ODS Reader

Spout can now read ODS files.
It's on par with the XLSX reader. The only difference is that the row iterator cannot be rewound.
It supports the different output formats from LibreOffice and Excel, skipping extra rows/cells if needed.

src/Spout/Reader/XLSX/Helper/SharedStringsHelper.php

@@ -83,7 +83,7 @@ class SharedStringsHelper
         $escaper = new \Box\Spout\Common\Escaper\XLSX();
 
         $sharedStringsFilePath = $this->getSharedStringsFilePath();
-        if ($xmlReader->open($sharedStringsFilePath, null, LIBXML_NONET) === false) {
+        if ($xmlReader->open($sharedStringsFilePath) === false) {
             throw new IOException('Could not open "' . self::SHARED_STRINGS_XML_FILE_PATH . '".');
         }
 

src/Spout/Reader/XLSX/Helper/SheetHelper.php

@@ -101,7 +101,6 @@ class SheetHelper
      */
     protected function getSheetFromXML($sheetDataXMLFilePath, $sheetIndexZeroBased)
     {
-        $sheetId = $sheetIndexZeroBased + 1;
         $sheetName = $this->getDefaultSheetName($sheetDataXMLFilePath);
 
         /*
@@ -123,7 +122,6 @@ class SheetHelper
 
             if (count($sheetNodes) === 1) {
                 $sheetNode = $sheetNodes[0];
-                $sheetId = (int) $sheetNode->getAttribute('sheetId');
                 $escapedSheetName = $sheetNode->getAttribute('name');
 
                 /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
@@ -132,7 +130,7 @@ class SheetHelper
             }
         }
 
-        return new Sheet($this->filePath, $sheetDataXMLFilePath, $this->sharedStringsHelper, $sheetId, $sheetIndexZeroBased, $sheetName);
+        return new Sheet($this->filePath, $sheetDataXMLFilePath, $this->sharedStringsHelper, $sheetIndexZeroBased, $sheetName);
     }
 
     /**

src/Spout/Reader/XLSX/RowIterator.php

@@ -77,6 +77,7 @@ class RowIterator implements IteratorInterface
         $this->sharedStringsHelper = $sharedStringsHelper;
 
         $this->xmlReader = new XMLReader();
+
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
         $this->escaper = new \Box\Spout\Common\Escaper\XLSX();
     }
@@ -143,7 +144,7 @@ class RowIterator implements IteratorInterface
 
         try {
             while ($this->xmlReader->read()) {
-                if ($this->xmlReader->nodeType === XMLReader::ELEMENT && $this->xmlReader->name === self::XML_NODE_DIMENSION) {
+                if ($this->xmlReader->isPositionedOnStartingNode(self::XML_NODE_DIMENSION)) {
                     // Read dimensions of the sheet
                     $dimensionRef = $this->xmlReader->getAttribute(self::XML_ATTRIBUTE_REF); // returns 'A1:M13' for instance (or 'A1' for empty sheet)
                     if (preg_match('/[A-Z\d]+:([A-Z\d]+)/', $dimensionRef, $matches)) {
@@ -151,7 +152,7 @@ class RowIterator implements IteratorInterface
                         $this->numColumns = CellHelper::getColumnIndexFromCellIndex($lastCellIndex) + 1;
                     }
 
-                } else if ($this->xmlReader->nodeType === XMLReader::ELEMENT && $this->xmlReader->name === self::XML_NODE_ROW) {
+                } else if ($this->xmlReader->isPositionedOnStartingNode(self::XML_NODE_ROW)) {
                     // Start of the row description
                     $isInsideRowTag = true;
 
@@ -164,7 +165,7 @@ class RowIterator implements IteratorInterface
                     }
                     $rowData = ($numberOfColumnsForRow !== 0) ? array_fill(0, $numberOfColumnsForRow, '') : [];
 
-                } else if ($isInsideRowTag && $this->xmlReader->nodeType === XMLReader::ELEMENT && $this->xmlReader->name === self::XML_NODE_CELL) {
+                } else if ($isInsideRowTag && $this->xmlReader->isPositionedOnStartingNode(self::XML_NODE_CELL)) {
                     // Start of a cell description
                     $currentCellIndex = $this->xmlReader->getAttribute(self::XML_ATTRIBUTE_CELL_INDEX);
                     $currentColumnIndex = CellHelper::getColumnIndexFromCellIndex($currentCellIndex);
@@ -172,16 +173,17 @@ class RowIterator implements IteratorInterface
                     $node = $this->xmlReader->expand();
                     $rowData[$currentColumnIndex] = $this->getCellValue($node);
 
-                } else if ($this->xmlReader->nodeType === XMLReader::END_ELEMENT && $this->xmlReader->name === self::XML_NODE_ROW) {
+                } else if ($this->xmlReader->isPositionedOnEndingNode(self::XML_NODE_ROW)) {
                     // End of the row description
                     // If needed, we fill the empty cells
                     $rowData = ($this->numColumns !== 0) ? $rowData : CellHelper::fillMissingArrayIndexes($rowData);
                     $this->numReadRows++;
                     break;
 
-                } else if ($this->xmlReader->nodeType === XMLReader::END_ELEMENT && $this->xmlReader->name === self::XML_NODE_WORKSHEET) {
+                } else if ($this->xmlReader->isPositionedOnEndingNode(self::XML_NODE_WORKSHEET)) {
                     // The closing "</worksheet>" marks the end of the file
                     $this->hasReachedEndOfFile = true;
+                    break;
                 }
             }
 
