Skip UTF8 byte order mark in CSV files

pimcore · May 20, 2024 · a31883f · a31883f
1 parent f25246e
commit a31883f
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 1 deletion.
diff --git a/src/DataSource/Interpreter/CsvFileInterpreter.php b/src/DataSource/Interpreter/CsvFileInterpreter.php
@@ -20,6 +20,8 @@
 
 class CsvFileInterpreter extends AbstractInterpreter
 {
+    private const UTF8_BOM = "\xEF\xBB\xBF";
+
     /**
      * @var bool
      */
@@ -43,6 +45,8 @@ class CsvFileInterpreter extends AbstractInterpreter
     protected function doInterpretFileAndCallProcessRow(string $path): void
     {
         if (($handle = fopen($path, 'r')) !== false) {
+            $this->skipByteOrderMark($handle);
+
             if ($this->skipFirstRow) {
                 //load first row and ignore it
                 $data = fgetcsv($handle, 0, $this->delimiter, $this->enclosure, $this->escape);
@@ -88,6 +92,8 @@ public function previewData(string $path, int $recordNumber = 0, array $mappedCo
         $readRecordNumber = -1;
 
         if ($this->fileValid($path) && ($handle = fopen($path, 'r')) !== false) {
+            $this->skipByteOrderMark($handle);
+
             if ($this->skipFirstRow) {
                 //load first row and ignore it
                 $data = fgetcsv($handle, 0, $this->delimiter, $this->enclosure, $this->escape);
@@ -127,4 +133,11 @@ public function previewData(string $path, int $recordNumber = 0, array $mappedCo
 
         return new PreviewData($columns, $previewData, $readRecordNumber, $mappedColumns);
     }
+    private function skipByteOrderMark($handle): void
+    {
+        $bom = fread($handle, strlen(self::UTF8_BOM));
+        if (0 !== strncmp(self::UTF8_BOM, $bom, strlen(self::UTF8_BOM))) {
+            rewind($handle);
+        }
+    }
 }
diff --git a/tests/bin/docker-compose.yml b/tests/bin/docker-compose.yml
@@ -22,7 +22,7 @@ services:
             PIMCORE_TEST: 1
             PIMCORE_TEST_DB_DSN: "mysql://pimcore:pimcore@db:3306/pimcore_test"
         depends_on:
-            - db            
+            - db
         volumes:
             - ../../.:/var/www/html
             - /var/www/html/vendor