| static public regexp_match_get ( $pattern, $subject, &$matches ) : integer | boolean | ||
| $pattern | string Regular expression | |
| $subject | string String to apply regular expression to | |
| $matches | array Reference to receive matches | |
| return | integer | boolean | Returns 1 if the pattern matches given subject, 0 if it does not, or FALSE if an error occurred. |
/**
* Retrieve the export as an XML string.
* @param $pluginUrl string the url to be requested for export.
* @param $postParams array additional post parameters
* @return string
*/
protected function getXmlOnExport($pluginUrl, $postParams = array())
{
// Prepare HTTP session.
$curlCh = curl_init();
curl_setopt($curlCh, CURLOPT_POST, true);
// Create a cookie file (required for log-in).
$cookies = tempnam(sys_get_temp_dir(), 'curlcookies');
// Log in.
$loginUrl = $this->baseUrl . '/index.php/test/login/signIn';
// Bug #8518 safety work-around
if ($this->password[0] == '@') {
die('CURL parameters may not begin with @.');
}
$loginParams = array('username' => 'admin', 'password' => $this->password);
curl_setopt($curlCh, CURLOPT_URL, $loginUrl);
curl_setopt($curlCh, CURLOPT_POSTFIELDS, $loginParams);
curl_setopt($curlCh, CURLOPT_COOKIEJAR, $cookies);
self::assertTrue(curl_exec($curlCh));
// Request export document.
$exportUrl = $this->baseUrl . '/index.php/test/manager/importexport/plugin/' . $pluginUrl;
curl_setopt($curlCh, CURLOPT_URL, $exportUrl);
// Bug #8518 safety work-around
foreach ($postParams as $paramValue) {
if ($paramValue[0] == '@') {
die('CURL parameters may not begin with @.');
}
}
curl_setopt($curlCh, CURLOPT_POSTFIELDS, $postParams);
curl_setopt($curlCh, CURLOPT_HTTPHEADER, array('Accept: application/xml, application/x-gtar, */*'));
curl_setopt($curlCh, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curlCh, CURLOPT_HEADER, true);
$response = curl_exec($curlCh);
do {
list($header, $response) = explode("\r\n\r\n", $response, 2);
} while (PKPString::regexp_match('#HTTP/.*100#', $header));
// Check whether we got a tar file.
if (PKPString::regexp_match('#Content-Type: application/x-gtar#', $header)) {
// Save the data to a temporary file.
$tempfile = tempnam(sys_get_temp_dir(), 'tst');
file_put_contents($tempfile, $response);
// Recursively extract tar file.
$result = $this->extractTarFile($tempfile);
unlink($tempfile);
} else {
$matches = null;
PKPString::regexp_match_get('#filename="([^"]+)"#', $header, $matches);
self::assertTrue(isset($matches[1]));
$result = array($matches[1] => $response);
}
// Destroy HTTP session.
curl_close($curlCh);
unlink($cookies);
return $result;
}
/**
* Extract and validate a plugin (prior to installation)
* @param $filePath string Full path to plugin archive
* @param $originalFileName string Original filename of plugin archive
* @return string|null Extracted plugin path on success; null on error
*/
function extractPlugin($filePath, $originalFileName, &$errorMsg)
{
$fileManager = new FileManager();
// tar archive basename (less potential version number) must
// equal plugin directory name and plugin files must be in a
// directory named after the plug-in (potentially with version)
$matches = array();
PKPString::regexp_match_get('/^[a-zA-Z0-9]+/', basename($originalFileName, '.tar.gz'), $matches);
$pluginShortName = array_pop($matches);
if (!$pluginShortName) {
$errorMsg = __('manager.plugins.invalidPluginArchive');
$fileManager->deleteFile($filePath);
return null;
}
// Create random dirname to avoid symlink attacks.
$pluginExtractDir = dirname($filePath) . DIRECTORY_SEPARATOR . $pluginShortName . substr(md5(mt_rand()), 0, 10);
mkdir($pluginExtractDir);
// Test whether the tar binary is available for the export to work
$tarBinary = Config::getVar('cli', 'tar');
if (!empty($tarBinary) && file_exists($tarBinary)) {
exec($tarBinary . ' -xzf ' . escapeshellarg($filePath) . ' -C ' . escapeshellarg($pluginExtractDir));
} else {
$errorMsg = __('manager.plugins.tarCommandNotFound');
}
$fileManager->deleteFile($filePath);
if (empty($errorMsg)) {
// Look for a directory named after the plug-in's short
// (alphanumeric) name within the extracted archive.
if (is_dir($tryDir = $pluginExtractDir . '/' . $pluginShortName)) {
return $tryDir;
// Success
}
// Failing that, look for a directory named after the
// archive. (Typically also contains the version number
// e.g. with github generated release archives.)
PKPString::regexp_match_get('/^[a-zA-Z0-9.-]+/', basename($originalFileName, '.tar.gz'), $matches);
if (is_dir($tryDir = $pluginExtractDir . '/' . array_pop($matches))) {
// We found a directory named after the archive
// within the extracted archive. (Typically also
// contains the version number, e.g. github
// generated release archives.)
return $tryDir;
}
$errorMsg = __('manager.plugins.invalidPluginArchive');
}
$fileManager->rmtree($pluginExtractDir);
return null;
}
/**
* Normalize incoming date string.
* @see Filter::process()
* @param $input string
* @return string
*/
function &process(&$input)
{
// FIXME: We have to i18nize this when expanding citation parsing to other languages
static $monthNames = array('Jan' => '01', 'Feb' => '02', 'Mar' => '03', 'Apr' => '04', 'May' => '05', 'Jun' => '06', 'Jul' => '07', 'Aug' => '08', 'Sep' => '09', 'Oct' => '10', 'Nov' => '11', 'Dec' => '12');
$dateExpressions = array('/(?P<year>\\d{4})-(?P<month>\\d{2})-(?P<day>\\d{2})/', '/(?P<year>\\d{4})(\\s|-)*(?P<monthName>[a-z]\\w+)?(\\s|-)*(?P<day>\\d+)?/i');
$normalizedDate = null;
foreach ($dateExpressions as $dateExpression) {
if (PKPString::regexp_match_get($dateExpression, $input, $parsedDate)) {
if (isset($parsedDate['year'])) {
$normalizedDate = $parsedDate['year'];
$month = '';
if (isset($parsedDate['monthName'])) {
$monthName = substr($parsedDate['monthName'], 0, 3);
if (isset($monthNames[$monthName])) {
// Convert the month name to a two digit numeric month representation.
$month = $monthNames[$monthName];
}
}
if (isset($parsedDate['month'])) {
// Convert month to a two digit representation.