@@ -192,6 +194,40 @@ class RowIterator implements IteratorInterface
         $this->rowDataBuffer = $rowData;
     }
 
+    /**
+     * Returns the (unescaped) correctly marshalled, cell value associated to the given XML node.
+     *
+     * @param \DOMNode $node
+     * @return string|int|float|bool|\DateTime|null The value associated with the cell (null when the cell has an error)
+     */
+    protected function getCellValue($node)
+    {
+        // Default cell type is "n"
+        $cellType = $node->getAttribute(self::XML_ATTRIBUTE_TYPE) ?: self::CELL_TYPE_NUMERIC;
+        $vNodeValue = $this->getVNodeValue($node);
+
+        if (($vNodeValue === '') && ($cellType !== self::CELL_TYPE_INLINE_STRING)) {
+            return $vNodeValue;
+        }
+
+        switch ($cellType) {
+            case self::CELL_TYPE_INLINE_STRING:
+                return $this->formatInlineStringCellValue($node);
+            case self::CELL_TYPE_SHARED_STRING:
+                return $this->formatSharedStringCellValue($vNodeValue);
+            case self::CELL_TYPE_STR:
+                return $this->formatStrCellValue($vNodeValue);
+            case self::CELL_TYPE_BOOLEAN:
+                return $this->formatBooleanCellValue($vNodeValue);
+            case self::CELL_TYPE_NUMERIC:
+                return $this->formatNumericCellValue($vNodeValue);
+            case self::CELL_TYPE_DATE:
+                return $this->formatDateCellValue($vNodeValue);
+            default:
+                return null;
+        }
+    }
+
     /**
      * Returns the cell's string value from a node's nested value node
      *
@@ -203,10 +239,7 @@ class RowIterator implements IteratorInterface
         // for cell types having a "v" tag containing the value.
         // if not, the returned value should be empty string.
         $vNode = $node->getElementsByTagName(self::XML_NODE_VALUE)->item(0);
-        if ($vNode !== null) {
-            return $vNode->nodeValue;
-        }
-        return "";
+        return ($vNode !== null) ? $vNode->nodeValue : '';
     }
 
     /**
@@ -296,40 +329,6 @@ class RowIterator implements IteratorInterface
         }
     }
 
-    /**
-     * Returns the (unescaped) correctly marshalled, cell value associated to the given XML node.
-     *
-     * @param \DOMNode $node
-     * @return string|int|float|bool|\DateTime|null The value associated with the cell (null when the cell has an error)
-     */
-    protected function getCellValue($node)
-    {
-        // Default cell type is "n"
-        $cellType = $node->getAttribute(self::XML_ATTRIBUTE_TYPE) ?: self::CELL_TYPE_NUMERIC;
-        $vNodeValue = $this->getVNodeValue($node);
-
-        if (($vNodeValue === '') && ($cellType !== self::CELL_TYPE_INLINE_STRING)) {
-            return $vNodeValue;
-        }
-
-        switch ($cellType) {
-            case self::CELL_TYPE_INLINE_STRING:
-                return $this->formatInlineStringCellValue($node);
-            case self::CELL_TYPE_SHARED_STRING:
-                return $this->formatSharedStringCellValue($vNodeValue);
-            case self::CELL_TYPE_STR:
-                return $this->formatStrCellValue($vNodeValue);
-            case self::CELL_TYPE_BOOLEAN:
-                return $this->formatBooleanCellValue($vNodeValue);
-            case self::CELL_TYPE_NUMERIC:
-                return $this->formatNumericCellValue($vNodeValue);
-            case self::CELL_TYPE_DATE:
-                return $this->formatDateCellValue($vNodeValue);
-            default:
-                return null;
-        }
-    }
-
     /**
      * Return the current element, from the buffer.
      * @link http://php.net/manual/en/iterator.current.php

src/Spout/Reader/XLSX/Sheet.php

@@ -15,9 +15,6 @@ class Sheet implements SheetInterface
     /** @var RowIterator To iterate over sheet's rows */
     protected $rowIterator;
 
-    /** @var int ID of the sheet */
-    protected $id;
-
     /** @var int Index of the sheet, based on order of creation (zero-based) */
     protected $index;
 
@@ -28,14 +25,12 @@ class Sheet implements SheetInterface
      * @param string $filePath Path of the XLSX file being read
      * @param string $sheetDataXMLFilePath Path of the sheet data XML file as in [Content_Types].xml
      * @param Helper\SharedStringsHelper Helper to work with shared strings
-     * @param int $sheetId ID of the sheet
      * @param int $sheetIndex Index of the sheet, based on order of creation (zero-based)
      * @param string $sheetName Name of the sheet
      */
-    public function __construct($filePath, $sheetDataXMLFilePath, $sharedStringsHelper, $sheetId, $sheetIndex, $sheetName)
+    public function __construct($filePath, $sheetDataXMLFilePath, $sharedStringsHelper, $sheetIndex, $sheetName)
     {
         $this->rowIterator = new RowIterator($filePath, $sheetDataXMLFilePath, $sharedStringsHelper);
-        $this->id = $sheetId;
         $this->index = $sheetIndex;
         $this->name = $sheetName;
     }
@@ -48,14 +43,6 @@ class Sheet implements SheetInterface
         return $this->rowIterator;
     }
 
-    /**
-     * @return int ID of the sheet
-     */
-    public function getId()
-    {
-        return $this->id;
-    }
-
     /**
      * @return int Index of the sheet, based on order of creation (zero-based)
      */