function process($currentXml)
{
$currentDirectory = dirname($this->docvertTransformDirectory);
$documentType = detectDocumentType($currentXml);
switch ($documentType) {
case 'OpenDocument1.0':
//text-colon-section text-colon-style-name="Sect2" text-colon-name="Section2">
$styles = $this->getStyles();
$currentXml = preg_replace('/<office:document-content[^>]*?>/s', "\$0" . $styles, $currentXml);
//displayXmlString($currentXml);
$currentXml = xsltTransform($currentXml, $this->docvertTransformDirectory . 'fix-opendocument-content.xsl');
//displayXmlString($currentXml);
$currentXml = xsltTransform($currentXml, $this->docvertTransformDirectory . 'fix-opendocument-content-stage2.xsl');
//displayXmlString($currentXml);
$currentXml = xsltTransform($currentXml, $currentDirectory . DIRECTORY_SEPARATOR . 'lib' . DIRECTORY_SEPARATOR . 'sun-xsl' . DIRECTORY_SEPARATOR . 'opendocument-content-to-docbook.xsl');
//displayXmlString($this->docvertTransformDirectory.'opendocument-content-to-docbook.xsl');
//displayXmlString($currentXml);
$metaData = $this->getMetaData();
$currentXml = preg_replace('/<db:book[^>]*?>/s', "\$0" . $metaData, $currentXml);
//displayXmlString($currentXml);
$currentXml = xsltTransform($currentXml, $this->docvertTransformDirectory . 'fix-docbook.xsl');
//displayXmlString($currentXml);
break;
case 'OpenOffice1.x':
webServiceError('&error-process-transformopendocumenttodocbook-openofficefile;');
break;
default:
webServiceError('&error-process-transformopendocumenttodocbook-unsupported-file; [' . revealXml($currentXml) . ']', 500, array('documentType' => $documentType));
break;
}
return $currentXml;
}
function getFtpConnection($host, $username, $password, $port)
{
$connectionId = ftp_connect($host);
if (!@ftp_login($connectionId, $username, $password)) {
webServiceError('&error-ftp-authentication;', 500, array('username' => $username, 'host' => $host));
}
return $connectionId;
}
public function process($currentXml)
{
if (DOCVERT_CLIENT_TYPE == 'web') {
displayXmlString($currentXml);
} else {
webServiceError($currentXml, 200);
}
}
public function process($currentXml)
{
if (!array_key_exists('withFile', $this->elementAttributes)) {
webServiceError('&error-process-transform-lacks-withfile;');
}
$xslPath = $this->pipelineDirectory . $this->elementAttributes['withFile'];
if (!file_exists($xslPath)) {
webServiceError('&error-process-transform-withfile-missing-file;', 500, array('xslPath' => $xslPath));
}
$this->elementAttributes['loopdepth'] = $this->loopDepth;
return xsltTransform($currentXml, $xslPath, $this->elementAttributes);
}
public function process($currentXml)
{
if (!array_key_exists('toFile', $this->elementAttributes)) {
webServiceError('&error-process-serialize-no-with-file;');
}
$toFile = $this->elementAttributes['toFile'];
$configFilenamesPath = dirname(dirname(__FILE__)) . DIRECTORY_SEPARATOR . 'custom-filenames.php';
include_once $configFilenamesPath;
$toFile = replaceCustomFilenamePlaceholders($toFile, $this->depthArray);
$destinationFilename = processDepthTemplate($toFile, $this->depthArray);
$destinationPath = $this->contentDirectory . DIRECTORY_SEPARATOR . $destinationFilename;
file_put_contents($destinationPath, $currentXml);
return $currentXml;
}
function xsltTransform($xmlString, $xsltPath, $xsltArguments = null)
{
if (!file_exists($xsltPath)) {
webserviceError('&error-xslt-path-not-found;', 500, array('path' => $xsltPath));
}
$result = null;
$xsltEnabledStatus = getXsltEnabledStatus();
switch ($xsltEnabledStatus) {
case 'php5':
$xslt = new XSLTProcessor();
$xsltDocument = new DOMDocument();
$xsltDocument->load($xsltPath);
$xslt->importStyleSheet($xsltDocument);
if (is_array($xsltArguments)) {
foreach ($xsltArguments as $key => $value) {
$xslt->setParameter('', $key, $value);
}
}
$errorLevelToDescribeMerelyDeprecatedWarnings = 999999;
$xmlDocument = new DOMDocument();
$xmlDocument->loadXML($xmlString);
$result = $xslt->transformToXML($xmlDocument);
break;
case 'php4':
$xsltproc = xslt_create();