$month = str_pad($parsedDate['month'], 2, '0', STR_PAD_LEFT);
}
if (!empty($month)) {
$normalizedDate .= '-' . $month;
if (isset($parsedDate['day'])) {
$normalizedDate .= '-' . str_pad($parsedDate['day'], 2, '0', STR_PAD_LEFT);
}
}
}
if (!empty($normalizedDate)) {
break;
}
}
}
return $normalizedDate;
}
/**
* @see Validator::isValid()
* @param $value mixed
* @return boolean
*/
function isValid($value)
{
return (bool) PKPString::regexp_match_get($this->_regExp, $value, $this->_matches);
}
/**
* @see OAIMetadataFormat#toXml
* TODO:
* <copyright-holder>
* In Isabelle's mapping document:
* Article order in the issue's Table of Contents
*/
function toXml(&$record, $format = null)
{
$article =& $record->getData('article');
$journal =& $record->getData('journal');
$section =& $record->getData('section');
$issue =& $record->getData('issue');
$galleys =& $record->getData('galleys');
$articleId = $article->getId();
// Cache issue ordering information.
static $issueId;
static $sectionSeq;
if (!isset($issueId) || $issueId != $issue->getId()) {
$sectionDao = DAORegistry::getDAO('SectionDAO');
$issueId = $issue->getId();
$sections = $sectionDao->getByIssueId($issueId);
$sectionSeq = array();
$i = 0;
foreach ($sections as $thisSection) {
$sectionSeq[$thisSection->getId()] = $i++;
}
unset($sections);
}
$abbreviation = $journal->getLocalizedSetting('abbreviation');
$printIssn = $journal->getSetting('printIssn');
$onlineIssn = $journal->getSetting('onlineIssn');
$primaryLocale = $article->getLanguage() != '' ? $article->getLanguage() : $journal->getPrimaryLocale();
$publisherInstitution = $journal->getSetting('publisherInstitution');
$datePublished = $article->getDatePublished();
if (!$datePublished) {
$datePublished = $issue->getDatePublished();
}
if ($datePublished) {
$datePublished = strtotime($datePublished);
}
$response = "<article\n" . "\txmlns=\"http://dtd.nlm.nih.gov/publishing/2.3\"\n" . "\txmlns:xlink=\"http://www.w3.org/1999/xlink\"\n" . "\txmlns:mml=\"http://www.w3.org/1998/Math/MathML\"\n" . "\txmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"\n" . "\txsi:schemaLocation=\"http://dtd.nlm.nih.gov/publishing/2.3\n" . "\thttp://dtd.nlm.nih.gov/publishing/2.3/xsd/journalpublishing.xsd\"\n" . (($s = $section->getLocalizedIdentifyType()) != '' ? "\tarticle-type=\"" . htmlspecialchars(Core::cleanVar($s)) . "\"" : '') . "\txml:lang=\"" . strtoupper(substr($primaryLocale, 0, 2)) . "\">\n" . "\t<front>\n" . "\t\t<journal-meta>\n" . "\t\t\t<journal-id journal-id-type=\"other\">" . htmlspecialchars(Core::cleanVar(($s = Config::getVar('oai', 'nlm_journal_id')) != '' ? $s : $journal->getPath())) . "</journal-id>\n" . "\t\t\t<journal-title>" . htmlspecialchars(Core::cleanVar($journal->getLocalizedName())) . "</journal-title>\n";
// Include translated journal titles
foreach ($journal->getName(null) as $locale => $title) {
if ($locale == $primaryLocale) {
continue;
}
$response .= "\t\t\t<trans-title xml:lang=\"" . strtoupper(substr($locale, 0, 2)) . "\">" . htmlspecialchars(Core::cleanVar($title)) . "</trans-title>\n";
}
$response .= (!empty($onlineIssn) ? "\t\t\t<issn pub-type=\"epub\">" . htmlspecialchars(Core::cleanVar($onlineIssn)) . "</issn>" : '') . (!empty($printIssn) ? "\t\t\t<issn pub-type=\"ppub\">" . htmlspecialchars(Core::cleanVar($printIssn)) . "</issn>" : '') . ($publisherInstitution != '' ? "\t\t\t<publisher><publisher-name>" . htmlspecialchars(Core::cleanVar($publisherInstitution)) . "</publisher-name></publisher>\n" : '') . "\t\t</journal-meta>\n" . "\t\t<article-meta>\n" . "\t\t\t<article-id pub-id-type=\"other\">" . htmlspecialchars(Core::cleanVar($article->getBestArticleId())) . "</article-id>\n" . (($s = $article->getPubId('doi')) ? "\t\t\t<article-id pub-id-type=\"doi\">" . htmlspecialchars(Core::cleanVar($s)) . "</article-id>\n" : '') . "\t\t\t<article-categories><subj-group subj-group-type=\"heading\"><subject>" . htmlspecialchars(Core::cleanVar($section->getLocalizedTitle())) . "</subject></subj-group></article-categories>\n" . "\t\t\t<title-group>\n" . "\t\t\t\t<article-title>" . htmlspecialchars(Core::cleanVar(strip_tags($article->getLocalizedTitle()))) . "</article-title>\n";
// Include translated journal titles
foreach ($article->getTitle(null) as $locale => $title) {
if ($locale == $primaryLocale) {
continue;
}
$response .= "\t\t\t\t<trans-title xml:lang=\"" . strtoupper(substr($locale, 0, 2)) . "\">" . htmlspecialchars(Core::cleanVar(strip_tags($title))) . "</trans-title>\n";
}
$response .= "\t\t\t</title-group>\n" . "\t\t\t<contrib-group>\n";
// Include authors
foreach ($article->getAuthors() as $author) {
$response .= "\t\t\t\t<contrib " . ($author->getPrimaryContact() ? 'corresp="yes" ' : '') . "contrib-type=\"author\">\n" . "\t\t\t\t\t<name name-style=\"western\">\n" . "\t\t\t\t\t\t<surname>" . htmlspecialchars(Core::cleanVar($author->getLastName())) . "</surname>\n" . "\t\t\t\t\t\t<given-names>" . htmlspecialchars(Core::cleanVar($author->getFirstName()) . (($s = $author->getMiddleName()) != '' ? " {$s}" : '')) . "</given-names>\n" . "\t\t\t\t\t</name>\n" . (($s = $author->getLocalizedAffiliation()) != '' ? "\t\t\t\t\t<aff>" . htmlspecialchars(Core::cleanVar($s)) . "</aff>\n" : '') . "\t\t\t\t\t<email>" . htmlspecialchars(Core::cleanVar($author->getEmail())) . "</email>\n" . (($s = $author->getUrl()) != '' ? "\t\t\t\t\t<uri>" . htmlspecialchars(Core::cleanVar($s)) . "</uri>\n" : '') . "\t\t\t\t</contrib>\n";
}
// Include editorships (optimized)
$response .= $this->getEditorialInfo($journal->getId());
$response .= "\t\t\t</contrib-group>\n";
if ($datePublished) {
$response .= "\t\t\t<pub-date pub-type=\"epub\">\n" . "\t\t\t\t<day>" . strftime('%d', $datePublished) . "</day>\n" . "\t\t\t\t<month>" . strftime('%m', $datePublished) . "</month>\n" . "\t\t\t\t<year>" . strftime('%Y', $datePublished) . "</year>\n" . "\t\t\t</pub-date>\n";
}
$response .= ($issue->getShowYear() ? "\t\t\t<pub-date pub-type=\"collection\"><year>" . htmlspecialchars(Core::cleanVar($issue->getYear())) . "</year></pub-date>\n" : '') . ($issue->getShowVolume() ? "\t\t\t<volume>" . htmlspecialchars(Core::cleanVar($issue->getVolume())) . "</volume>\n" : '') . ($issue->getShowNumber() ? "\t\t\t<issue seq=\"" . htmlspecialchars(Core::cleanVar($sectionSeq[$section->getId()] * 100 + $article->getSequence())) . "\">" . htmlspecialchars(Core::cleanVar($issue->getNumber())) . "</issue>\n" : '') . "\t\t\t<issue-id pub-id-type=\"other\">" . htmlspecialchars(Core::cleanVar($issue->getBestIssueId())) . "</issue-id>\n" . ($issue->getShowTitle() ? "\t\t\t<issue-title>" . htmlspecialchars(Core::cleanVar($issue->getLocalizedTitle())) . "</issue-title>\n" : '');
// Include page info, if available and parseable.
