/**
* Add a file to the search index.
* @param $articleId int
* @param $type int
* @param $fileId int
*/
function updateFileIndex($articleId, $type, $fileId)
{
import('classes.file.ArticleFileManager');
$fileMgr = new ArticleFileManager($articleId);
$file =& $fileMgr->getFile($fileId);
if (isset($file)) {
$parser =& SearchFileParser::fromFile($file);
}
if (isset($parser)) {
if ($parser->open()) {
$searchDao =& DAORegistry::getDAO('ArticleSearchDAO');
$objectId = $searchDao->insertObject($articleId, $type, $fileId);
$position = 0;
while (($text = $parser->read()) !== false) {
ArticleSearchIndex::indexObjectKeywords($objectId, $text, $position);
}
$parser->close();
}
}
}
/**
* Add a file to the search index.
* @param $monographId int
* @param $type int
* @param $fileId int
*/
function updateFileIndex($monographId, $type, $fileId)
{
$submissionFileDao =& DAORegistry::getDAO('SubmissionFileDAO');
/* @var $submissionFileDao SubmissionFileDAO */
$file =& $submissionFileDao->getLatestRevision($fileId);
if (isset($file)) {
$parser =& SearchFileParser::fromFile($file);
}
if (isset($parser)) {
if ($parser->open()) {
$searchDao =& DAORegistry::getDAO('MonographSearchDAO');
$objectId = $searchDao->insertObject($monographId, $type, $fileId);
$position = 0;
while (($text = $parser->read()) !== false) {
MonographSearchIndex::indexObjectKeywords($objectId, $text, $position);
}
$parser->close();
}
}
}
/**
* @see OAIMetadataFormat#toXml
* TODO:
* <copyright-holder>
* In Isabelle's mapping document:
* Article order in the issue's Table of Contents
*/
function toXml(&$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 $response;
}
function &generateArticleDom(&$doc, &$journal, &$issue, &$article, &$galley)
{
$unavailableString = Locale::translate('plugins.importexport.erudit.unavailable');
$root =& XMLCustomWriter::createElement($doc, 'article');
XMLCustomWriter::setAttribute($root, 'idprop', $journal->getJournalId() . '-' . $issue->getIssueId() . '-' . $article->getArticleId() . '-' . $galley->getGalleyId(), false);
XMLCustomWriter::setAttribute($root, 'arttype', 'article');
$lang = $article->getLanguage();
XMLCustomWriter::setAttribute($root, 'lang', isset($lang) ? $lang : 'en');
XMLCustomWriter::setAttribute($root, 'processing', 'cart');
/* --- admin --- */
$adminNode =& XMLCustomWriter::createElement($doc, 'admin');
XMLCustomWriter::appendChild($root, $adminNode);
/* --- articleinfo --- */
$articleInfoNode =& XMLCustomWriter::createElement($doc, 'articleinfo');
XMLCustomWriter::appendChild($adminNode, $articleInfoNode);
// The first public ID should be a full URL to the article.
$urlIdNode =& XMLCustomWriter::createChildWithText($doc, $articleInfoNode, 'idpublic', Request::url($journal->getPath(), 'article', 'view', array($article->getArticleId(), $galley->getGalleyId())));
XMLCustomWriter::setAttribute($urlIdNode, 'scheme', 'sici');
/* --- journal --- */
$journalNode =& XMLCustomWriter::createElement($doc, 'journal');
XMLCustomWriter::appendChild($adminNode, $journalNode);
XMLCustomWriter::setAttribute($journalNode, 'id', 'ojs-' . $journal->getPath());
XMLCustomWriter::createChildWithText($doc, $journalNode, 'jtitle', $journal->getJournalTitle());
XMLCustomWriter::createChildWithText($doc, $journalNode, 'jshorttitle', $journal->getLocalizedSetting('initials'), false);
if (!($printIssn = $journal->getSetting('printIssn'))) {
$printIssn = $unavailableString;
}
XMLCustomWriter::createChildWithText($doc, $journalNode, 'idissn', $printIssn);
if (!($onlineIssn = $journal->getSetting('onlineIssn'))) {
$onlineIssn = $unavailableString;
}
XMLCustomWriter::createChildWithText($doc, $journalNode, 'iddigissn', $onlineIssn);
/* --- issue --- */