$xmlString = array('/_xml' => $xmlString);
$xsltPath = 'file://' . $xsltPath;
$result = @xslt_process($xsltproc, 'arg:/_xml', $xsltPath, NULL, $xmlString, $xsltArguments) or webServiceError('&error-xslt-processor-error;', 500, array('path' => $xsltPath, 'errorMessage' => xslt_error($xsltproc)));
if (empty($result) or xslt_error($xsltproc) != null) {
webServiceError('&error-xslt-processor-error;', 500, array('path' => $xsltPath, 'errorMessage' => xslt_error($xsltproc)));
}
xslt_free($xsltproc);
break;
default:
$commandLineMessage = '';
$phpVersion = getPhpVersion();
if ($phpVersion >= 5) {
webServiceError('&error-xslt-not-available;');
} else {
webServiceError('&error-php5-required;', 500, array('phpVersion' => $phpVersion));
}
}
return $result;
}
function performTest(&$currentXml)
{
if (!array_key_exists('withFile', $this->elementAttributes)) {
webServiceError('&error-process-test-withfile;');
}
$xslPath = null;
if (stripos($this->elementAttributes['withFile'], 'internal://') !== FALSE) {
$xslPath = $this->docvertTransformDirectory . str_ireplace('internal://', '', $this->elementAttributes['withFile']);
} else {
$xslPath = $this->pipelineDirectory . $this->elementAttributes['withFile'];
}
if (!file_exists($xslPath)) {
webServiceError('&error-process-test-missing-xsl;', 500, array('xslPath' => $xslPath));
}
$xslAttributes = array('loopdepth' => $this->loopDepth);
$testResults = xsltTransform($currentXml, $xslPath, $xslAttributes);
if (trim($testResults)) {
$this->logError($testResults, 'raw');
}
}
function uploadToUploadLocation($uploadLocation, $previewDirectory, $remoteDirectory)
{
switch ($uploadLocation['protocol']) {
case 'ftp':
return copyViaFtpRecursively($uploadLocation, $previewDirectory, $remoteDirectory, "active");
break;
case 'ftp-pasv':
return copyViaFtpRecursively($uploadLocation, $previewDirectory, $remoteDirectory, "passive");
break;
case 'webdav':
case 'webdav-tls':
return copyViaWebDAVRecursively($uploadLocation, $previewDirectory, $remoteDirectory);
break;
case 'bloggerapi':
case 'bloggerapi-ssl':
return copyViaBloggerAPI($uploadLocation, $previewDirectory, $remoteDirectory);
break;
default:
webServiceError('&error-unknown-protocol; ' . $uploadLocation['protocol']);
break;
}
}
public function process($currentXml)
{
if (!array_key_exists('withFiles', $this->elementAttributes)) {
webServiceError('&error-process-includedependentfiles-withfiles;');
}
$pathsToCopy = explode(',', $this->elementAttributes['withFiles']);
foreach ($pathsToCopy as $pathToCopy) {
$trimmedPathToCopy = trim($pathToCopy);
if ($trimmedPathToCopy) {
$pathWildCard = $this->pipelineDirectory . $trimmedPathToCopy;
$pathMatches = glob($pathWildCard);
if ($pathMatches !== False) {
foreach ($pathMatches as $pathMatch) {
$destinationPath = $this->contentDirectory . DIRECTORY_SEPARATOR . basename($pathMatch);
$this->copyRecursively($pathMatch, $destinationPath);
}
} else {
$this->logError(array('&error-unable-to-read-directory;', array('path' => $pathWildCard)), 'error');
}
}
}
return $currentXml;
}
function getOperatingSystemsTemporaryDirectory()
{
if (defined('OPERATING_SYSTEM_TEMPORARY_DIRECTORY')) {
return OPERATING_SYSTEM_TEMPORARY_DIRECTORY;
}
$directoriesToCheck = array();
if (isset($_ENV)) {
if (isset($_ENV['TMPDIR'])) {
$directoriesToCheck[] = $_ENV['TMPDIR'];
}
if (isset($_ENV['TMP'])) {
$directoriesToCheck[] = $_ENV['TMP'];
}
}
if (DIRECTORY_SEPARATOR == '/') {
$directoriesToCheck[] = '/tmp/';
} else {
$directoriesToCheck[] = '\\temp\\';
$directoriesToCheck[] = '\\windows\\temp\\';
}
foreach ($directoriesToCheck as $directoryToCheck) {
if (is_writable($directoryToCheck)) {
define('OPERATING_SYSTEM_TEMPORARY_DIRECTORY', $directoryToCheck);
return OPERATING_SYSTEM_TEMPORARY_DIRECTORY;
}
}
webServiceError('&error-config-file-not-writable;', 500, array('path' => implode(', ', $directoriesToCheck)));
}
$docbookBody = preg_replace("/<docvert-remove-me[^>]*?>/", '', $docbookBody);
$docbookBody = preg_replace("/<\\/docvert-remove-me[^>]*?>/", '', $docbookBody);
