/**
* Split a string into a clean array of keywords
* @param $text string
* @param $allowWildcards boolean
* @return array of keywords
*/
static function filterKeywords($text, $allowWildcards = false)
{
$minLength = Config::getVar('search', 'min_word_length');
$stopwords = self::_loadStopwords();
// Join multiple lines into a single string
if (is_array($text)) {
$text = join("\n", $text);
}
$cleanText = Core::cleanVar($text);
// Remove punctuation
$cleanText = String::regexp_replace('/[!"\\#\\$%\'\\(\\)\\.\\?@\\[\\]\\^`\\{\\}~]/', '', $cleanText);
$cleanText = String::regexp_replace('/[\\+,:;&\\/<=>\\|\\\\]/', ' ', $cleanText);
$cleanText = String::regexp_replace('/[\\*]/', $allowWildcards ? '%' : ' ', $cleanText);
$cleanText = String::strtolower($cleanText);
// Split into words
$words = String::regexp_split('/\\s+/', $cleanText);
// FIXME Do not perform further filtering for some fields, e.g., author names?
// Remove stopwords
$keywords = array();
foreach ($words as $k) {
if (!isset($stopwords[$k]) && String::strlen($k) >= $minLength && !is_numeric($k)) {
$keywords[] = String::substr($k, 0, SEARCH_KEYWORD_MAX_LENGTH);
}
}
return $keywords;
}
/**
* Display user login form.
* Redirect to user index page if user is already validated.
*/
function index($args, &$request)
{
$this->validate();
$this->setupTemplate($request);
if (Validation::isLoggedIn()) {
$request->redirect(null, 'user');
}
if (Config::getVar('security', 'force_login_ssl') && $request->getProtocol() != 'https') {
// Force SSL connections for login
$request->redirectSSL();
}
$sessionManager =& SessionManager::getManager();
$session =& $sessionManager->getUserSession();
$templateMgr =& TemplateManager::getManager();
// If the user wasn't expecting a login page, i.e. if they're new to the
// site and want to submit a paper, it helps to explain why they need to
// register.
if ($request->getUserVar('loginMessage')) {
$templateMgr->assign('loginMessage', $request->getUserVar('loginMessage'));
}
$templateMgr->assign('username', $session->getSessionVar('username'));
$templateMgr->assign('remember', $request->getUserVar('remember'));
$templateMgr->assign('source', $request->getUserVar('source'));
$templateMgr->assign('showRemember', Config::getVar('general', 'session_lifetime') > 0);
// For force_login_ssl with base_url[...]: make sure SSL used for login form
$loginUrl = $this->_getLoginUrl($request);
if (Config::getVar('security', 'force_login_ssl')) {
$loginUrl = String::regexp_replace('/^http:/', 'https:', $loginUrl);
}
$templateMgr->assign('loginUrl', $loginUrl);
$templateMgr->display('user/login.tpl');
}
/**
* @see Filter::process()
* @param $citationString string
* @return MetadataDescription
*/
function &process($citationString)
{
$nullVar = null;
$queryParams = array('demo' => '3', 'textlines' => $citationString);
// Parscit web form - the result is (mal-formed) HTML
if (is_null($result = $this->callWebService(PARSCIT_WEBSERVICE, $queryParams, XSL_TRANSFORMER_DOCTYPE_STRING, 'POST'))) {
return $nullVar;
}
$result = html_entity_decode($result);
// Detect errors.
if (!String::regexp_match('/.*<algorithm[^>]+>.*<\\/algorithm>.*/s', $result)) {
$translationParams = array('filterName' => $this->getDisplayName());
$this->addError(Locale::translate('submission.citations.filter.webserviceResultTransformationError', $translationParams));
return $nullVar;
}
// Screen-scrape the tagged portion and turn it into XML.
$xmlResult = String::regexp_replace('/.*<algorithm[^>]+>(.*)<\\/algorithm>.*/s', '\\1', $result);
$xmlResult = String::regexp_replace('/&/', '&', $xmlResult);
// Transform the result into an array of meta-data.
if (is_null($metadata = $this->transformWebServiceResults($xmlResult, dirname(__FILE__) . DIRECTORY_SEPARATOR . 'parscit.xsl'))) {
return $nullVar;
}
// Extract a publisher from the place string if possible.