$issueNode =& XMLCustomWriter::createElement($doc, 'issue');
XMLCustomWriter::appendChild($adminNode, $issueNode);
XMLCustomWriter::setAttribute($issueNode, 'id', 'ojs-' . $issue->getBestIssueId());
XMLCustomWriter::createChildWithText($doc, $issueNode, 'volume', $issue->getVolume(), false);
XMLCustomWriter::createChildWithText($doc, $issueNode, 'issueno', $issue->getNumber(), false);
$pubNode =& XMLCustomWriter::createElement($doc, 'pub');
XMLCustomWriter::appendChild($issueNode, $pubNode);
XMLCustomWriter::createChildWithText($doc, $pubNode, 'year', $issue->getYear());
$digPubNode =& XMLCustomWriter::createElement($doc, 'digpub');
XMLCustomWriter::appendChild($issueNode, $digPubNode);
XMLCustomWriter::createChildWithText($doc, $digPubNode, 'date', EruditExportDom::formatDate($issue->getDatePublished()));
/* --- Publisher & DTD --- */
$publisherInstitution =& $journal->getSetting('publisherInstitution');
$publisherNode =& XMLCustomWriter::createElement($doc, 'publisher');
XMLCustomWriter::setAttribute($publisherNode, 'id', 'ojs-' . $journal->getJournalId() . '-' . $issue->getIssueId() . '-' . $article->getArticleId());
XMLCustomWriter::appendChild($adminNode, $publisherNode);
$publisherInstitution = $unavailableString;
if (empty($publisherInstitution)) {
$publisherInstitution = $unavailableString;
}
XMLCustomWriter::createChildWithText($doc, $publisherNode, 'orgname', $publisherInstitution);
$digprodNode =& XMLCustomWriter::createElement($doc, 'digprod');
XMLCustomWriter::createChildWithText($doc, $digprodNode, 'orgname', $publisherInstitution);
XMLCustomWriter::setAttribute($digprodNode, 'id', 'ojs-prod-' . $journal->getJournalId() . '-' . $issue->getIssueId() . '-' . $article->getArticleId());
XMLCustomWriter::appendChild($adminNode, $digprodNode);
$digdistNode =& XMLCustomWriter::createElement($doc, 'digdist');
XMLCustomWriter::createChildWithText($doc, $digdistNode, 'orgname', $publisherInstitution);
XMLCustomWriter::setAttribute($digdistNode, 'id', 'ojs-dist-' . $journal->getJournalId() . '-' . $issue->getIssueId() . '-' . $article->getArticleId());
XMLCustomWriter::appendChild($adminNode, $digdistNode);
$dtdNode =& XMLCustomWriter::createElement($doc, 'dtd');
XMLCustomWriter::appendChild($adminNode, $dtdNode);
XMLCustomWriter::setAttribute($dtdNode, 'name', 'Erudit Article');
XMLCustomWriter::setAttribute($dtdNode, 'version', '3.0.0');
/* --- copyright --- */
$copyright = $journal->getLocalizedSetting('copyrightNotice');
XMLCustomWriter::createChildWithText($doc, $adminNode, 'copyright', empty($copyright) ? $unavailableString : $copyright);
/* --- frontmatter --- */
$frontMatterNode =& XMLCustomWriter::createElement($doc, 'frontmatter');
XMLCustomWriter::appendChild($root, $frontMatterNode);
$titleGroupNode =& XMLCustomWriter::createElement($doc, 'titlegr');
XMLCustomWriter::appendChild($frontMatterNode, $titleGroupNode);
XMLCustomWriter::createChildWithText($doc, $titleGroupNode, 'title', strip_tags($article->getArticleTitle()));
/* --- authorgr --- */
$authorGroupNode =& XMLCustomWriter::createElement($doc, 'authorgr');
XMLCustomWriter::appendChild($frontMatterNode, $authorGroupNode);
$authorNum = 1;
foreach ($article->getAuthors() as $author) {
$authorNode =& XMLCustomWriter::createElement($doc, 'author');
XMLCustomWriter::appendChild($authorGroupNode, $authorNode);
XMLCustomWriter::setAttribute($authorNode, 'id', 'ojs-' . $journal->getJournalId() . '-' . $issue->getIssueId() . '-' . $article->getArticleId() . '-' . $galley->getGalleyId() . '-' . $authorNum);
$persNameNode =& XMLCustomWriter::createElement($doc, 'persname');
XMLCustomWriter::appendChild($authorNode, $persNameNode);
// Opatan Inc.
