CM-873 fix test and implementation

keboola · Jan 10, 2024 · 362e56f · 362e56f
1 parent 76afdee
commit 362e56f
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 2 deletions.
diff --git a/src/CsvReader.php b/src/CsvReader.php
@@ -13,6 +13,7 @@ class CsvReader extends AbstractCsvFile implements Iterator
      * @deprecated use Keboola\Csv\CsvOptions::DEFAULT_ENCLOSURE
      */
     const DEFAULT_ESCAPED_BY = CsvOptions::DEFAULT_ESCAPED_BY;
+    const SAMPLE_SIZE = 10000;
 
     /**
      * @var int
@@ -124,7 +125,12 @@ protected function openCsvFile($fileName)
     protected function detectLineBreak()
     {
         @rewind($this->getFilePointer());
-        $sample = @fread($this->getFilePointer(), 10000);
+        $sample = @fread($this->getFilePointer(), self::SAMPLE_SIZE);
+        if (substr($sample, -1) === "\r") {
+            // we might have hit the file in the middle of CR+LF, only getting CR
+            @rewind($this->getFilePointer());
+            $sample = @fread($this->getFilePointer(), self::SAMPLE_SIZE+1);
+        }
 
         return LineBreaksHelper::detectLineBreaks($sample, $this->getEnclosure(), $this->getEscapedBy());
     }

diff --git a/tests/CsvReadTest.php b/tests/CsvReadTest.php
@@ -33,7 +33,8 @@ public function testColumnsCount()
 
     public function testNewlineDetectionEdgecaseWithCrLf()
     {
-        $this->expectExceptionMessage('Invalid line break. Please use unix \n or win \r\n line breaks.');
+        $this->expectNotToPerformAssertions();
+        // this used to throw "Invalid line break. Please use unix \n or win \r\n line breaks." before the fix
         new CsvReader(__DIR__ . '/data/test-input-edgecase.crlf.csv');
     }