$matches = null;
if (PKPString::regexp_match_get('/^[Pp][Pp]?[.]?[ ]?(\\d+)$/', $article->getPages(), $matches)) {
$matchedPage = htmlspecialchars(Core::cleanVar($matches[1]));
$response .= "\t\t\t\t<fpage>{$matchedPage}</fpage><lpage>{$matchedPage}</lpage>\n";
$pageCount = 1;
} elseif (PKPString::regexp_match_get('/^[Pp][Pp]?[.]?[ ]?(\\d+)[ ]?(-|–)[ ]?([Pp][Pp]?[.]?[ ]?)?(\\d+)$/', $article->getPages(), $matches)) {
$matchedPageFrom = htmlspecialchars(Core::cleanVar($matches[1]));
$matchedPageTo = htmlspecialchars(Core::cleanVar($matches[4]));
$response .= "\t\t\t\t<fpage>{$matchedPageFrom}</fpage>\n" . "\t\t\t\t<lpage>{$matchedPageTo}</lpage>\n";
$pageCount = $matchedPageTo - $matchedPageFrom + 1;
}
$response .= "\t\t\t<permissions>\n" . "\t\t\t\t<copyright-statement>" . htmlspecialchars(__('submission.copyrightStatement', array('copyrightYear' => $article->getCopyrightYear(), 'copyrightHolder' => $article->getLocalizedCopyrightHolder()))) . "</copyright-statement>\n" . ($datePublished ? "\t\t\t\t<copyright-year>" . $article->getCopyrightYear() . "</copyright-year>\n" : '') . "\t\t\t\t<license xlink:href=\"" . $article->getLicenseURL() . "\">\n" . (($s = Application::getCCLicenseBadge($article->getLicenseURL())) ? "\t\t\t\t\t<license-p>" . strip_tags($s) . "</license-p>\n" : '') . "\t\t\t\t</license>\n" . "\t\t\t</permissions>\n" . "\t\t\t<self-uri xlink:href=\"" . htmlspecialchars(Core::cleanVar(Request::url($journal->getPath(), 'article', 'view', $article->getBestArticleId()))) . "\" />\n";
// Include galley links
foreach ($article->getGalleys() as $galley) {
$response .= "\t\t\t<self-uri content-type=\"" . htmlspecialchars(Core::cleanVar($galley->getFileType())) . "\" xlink:href=\"" . htmlspecialchars(Core::cleanVar(Request::url($journal->getPath(), 'article', 'view', array($article->getBestArticleId(), $galley->getId())))) . "\" />\n";
}
// Include abstract(s)
$abstract = htmlspecialchars(Core::cleanVar(strip_tags($article->getLocalizedAbstract())));
if (!empty($abstract)) {
$abstract = "<p>{$abstract}</p>";
// $abstract = '<p>' . PKPString::regexp_replace('/\n+/', '</p><p>', $abstract) . '</p>';
$response .= "\t\t\t<abstract xml:lang=\"" . strtoupper(substr($primaryLocale, 0, 2)) . "\">{$abstract}</abstract>\n";
}
if (is_array($article->getAbstract(null))) {
foreach ($article->getAbstract(null) as $locale => $abstract) {
if ($locale == $primaryLocale || empty($abstract)) {
continue;
}
$abstract = htmlspecialchars(Core::cleanVar(strip_tags($abstract)));
if (empty($abstract)) {
continue;
}
$abstract = "<p>{$abstract}</p>";
//$abstract = '<p>' . PKPString::regexp_replace('/\n+/', '</p><p>', $abstract) . '</p>';
$response .= "\t\t\t<abstract-trans xml:lang=\"" . strtoupper(substr($locale, 0, 2)) . "\">{$abstract}</abstract-trans>\n";
}
}
$subjects = array();
if (is_array($article->getSubject(null))) {
foreach ($article->getSubject(null) as $locale => $subject) {
$s = array_map('trim', explode(';', Core::cleanVar($subject)));
if (!empty($s)) {
$subjects[$locale] = $s;
}
}
}
if (!empty($subjects)) {
foreach ($subjects as $locale => $s) {
$response .= "\t\t\t<kwd-group xml:lang=\"" . strtoupper(substr($locale, 0, 2)) . "\">\n";
foreach ($s as $subject) {
$response .= "\t\t\t\t<kwd>" . htmlspecialchars($subject) . "</kwd>\n";
}
$response .= "\t\t\t</kwd-group>\n";
}
}
$response .= (isset($pageCount) ? "\t\t\t<counts><page-count count=\"" . (int) $pageCount . "\" /></counts>\n" : '') . "\t\t</article-meta>\n" . "\t</front>\n";
// Include body text (for search indexing only)
import('classes.search.ArticleSearchIndex');
$text = '';
$galleys = $article->getGalleys();
// Give precedence to HTML galleys, as they're quickest to parse
usort($galleys, create_function('$a, $b', 'return $a->getFileType()==\'text/html\')?-1:1;'));
// Determine any access limitations. If there are, do not
// provide the full-text.
import('classes.issue.IssueAction');
$issueAction = new IssueAction();
$subscriptionRequired = $issueAction->subscriptionRequired($issue);
$isSubscribedDomain = $issueAction->subscribedDomain($journal, $issue->getId(), $article->getId());
if (!$subscriptionRequired || $isSubscribedDomain) {
foreach ($galleys as $galley) {
$parser =& SearchFileParser::fromFile($galley);
if ($parser && $parser->open()) {
while (($s = $parser->read()) !== false) {
$text .= $s;
}
$parser->close();
}
if ($galley->getFileType() == 'text/html') {
$text = strip_tags($text);
}
unset($galley);
// Use the first parseable galley.
if (!empty($text)) {
break;
}
}
}
if (!empty($text)) {
$response .= "\t<body><p>" . htmlspecialchars(Core::cleanVar(Core::cleanVar($text))) . "</p></body>\n";
}
// Add NLM citation info
$filterDao = DAORegistry::getDAO('FilterDAO');
/* @var $filterDao FilterDAO */
$nlmFilters = $filterDao->getObjectsByGroup('submission=>nlm23-article-xml');
assert(count($nlmFilters) == 1);
$nlmFilter = array_pop($nlmFilters);
$nlmXmlDom = new DOMDocument();
$nlmXmlDom->loadXML($nlmFilter->execute($article));
$documentElement =& $nlmXmlDom->documentElement;
// Work-around for hasChildNodes being stupid about whitespace.