foreach ((array) $author->getFirstName(null) as $locale => $firstName) {
$firstName = strip_tags($firstName);
$firstNameNode =& XMLCustomWriter::createElement($doc, 'firstname');
XMLCustomWriter::setAttribute($firstNameNode, 'lang', $locale);
XMLCustomWriter::appendChild($persNameNode, $firstNameNode);
XMLCustomWriter::createChildWithText($doc, $firstNameNode, 'blocktext', $firstName);
unset($firstNameNode);
}
// Opatan Inc.
foreach ((array) $author->getMiddleName(null) as $locale => $middleName) {
$middleName = strip_tags($middleName);
$middleNameNode =& XMLCustomWriter::createElement($doc, 'middlename');
XMLCustomWriter::setAttribute($middleNameNode, 'lang', $locale);
XMLCustomWriter::appendChild($persNameNode, $middleNameNode);
XMLCustomWriter::createChildWithText($doc, $middleNameNode, 'blocktext', $middleName);
unset($middleNameNode);
}
// Opatan Inc.
foreach ((array) $author->getLastName(null) as $locale => $lastName) {
$lastName = strip_tags($lastName);
$lastNameNode =& XMLCustomWriter::createElement($doc, 'familyname');
XMLCustomWriter::setAttribute($lastNameNode, 'lang', $locale);
XMLCustomWriter::appendChild($persNameNode, $lastNameNode);
XMLCustomWriter::createChildWithText($doc, $lastNameNode, 'blocktext', $lastName);
unset($lastNameNode);
}
// Opatan Inc.
foreach ((array) $author->getAffiliation(null) as $locale => $affiliation) {
$affiliation = strip_tags($affiliation);
$affiliationNode =& XMLCustomWriter::createElement($doc, 'affiliation');
XMLCustomWriter::setAttribute($affiliationNode, 'lang', $locale);
XMLCustomWriter::appendChild($authorNode, $affiliationNode);
XMLCustomWriter::createChildWithText($doc, $affiliationNode, 'blocktext', $affiliation, false);
}
$authorNum++;
}
/* --- abstract and keywords --- */
foreach ((array) $article->getAbstract(null) as $locale => $abstract) {
$abstract = strip_tags($abstract);
$abstractNode =& XMLCustomWriter::createElement($doc, 'abstract');
XMLCustomWriter::setAttribute($abstractNode, 'lang', $locale);
XMLCustomWriter::appendChild($frontMatterNode, $abstractNode);
XMLCustomWriter::createChildWithText($doc, $abstractNode, 'blocktext', $abstract);
unset($abstractNode);
}
if ($keywords = $article->getArticleSubject()) {
$keywordGroupNode =& XMLCustomWriter::createElement($doc, 'keywordgr');
XMLCustomWriter::setAttribute($keywordGroupNode, 'lang', ($language = $article->getLanguage()) ? $language : 'en');
foreach (explode(';', $keywords) as $keyword) {
XMLCustomWriter::createChildWithText($doc, $keywordGroupNode, 'keyword', trim($keyword), false);
}
XMLCustomWriter::appendChild($frontMatterNode, $keywordGroupNode);
}
/* --- body --- */
$bodyNode =& XMLCustomWriter::createElement($doc, 'body');
XMLCustomWriter::appendChild($root, $bodyNode);
import('file.ArticleFileManager');
$articleFileManager =& new ArticleFileManager($article->getArticleId());
$file =& $articleFileManager->getFile($galley->getFileId());
$parser =& SearchFileParser::fromFile($file);
if (isset($parser)) {
if ($parser->open()) {
// File supports text indexing.
$textNode =& XMLCustomWriter::createElement($doc, 'text');
XMLCustomWriter::appendChild($bodyNode, $textNode);
while (($line = $parser->read()) !== false) {
$line = trim($line);
if ($line != '') {
XMLCustomWriter::createChildWithText($doc, $textNode, 'blocktext', $line, false);
}
}
$parser->close();
}
}
return $root;
}
function SearchHelperParser($type, $filePath)
{
parent::SearchFileParser($filePath);
$this->type = $type;
}
/**
* Signal to the indexing back-end that an article file changed.
*
* @see ArticleSearchIndex::articleMetadataChanged() above for more
* comments.
*
* @param $articleId int
* @param $type int
* @param $fileId int
*/
function articleFileChanged($articleId, $type, $fileId)
{
// Check whether a search plug-in jumps in.
