Method to Test Whether Csv Will Be Affected by Php9

See PHPOffice#4161. The best way to future-proof is to set the escape character to null string, and set testAutoDetect to false before reading. However, depending on the file being read, this may lead to different results than expected. This will be unavoidable because Php itself will change. This PR adds a new static method `affectedByPhp9` to Csv Reader. This can be used to identify in advance whether an input file will be affected by the changes. This will allow users to identify problems in advance, and prepare for how they might be handled.

src/PhpSpreadsheet/Reader/Csv.php

@@ -10,6 +10,7 @@
 use PhpOffice\PhpSpreadsheet\Shared\StringHelper;
 use PhpOffice\PhpSpreadsheet\Spreadsheet;
 use PhpOffice\PhpSpreadsheet\Worksheet\Worksheet;
+use Throwable;
 
 class Csv extends BaseReader
 {
@@ -74,7 +75,7 @@ class Csv extends BaseReader
      * It is anticipated that it will conditionally be set
      * to null-string for Php9 and above.
      */
-    private static string $defaultEscapeCharacter = '\\';
+    private static string $defaultEscapeCharacter = PHP_VERSION_ID < 90000 ? '\\' : '';
 
     /**
      * Callback for setting defaults in construction.
@@ -286,6 +287,12 @@ private function openFileOrMemory(string $filename): void
         if (!$fhandle) {
             throw new ReaderException($filename . ' is an Invalid Spreadsheet file.');
         }
+        if ($this->inputEncoding === 'UTF-8') {
+            $encoding = self::guessEncodingBom($filename);
+            if ($encoding !== '') {
+                $this->inputEncoding = $encoding;
+            }
+        }
         if ($this->inputEncoding === self::GUESS_ENCODING) {
             $this->inputEncoding = self::guessEncoding($filename, $this->fallbackEncoding);
         }
@@ -313,7 +320,7 @@ public function setTestAutoDetect(bool $value): self
     private function setAutoDetect(?string $value): ?string
     {
         $retVal = null;
-        if ($value !== null && $this->testAutodetect) {
+        if ($value !== null && $this->testAutodetect && PHP_VERSION_ID < 90000) {
             $retVal2 = @ini_set('auto_detect_line_endings', $value);
             if (is_string($retVal2)) {
                 $retVal = $retVal2;
@@ -362,6 +369,21 @@ private function loadStringOrFile(string $filename, Spreadsheet $spreadsheet, bo
         // Deprecated in Php8.1
         $iniset = $this->setAutoDetect('1');
 
+        try {
+            $this->loadStringOrFile2($filename, $spreadsheet, $dataUri);
+            $this->setAutoDetect($iniset);
+        } catch (Throwable $e) {
+            $this->setAutoDetect($iniset);
+
+            throw $e;
+        }
+
+        return $spreadsheet;
+    }
+
+    private function loadStringOrFile2(string $filename, Spreadsheet $spreadsheet, bool $dataUri): void
+    {
+
         // Open file
         if ($dataUri) {
             $this->openDataUri($filename);
@@ -433,11 +455,6 @@ private function loadStringOrFile(string $filename, Spreadsheet $spreadsheet, bo
 
         // Close file
         fclose($fileHandle);
-
-        $this->setAutoDetect($iniset);
-
-        // Return
-        return $spreadsheet;
     }
 
     /**
@@ -545,7 +562,9 @@ public function getContiguous(): bool
      */
     public function setEscapeCharacter(string $escapeCharacter): self
     {
-        $this->escapeCharacter = $escapeCharacter;
+        if (PHP_VERSION_ID < 90000) {
+            $this->escapeCharacter = $escapeCharacter;
+        }
 
         return $this;
     }
@@ -689,4 +708,39 @@ private static function getCsv(
 
         return fgetcsv($stream, $length, $separator, $enclosure, $escape);
     }
+
+    public static function affectedByPhp9(
+        string $filename,
+        string $inputEncoding = 'UTF-8',
+        ?string $delimiter = null,
+        string $enclosure = '"',
+        string $escapeCharacter = '\\'
+    ): bool {
+        if (PHP_VERSION_ID < 70400 || PHP_VERSION_ID >= 90000) {
+            throw new ReaderException('Function valid only for Php7.4 or Php8'); // @codeCoverageIgnore
+        }
+        $reader1 = new self();
+        $reader1->setInputEncoding($inputEncoding)
+            ->setTestAutoDetect(true)
+            ->setEscapeCharacter($escapeCharacter)
+            ->setDelimiter($delimiter)
+            ->setEnclosure($enclosure);
+        $spreadsheet1 = $reader1->load($filename);
+        $sheet1 = $spreadsheet1->getActiveSheet();
+        $array1 = $sheet1->toArray(null, false, false);
+        $spreadsheet1->disconnectWorksheets();
+
+        $reader2 = new self();
+        $reader2->setInputEncoding($inputEncoding)
+            ->setTestAutoDetect(false)
+            ->setEscapeCharacter('')
+            ->setDelimiter($delimiter)
+            ->setEnclosure($enclosure);
+        $spreadsheet2 = $reader2->load($filename);
+        $sheet2 = $spreadsheet2->getActiveSheet();
+        $array2 = $sheet2->toArray(null, false, false);
+        $spreadsheet2->disconnectWorksheets();
+
+        return $array1 !== $array2;
+    }
 }

tests/PhpSpreadsheetTests/Reader/Csv/CsvLineEndingTest.php

@@ -25,6 +25,9 @@ protected function tearDown(): void
      */
     public function testEndings(string $ending): void
     {
+        if ($ending === "\r" && PHP_VERSION_ID >= 90000) {
+            self::markTestSkipped('Mac line endings not supported for Php9+');
+        }
         $this->tempFile = $filename = File::temporaryFilename();
         $data = ['123', '456', '789'];
         file_put_contents($filename, implode($ending, $data));

tests/PhpSpreadsheetTests/Reader/Csv/Php9Test.php

@@ -0,0 +1,37 @@
+<?php
+
+declare(strict_types=1);
+
+namespace PhpOffice\PhpSpreadsheetTests\Reader\Csv;
+
+use PhpOffice\PhpSpreadsheet\Reader\Csv;
+use PhpOffice\PhpSpreadsheet\Reader\Exception as ReaderException;
+use PHPUnit\Framework\TestCase;
+
+class Php9Test extends TestCase
+{
+    public function testAffectedByPhp9(): void
+    {
+        if (PHP_VERSION_ID >= 90000) {
+            $this->expectException(ReaderException::class);
+            $this->expectExceptionMessage('Php7.4 or Php8');
+        }
+        $dir = 'tests/data/Reader/CSV';
+        $files = glob("$dir/*");
+        self::assertNotFalse($files);
+        $affected = [];
+        foreach ($files as $file) {
+            $base = basename($file);
+            $encoding = 'UTF-8';
+            if (str_contains($base, 'utf') && !str_contains($base, 'bom')) {
+                $encoding = 'guess';
+            }
+            $result = Csv::affectedByPhp9($file, $encoding);
+            if ($result) {
+                $affected[] = $base;
+            }
+        }
+        $expected = ['backslash.csv', 'escape.csv', 'linend.mac.csv'];
+        self::assertSame($expected, $affected);
+    }
+}

tests/data/Reader/CSV/linend.mac.csv

@@ -0,0 +1 @@
+A,12,3

tests/data/Reader/CSV/linend.unix.csv

@@ -0,0 +1,2 @@
+A,1
+2,3

tests/data/Reader/CSV/linend.win.csv

@@ -0,0 +1,2 @@
+A,1
+2,3