$hasChildren = false;
if (isset($documentElement->childNodes)) {
foreach ($documentElement->childNodes as $c) {
if ($c->nodeType == XML_ELEMENT_NODE) {
$hasChildren = true;
}
}
}
// If there were any citations, include them.
if ($hasChildren) {
$innerXml = $nlmXmlDom->saveXML($documentElement);
$response .= "<back>{$innerXml}</back>\n";
}
$response .= "</article>";
return $response;
}
/**
* Submit a new test article.
* @param $title string
* @return integer the id of the new article
*/
protected function submitArticle($title = 'Editing test article')
{
// We need to be logged in to submit an article.
$this->logIn();
$submissionPage = $this->baseUrl . '/index.php/lucene-test/author/submit/';
//
// First submission page.
//
$this->verifyAndOpen($submissionPage . '1');
$this->waitForElementPresent('css=input.defaultButton');
// Set Section.
$this->select('sectionId', 'value=3');
// Accept submission conditions.
$checkboxId = 1;
while ($this->isElementPresent("checklist-{$checkboxId}")) {
$this->check("checklist-{$checkboxId}");
$checkboxId++;
}
// Submit first submission page.
$this->clickAndWait('css=input.defaultButton');
//
// Second submission page.
//
$this->waitForLocation($submissionPage . '2*');
$this->waitForElementPresent('css=input.defaultButton');
// We should now have the article ID in the URL.
$url = $this->getLocation();
$matches = null;
PKPString::regexp_match_get('/articleId=([0-9]+)/', $url, $matches);
self::assertTrue(count($matches) == 2);
$articleId = $matches[1];
self::assertTrue(is_numeric($articleId));
$articleId = (int) $articleId;
// Submit the second submission page.
$this->clickAndWait('css=input.defaultButton');
$this->waitForConfirmation('*Are you sure you wish to continue*');
//
// Third submission page.
//
$this->waitForLocation($submissionPage . '3*');
$this->waitForElementPresent('css=input.defaultButton');
// Fill in article metadata.
$this->type('authors-0-firstName', 'Arthur');
$this->type('authors-0-lastName', 'McAutomatic');
$this->type('title', $title);
$this->verifyElementPresent('id=abstract_ifr');
if ($this->verified()) {
// TinyMCE hack.
$jsScript = "selenium.browserbot.getCurrentWindow().document" . ".getElementById('abstract_ifr').contentDocument.body.innerHTML = " . "'{$title} abstract'";
$this->getEval($jsScript);
} else {
$this->type('abstract', $title . ' abstract');
}
// Submit metadata.
$this->clickAndWait('css=input.defaultButton');
//
// Fourth and fifth submission page.
//
$this->waitForLocation($submissionPage . '4*');
$this->waitForElementPresent('css=input.defaultButton');
// Do not upload any supplementary file and continue.
$this->clickAndWait('css=input.defaultButton');
$this->waitForLocation($submissionPage . '5*');
$this->waitForElementPresent('css=input.defaultButton');
// Confirm the submission.
$this->clickAndWait('css=input.defaultButton');
return $articleId;
}
/**
* Do the actual web service request.
* @param $action string
* @param $arg string
* @param $attachment array
* @return boolean|string True for success, an error message otherwise.
*/
function _doRequest($action, $arg, $attachment = null)
{
// Build the multipart SOAP message from scratch.
$soapMessage = '<SOAP-ENV:Envelope xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/" ' . 'xmlns:med="http://www.medra.org">' . '<SOAP-ENV:Header/>' . '<SOAP-ENV:Body>' . "<med:{$action}>{$arg}</med:{$action}>" . '</SOAP-ENV:Body>' . '</SOAP-ENV:Envelope>';
$soapMessageId = $this->_getContentId($action);
if ($attachment) {
assert(count($attachment) == 1);
$request = "--MIME_boundary\r\n" . $this->_getMimePart($soapMessageId, $soapMessage) . "--MIME_boundary\r\n" . $this->_getMimePart(key($attachment), current($attachment)) . "--MIME_boundary--\r\n";
$contentType = 'multipart/related; type="text/xml"; boundary="MIME_boundary"';
} else {
$request = $soapMessage;
$contentType = 'text/xml';
}
// Prepare HTTP session.
$curlCh = curl_init();
curl_setopt($curlCh, CURLOPT_RETURNTRANSFER, true);
curl_setopt($curlCh, CURLOPT_POST, true);
// Set up basic authentication.
curl_setopt($curlCh, CURLOPT_HTTPAUTH, CURLAUTH_BASIC);
curl_setopt($curlCh, CURLOPT_USERPWD, $this->_auth);
// Set up SSL.
curl_setopt($curlCh, CURLOPT_SSL_VERIFYPEER, false);
// Make SOAP request.
curl_setopt($curlCh, CURLOPT_URL, $this->_endpoint);
$extraHeaders = array('SOAPAction: "' . $action . '"', 'Content-Type: ' . $contentType, 'UserAgent: OJS-mEDRA');
curl_setopt($curlCh, CURLOPT_HTTPHEADER, $extraHeaders);
curl_setopt($curlCh, CURLOPT_POSTFIELDS, $request);
$result = true;
$response = curl_exec($curlCh);
// We do not localize our error messages as they are all
// fatal errors anyway and must be analyzed by technical staff.
if ($response === false) {
$result = 'OJS-mEDRA: Expected string response.';
}
if ($result === true && ($status = curl_getinfo($curlCh, CURLINFO_HTTP_CODE)) != MEDRA_WS_RESPONSE_OK) {
$result = 'OJS-mEDRA: Expected ' . MEDRA_WS_RESPONSE_OK . ' response code, got ' . $status . ' instead.';
}
curl_close($curlCh);
// Check SOAP response by simple string manipulation rather
// than instantiating a DOM.
if (is_string($response)) {
$matches = array();
PKPString::regexp_match_get('#<faultstring>([^<]*)</faultstring>#', $response, $matches);
if (empty($matches)) {
if ($attachment) {
assert(PKPString::regexp_match('#<returnCode>success</returnCode>#', $response));
} else {
$parts = explode("\r\n\r\n", $response);
$result = array_pop($parts);
$result = PKPString::regexp_replace('/>[^>]*$/', '>', $result);
}
} else {
$result = 'mEDRA: ' . $status . ' - ' . $matches[1];
}
} else {
$result = 'OJS-mEDRA: Expected string response.';
}
return $result;
}
/**
* Converts a string with a single person
* to an NLM name description.