//displayXmlString($docbookBody);
//displayXmlString(file_get_contents($docbookBodyXsltPath));
$allDocumentsPreviewDirectory = dirname(__FILE__) . DIRECTORY_SEPARATOR . 'writable' . DIRECTORY_SEPARATOR . $documentPathParts[0] . DIRECTORY_SEPARATOR;
$previewDirectory = $allDocumentsPreviewDirectory . $documentPathParts[1] . DIRECTORY_SEPARATOR;
$unitTestResults = $previewDirectory . 'test.html';
if (file_exists($unitTestResults)) {
silentlyUnlink($unitTestResults);
}
$zipsInPreviewDirectory = glob($allDocumentsPreviewDirectory . '*.zip');
if (count($zipsInPreviewDirectory)) {
foreach ($zipsInPreviewDirectory as $zipInPreviewDirectory) {
silentlyUnlink($zipInPreviewDirectory);
if (file_exists($zipInPreviewDirectory)) {
webServiceError('Docvert internal error: unable to remove ZIP file at "' . $zipInPreviewDirectory . '"');
}
}
$zipFilePath = $zipsInPreviewDirectory[0];
} else {
$zipFileName = chooseNameOfZipFile($allDocumentsPreviewDirectory);
$zipFilePath = $allDocumentsPreviewDirectory . $zipFileName;
}
$filesInPreviewDirectory = glob($previewDirectory . '*');
foreach ($filesInPreviewDirectory as $fileInPreviewDirectory) {
if (!stringStartsWith(basename($fileInPreviewDirectory), "docvert") && !stringEndsWith(basename($fileInPreviewDirectory), "wmf") && !stringEndsWith(basename($fileInPreviewDirectory), "gif") && !stringEndsWith(basename($fileInPreviewDirectory), "png") && !stringEndsWith(basename($fileInPreviewDirectory), "jpeg") && !stringEndsWith(basename($fileInPreviewDirectory), "jpg") && !stringEndsWith(basename($fileInPreviewDirectory), "svg")) {
//print 'Delete: '.$fileInPreviewDirectory.'<br />';
silentlyUnlink($fileInPreviewDirectory);
} else {
//print 'Retain: '.$fileInPreviewDirectory.'<br />';
}
function process($currentXml)
{
$extractImagesPath = $this->docvertTransformDirectory . 'extract-pages-html-images-and-links.xsl';
$htmlUrls = trim(xsltTransform($currentXml, $extractImagesPath));
$htmlUrlLines = explode("\n", $htmlUrls);
$imageUrls = array();
foreach ($htmlUrlLines as $htmlUrlLine) {
if (trim($htmlUrlLine) == '') {
continue;
}
$urlLineParts = explode("\t", $htmlUrlLine);
$urlType = $urlLineParts[0];
$baseUrl = $urlLineParts[1];
$possiblyRelativeUrl = $urlLineParts[2];
$fullUrl = '';
if (stringStartsWith($possiblyRelativeUrl, "http://") || stringStartsWith($possiblyRelativeUrl, "https://") || stringStartsWith($possiblyRelativeUrl, "mailto:")) {
$fullUrl = $possiblyRelativeUrl;
} else {
$connectionPart = getUrlConnectionPart($baseUrl);
$getUrlLocalPart = getUrlLocalPart($baseUrl);
$localPartDirectory = getUrlLocalPartDirectory($baseUrl);
if (stringStartsWith($possiblyRelativeUrl, '/')) {
$fullUrl = $connectionPart . $possiblyRelativeUrl;
} else {
$relativePath = resolveRelativeUrl($localPartDirectory . $possiblyRelativeUrl);
if (!stringStartsWith($relativePath, '/')) {
$relativePath = '/' . $relativePath;
}
$fullUrl = $connectionPart . $relativePath;
}
}
$missingImagePlaceholderImagePath = dirname(dirname(__FILE__)) . DIRECTORY_SEPARATOR . 'files' . DIRECTORY_SEPARATOR . '404image.gif';
if (!file_exists($missingImagePlaceholderImagePath)) {
webServiceError('&dynamic-error-process-downloadimagesandsetlinks-missing-placeholder;', 500, array('fourOhFourImagePath' => $fourOhFourImagePath));
}
$fullUrl = html_entity_decode($fullUrl);
switch ($urlType) {
case 'image':
$imageData = file_get_contents($fullUrl);
if ($imageData == null) {
$imageData = file_get_contents($missingImagePlaceholderImagePath);
}
$picturesDirectory = $this->contentDirectory . DIRECTORY_SEPARATOR . 'Pictures';
if (!file_exists($picturesDirectory)) {
mkdir($picturesDirectory);
}
if (!function_exists('imagecreatefromstring')) {
webServiceError('&error-process-downloadimagesandsetlinks-missing-gd;');
}
$imageResource = imagecreatefromstring($imageData);
if (!$imageResource) {
$imageResource = imagecreatefromstring(file_get_contents($missingImagePlaceholderImagePath));
}
$imageWidth = imagesx($imageResource);
$imageHeight = imagesy($imageResource);