$metadata =& $this->fixPublisherNameAndLocation($metadata);
return $this->getNlm30CitationDescriptionFromMetadataArray($metadata);
}
/**
* @see Filter::process()
* @param $isbn string
* @return MetadataDescription a looked up citation description
* or null if the filter fails
*/
function &process($isbn)
{
$nullVar = null;
// Instantiate the web service request
$lookupParams = array('access_key' => $this->getApiKey(), 'index1' => 'isbn', 'results' => 'details,authors', 'value1' => $isbn);
// Call the web service
if (is_null($resultDOM =& $this->callWebService(ISBNDB_WEBSERVICE_URL, $lookupParams))) {
return $nullVar;
}
// Transform and pre-process the web service result
if (is_null($metadata =& $this->transformWebServiceResults($resultDOM, dirname(__FILE__) . DIRECTORY_SEPARATOR . 'isbndb.xsl'))) {
return $nullVar;
}
// Extract place and publisher from the combined entry.
$metadata['publisher-loc'] = String::trimPunctuation(String::regexp_replace('/^(.+):.*/', '\\1', $metadata['place-publisher']));
$metadata['publisher-name'] = String::trimPunctuation(String::regexp_replace('/.*:([^,]+),?.*/', '\\1', $metadata['place-publisher']));
unset($metadata['place-publisher']);
// Reformat the publication date
$metadata['date'] = String::regexp_replace('/^[^\\d{4}]+(\\d{4}).*/', '\\1', $metadata['date']);
// Clean non-numerics from ISBN
$metadata['isbn'] = String::regexp_replace('/[^\\dX]*/', '', $isbn);
// Set the publicationType
$metadata['[@publication-type]'] = NLM_PUBLICATION_TYPE_BOOK;
return $this->getNlmCitationDescriptionFromMetadataArray($metadata);
}
/**
* Format XML for single DC element.
* @param $propertyName string
* @param $value array
* @param $multilingual boolean optional
*/
function formatElement($propertyName, $values, $multilingual = false)
{
if (!is_array($values)) {
$values = array($values);
}
// Translate the property name to XML syntax.
$openingElement = str_replace(array('[@', ']'), array(' ', ''), $propertyName);
$closingElement = String::regexp_replace('/\\[@.*/', '', $propertyName);
// Create the actual XML entry.
$response = '';
foreach ($values as $key => $value) {
if ($multilingual) {
$key = str_replace('_', '-', $key);
assert(is_array($value));
foreach ($value as $subValue) {
if ($key == METADATA_DESCRIPTION_UNKNOWN_LOCALE) {
$response .= "\t<{$openingElement}>" . OAIUtils::prepOutput($subValue) . "</{$closingElement}>\n";
} else {
$response .= "\t<{$openingElement} xml:lang=\"{$key}\">" . OAIUtils::prepOutput($subValue) . "</{$closingElement}>\n";
}
}
} else {
assert(is_scalar($value));
$response .= "\t<{$openingElement}>" . OAIUtils::prepOutput($value) . "</{$closingElement}>\n";
}
}
return $response;
}
/**
* @see Filter::process()
* @param $citationDescription MetadataDescription
* @return MetadataDescription
*/
function &process(&$citationDescription)
{
$nullVar = null;
$searchParams = array('pid' => $this->getEmail(), 'noredirect' => 'true', 'format' => 'unixref');
$doi = $citationDescription->getStatement('pub-id[@pub-id-type="doi"]');
if (!empty($doi)) {
// Directly look up the DOI with OpenURL 0.1
$searchParams['id'] = 'doi:' . $doi;
} else {
// Use OpenURL meta-data to search for the entry
if (is_null($openUrlMetadata = $this->_prepareOpenUrlSearch($citationDescription))) {
return $nullVar;
}
$searchParams += $openUrlMetadata;
}
// Call the CrossRef web service
if (is_null($resultXml =& $this->callWebService(CROSSREF_WEBSERVICE_URL, $searchParams, XSL_TRANSFORMER_DOCTYPE_STRING))) {
return $nullVar;
}
// Remove default name spaces from XML as CrossRef doesn't
// set them reliably and element names are unique anyway.
$resultXml = String::regexp_replace('/ xmlns="[^"]+"/', '', $resultXml);
// Transform and process the web service result
if (is_null($metadata =& $this->transformWebServiceResults($resultXml, dirname(__FILE__) . DIRECTORY_SEPARATOR . 'crossref.xsl'))) {
return $nullVar;
}
return $this->addMetadataArrayToNlmCitationDescription($metadata, $citationDescription);
}
/**
* @see Filter::process()
* @param $input string
* @return mixed array
*/
function &process(&$input)
{
// The default implementation assumes that raw citations are
// separated with line endings.