*
* TODO: add initials from all given names to initials
* element
*
* @param $personString string
* @param $title boolean true to parse for title
* @param $degrees boolean true to parse for degrees
* @return MetadataDescription an NLM name description or null
* if the string could not be converted
*/
function &_parsePersonString($personString, $title, $degrees)
{
// Expressions to parse person strings, ported from CiteULike person
// plugin, see http://svn.citeulike.org/svn/plugins/person.tcl
static $personRegex = array('title' => '(?:His (?:Excellency|Honou?r)\\s+|Her (?:Excellency|Honou?r)\\s+|The Right Honou?rable\\s+|The Honou?rable\\s+|Right Honou?rable\\s+|The Rt\\.? Hon\\.?\\s+|The Hon\\.?\\s+|Rt\\.? Hon\\.?\\s+|Mr\\.?\\s+|Ms\\.?\\s+|M\\/s\\.?\\s+|Mrs\\.?\\s+|Miss\\.?\\s+|Dr\\.?\\s+|Sir\\s+|Dame\\s+|Prof\\.?\\s+|Professor\\s+|Doctor\\s+|Mister\\s+|Mme\\.?\\s+|Mast(?:\\.|er)?\\s+|Lord\\s+|Lady\\s+|Madam(?:e)?\\s+|Priv\\.-Doz\\.\\s+)+', 'degrees' => '(,\\s+(?:[A-Z\\.]+))+', 'initials' => '(?:(?:[A-Z]\\.){1,3}[A-Z]\\.?)|(?:(?:[A-Z]\\.\\s){1,3}[A-Z]\\.?)|(?:[A-Z]{1,4})|(?:(?:[A-Z]\\.-?){1,4})|(?:(?:[A-Z]\\.-?){1,3}[A-Z]\\.?)|(?:(?:[A-Z]-){1,3}[A-Z])|(?:(?:[A-Z]\\s){1,3}[A-Z]\\.?)|(?:(?:[A-Z]-){1,3}[A-Z]\\.?)', 'prefix' => 'Dell(?:[a|e])?(?:\\s|$)|Dalle(?:\\s|$)|D[a|e]ll\'(?:\\s|$)|Dela(?:\\s|$)|Del(?:\\s|$)|[Dd]e(?:\\s|$)(?:La(?:\\s|$)|Los(?:\\s|$))?|[Dd]e(?:\\s|$)|[Dd][a|i|u](?:\\s|$)|L[a|e|o](?:\\s|$)|[D|L|O]\'|St\\.?(?:\\s|$)|San(?:\\s|$)|[Dd]en(?:\\s|$)|[Vv]on(?:\\s|$)(?:[Dd]er(?:\\s|$))?|(?:[Ll][ea](?:\\s|$))?[Vv]an(?:\\s|$)(?:[Dd]e(?:n|r)?(?:\\s|$))?', 'givenName' => '(?:[^ \\t\\n\\r\\f\\v,.;()]{2,}|[^ \\t\\n\\r\\f\\v,.;()]{2,}\\-[^ \\t\\n\\r\\f\\v,.;()]{2,})');
// The expressions for given name, suffix and surname are the same
$personRegex['surname'] = $personRegex['suffix'] = $personRegex['givenName'];
$personRegex['double-surname'] = "(?:" . $personRegex['surname'] . "\\s)*" . $personRegex['surname'];
// Shortcut for prefixed surname
$personRegexPrefixedSurname = "(?P<prefix>(?:" . $personRegex['prefix'] . ")?)(?P<surname>" . $personRegex['surname'] . ")";
$personRegexPrefixedDoubleSurname = "(?P<prefix>(?:" . $personRegex['prefix'] . ")?)(?P<surname>" . $personRegex['double-surname'] . ")";
// Instantiate the target person description
$personDescription = new MetadataDescription('lib.pkp.plugins.metadata.nlm30.schema.Nlm30NameSchema', $this->_assocType);
// Clean the person string
$personString = trim($personString);
// 1. Extract title and degree from the person string and use this as suffix
$suffixString = '';
$results = array();
if ($title && PKPString::regexp_match_get('/^(' . $personRegex['title'] . ')/i', $personString, $results)) {
$suffixString = trim($results[1], ',:; ');
$personString = PKPString::regexp_replace('/^(' . $personRegex['title'] . ')/i', '', $personString);
}
if ($degrees && PKPString::regexp_match_get('/(' . $personRegex['degrees'] . ')$/i', $personString, $results)) {
$degreesArray = explode(',', trim($results[1], ','));
foreach ($degreesArray as $key => $degree) {
$degreesArray[$key] = PKPString::trimPunctuation($degree);
}
$suffixString .= ' - ' . implode('; ', $degreesArray);
$personString = PKPString::regexp_replace('/(' . $personRegex['degrees'] . ')$/i', '', $personString);
}
if (!empty($suffixString)) {
$personDescription->addStatement('suffix', $suffixString);
}
// Space initials when followed by a given name or last name.
$personString = PKPString::regexp_replace('/([A-Z])\\.([A-Z][a-z])/', '\\1. \\2', $personString);
// 2. Extract names and initials from the person string
// The parser expressions are ordered by specificity. The most specific expressions
// come first. Only if these specific expressions don't work will we turn to less
// specific ones. This avoids parsing errors. It also explains why we don't use the
// ?-quantifier for optional elements like initials or middle name where they could
// be misinterpreted.