$fileExtension = substr($fullUrl, strrpos($fullUrl, '.') + 1);
switch ($fileExtension) {
case 'jpg':
case 'jpeg':
case 'gif':
case 'png':
break;
default:
$fileExtension = 'jpg';
}
$openDocumentPath = 'Pictures/' . md5($fullUrl) . '.' . $fileExtension;
file_put_contents($this->contentDirectory . DIRECTORY_SEPARATOR . $openDocumentPath, $imageData);
$imageUrlReplacement = $openDocumentPath . '" c:width="' . $imageWidth . '" c:height="' . $imageHeight;
//FIXME: assumes image @src has double-quote and not single
$currentXml = str_replace('"' . $possiblyRelativeUrl . '"', '"' . $imageUrlReplacement . '"', $currentXml);
break;
case 'link':
$linkUrl = $urlLineParts[2];
//print '"'.$possiblyRelativeUrl.' vs '.$fullUrl.'<hr />';
$currentXml = str_replace('"' . $possiblyRelativeUrl . '"', '"' . htmlentities($fullUrl) . '"', $currentXml);
break;
}
}
return $currentXml;
}
function followUrlRedirects($url, $maximumNumberOfOfRedirects = false)
{
if ($maximumNumberOfOfRedirects === false) {
$maximumNumberOfOfRedirects = 10;
}
$finalRedirectionUrl = false;
$redirectPath = array();
$numberOfRedirectsRemaining = $maximumNumberOfOfRedirects;
$timeOutInSeconds = 5;
while ($finalRedirectionUrl == false) {
$previousUrl = $url;
$wasRedirectedThisTime = false;
$domainAndPort = getUrlDomainAndPortPart($url);
$redirectPath[] = $url;
$localPart = getUrlLocalPart($url);
$result = pullpage('HEAD', $domainAndPort[0], $domainAndPort[1], $localPart, false, false, false, false, null, null, false, $timeOutInSeconds);
if ($result === false) {
return null;
}
$result = explode("\n", $result);
foreach ($result as $line) {
$headParts = explode(':', $line);
if (count($headParts) > 1) {
$subject = array_shift($headParts);
$value = implode(':', $headParts);
if (strtolower(trim($subject)) == "location") {
$url = trim($value);
if (!stringStartsWith($url, 'http://') && !stringStartsWith($url, 'https://')) {
$url = 'http://' . $domainAndPort[0] . $url;
}
//print "Was redirected to ".$url."<br />";
if ($previousUrl != $url) {
$wasRedirectedThisTime = true;
}
}
}
}
if ($wasRedirectedThisTime == false) {
$finalRedirectionUrl = true;
}
if ($numberOfRedirectsRemaining <= 0) {
webServiceError("&error-maximum-number-of-redirects-followed;", 500, array('redirectPaths' => implode("</li><li>", $redirectPath)));
}
$numberOfRedirectsRemaining--;
}
return $url;
}
function wmfOrEmfToPdf($imagePath, &$currentXml)
{
//Step 1. Detect width/height of image.
$imageName = basename($imagePath);
$imageOffset = strpos($currentXml, $imageName);
if ($imageOffset === False) {
return False;
}
//image not in document, don't worry about it.
//header('Content-type: text/xml'); die($currentXml);
$dom = simplexml_load_string($currentXml);
$xpath = "//*[@xlink:href='" . $imageName . "']//parent::draw:frame";
$imageMatch = $dom->xpath($xpath);
if (count($imageMatch) == 0) {
webServiceError('&error-process-convertimages-no-dom;', 500, array('xpath' => $xpath));
}
$imageMatch = $imageMatch[0];
$attributes = $imageMatch->attributes('svg', true);
$width = (string) $attributes['width'];
$height = (string) $attributes['height'];
//Step 2. Make an ODT file containing only the WMF/EMF
// (ugh.. I know, but it works and it's reliable because we benefit from OpenOffice's years of
// reverse-engineering the EMF/WMF formats so really we should get over it)
//step 2a -- make a working directory for our OpenDocument file and copy the files in
$workingDirectory = getTemporaryDirectoryInsideDirectory($this->contentDirectory);
mkdir($workingDirectory . DIRECTORY_SEPARATOR . 'Pictures');
$destinationImagePath = $workingDirectory . DIRECTORY_SEPARATOR . 'Pictures' . DIRECTORY_SEPARATOR . basename($imagePath);
copy($imagePath, $destinationImagePath);
$odtTemplateDirectory = DOCVERT_DIR . 'core' . DIRECTORY_SEPARATOR . 'files' . DIRECTORY_SEPARATOR;
$stylesXml = file_get_contents($odtTemplateDirectory . 'styles.xml');
$stylesXml = str_replace('{{page-width}}', $width, $stylesXml);
$stylesXml = str_replace('{{page-height}}', $height, $stylesXml);
