public function testAddFieldMethodChaining()
{
$document = new Document();
$this->assertTrue($document->addField(Document\Field::Text('title', 'Title')) instanceof Document);
$document = new Document();
$document->addField(Document\Field::Text('title', 'Title'))->addField(Document\Field::Text('annotation', 'Annotation'))->addField(Document\Field::Text('body', 'Document body, document body, document body...'));
}
public function testEncoding()
{
// forcing filter to UTF-8
$utf8text = iconv('UTF-8', 'UTF-8', 'Words with umlauts: åãü...');
$iso8859_1 = iconv('UTF-8', 'ISO-8859-1', $utf8text);
$field = Document\Field::Text('field', $iso8859_1, 'ISO-8859-1');
$this->assertEquals($field->encoding, 'ISO-8859-1');
$this->assertEquals($field->name, 'field');
$this->assertEquals($field->value, $iso8859_1);
$this->assertEquals($field->getUtf8Value(), $utf8text);
}
/**
* Object constructor
*
* @param string $fileName
* @param boolean $storeContent
* @throws \Zend\Search\Lucene\Exception\ExtensionNotLoadedException
* @throws \Zend\Search\Lucene\Exception\RuntimeException
*/
private function __construct($fileName, $storeContent)
{
if (!class_exists('ZipArchive', false)) {
throw new ExtensionNotLoadedException('MS Office documents processing functionality requires Zip extension to be loaded');
}
// Document data holders
$slides = array();
$slideNotes = array();
$documentBody = array();
$coreProperties = array();
// Open AbstractOpenXML package
$package = new \ZipArchive();
$package->open($fileName);
// Read relations and search for officeDocument
$relationsXml = $package->getFromName('_rels/.rels');
if ($relationsXml === false) {
throw new RuntimeException('Invalid archive or corrupted .pptx file.');
}
$relations = simplexml_load_string($relationsXml);
foreach ($relations->Relationship as $rel) {
if ($rel["Type"] == AbstractOpenXML::SCHEMA_OFFICEDOCUMENT) {
// Found office document! Search for slides...
$slideRelations = simplexml_load_string($package->getFromName($this->absoluteZipPath(dirname($rel["Target"]) . "/_rels/" . basename($rel["Target"]) . ".rels")));
foreach ($slideRelations->Relationship as $slideRel) {
if ($slideRel["Type"] == self::SCHEMA_SLIDERELATION) {
// Found slide!
$slides[str_replace('rId', '', (string) $slideRel["Id"])] = simplexml_load_string($package->getFromName($this->absoluteZipPath(dirname($rel["Target"]) . "/" . dirname($slideRel["Target"]) . "/" . basename($slideRel["Target"]))));
// Search for slide notes
$slideNotesRelations = simplexml_load_string($package->getFromName($this->absoluteZipPath(dirname($rel["Target"]) . "/" . dirname($slideRel["Target"]) . "/_rels/" . basename($slideRel["Target"]) . ".rels")));
foreach ($slideNotesRelations->Relationship as $slideNoteRel) {
if ($slideNoteRel["Type"] == self::SCHEMA_SLIDENOTESRELATION) {
// Found slide notes!
$slideNotes[str_replace('rId', '', (string) $slideRel["Id"])] = simplexml_load_string($package->getFromName($this->absoluteZipPath(dirname($rel["Target"]) . "/" . dirname($slideRel["Target"]) . "/" . dirname($slideNoteRel["Target"]) . "/" . basename($slideNoteRel["Target"]))));
break;
}
}
}
}
break;
}
}
// Sort slides
ksort($slides);
ksort($slideNotes);
// Extract contents from slides
foreach ($slides as $slideKey => $slide) {
// Register namespaces
$slide->registerXPathNamespace("p", self::SCHEMA_PRESENTATIONML);
$slide->registerXPathNamespace("a", self::SCHEMA_DRAWINGML);
// Fetch all text
$textElements = $slide->xpath('//a:t');
foreach ($textElements as $textElement) {
$documentBody[] = (string) $textElement;
}
// Extract contents from slide notes
if (isset($slideNotes[$slideKey])) {
// Fetch slide note
$slideNote = $slideNotes[$slideKey];
// Register namespaces
$slideNote->registerXPathNamespace("p", self::SCHEMA_PRESENTATIONML);
$slideNote->registerXPathNamespace("a", self::SCHEMA_DRAWINGML);
// Fetch all text
$textElements = $slideNote->xpath('//a:t');
foreach ($textElements as $textElement) {
$documentBody[] = (string) $textElement;
}
}
}
// Read core properties
$coreProperties = $this->extractMetaData($package);
// Close file
$package->close();
// Store filename
$this->addField(Field::Text('filename', $fileName, 'UTF-8'));
// Store contents
if ($storeContent) {
$this->addField(Field::Text('body', implode(' ', $documentBody), 'UTF-8'));
} else {
$this->addField(Field::UnStored('body', implode(' ', $documentBody), 'UTF-8'));
}
// Store meta data properties
foreach ($coreProperties as $key => $value) {
$this->addField(Field::Text($key, $value, 'UTF-8'));
}
// Store title (if not present in meta data)
if (!isset($coreProperties['title'])) {
$this->addField(Field::Text('title', $fileName, 'UTF-8'));
}
}
/**
* Object constructor
*
* @param string $data HTML string (may be HTML fragment, )
* @param boolean $isFile
* @param boolean $storeContent
* @param string $defaultEncoding HTML encoding, is used if it's not specified using Content-type HTTP-EQUIV meta tag.