$hookResult =& HookRegistry::call('ArticleSearchIndex::articleFileChanged', array($articleId, $type, $fileId));
// If no search plug-in is activated then fall back to the
// default database search implementation.
if ($hookResult === false || is_null($hookResult)) {
import('classes.file.ArticleFileManager');
$fileManager = new ArticleFileManager($articleId);
$file =& $fileManager->getFile($fileId);
if (isset($file)) {
$parser =& SearchFileParser::fromFile($file);
}
if (isset($parser)) {
if ($parser->open()) {
$searchDao =& DAORegistry::getDAO('ArticleSearchDAO');
$objectId = $searchDao->insertObject($articleId, $type, $fileId);
$position = 0;
while (($text = $parser->read()) !== false) {
$this->_indexObjectKeywords($objectId, $text, $position);
}
$parser->close();
}
}
}
}
/**
* Create a text parser for a file.
* @param $file [Article|Paper]File
* @return SearchFileParser
*/
function &fromFile(&$file)
{
$returner =& SearchFileParser::fromFileType($file->getFileType(), $file->getFilePath());
return $returner;
}
/**
* Signal to the indexing back-end that an article file changed.
*
* @see ArticleSearchIndex::articleMetadataChanged() above for more
* comments.
*
* @param $articleId int
* @param $type int
* @param $fileId int
*/
function submissionFileChanged($articleId, $type, $fileId)
{
// Check whether a search plug-in jumps in.
$hookResult = HookRegistry::call('ArticleSearchIndex::submissionFileChanged', array($articleId, $type, $fileId));
// If no search plug-in is activated then fall back to the
// default database search implementation.
if ($hookResult === false || is_null($hookResult)) {
$submissionFileDao = DAORegistry::getDAO('SubmissionFileDAO');
/* @var $submissionFileDao SubmissionFileDAO */
$file = $submissionFileDao->getLatestRevision($fileId);
if (isset($file)) {
$parser = SearchFileParser::fromFile($file);
}
if (isset($parser) && $parser->open()) {
$searchDao = DAORegistry::getDAO('ArticleSearchDAO');
$objectId = $searchDao->insertObject($articleId, $type, $fileId);
$position = 0;
while (($text = $parser->read()) !== false) {
self::_indexObjectKeywords($objectId, $text, $position);
}
$parser->close();
}
}
}
function __construct($type, $filePath)
{
parent::__construct($filePath);
$this->type = $type;
}
/**
* @see OAIMetadataFormat#toXml
*/
function toXml(&$record, $format = null)
{
$conference =& $record->getData('conference');
$schedConf =& $record->getData('schedConf');
$paper =& $record->getData('paper');
$track =& $record->getData('track');
$galleys =& $record->getData('galleys');
$paperId = $paper->getId();
$primaryLocale = $conference->getPrimaryLocale();
// If possible, use the paper presentation date for the paper date fields.
// Otherwise, use the date published (i.e. the date it was marked "completed"
// in the workflow).
if ($datePublished = $paper->getStartTime()) {
$datePublished = strtotime($datePublished);
} else {
$datePublished = strtotime($paper->getDatePublished());
}
$response = "<article\n" . "\txmlns=\"http://dtd.nlm.nih.gov/publishing/2.3\"\n" . "\txmlns:xlink=\"http://www.w3.org/1999/xlink\"\n" . "\txmlns:mml=\"http://www.w3.org/1998/Math/MathML\"\n" . "\txmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"\n" . "\txsi:schemaLocation=\"http://dtd.nlm.nih.gov/publishing/2.3\n" . "\thttp://dtd.nlm.nih.gov/publishing/2.3/xsd/journalpublishing.xsd\"\n" . (($s = $track->getLocalizedIdentifyType()) != '' ? "\tarticle-type=\"" . htmlspecialchars(Core::cleanVar($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 : $conference->getPath() . '-' . $schedConf->getPath())) . "</journal-id>\n" . "\t\t\t<journal-title>" . htmlspecialchars(Core::cleanVar($schedConf->getLocalizedName())) . "</journal-title>\n";
// Include translated scheduled conference titles
foreach ($schedConf->getTitle(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";
}