file_put_contents($workingDirectory . DIRECTORY_SEPARATOR . 'styles.xml', $stylesXml);
copy($odtTemplateDirectory . 'settings.xml', $workingDirectory . DIRECTORY_SEPARATOR . 'settings.xml');
copy($odtTemplateDirectory . 'meta.xml', $workingDirectory . DIRECTORY_SEPARATOR . 'meta.xml');
copy($odtTemplateDirectory . 'manifest.rdf', $workingDirectory . DIRECTORY_SEPARATOR . 'manifest.rdf');
copy($odtTemplateDirectory . 'mimetype', $workingDirectory . DIRECTORY_SEPARATOR . 'mimetype');
$contentXml = file_get_contents($odtTemplateDirectory . 'content.xml');
$imageTemplate = '<text:p><draw:frame text:anchor-type="as-char" svg:width="{{width}}" svg:height="{{height}}" draw:z-index="1"><draw:image xlink:href="{{path}}" xlink:type="simple" xlink:show="embed" xlink:actuate="onLoad"/></draw:frame></text:p>';
$imageString = str_replace('{{width}}', $width, $imageTemplate);
$imageString = str_replace('{{height}}', $height, $imageString);
$imageString = str_replace('{{path}}', 'Pictures/' . basename($destinationImagePath), $imageString);
$contentXml = str_replace('<!--{{content}}-->', $imageString, $contentXml);
file_put_contents($workingDirectory . DIRECTORY_SEPARATOR . 'content.xml', $contentXml);
mkdir($workingDirectory . DIRECTORY_SEPARATOR . 'META-INF');
$manifestXml = file_get_contents($odtTemplateDirectory . 'manifest.xml');
$manifestItemTemplate = ' <manifest:file-entry manifest:media-type="" manifest:full-path="{{path}}"/>';
$manifestItem = str_replace('{{path}}', 'Pictures/' . basename($imagePath), $manifestItemTemplate);
$manifestXml = str_replace('<!--{{content}}-->', $manifestItem, $manifestXml);
file_put_contents($workingDirectory . DIRECTORY_SEPARATOR . 'META-INF' . DIRECTORY_SEPARATOR . 'manifest.xml', $manifestXml);
//step 2b zip it into an ODT
$zipPath = $this->contentDirectory . DIRECTORY_SEPARATOR . basename($imagePath) . '.odt';
$zipPath = zipFiles($workingDirectory, $zipPath);
$zipData = file_get_contents($zipPath);
silentlyUnlink($zipPath);
silentlyUnlink($workingDirectory);
//Step 3 . Stream it to PyODConverter. Make a PDF and save it.
$pyodConverterPath = DOCVERT_DIR . 'core' . DIRECTORY_SEPARATOR . 'lib' . DIRECTORY_SEPARATOR . 'pyodconverter' . DIRECTORY_SEPARATOR . 'pyodconverter.py';
if (!file_exists($pyodConverterPath)) {
die("Can't find PyODconverter at " . htmlentities($pyodConverterPath));
}
$command = $pyodConverterPath . ' --stream --pdf';
$response = shellCommand($command, 20, $zipData, false);
$pdfMagicBytes = '%PDF';
if (substr($response['stdOut'], 0, strlen($pdfMagicBytes)) != $pdfMagicBytes) {
die("Expected a PDF response was didn't receive one. Received back " . htmlentities(print_r($response, true)));
}
$imagePathInfo = pathinfo($imagePath);
$pdfPath = dirname($imagePath) . DIRECTORY_SEPARATOR . basename($imagePath, '.' . $imagePathInfo['extension']) . '.pdf';
file_put_contents($pdfPath, $response['stdOut']);
return array('width' => $width, 'height' => $height, 'path' => $pdfPath);
}
function generateDocument($pages, $generatorPipeline)
{
if (preg_match('/.\\//s', $generatorPipeline)) {
webServiceError('&error-disallowed-characters;');
}
$userAgent = 'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:bignumber) Docvert';
$httpContextOptions = array('http' => array('header' => 'User-Agent: ' . $userAgent));
$httpContext = stream_context_create($httpContextOptions);
$docvertDir = dirname(dirname(__FILE__)) . DIRECTORY_SEPARATOR;
$docvertWritableDir = getWritableDirectory();
$disallowDocumentGeneration = getGlobalConfigItem('doNotAllowDocumentGeneration');
if ($disallowDocumentGeneration == 'true') {
webServiceError('&document-generation-disabled;');
}
$pageXml = '<c:document xmlns="http://www.w3.org/1999/xhtml" xmlns:c="container">' . "\n";
$pageTemplate = "\n\t" . '<c:page url="{{url}}" {{baseUrl}}>{{page}}</c:page>' . "\n";
$config = array('indent' => true, 'output-xhtml' => true, 'wrap' => 200);
if (!class_exists('tidy')) {
webServiceError('&tidy-is-not-installed;');
}