$personExpressions = array('/^' . $personRegexPrefixedSurname . '$/i', '/^(?P<initials>' . $personRegex['initials'] . ')\\s' . $personRegexPrefixedSurname . '$/', '/^' . $personRegexPrefixedSurname . ',?\\s(?P<initials>' . $personRegex['initials'] . ')$/', '/^' . $personRegexPrefixedDoubleSurname . ',\\s(?P<givenName>' . $personRegex['givenName'] . ')\\s(?P<initials>' . $personRegex['initials'] . ')$/', '/^(?P<givenName>' . $personRegex['givenName'] . ')\\s(?P<initials>' . $personRegex['initials'] . ')\\s' . $personRegexPrefixedSurname . '$/', '/^' . $personRegexPrefixedDoubleSurname . ',\\s(?P<givenName>(?:' . $personRegex['givenName'] . '\\s)+)(?P<initials>' . $personRegex['initials'] . ')$/', '/^(?P<givenName>(?:' . $personRegex['givenName'] . '\\s)+)(?P<initials>' . $personRegex['initials'] . ')\\s' . $personRegexPrefixedSurname . '$/', '/^' . $personRegexPrefixedDoubleSurname . ',(?P<givenName>(?:\\s' . $personRegex['givenName'] . ')+)$/', '/^(?P<givenName>(?:' . $personRegex['givenName'] . '\\s)+)' . $personRegexPrefixedSurname . '$/', '/^\\s*(?P<surname>' . $personRegex['surname'] . ')(?P<suffix>(?:\\s+' . $personRegex['suffix'] . ')?)\\s*,\\s*(?P<initials>(?:' . $personRegex['initials'] . ')?)\\s*\\((?P<givenName>(?:\\s*' . $personRegex['givenName'] . ')+)\\s*\\)\\s*(?P<prefix>(?:' . $personRegex['prefix'] . ')?)$/', '/^(?P<givenName>' . $personRegex['givenName'] . ')\\.(?P<surname>' . $personRegex['double-surname'] . ')$/', '/^(?P<surname>.*)$/');
$results = array();
foreach ($personExpressions as $expressionId => $personExpression) {
if ($nameFound = PKPString::regexp_match_get($personExpression, $personString, $results)) {
// Given names
if (!empty($results['givenName'])) {
// Split given names
$givenNames = explode(' ', trim($results['givenName']));
foreach ($givenNames as $givenName) {
$personDescription->addStatement('given-names', $givenName);
unset($givenName);
}
}
// Initials (will also be saved as given names)
if (!empty($results['initials'])) {
$results['initials'] = str_replace(array('.', '-', ' '), array('', '', ''), $results['initials']);
for ($initialNum = 0; $initialNum < PKPString::strlen($results['initials']); $initialNum++) {
$initial = $results['initials'][$initialNum];
$personDescription->addStatement('given-names', $initial);
unset($initial);
}
}
// Surname
if (!empty($results['surname'])) {
// Correct all-upper surname
if (strtoupper($results['surname']) == $results['surname']) {
$results['surname'] = ucwords(strtolower($results['surname']));
}
$personDescription->addStatement('surname', $results['surname']);
}
// Prefix/Suffix
foreach (array('prefix', 'suffix') as $propertyName) {
if (!empty($results[$propertyName])) {
$results[$propertyName] = trim($results[$propertyName]);
$personDescription->addStatement($propertyName, $results[$propertyName]);
}
}
break;
}
}
return $personDescription;
}
/**
* Processes form-submitted addresses for inclusion in
* the recipient list
* @param $currentList array Current recipient/cc/bcc list
* @param $newAddresses array "Raw" form parameter for additional addresses
*/
function &processAddresses($currentList, &$newAddresses)
{
foreach ($newAddresses as $newAddress) {
$regs = array();
// Match the form "My Name <*****@*****.**>"
if (PKPString::regexp_match_get('/^([^<>' . "\n" . ']*[^<> ' . "\n" . '])[ ]*<(?P<email>' . PCRE_EMAIL_ADDRESS . ')>$/i', $newAddress, $regs)) {
$currentList[] = array('name' => $regs[1], 'email' => $regs['email']);
} elseif (PKPString::regexp_match_get('/^<?(?P<email>' . PCRE_EMAIL_ADDRESS . ')>?$/i', $newAddress, $regs)) {
$currentList[] = array('name' => '', 'email' => $regs['email']);
} elseif ($newAddress != '') {
$this->errorMessages[] = array('type' => MAIL_ERROR_INVALID_EMAIL, 'address' => $newAddress);
}
}
return $currentList;
}
/**
* @copydoc Filter::process()
* @param $input string
* @return MetadataDescription
*/
function &process(&$input)
{
$citationString =& $input;
// Initialize the parser result array
$matches = array();
$metadata = array();
// Parse out any embedded URLs
$urlPattern = '(<?(https?://([-\\w\\.]+)+(:\\d+)?(/([\\w/_\\.,]*(\\?[^\\s>]+)?)?)?)>?)';
if (PKPString::regexp_match_get($urlPattern, $citationString, $matches)) {
// Assume that the URL is a link to the resource.
$metadata['uri'] = $matches[1];
// Remove the URL from the citation string
$citationString = PKPString::regexp_replace($urlPattern, '', $citationString);
// If the URL is a link to PubMed, save the PMID
$pmIdExpressions = array('/list_uids=(?P<pmId>\\d+)/i', '/pubmed.*details_term=(?P<pmId>\\d+)/i', '/pubmedid=(?P<pmId>\\d+)/i');
foreach ($pmIdExpressions as $pmIdExpression) {
if (PKPString::regexp_match_get($pmIdExpression, $matches[1], $pmIdMatches)) {
$metadata['pub-id[@pub-id-type="pmid"]'] = $pmIdMatches['pmId'];
break;
}
}
}
// Parse out an embedded PMID and remove from the citation string
$pmidPattern = '/pmid:?\\s*(\\d+)/i';
if (PKPString::regexp_match_get($pmidPattern, $citationString, $matches)) {
$metadata['pub-id[@pub-id-type="pmid"]'] = $matches[1];
$citationString = PKPString::regexp_replace($pmidPattern, '', $citationString);
}
// Parse out an embedded DOI and remove it from the citation string
$doiPattern = '/doi:?\\s*(\\S+)/i';
if (PKPString::regexp_match_get($doiPattern, $citationString, $matches)) {
$metadata['pub-id[@pub-id-type="doi"]'] = $matches[1];
$citationString = PKPString::regexp_replace($doiPattern, '', $citationString);
}
// Parse out the access date if we have one and remove it from the citation string
$accessDatePattern = '/accessed:?\\s*([\\s\\w]+)/i';
if (PKPString::regexp_match_get($accessDatePattern, $citationString, $matches)) {
$metadata['access-date'] = $matches[1];
$citationString = PKPString::regexp_replace($accessDatePattern, '', $citationString);
}
// Clean out square brackets
$citationString = PKPString::regexp_replace('/\\[(\\s*(pubmed|medline|full text)\\s*)*]/i', '', $citationString);
// Book citation
$unparsedTail = '';
if (PKPString::regexp_match_get("/\\s*(?P<authors>[^\\.]+)\\.\\s*(?P<source>.*?)\\s*(?P<publisherLoc>[^\\.]*):\\s*(?P<publisherName>[^:]*?);\\s*(?P<date>\\d\\d\\d\\d.*?)(?P<tail>.*)/", $citationString, $matches)) {
$metadata['[@publication-type]'] = NLM30_PUBLICATION_TYPE_BOOK;
$metadata['author'] = $matches['authors'];
$metadata['source'] = $matches['source'];
$metadata['publisher-loc'] = $matches['publisherLoc'];
$metadata['publisher-name'] = $matches['publisherName'];
$metadata['date'] = $matches['date'];
$unparsedTail = $matches['tail'];
// Journal citation
} elseif (PKPString::regexp_match_get("/\\s*(?P<authors>[^\\.]+)\\.\\s*(?P<titleSource>.*)\\s*(?P<date>\\d\\d\\d\\d.*?);(?P<volumeAndIssue>[^:]+):(?P<tail>.*)/", $citationString, $matches)) {
$metadata['[@publication-type]'] = NLM30_PUBLICATION_TYPE_JOURNAL;
$metadata['author'] = $matches['authors'];
$titleSource = array();
if (PKPString::regexp_match_get("/(.*[\\.!\\?])(.*)/", trim($matches['titleSource'], " ."), $titleSource)) {
$metadata['article-title'] = $titleSource[1];
$metadata['source'] = $titleSource[2];
}
$metadata['date'] = $matches['date'];
$volumeAndIssue = array();
if (PKPString::regexp_match_get("/([^\\(]+)(\\(([^\\)]+)\\))?/", $matches['volumeAndIssue'], $volumeAndIssue)) {
$metadata['volume'] = $volumeAndIssue[1];
if (isset($volumeAndIssue[3])) {
$metadata['issue'] = $volumeAndIssue[3];
}
}
$unparsedTail = $matches['tail'];
// Web citation with or without authors
} elseif (PKPString::regexp_match_get("/\\s*(?P<citationSource>.*?)\\s*URL:\\s*(?P<tail>.*)/", $citationString, $matches)) {
$unparsedTail = $matches['tail'];
$citationParts = explode(".", trim($matches['citationSource'], '. '));
switch (count($citationParts)) {
case 0:
// This case should never occur...
assert(false);
break;
case 1:
// Assume this to be a title for the web site.
