF parser CFF/Custom encoding extraction'
+creationDate: 2026-06-04T11:31:22-05:00
+modificationDate: 2026-06-04T11:31:22-05:00
+pages:
+  -
+    content: |-
+      HELLO PDF WORLD.
+      THIS PDF USES CUSTOM ENCODING.
+      VISIBLE TEXT IS WHAT YOU READ ABOVE.
+      NAIVE PARSERS THAT IGNORE THE
+      DIFFERENCES ARRAY AND TOUNICODE CMAP
+      WILL OUTPUT GARBAGE FOR EVERY LINE.
+      EVERY VISIBLE CHARACTER MAPS THROUGH A
+      DIFFERENCES ARRAY AT WINANSI CONTROL
+      POSITIONS, FORCING TOUNICODE LOOKUP.