/**
* @expectedException \Box\Spout\Reader\Exception\XMLProcessingException
*
* @return void
*/
public function testNextShouldThrowExceptionOnError()
{
// The sharedStrings.xml file in "attack_billion_laughs.xlsx" contains
// a doctype element that causes read errors
$resourcePath = $this->getResourcePath('attack_billion_laughs.xlsx');
$sheetDataXMLFilePath = 'zip://' . $resourcePath . '#xl/sharedStrings.xml';
$xmlReader = new XMLReader();
if ($xmlReader->open($sheetDataXMLFilePath) !== false) {
@$xmlReader->next('sst');
}
}
/**
* Reads the styles.xml file and extract the relevant information from the file.
*
* @return void
*/
protected function extractRelevantInfo()
{
$this->customNumberFormats = [];
$this->stylesAttributes = [];
$xmlReader = new XMLReader();
if ($xmlReader->openFileInZip($this->filePath, self::STYLES_XML_FILE_PATH)) {
while ($xmlReader->read()) {
if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_NUM_FMTS)) {
$numFmtsNode = new SimpleXMLElement($xmlReader->readOuterXml());
$this->extractNumberFormats($numFmtsNode);
} else {
if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_CELL_XFS)) {
$cellXfsNode = new SimpleXMLElement($xmlReader->readOuterXml());
$this->extractStyleAttributes($cellXfsNode);
}
}
}
$xmlReader->close();
}
}
/**
* Returns a SimpleXMLElement node from the current node in the given XMLReader instance.
* This is to simplify the parsing of the subtree.
*
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader
* @return \Box\Spout\Reader\Wrapper\SimpleXMLElement
* @throws \Box\Spout\Common\Exception\IOException If the current node cannot be read
*/
protected function getSimpleXmlElementNodeFromXMLReader($xmlReader)
{
$node = null;
try {
$node = new SimpleXMLElement($xmlReader->readOuterXml());
} catch (XMLProcessingException $exception) {
throw new IOException("The sharedStrings.xml file contains unreadable data [{$exception->getMessage()}].");
}
return $node;
}
/**
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object, positioned on a "<c>" tag
* @return int
* @throws \Box\Spout\Common\Exception\InvalidArgumentException When the given cell index is invalid
*/
protected function getCellIndex($xmlReader)
{
// Get "r" attribute if present (from something like <c r="A1"...>
$currentCellIndex = $xmlReader->getAttribute(self::XML_ATTRIBUTE_CELL_INDEX);
return $currentCellIndex !== null ? CellHelper::getColumnIndexFromCellIndex($currentCellIndex) : $this->lastColumnIndexProcessed + 1;
}
/**
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object, positioned on a "<table:table-cell>" starting node
* @return int The value of "table:number-columns-repeated" attribute of the current node, or 1 if attribute missing
*/
protected function getNumColumnsRepeatedForCurrentNode($xmlReader)
{
$numColumnsRepeated = $xmlReader->getAttribute(self::XML_ATTRIBUTE_NUM_COLUMNS_REPEATED);
return $numColumnsRepeated !== null ? intval($numColumnsRepeated) : 1;
}
/**
* @param string $sheetId The sheet ID, as defined in "workbook.xml"
* @return string The XML file path describing the sheet inside "workbook.xml.res", for the given sheet ID
*/
protected function getSheetDataXMLFilePathForSheetId($sheetId)
{
$sheetDataXMLFilePath = '';
// find the file path of the sheet, by looking at the "workbook.xml.res" file
$xmlReader = new XMLReader();
if ($xmlReader->openFileInZip($this->filePath, self::WORKBOOK_XML_RELS_FILE_PATH)) {
while ($xmlReader->read()) {
if ($xmlReader->isPositionedOnStartingNode('Relationship')) {
$relationshipSheetId = $xmlReader->getAttribute('Id');
if ($relationshipSheetId === $sheetId) {
// In workbook.xml.rels, it is only "worksheets/sheet1.xml"
// In [Content_Types].xml, the path is "/xl/worksheets/sheet1.xml"
$sheetDataXMLFilePath = $xmlReader->getAttribute('Target');
// sometimes, the sheet data file path already contains "/xl/"...