$tidy = new tidy();
$baseTagPattern = "/<base[^>]*?href=([^>]*?)>/is";
foreach ($pages as $page) {
if (trim($page) != '' && (stringStartsWith($page, 'http://') || stringStartsWith($page, 'https://'))) {
$pageHtml = file_get_contents($page, null, $httpContext);
$tidy->parseString($pageHtml, $config, 'utf8');
$tidy->cleanRepair();
$thisPage = str_replace('{{url}}', $page, $pageTemplate);
$baseUrl = '';
//supporting that ugly old hack of <base>
preg_match($baseTagPattern, $pageHtml, $matches);
if (count($matches) > 0) {
$baseUrl = 'baseUrl="' . substr($matches[1], 1, -2) . '"';
}
$thisPage = str_replace('{{baseUrl}}', $baseUrl, $thisPage);
$tidiedPageContents = characterEntityToNCR(removeDoctype(removeXmlComments($tidy)));
$styleTagPattern = "/<style.*?<\\/style>/is";
$tidiedPageContents = preg_replace($styleTagPattern, '', $tidiedPageContents);
$scriptTagPattern = "/<script.*?<\\/script>/is";
$tidiedPageContents = preg_replace($scriptTagPattern, '', $tidiedPageContents);
$questionMarkPattern = "/<\\?.*?\\?>/is";
//as strangely used on news.yahoo.com
$tidiedPageContents = preg_replace($questionMarkPattern, '', $tidiedPageContents);
$thisPage = str_replace('{{page}}', $tidiedPageContents, $thisPage);
$pageXml .= $thisPage;
}
}
$pageXml .= '</c:document>';
$temporaryDirectory = getTemporaryDirectory();
$pipelineDirectory = dirname(dirname(__FILE__)) . DIRECTORY_SEPARATOR . 'generator-pipeline' . DIRECTORY_SEPARATOR . $generatorPipeline . DIRECTORY_SEPARATOR;
$pipelinePath = $pipelineDirectory . 'pipeline.xml';
if (!file_exists($pipelinePath)) {
webServiceError('&generation-pipeline-not-found; ' . revealXml($pipelinePath));
}
$pipelineString = file_get_contents($pipelinePath);
$pipelineString = substr($pipelineString, strpos($pipelineString, '<pipeline>') + 10);
$pipelineString = substr($pipelineString, 0, strpos($pipelineString, '</pipeline>'));
$pipelineStages = xmlStringToArray($pipelineString);
$pipelineSettings = array("pipeline" => $generatorPipeline, "autopipeline" => $generatorPipeline);
processAPipelineLevel($pipelineStages, $pageXml, $pipelineDirectory, $temporaryDirectory, $temporaryDirectory, $pipelineSettings);
$openDocumentPath = $temporaryDirectory . 'output.odt';
zipFiles($temporaryDirectory, $openDocumentPath);
header('Content-disposition: attachment; filename=' . basename($openDocumentPath));
header('Content-type: application/vnd.oasis.opendocument.text');
readfile($openDocumentPath);
}
<?php
if (substr(PHP_VERSION, 0, 1) == "4") {
webServiceError('&error-ensure-php5;', 500, array('phpVersion' => PHP_VERSION));
}
<?php
ob_start();
$appDir = dirname(__FILE__) . DIRECTORY_SEPARATOR;
define('DOCVERT_DIR', $appDir);
define('DOCVERT_CLIENT_TYPE', 'web');
include_once DOCVERT_DIR . 'core/lib.php';
if (!isset($_REQUEST['pages'])) {
webServiceError('&error-generator-expected-pages-parameter;');
} elseif (count($_REQUEST['pages']) == 0) {
webServiceError('&error-generator-expected-pages-parameter;');
}
if (!isset($_REQUEST['generatorPipeline'])) {
webServiceError('&error-generator-expected-generatorpipeline-parameter;');
}
return generateDocument($_REQUEST['pages'], $_REQUEST['generatorPipeline']);
function showGenerationStep()
{
$docvertDir = dirname(dirname(__FILE__)) . DIRECTORY_SEPARATOR;
$disallowDocumentGeneration = getGlobalConfigItem('doNotAllowDocumentGeneration');
if ($disallowDocumentGeneration == 'true') {
return $this->getThemeFragment('generation-disabled.htmlf');
}
if (isset($_REQUEST['step'])) {
switch ($_REQUEST['step']) {
case '4':
if (!isset($_REQUEST['pages'])) {
webServiceError('&error-webpage-generation-no-pages;');
}
$template = $this->getThemeFragment('generation-step4.htmlf');
$hiddenFormChosenPages = array();
$listItems = array();
foreach ($_REQUEST['pages'] as $page) {
$listItems[] = "\n\t\t\t\t" . '<li>' . $page . '</li>';
$hiddenFormChosenPages[] = "\n\t\t\t\t" . '<input type="hidden" name="pages[]" value="' . $page . '"/>';
}
$template = str_replace('{{page-order}}', implode($listItems), $template);