$metadata['article-title'] = $citationParts[0];
break;
case 2:
// Assume the format: Authors. Title.
$metadata['author'] = $citationParts[0];
$metadata['article-title'] = $citationParts[1];
break;
default:
// Assume the format: Authors. Article Title. Journal Title.
$metadata['author'] = array_shift($citationParts);
// The last part is assumed to be the journal title
$metadata['source'] = array_pop($citationParts);
// Everything in between is assumed to belong to the article title
$metadata['article-title'] = implode('.', $citationParts);
}
}
// TODO: Handle in-ref titles, eg. with editor lists
// Extract page numbers if possible
$pagesPattern = "/^[:p\\.\\s]*(?P<fpage>[Ee]?\\d+)(-(?P<lpage>\\d+))?/";
if (!empty($unparsedTail) && PKPString::regexp_match_get($pagesPattern, $unparsedTail, $matches)) {
$metadata['fpage'] = $matches['fpage'];
if (isset($matches['lpage'])) {
$metadata['lpage'] = $matches['lpage'];
}
// Add the unparsed part of the citation string as a comment so it doesn't get lost.
$comment = PKPString::trimPunctuation(PKPString::regexp_replace($pagesPattern, '', $unparsedTail));
if (!empty($comment)) {
$metadata['comment'] = $comment;
}
}
// Make the meta-data fully NLM citation compliant
$metadata =& $this->postProcessMetadataArray($metadata);
// Create the NLM citation description
return $this->getNlm30CitationDescriptionFromMetadataArray($metadata);
}
/**
* Handle fetch requests for this plugin.
*/
function fetch($args, $request)
{
if (!$this->getEnabled()) {
return false;
}
$scheme = array_shift($args);
switch ($scheme) {
case 'doi':
$doi = implode('/', $args);
$journal = $request->getJournal();
$publishedArticleDao = DAORegistry::getDAO('PublishedArticleDAO');
/* @var $publishedArticleDao PublishedArticleDAO */
$article = $publishedArticleDao->getPublishedArticleByPubId('doi', $doi, $journal ? $journal->getId() : null);
if (is_a($article, 'PublishedArticle')) {
$request->redirect(null, 'article', 'view', $article->getBestArticleId());
}
break;
case 'vnp':
// Volume, number, page
// Volume, number, page
case 'ynp':
// Volume, number, year, page
// This can only be used from within a journal context
$journal = $request->getJournal();
if (!$journal) {
break;
}
if ($scheme == 'vnp') {
$volume = (int) array_shift($args);
$year = null;
} elseif ($scheme == 'ynp') {
$year = (int) array_shift($args);
$volume = null;
} else {
return;
// Suppress scrutinizer warn
}
$number = array_shift($args);
$page = (int) array_shift($args);
$issueDao = DAORegistry::getDAO('IssueDAO');
$issues = $issueDao->getPublishedIssuesByNumber($journal->getId(), $volume, $number, $year);
// Ensure only one issue matched, and fetch it.
$issue = $issues->next();
if (!$issue || $issues->next()) {
break;
}
unset($issues);
$publishedArticleDao = DAORegistry::getDAO('PublishedArticleDAO');
$articles =& $publishedArticleDao->getPublishedArticles($issue->getId());
foreach ($articles as $article) {
// Look for the correct page in the list of articles.
$matches = null;
if (PKPString::regexp_match_get('/^[Pp][Pp]?[.]?[ ]?(\\d+)$/', $article->getPages(), $matches)) {
$matchedPage = $matches[1];
if ($page == $matchedPage) {
$request->redirect(null, 'article', 'view', $article->getBestArticleId());
}
}
if (PKPString::regexp_match_get('/^[Pp][Pp]?[.]?[ ]?(\\d+)[ ]?-[ ]?([Pp][Pp]?[.]?[ ]?)?(\\d+)$/', $article->getPages(), $matches)) {
$matchedPageFrom = $matches[1];
$matchedPageTo = $matches[3];
if ($page >= $matchedPageFrom && ($page < $matchedPageTo || $page == $matchedPageTo && ($matchedPageFrom = $matchedPageTo))) {
$request->redirect(null, 'article', 'view', $article->getBestArticleId());
}
}
unset($article);
}
}
// Failure.
header("HTTP/1.0 500 Internal Server Error");
$templateMgr = TemplateManager::getManager($request);
AppLocale::requireComponents(LOCALE_COMPONENT_APP_COMMON);
$templateMgr->assign('message', 'plugins.gateways.resolver.errors.errorMessage');
$templateMgr->display('frontend/pages/message.tpl');
exit;
}
/**
* Fills the given citation object with
* meta-data retrieved from PubMed.