// 1) Remove empty lines
$input = String::regexp_replace('/[\\r\\n]+/s', "\n", $input);
// 2) Break up at line endings
$output = explode("\n", $input);
// TODO: Implement more complex treatment, e.g. filtering of
// number strings at the beginning of each string, etc.
return $output;
}
/**
* Get the HTML contents for this block.
* @param $templateMgr object
* @param $request PKPRequest
* @return $string
*/
function getContents(&$templateMgr, $request = null)
{
if (!defined('SESSION_DISABLE_INIT')) {
$session =& $request->getSession();
$templateMgr->assign_by_ref('userSession', $session);
$templateMgr->assign('loggedInUsername', $session->getSessionVar('username'));
$loginUrl = $request->url(null, null, 'login', 'signIn');
if (Config::getVar('security', 'force_login_ssl')) {
$loginUrl = String::regexp_replace('/^http:/', 'https:', $loginUrl);
}
$templateMgr->assign('userBlockLoginUrl', $loginUrl);
}
return parent::getContents($templateMgr, $request);
}
function getContents(&$templateMgr)
{
if (!defined('SESSION_DISABLE_INIT')) {
$session =& Request::getSession();
$templateMgr->assign_by_ref('userSession', $session);
$templateMgr->assign('loggedInUsername', $session->getSessionVar('username'));
$loginUrl = Request::url(null, 'login', 'signIn');
$templateMgr->assign('paperPackageUpPlugin', "/index.php/mr2/PaperPackageUpPlugin/view/");
if (Config::getVar('security', 'force_login_ssl')) {
$loginUrl = String::regexp_replace('/^http:/', 'https:', $loginUrl);
}
$templateMgr->assign('userBlockLoginUrl', $loginUrl);
}
return parent::getContents($templateMgr);
}
/**
* @see Filter::process()
* @param $citationString string
* @return MetadataDescription
*/
function &process($citationString)
{
$nullVar = null;
$queryParams = array('textlines' => $citationString);
// Parscit web form - the result is (mal-formed) HTML
if (is_null($result = $this->callWebService(PARSCIT_WEBSERVICE, $queryParams, XSL_TRANSFORMER_DOCTYPE_STRING))) {
return $nullVar;
}
// Screen-scrape the tagged portion and turn it into XML
$xmlResult = String::regexp_replace('/.*<algorithm[^>]+>(.*)<\\/algorithm>.*/s', '\\1', html_entity_decode($result));
$xmlResult = String::regexp_replace('/&/', '&', $xmlResult);
// Transform the result into an array of meta-data
if (is_null($metadata = $this->transformWebServiceResults($xmlResult, dirname(__FILE__) . DIRECTORY_SEPARATOR . 'parscit.xsl'))) {
return $nullVar;
}
// Extract a publisher from the place string if possible
$metadata =& $this->fixPublisherNameAndLocation($metadata);
return $this->addMetadataArrayToNlmCitationDescription($metadata);
}
/**
* @see Filter::process()
* @param $input string
* @return mixed array
*/
function &process(&$input)
{
// The default implementation assumes that raw citations are
// separated with line endings.
// 1) Remove empty lines and normalize line endings.
$input = String::regexp_replace('/[\\r\\n]+/s', "\n", $input);
// 2) Remove trailing/leading line breaks.
$input = trim($input, "\n");
// 3) Break up at line endings.
if (empty($input)) {
$citations = array();
} else {
$citations = explode("\n", $input);
}
// 4) Remove numbers from the beginning of each citation.
foreach ($citations as $index => $citation) {
$citations[$index] = String::regexp_replace('/^\\s*[\\[#]?[0-9]+[.)\\]]?\\s*/', '', $citation);
}
return $citations;
}
function getContents(&$templateMgr)
{
if (!defined('SESSION_DISABLE_INIT')) {
$session =& Request::getSession();
$templateMgr->assign_by_ref('userSession', $session);
$templateMgr->assign('loggedInUsername', $session->getSessionVar('username'));
$loginUrl = Request::url(null, 'login', 'signIn');
// if the page is not ssl enabled, and force_login_ssl is set, this flag will present a link instead of the form
$forceSSL = false;
if (Config::getVar('security', 'force_login_ssl')) {
if (Request::getProtocol() != 'https') {
$loginUrl = Request::url(null, 'login');
$forceSSL = true;
}
$loginUrl = String::regexp_replace('/^http:/', 'https:', $loginUrl);
}
$templateMgr->assign('userBlockLoginSSL', $forceSSL);
$templateMgr->assign('userBlockLoginUrl', $loginUrl);
}
return parent::getContents($templateMgr);
}
/**
* @see PubIdPlugin::getPubId()
*/
function getPubId($pubObject, $preview = false)
{
// Determine the type of the publishing object.