if (strpos($sheetDataXMLFilePath, '/xl/') !== 0) {
$sheetDataXMLFilePath = '/xl/' . $sheetDataXMLFilePath;
break;
}
}
}
}
$xmlReader->close();
}
return $sheetDataXMLFilePath;
}
/**
* @param string $fileName
* @return \DOMNode[]
*/
private function getCellElementsFromSheetXmlFile($fileName)
{
$cellElements = [];
$resourcePath = $this->getGeneratedResourcePath($fileName);
$pathToStylesXmlFile = $resourcePath . '#xl/worksheets/sheet1.xml';
$xmlReader = new \XMLReader();
$xmlReader->open('zip://' . $pathToStylesXmlFile);
while ($xmlReader->read()) {
if ($xmlReader->nodeType === \XMLReader::ELEMENT && $xmlReader->name === 'c') {
$cellElements[] = $xmlReader->expand();
}
}
return $cellElements;
}
/**
* @param string $fileName
* @param string $section
* @return \DomElement
*/
private function getXmlSectionFromStylesXmlFile($fileName, $section)
{
$resourcePath = $this->getGeneratedResourcePath($fileName);
$pathToStylesXmlFile = $resourcePath . '#styles.xml';
$xmlReader = new XMLReader();
$xmlReader->open('zip://' . $pathToStylesXmlFile);
$xmlReader->readUntilNodeFound($section);
return $xmlReader->expand();
}
/**
* @param string $fileName
* @param int $sheetIndex
* @return XMLReader
*/
private function moveReaderToCorrectTableNode($fileName, $sheetIndex)
{
$resourcePath = $this->getGeneratedResourcePath($fileName);
$pathToSheetFile = $resourcePath . '#content.xml';
$xmlReader = new XMLReader();
$xmlReader->open('zip://' . $pathToSheetFile);
$xmlReader->readUntilNodeFound('table:table');
for ($i = 1; $i < $sheetIndex; $i++) {
$xmlReader->readUntilNodeFound('table:table');
}
return $xmlReader;
}
/**
* Extracts style attributes from the "xf" nodes, inside the "cellXfs" section.
* For simplicity, the styles attributes are kept in memory. This is possible thanks
* to the reuse of styles. So 1 million cells should not use 1 million styles.
*
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XML Reader positioned on the "cellXfs" node
* @return void
*/
protected function extractStyleAttributes($xmlReader)
{
while ($xmlReader->read()) {
if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_XF)) {
$numFmtId = $xmlReader->getAttribute(self::XML_ATTRIBUTE_NUM_FMT_ID);
$normalizedNumFmtId = $numFmtId !== null ? intval($numFmtId) : null;
$applyNumberFormat = $xmlReader->getAttribute(self::XML_ATTRIBUTE_APPLY_NUMBER_FORMAT);
$normalizedApplyNumberFormat = $applyNumberFormat !== null ? !!$applyNumberFormat : null;
$this->stylesAttributes[] = [self::XML_ATTRIBUTE_NUM_FMT_ID => $normalizedNumFmtId, self::XML_ATTRIBUTE_APPLY_NUMBER_FORMAT => $normalizedApplyNumberFormat];
} else {
if ($xmlReader->isPositionedOnEndingNode(self::XML_NODE_CELL_XFS)) {
// Once done reading "cellXfs" node's children
break;
}
}
}
}
/**
* Extracts style attributes from the "xf" nodes, inside the "cellXfs" section.
* For simplicity, the styles attributes are kept in memory. This is possible thanks
* to the reuse of styles. So 1 million cells should not use 1 million styles.
*
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XML Reader positioned on the "cellXfs" node
* @return void
*/
protected function extractStyleAttributes($xmlReader)
{
while ($xmlReader->read()) {
if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_XF)) {
$this->stylesAttributes[] = [self::XML_ATTRIBUTE_NUM_FMT_ID => intval($xmlReader->getAttribute(self::XML_ATTRIBUTE_NUM_FMT_ID)), self::XML_ATTRIBUTE_APPLY_NUMBER_FORMAT => !!$xmlReader->getAttribute(self::XML_ATTRIBUTE_APPLY_NUMBER_FORMAT)];
} else {
if ($xmlReader->isPositionedOnEndingNode(self::XML_NODE_CELL_XFS)) {
// Once done reading "cellXfs" node's children
break;
}
}
}
}
