Writer Xls Handle Characters Outside Unicode BMP

samples/Basic/27_Images_Xls.php

@@ -5,7 +5,7 @@
 require __DIR__ . '/../Header.php';
 
 // Read from Xls (.xls) template
-$helper->log('Load Xlsx template file');
+$helper->log('Load Xls template file');
 $reader = IOFactory::createReader('Xls');
 $spreadsheet = $reader->load(__DIR__ . '/../templates/27template.xls');
 

src/PhpSpreadsheet/Shared/StringHelper.php

@@ -412,11 +412,9 @@ public static function UTF8toBIFF8UnicodeShort(string $textValue, array $arrcRun
      */
     public static function UTF8toBIFF8UnicodeLong(string $textValue): string
     {
-        // character count
-        $ln = self::countCharacters($textValue, 'UTF-8');
-
         // characters
         $chars = self::convertEncoding($textValue, 'UTF-16LE', 'UTF-8');
+        $ln = (int) (strlen($chars) / 2);  // N.B. - strlen, not mb_strlen issue #642
 
         return pack('vC', $ln, 0x0001) . $chars;
     }

tests/PhpSpreadsheetTests/Writer/Xls/Issue642Test.php

@@ -0,0 +1,32 @@
+<?php
+
+namespace PhpOffice\PhpSpreadsheetTests\Writer\Xls;
+
+use PhpOffice\PhpSpreadsheet\Shared\File;
+use PhpOffice\PhpSpreadsheet\Spreadsheet;
+use PhpOffice\PhpSpreadsheet\Writer\Xls;
+use PHPUnit\Framework\TestCase;
+
+class Issue642Test extends TestCase
+{
+    public function testCharOutsideBMP(): void
+    {
+        $spreadsheet = new Spreadsheet();
+        $sheet = $spreadsheet->getActiveSheet();
+        $stringUtf8 = "Hello\u{1f600}goodbye";
+        self::assertSame(13, mb_strlen($stringUtf8));
+        $stringUtf16 = (string) iconv('UTF-8', 'UTF-16LE', $stringUtf8);
+        self::assertSame(28, strlen($stringUtf16)); // each character requires 2 bytes except for non-BMP which requires 4
+        $sheet->getCell('A1')->setValue($stringUtf8);
+        $outputFilename = File::temporaryFilename();
+        $writer = new Xls($spreadsheet);
+        $writer->save($outputFilename);
+        $spreadsheet->disconnectWorksheets();
+        $contents = (string) file_get_contents($outputFilename);
+        unlink($outputFilename);
+        $expected = "\x00\x0e\x00\x01" . $stringUtf16; // length is 14 (0e), not 13
+        self::assertStringContainsString($expected, $contents);
+        $unexpected = "\x00\x0d\x00\x01" . $stringUtf16; // length is 14 (0e), not 13
+        self::assertStringNotContainsString($unexpected, $contents);
+    }
+}