*/
private function __construct($data, $isFile, $storeContent, $defaultEncoding = '')
{
$this->_doc = new \DOMDocument();
$this->_doc->substituteEntities = true;
if ($isFile) {
$htmlData = file_get_contents($data);
} else {
$htmlData = $data;
}
@$this->_doc->loadHTML($htmlData);
if ($this->_doc->encoding === null) {
// Document encoding is not recognized
/** @todo improve HTML vs HTML fragment recognition */
if (preg_match('/<html>/i', $htmlData, $matches, PREG_OFFSET_CAPTURE)) {
// It's an HTML document
// Add additional HEAD section and recognize document
$htmlTagOffset = $matches[0][1] + strlen($matches[0][0]);
@$this->_doc->loadHTML(iconv($defaultEncoding, 'UTF-8//IGNORE', substr($htmlData, 0, $htmlTagOffset)) . '<head><META HTTP-EQUIV="Content-type" CONTENT="text/html; charset=UTF-8"/></head>' . iconv($defaultEncoding, 'UTF-8//IGNORE', substr($htmlData, $htmlTagOffset)));
// Remove additional HEAD section
$xpath = new \DOMXPath($this->_doc);
$head = $xpath->query('/html/head')->item(0);
$head->parentNode->removeChild($head);
} else {
// It's an HTML fragment
@$this->_doc->loadHTML('<html><head><META HTTP-EQUIV="Content-type" CONTENT="text/html; charset=UTF-8"/></head><body>' . iconv($defaultEncoding, 'UTF-8//IGNORE', $htmlData) . '</body></html>');
}
}
/** @todo Add correction of wrong HTML encoding recognition processing
* The case is:
* Content-type HTTP-EQUIV meta tag is presented, but ISO-8859-5 encoding is actually used,
* even $this->_doc->encoding demonstrates another recognized encoding
*/
$xpath = new \DOMXPath($this->_doc);
$docTitle = '';
$titleNodes = $xpath->query('/html/head/title');
foreach ($titleNodes as $titleNode) {
// title should always have only one entry, but we process all nodeset entries
$docTitle .= $titleNode->nodeValue . ' ';
}
$this->addField(Field::Text('title', $docTitle, 'UTF-8'));
$metaNodes = $xpath->query('/html/head/meta[@name]');
foreach ($metaNodes as $metaNode) {
$this->addField(Field::Text($metaNode->getAttribute('name'), $metaNode->getAttribute('content'), 'UTF-8'));
}
$docBody = '';
$bodyNodes = $xpath->query('/html/body');
foreach ($bodyNodes as $bodyNode) {
// body should always have only one entry, but we process all nodeset entries
$this->_retrieveNodeText($bodyNode, $docBody);
}
if ($storeContent) {
$this->addField(Field::Text('body', $docBody, 'UTF-8'));
} else {
$this->addField(Field::UnStored('body', $docBody, 'UTF-8'));
}
$linkNodes = $this->_doc->getElementsByTagName('a');
foreach ($linkNodes as $linkNode) {
if (($href = $linkNode->getAttribute('href')) != '' && (!self::$_excludeNoFollowLinks || strtolower($linkNode->getAttribute('rel')) != 'nofollow')) {
$this->_links[] = $href;
}
}
$linkNodes = $this->_doc->getElementsByTagName('area');
foreach ($linkNodes as $linkNode) {
if (($href = $linkNode->getAttribute('href')) != '' && (!self::$_excludeNoFollowLinks || strtolower($linkNode->getAttribute('rel')) != 'nofollow')) {
$this->_links[] = $href;
}
}
$this->_links = array_unique($this->_links);
$linkNodes = $xpath->query('/html/head/link');
foreach ($linkNodes as $linkNode) {
if (($href = $linkNode->getAttribute('href')) != '') {
$this->_headerLinks[] = $href;
}
}
$this->_headerLinks = array_unique($this->_headerLinks);
}
/**
* Object constructor
*
* @param string $fileName
* @param boolean $storeContent