$template = str_replace('{{hidden-form-chosen-pages}}', implode($hiddenFormChosenPages), $template);
$generatorPipelines = glob($this->docvertRootDirectory . 'generator-pipeline' . DIRECTORY_SEPARATOR . '*');
$generatorPipelinesArray = array();
foreach ($generatorPipelines as $generatorPipeline) {
$generatorName = basename($generatorPipeline);
$generatorPipelinesArray[] = '<option value="' . $generatorName . '">' . $generatorName . '</option>';
}
return str_replace('{{generator-pipelines}}', implode('', $generatorPipelinesArray), $template);
case '3':
$template = $this->getThemeFragment('generation-step3.htmlf');
$listItems = array();
foreach ($_REQUEST['pages'] as $page) {
$listItems[] = "\n\t\t\t\t" . '<option value="' . $page . '">' . $page . '</option>';
}
return str_replace('{{chosen-scrape-urls}}', implode($listItems), $template);
case '2':
if (!isset($_REQUEST['url'])) {
webServiceError('&error-webpage-generation-url;');
}
$originalUrl = $_REQUEST['url'];
if (trim($originalUrl) == '') {
webServiceError('&error-webpage-generation-no-url-given;');
}
if (!stringStartsWith($originalUrl, 'http')) {
$originalUrl = 'http://' . $originalUrl;
}
$originalUrl = str_replace(array("\n", "\r", "\t", " "), '', $originalUrl);
include_once dirname(__FILE__) . '/http.php';
if (trim(getUrlLocalPart($originalUrl)) == '') {
$originalUrl = followUrlRedirects($originalUrl . '/');
} else {
$originalUrl = followUrlRedirects($originalUrl);
}
if ($originalUrl === false) {
webServiceError('&error-webpage-cannot-get-url;', 500, array('url' => $originalUrl));
}
$page = file_get_contents($originalUrl);
$baseTagPattern = "/<base[^>]*?href=([^>]*?)>/is";
preg_match($baseTagPattern, $page, $matches);
if (count($matches) > 0) {
$originalUrl = trim($matches[1]);
$originalUrl = substr($originalUrl, 1, strlen($originalUrl) - 2);
}
$url = $originalUrl;
$connectionPart = getUrlConnectionPart($url);
$getUrlLocalPart = getUrlLocalPart($url);
$localPartDirectory = getUrlLocalPartDirectory($url);
$links = array();
$matches = null;
preg_match_all('/href="(.*?)"/', $page, $matches);
$matches = $matches[1];
$urls = array();
$urls[$originalUrl] = 'value that does not matter';
foreach ($matches as $match) {
$link = $match;
if (stringStartsWith($link, '/')) {
$link = $connectionPart . $link;
} elseif (stringStartsWith($link, "http://") || stringStartsWith($link, "https://")) {
} elseif (stringStartsWith($link, "mailto:")) {
} else {
$link = $connectionPart . resolveRelativeUrl($localPartDirectory . $link);
}
if (containsString($link, '#')) {
$link = substringBefore($link, '#');
}
if (stringEndsWith($link, '?')) {
$link = substringBefore($link, '?');
}
if (stringStartsWith($link, 'http')) {
$fileExtension = substr($link, strrpos($link, '.') + 1);
switch ($fileExtension) {
case 'avi':
case 'mov':
case 'mpg':
case 'css':
case 'jpeg':
case 'jpg':
case 'gif':
case 'png':
case 'bmp':
case 'apng':
case 'tiff':
case 'ico':
case 'js':
case 'gz':
case 'tar':
case 'zip':
case 'bin':
case 'sit':
case 'mp3':
case 'mp4':
case 'wav':
case 'swf':
case 'fla':
case 'rss':
case 'atom':
case 'pdf':
case 'xls':
case 'doc':
case 'txt':
case 'pps':
break;
default:
$urls[$link] = 'value that does not matter';
}
}
}
$urls = array_keys($urls);
$mostLikelyUrls = array();
$possibleUrls = array();
$unlikelyUrls = array();
$numberOfSlashesInOriginalUrl = strlen($originalUrl) - strlen(str_replace('/', '', $originalUrl));
foreach ($urls as $url) {
$url = followUrlRedirects($url);
if (trim($url) != '') {
$numberOfSlashesInUrl = strlen($url) - strlen(str_replace('/', '', $url));
if (stringStartsWith($url, $connectionPart . $localPartDirectory) && $numberOfSlashesInUrl == $numberOfSlashesInOriginalUrl) {
$mostLikelyUrls[] = $url;
} elseif (stringStartsWith($url, $connectionPart)) {
$possibleUrls[] = $url;
} else {
$unlikelyUrls[] = $url;
}
}
}
asort($unlikelyUrls);
$itemId = 0;
foreach ($mostLikelyUrls as $url) {