* @param $pmid string
* @return MetadataDescription
*/
function &_lookup($pmid)
{
$nullVar = null;
// Use eFetch to get XML metadata for the given PMID
$lookupParams = array('db' => 'pubmed', 'mode' => 'xml', 'tool' => 'pkp-wal', 'id' => $pmid);
if (!is_null($this->getEmail())) {
$lookupParams['email'] = $this->getEmail();
}
// Call the eFetch URL and get an XML result
if (is_null($resultDOM = $this->callWebService(PUBMED_WEBSERVICE_EFETCH, $lookupParams))) {
return $nullVar;
}
$articleTitleNodes = $resultDOM->getElementsByTagName('ArticleTitle');
$articleTitleFirstNode = $articleTitleNodes->item(0);
$medlineTaNodes = $resultDOM->getElementsByTagName('MedlineTA');
$medlineTaFirstNode = $medlineTaNodes->item(0);
$metadata = array('pub-id[@pub-id-type="pmid"]' => $pmid, 'article-title' => $articleTitleFirstNode->textContent, 'source' => $medlineTaFirstNode->textContent);
$volumeNodes = $resultDOM->getElementsByTagName('Volume');
$issueNodes = $resultDOM->getElementsByTagName('Issue');
if ($volumeNodes->length > 0) {
$volumeFirstNode = $volumeNodes->item(0);
}
$metadata['volume'] = $volumeFirstNode->textContent;
if ($issueNodes->length > 0) {
$issueFirstNode = $issueNodes->item(0);
}
$metadata['issue'] = $issueFirstNode->textContent;
// Get list of author full names
foreach ($resultDOM->getElementsByTagName("Author") as $authorNode) {
if (!isset($metadata['person-group[@person-group-type="author"]'])) {
$metadata['person-group[@person-group-type="author"]'] = array();
}
// Instantiate an NLM name description
$authorDescription = new MetadataDescription('lib.pkp.plugins.metadata.nlm30.schema.Nlm30NameSchema', ASSOC_TYPE_AUTHOR);
// Surname
$lastNameNodes = $authorNode->getElementsByTagName('LastName');
$lastNameFirstNode = $lastNameNodes->item(0);
$authorDescription->addStatement('surname', $lastNameFirstNode->textContent);
// Given names
$givenNamesString = '';
$firstNameNodes = $authorNode->getElementsByTagName('FirstName');
if ($firstNameNodes->length > 0) {
$firstNameFirstNode = $firstNameNodes->item(0);
$givenNamesString = $firstNameFirstNode->textContent;
} else {
$foreNameNodes = $authorNode->getElementsByTagName('ForeName');
if ($foreNameNodes->length > 0) {
$foreNameFirstNode = $foreNameNodes->item(0);
$givenNamesString = $foreNameFirstNode->textContent;
}
}
if (!empty($givenNamesString)) {
foreach (explode(' ', $givenNamesString) as $givenName) {
$authorDescription->addStatement('given-names', PKPString::trimPunctuation($givenName));
}
}
// Suffix
$suffixNodes = $authorNode->getElementsByTagName('Suffix');
if ($suffixNodes->length > 0) {
$suffixFirstNode = $suffixNodes->item(0);
$authorDescription->addStatement('suffix', $suffixFirstNode->textContent);
}
// Include collective names
// FIXME: This corresponds to an NLM-citation <collab> tag and should be part of the Metadata implementation
/*if ($resultDOM->getElementsByTagName("CollectiveName")->length > 0 && $authorNode->getElementsByTagName("CollectiveName")->item(0)->textContent != '') {
}*/
$metadata['person-group[@person-group-type="author"]'][] =& $authorDescription;
unset($authorDescription);
}
// Extract pagination
$medlinePgnNodes = $resultDOM->getElementsByTagName('MedlinePgn');
$medlinePgnFirstNode = $medlinePgnNodes->item(0);
if (PKPString::regexp_match_get("/^[:p\\.\\s]*(?P<fpage>[Ee]?\\d+)(-(?P<lpage>\\d+))?/", $medlinePgnFirstNode->textContent, $pages)) {
$fPage = (int) $pages['fpage'];
$metadata['fpage'] = $fPage;
if (!empty($pages['lpage'])) {
$lPage = (int) $pages['lpage'];
// Deal with shortcuts like '382-7'
if ($lPage < $fPage) {
$lPage = (int) (PKPString::substr($pages['fpage'], 0, -PKPString::strlen($pages['lpage'])) . $pages['lpage']);
}
$metadata['lpage'] = $lPage;
}
}
// Get publication date (can be in several places in PubMed).
$dateNode = null;
$articleDateNodes = $resultDOM->getElementsByTagName('ArticleDate');
if ($articleDateNodes->length > 0) {
$dateNode = $articleDateNodes->item(0);
} else {
$pubDateNodes = $resultDOM->getElementsByTagName('PubDate');
if ($pubDateNodes->length > 0) {
$dateNode = $pubDateNodes->item(0);
}
}
// Retrieve the data parts and assemble date.
if (!is_null($dateNode)) {
$publicationDate = '';
$requiresNormalization = false;
foreach (array('Year' => 4, 'Month' => 2, 'Day' => 2) as $dateElement => $padding) {
$dateElementNodes = $dateNode->getElementsByTagName($dateElement);
if ($dateElementNodes->length > 0) {
if (!empty($publicationDate)) {
$publicationDate .= '-';
}
$dateElementFirstNode = $dateElementNodes->item(0);
$datePart = str_pad($dateElementFirstNode->textContent, $padding, '0', STR_PAD_LEFT);
if (!is_numeric($datePart)) {
$requiresNormalization = true;
}
$publicationDate .= $datePart;
} else {
break;
}
}
// Normalize the date to NLM standard if necessary.
if ($requiresNormalization) {
$dateFilter = new DateStringNormalizerFilter();
$publicationDate = $dateFilter->execute($publicationDate);
}
if (!empty($publicationDate)) {
$metadata['date'] = $publicationDate;
}
}
// Get publication type
$publicationTypeNodes = $resultDOM->getElementsByTagName('PublicationType');
if ($publicationTypeNodes->length > 0) {
foreach ($publicationTypeNodes as $publicationType) {
// The vast majority of items on PubMed are articles so catch these...
if (PKPString::strpos(PKPString::strtolower($publicationType->textContent), 'article') !== false) {
$metadata['[@publication-type]'] = NLM30_PUBLICATION_TYPE_JOURNAL;
break;
}
}
}
// Get DOI if it exists
$articleIdNodes = $resultDOM->getElementsByTagName('ArticleId');
foreach ($articleIdNodes as $idNode) {
if ($idNode->getAttribute('IdType') == 'doi') {
$metadata['pub-id[@pub-id-type="doi"]'] = $idNode->textContent;
}
}
// Use eLink utility to find fulltext links
$lookupParams = array('dbfrom' => 'pubmed', 'cmd' => 'llinks', 'tool' => 'pkp-wal', 'id' => $pmid);
if (!is_null($resultDOM = $this->callWebService(PUBMED_WEBSERVICE_ELINK, $lookupParams))) {
// Get a list of possible links
foreach ($resultDOM->getElementsByTagName("ObjUrl") as $linkOut) {
$attributes = '';
foreach ($linkOut->getElementsByTagName("Attribute") as $attribute) {
$attributes .= PKPString::strtolower($attribute->textContent) . ' / ';
}
// Only add links to open access resources
if (PKPString::strpos($attributes, "subscription") === false && PKPString::strpos($attributes, "membership") === false && PKPString::strpos($attributes, "fee") === false && $attributes != "") {
$urlNodes = $linkOut->getElementsByTagName('Url');
$urlFirstNode = $urlNodes->item(0);
$links[] = $urlFirstNode->textContent;
}
}
// Take the first link if we have any left (presumably pubmed returns them in preferential order)
if (isset($links[0])) {
$metadata['uri'] = $links[0];
}
}
return $this->getNlm30CitationDescriptionFromMetadataArray($metadata);
}
/**
* Convert a Windows path to a cygwin path.
* @param string $path Windows path
* @return string Cygwin path.
*/
function cygwinConversion($path)
{
$path = str_replace('\\', '/', $path);
$matches = null;
PKPString::regexp_match_get('/^([A-Z]):/i', $path, $matches);
if (isset($matches[1]) && strlen($matches[1]) === 1) {
$path = PKPString::regexp_replace('/^[A-Z]:/i', '/cygdrive/' . strtolower($matches[1]), $path);
}
return $path;
}