* @throws \Zend\Search\Lucene\Exception\ExtensionNotLoadedException
* @throws \Zend\Search\Lucene\Exception\RuntimeException
*/
private function __construct($fileName, $storeContent)
{
if (!class_exists('ZipArchive', false)) {
throw new ExtensionNotLoadedException('MS Office documents processing functionality requires Zip extension to be loaded');
}
// Document data holders
$documentBody = array();
$coreProperties = array();
// Open AbstractOpenXML package
$package = new \ZipArchive();
$package->open($fileName);
// Read relations and search for officeDocument
$relationsXml = $package->getFromName('_rels/.rels');
if ($relationsXml === false) {
throw new RuntimeException('Invalid archive or corrupted .docx file.');
}
$relations = simplexml_load_string($relationsXml);
foreach ($relations->Relationship as $rel) {
if ($rel["Type"] == AbstractOpenXML::SCHEMA_OFFICEDOCUMENT) {
// Found office document! Read in contents...
$contents = simplexml_load_string($package->getFromName($this->absoluteZipPath(dirname($rel['Target']) . '/' . basename($rel['Target']))));
$contents->registerXPathNamespace('w', self::SCHEMA_WORDPROCESSINGML);
$paragraphs = $contents->xpath('//w:body/w:p');
foreach ($paragraphs as $paragraph) {
$runs = $paragraph->xpath('.//w:r/*[name() = "w:t" or name() = "w:br"]');
if ($runs === false) {
// Paragraph doesn't contain any text or breaks
continue;
}
foreach ($runs as $run) {
if ($run->getName() == 'br') {
// Break element
$documentBody[] = ' ';
} else {
$documentBody[] = (string) $run;
}
}
// Add space after each paragraph. So they are not bound together.
$documentBody[] = ' ';
}
break;
}
}
// Read core properties
$coreProperties = $this->extractMetaData($package);
// Close file
$package->close();
// Store filename
$this->addField(Field::Text('filename', $fileName, 'UTF-8'));
// Store contents
if ($storeContent) {
$this->addField(Field::Text('body', implode('', $documentBody), 'UTF-8'));
} else {
$this->addField(Field::UnStored('body', implode('', $documentBody), 'UTF-8'));
}
// Store meta data properties
foreach ($coreProperties as $key => $value) {
$this->addField(Field::Text($key, $value, 'UTF-8'));
}
// Store title (if not present in meta data)
if (!isset($coreProperties['title'])) {
$this->addField(Field::Text('title', $fileName, 'UTF-8'));
}
}
/**
* Object constructor
*
* @param string $fileName
* @param boolean $storeContent
* @throws \Zend\Search\Lucene\Exception\ExtensionNotLoadedException
* @throws \Zend\Search\Lucene\Exception\RuntimeException
*/
private function __construct($fileName, $storeContent)
{
if (!class_exists('ZipArchive', false)) {
throw new ExtensionNotLoadedException(
'MS Office documents processing functionality requires Zip extension to be loaded'
);
}
// Document data holders
$sharedStrings = array();
$worksheets = array();
$documentBody = array();
$coreProperties = array();
// Open AbstractOpenXML package
$package = new \ZipArchive();
$package->open($fileName);
// Read relations and search for officeDocument
$relationsXml = $package->getFromName('_rels/.rels');
if ($relationsXml === false) {
throw new RuntimeException('Invalid archive or corrupted .xlsx file.');
}
$relations = simplexml_load_string($relationsXml);
foreach ($relations->Relationship as $rel) {
if ($rel["Type"] == AbstractOpenXML::SCHEMA_OFFICEDOCUMENT) {
// Found office document! Read relations for workbook...