$response .= "\t\t</journal-meta>\n" . "\t\t<article-meta>\n" . "\t\t\t<article-id pub-id-type=\"other\">" . htmlspecialchars(Core::cleanVar($paper->getId())) . "</article-id>\n" . "\t\t\t<article-categories><subj-group subj-group-type=\"heading\"><subject>" . htmlspecialchars(Core::cleanVar($track->getLocalizedTitle())) . "</subject></subj-group></article-categories>\n" . "\t\t\t<title-group>\n" . "\t\t\t\t<article-title>" . htmlspecialchars(Core::cleanVar(strip_tags($paper->getLocalizedTitle()))) . "</article-title>\n";
// Include translated journal titles
foreach ($paper->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";
}
$response .= "\t\t\t</title-group>\n" . "\t\t\t<contrib-group>\n";
// Include authors
foreach ($paper->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 .= $this->getEditorialInfo($conference->getId());
$response .= "\t\t\t</contrib-group>\n" . "\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";
$response .= "\t\t\t<permissions>\n" . (($s = $conference->getLocalizedSetting('copyrightNotice')) != '' ? "\t\t\t\t<copyright-statement>" . htmlspecialchars(Core::cleanVar($s)) . "</copyright-statement>\n" : '') . "\t\t\t\t<copyright-year>" . strftime('%Y', $datePublished) . "</copyright-year>\n" . "\t\t\t</permissions>\n" . "\t\t\t<self-uri xlink:href=\"" . htmlspecialchars(Core::cleanVar(Request::url($conference->getPath(), $schedConf->getPath(), 'paper', 'view', $paper->getId()))) . "\" />\n";
// Include galley links
foreach ($paper->getGalleys() as $galley) {
$response .= "\t\t\t<self-uri content-type=\"" . htmlspecialchars(Core::cleanVar($galley->getFileType())) . "\" xlink:href=\"" . htmlspecialchars(Core::cleanVar(Request::url($conference->getPath(), $schedConf->getPath(), 'paper', 'view', array($paper->getId(), $galley->getId())))) . "\" />\n";
}
// Include abstract(s)
$abstract = htmlspecialchars(Core::cleanVar(strip_tags($paper->getLocalizedAbstract())));
if (!empty($abstract)) {
$abstract = "<p>{$abstract}</p>";
$response .= "\t\t\t<abstract xml:lang=\"" . strtoupper(substr($primaryLocale, 0, 2)) . "\">{$abstract}</abstract>\n";
}
if (is_array($paper->getAbstract(null))) {
foreach ($paper->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>";
$response .= "\t\t\t<abstract-trans xml:lang=\"" . strtoupper(substr($locale, 0, 2)) . "\">{$abstract}</abstract-trans>\n";
}
}
$subjects = array();
if (is_array($paper->getSubject(null))) {
foreach ($paper->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";
}
}
$locationCity = $schedConf->getSetting('locationCity');
$locationCountry = $schedConf->getSetting('locationCountry');
if (empty($locationCity) && empty($locationCountry)) {
$confLoc = '';
} elseif (empty($locationCity) && !empty($locationCountry)) {
$confLoc = $locationCountry;
} elseif (empty($locationCountry)) {
$confLoc = $locationCity;
} else {
$confLoc = "{$locationCity}, {$locationCountry}";
}
$response .= "\t\t\t<conference>\n" . "\t\t\t\t<conf-date>" . strftime('%Y-%m-%d', $schedConf->getSetting('startDate')) . "</conf-date>\n" . "\t\t\t\t<conf-name>" . htmlspecialchars(Core::cleanVar($schedConf->getLocalizedName())) . "</conf-name>\n" . "\t\t\t\t<conf-acronym>" . htmlspecialchars(Core::cleanVar($schedConf->getLocalizedAcronym())) . "</conf-acronym>\n" . (!empty($confLoc) ? "\t\t\t\t<conf-loc>" . htmlspecialchars(Core::cleanVar($confLoc)) . "</conf-loc>\n" : '') . "\t\t\t</conference>\n" . "\t\t</article-meta>\n" . "\t</front>\n";
// Include body text (for search indexing only)
import('classes.search.PaperSearchIndex');
$text = '';
// $galleys = $paper->getGalleys();
// Give precedence to HTML galleys, as they're quickest to parse
usort($galleys, create_function('$a, $b', 'return $a->isHtmlGalley()?-1:1;'));
// Determine any access limitations. If there are, do not
// provide the full-text.
import('classes.schedConf.SchedConfAction');
$mayViewProceedings = SchedConfAction::mayViewProceedings($schedConf);
if ($mayViewProceedings) {
foreach ($galleys as $galley) {
$parser =& SearchFileParser::fromFile($galley);
if ($parser && $parser->open()) {
while (($s = $parser->read()) !== false) {
$text .= $s;
}
$parser->close();
}
if ($galley->isHtmlGalley()) {
$text = strip_tags($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";
}
$response .= "</article>";
return $response;
}