$links[] = '<li class="orderingItem"><label for="urlId' . $itemId . '"><input type="checkbox" name="pages[]" value="' . $url . '" id="urlId' . $itemId . '" checked="checked"/><span class="title">' . $url . '</label></span></li>' . "\n";
$itemId++;
}
foreach ($possibleUrls as $url) {
$links[] = '<li class="orderingItem"><label for="urlId' . $itemId . '"><input type="checkbox" name="pages[]" value="' . $url . '" id="urlId' . $itemId . '"/><span class="title">' . $url . '</label></span></li>' . "\n";
$itemId++;
}
foreach ($unlikelyUrls as $url) {
$links[] = '<li class="orderingItem"><label for="urlId' . $itemId . '"><input type="checkbox" name="pages[]" value="' . $url . '" id="urlId' . $itemId . '"/><span class="title">' . $url . '</label></span></li>' . "\n";
$itemId++;
}
$step2Template = $this->getThemeFragment('generation-step2.htmlf');
$step2Template = str_replace('{{scrape-results}}', implode('', $links), $step2Template);
$step2Template = str_replace('{{scrape-url}}', $url, $step2Template);
return $step2Template;
default:
return $this->getThemeFragment('generation-step1.htmlf');
}
} else {
return $this->getThemeFragment('generation-step1.htmlf');
}
}
/**
* Run a shell command. This function returns STD_ERR as well unlike
* PHPs inbuilt shell commands like shell_exec() or passthru() or system().
* @return string
*/
function shellCommand($command, $timeoutInSeconds = null, $dataToStdIn = null, $haltOnError = false)
{
if ($timeoutInSeconds === null) {
$timeoutInSeconds = 120;
}
$pipes = null;
if ($dataToStdIn) {
$descriptor = array(0 => array("pipe", "r"), 1 => array("pipe", "w"), 2 => array("pipe", "w"));
$currentWorkingDirectory = getOperatingSystemsTemporaryDirectory();
$envionmentVariables = array();
$process = proc_open($command, $descriptor, $pipes, $currentWorkingDirectory, $envionmentVariables);
fwrite($pipes[0], $dataToStdIn);
fclose($pipes[0]);
stream_set_timeout($pipes[1], $timeoutInSeconds);
stream_set_timeout($pipes[2], $timeoutInSeconds);
} else {
$process = popen("({$command})2>&1&", "r");
if ($timeoutInSeconds == 0) {
pclose($process);
return;
}
//stream_set_timeout($process, $timeoutInSeconds);
$pipes[] = $process;
}
if (!is_resource($process)) {
if ($haltOnError) {
webServiceError($command);
}
if (!$dataToStdIn) {
return null;
}
return array('stdOut' => null, 'statusCode' => -1, 'stdErr' => null);
}
$response = array();
$endTime = microtime(true) + (double) $timeoutInSeconds;
foreach ($pipes as $pipe) {
if (!is_resource($pipe)) {
continue;
}
$returnValue = null;
while (!feof($pipe)) {
$returnValue .= fgets($pipe, 8);
$streamInfo = stream_get_meta_data($pipe);
if ($streamInfo['timed_out'] === true || microtime(true) > $endTime) {
$returnValue .= 'Docvert timeout';
break;
}
}
$response[] = $returnValue;
pclose($pipe);
}
if (!$dataToStdIn) {
return $response[0];
} else {
$statusCode = proc_close($process);
if ($statusCode !== 0 && $haltOnError) {
webServiceError($statusCode . ' ' . implode(' ', $pipes));
}
return array('stdOut' => $response[0], 'statusCode' => $statusCode, 'stdErr' => $response[1]);
}
}
}
}
if (!$files || !$converter || !$pipeline || !$outputZip || $extraParameters) {
$commandLineHelp = "\n" . $extraParameters;
$commandLineHelp .= '&error-command-line-help;';
$errorPlaceholders['commandLineFiles'] = '';
$errorPlaceholders['commandLineConverter'] = '';
$errorPlaceholders['commandLinePipeline'] = '';
$errorPlaceholders['commandLineOutputZip'] = '';
if (!$files) {
$errorPlaceholders['commandLineFiles'] = ' --input-files ';
}
if (!$converter) {
$errorPlaceholders['commandLineConverter'] = ' --converter';
}
if (!$pipeline) {
$errorPlaceholders['commandLinePipeline'] = ' --pipeline';
}
if (!$outputZip) {
$errorPlaceholders['commandLineOutputZip'] = ' --output-zip';
}
webServiceError($commandLineHelp, 400, $errorPlaceholders);
}
//print_r($files);
//print $converter;
//print $pipeline;
//print $autoPipeline;
//print $afterConversion;
//print $setupOpenOfficeOrg;
//print $outputZip;
processConversion($files, $converter, $pipeline, $autoPipeline, $afterConversion, $setupOpenOfficeOrg, $outputZip);