$workbookRelations = simplexml_load_string($package->getFromName( $this->absoluteZipPath(dirname($rel["Target"]) . "/_rels/" . basename($rel["Target"]) . ".rels")) );
$workbookRelations->registerXPathNamespace("rel", AbstractOpenXML::SCHEMA_RELATIONSHIP);
// Read shared strings
$sharedStringsPath = $workbookRelations->xpath("rel:Relationship[@Type='" . self::SCHEMA_SHAREDSTRINGS . "']");
$sharedStringsPath = (string)$sharedStringsPath[0]['Target'];
$xmlStrings = simplexml_load_string($package->getFromName( $this->absoluteZipPath(dirname($rel["Target"]) . "/" . $sharedStringsPath)) );
if (isset($xmlStrings) && isset($xmlStrings->si)) {
foreach ($xmlStrings->si as $val) {
if (isset($val->t)) {
$sharedStrings[] = (string)$val->t;
} elseif (isset($val->r)) {
$sharedStrings[] = $this->_parseRichText($val);
}
}
}
// Loop relations for workbook and extract worksheets...
foreach ($workbookRelations->Relationship as $workbookRelation) {
if ($workbookRelation["Type"] == self::SCHEMA_WORKSHEETRELATION) {
$worksheets[ str_replace( 'rId', '', (string)$workbookRelation["Id"]) ] = simplexml_load_string(
$package->getFromName( $this->absoluteZipPath(dirname($rel["Target"]) . "/" . dirname($workbookRelation["Target"]) . "/" . basename($workbookRelation["Target"])) )
);
}
}
break;
}
}
// Sort worksheets
ksort($worksheets);
// Extract contents from worksheets
foreach ($worksheets as $sheetKey => $worksheet) {
foreach ($worksheet->sheetData->row as $row) {
foreach ($row->c as $c) {
// Determine data type
$dataType = (string)$c["t"];
switch ($dataType) {
case "s":
// Value is a shared string
if ((string)$c->v != '') {
$value = $sharedStrings[intval($c->v)];
} else {
$value = '';
}
break;
case "b":
// Value is boolean
$value = (string)$c->v;
if ($value == '0') {
$value = false;
} else if ($value == '1') {
$value = true;
} else {
$value = (bool)$c->v;
}
break;
case "inlineStr":
// Value is rich text inline
$value = $this->_parseRichText($c->is);
break;
case "e":
// Value is an error message
if ((string)$c->v != '') {
$value = (string)$c->v;
} else {
$value = '';
}
break;
default:
// Value is a string
$value = (string)$c->v;
// Check for numeric values
if (is_numeric($value) && $dataType != 's') {
if ($value == (int)$value) $value = (int)$value;
elseif ($value == (float)$value) $value = (float)$value;
elseif ($value == (double)$value) $value = (double)$value;
}
}
$documentBody[] = $value;
}
}
}
// Read core properties
$coreProperties = $this->extractMetaData($package);
// Close file
$package->close();
// Store filename
$this->addField(Field::Text('filename', $fileName, 'UTF-8'));
// Store contents
if ($storeContent) {
$this->addField(Field::Text('body', implode(' ', $documentBody), 'UTF-8'));
} else {
$this->addField(Field::UnStored('body', implode(' ', $documentBody), 'UTF-8'));
}
// Store meta data properties
foreach ($coreProperties as $key => $value)
{
$this->addField(Field::Text($key, $value, 'UTF-8'));
}
// Store title (if not present in meta data)
if (!isset($coreProperties['title']))
{
$this->addField(Field::Text('title', $fileName, 'UTF-8'));
}
}
public function testTermsStreamInterfaceSkipToTermsRetrievingTwoTermsCase()
{
$index = Lucene\Lucene::create(__DIR__ . '/_index/_files');
// Zero terms
$doc = new Document();
$doc->addField(Document\Field::Text('contents', 'someterm word'));
$index->addDocument($doc);
unset($index);
$index = Lucene\Lucene::open(__DIR__ . '/_index/_files');
$index->resetTermsStream();
$index->skipTo(new Index\Term('term', 'contents'));
$this->assertTrue($index->currentTerm() == new Index\Term('word', 'contents'));
$index->closeTermsStream();
}
public function testEncoding()
{
$field = Document\Field::Text('field', 'Words with umlauts: εγό...', 'ISO-8859-1');
$this->assertEquals($field->encoding, 'ISO-8859-1');
$this->assertEquals($field->name, 'field');
$this->assertEquals($field->value, 'Words with umlauts: εγό...');
$this->assertEquals($field->getUtf8Value(), 'Words with umlauts: Γ₯ãü...');
}