$pubObjectType = $this->getPubObjectType($pubObject);
// Initialize variables for publication objects.
$publicationFormat = $pubObjectType == 'PublicationFormat' ? $pubObject : null;
$monograph = $pubObjectType == 'Monograph' ? $pubObject : null;
// Get the press id of the object.
if (in_array($pubObjectType, array('PublicationFormat', 'Monograph'))) {
$pressId = $pubObject->getContextId();
} else {
return null;
}
$press = $this->_getPress($pressId);
if (!$press) {
return null;
}
$pressId = $press->getId();
// If we already have an assigned DOI, use it.
$storedDOI = $pubObject->getStoredPubId('doi');
if ($storedDOI) {
return $storedDOI;
}
// Retrieve the DOI prefix.
$doiPrefix = $this->getSetting($pressId, 'doiPrefix');
if (empty($doiPrefix)) {
return null;
}
// Generate the DOI suffix.
$doiSuffixGenerationStrategy = $this->getSetting($pressId, 'doiSuffix');
switch ($doiSuffixGenerationStrategy) {
case 'customId':
$doiSuffix = $pubObject->getData('doiSuffix');
break;
case 'pattern':
$doiSuffix = $this->getSetting($pressId, "doi{$pubObjectType}SuffixPattern");
// %p - press initials
$doiSuffix = String::regexp_replace('/%p/', String::strtolower($press->getPath()), $doiSuffix);
if ($publicationFormat) {
// %m - monograph id, %f - publication format id
$doiSuffix = String::regexp_replace('/%m/', $publicationFormat->getMonographId(), $doiSuffix);
$doiSuffix = String::regexp_replace('/%f/', $publicationFormat->getId(), $doiSuffix);
}
if ($monograph) {
// %m - monograph id
$doiSuffix = String::regexp_replace('/%m/', $monograph->getId(), $doiSuffix);
}
break;
default:
$doiSuffix = String::strtolower($press->getPath());
if ($publicationFormat) {
$doiSuffix .= '.' . $publicationFormat->getMonographId();
$doiSuffix .= '.' . $publicationFormat->getId();
}
if ($monograph) {
$doiSuffix .= '.' . $monograph->getId();
}
}
if (empty($doiSuffix)) {
return null;
}
// Join prefix and suffix.
$doi = $doiPrefix . '/' . $doiSuffix;
if (!$preview) {
// Save the generated DOI.
$this->setStoredPubId($pubObject, $pubObjectType, $doi);
}
return $doi;
}
/**
* 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 && String::regexp_match_get('/^(' . $personRegex['title'] . ')/i', $personString, $results)) {
$suffixString = trim($results[1], ',:; ');
$personString = String::regexp_replace('/^(' . $personRegex['title'] . ')/i', '', $personString);
}
if ($degrees && String::regexp_match_get('/(' . $personRegex['degrees'] . ')$/i', $personString, $results)) {
$degreesArray = explode(',', trim($results[1], ','));
foreach ($degreesArray as $key => $degree) {
$degreesArray[$key] = String::trimPunctuation($degree);
}
$suffixString .= ' - ' . implode('; ', $degreesArray);
$personString = String::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 = String::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 = String::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; $initialNum < String::strlen($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 $personDescription;
}
/**
* @see Form::fetch()
*/
function fetch($request)
{
$templateMgr = TemplateManager::getManager($request);
// The form description depends on the current state
// of the selection process: do we select a filter template
// or configure the settings of a selected template?
$filter =& $this->getFilter();
if (is_a($filter, 'Filter')) {
$displayName = $filter->getDisplayName();
$templateMgr->assign('filterDisplayName', $displayName);
if (count($filter->getSettings())) {
// We need a filter specific translation key so that we
// can explain the filter's configuration options.
// We use the display name to generate such a key as this
// is probably easiest for translators to understand.
// This also has the advantage that we can explain
// composite filters individually.
// FIXME: When we start to translate display names then
// please make sure that you use the en-US key for this
// processing. Alternatively we might want to introduce
// an alphanumeric "filter key" to the filters table.
$filterKey = String::regexp_replace('/[^a-zA-Z0-9]/', '', $displayName);
$filterKey = strtolower(substr($filterKey, 0, 1)) . substr($filterKey, 1);
$formDescriptionKey = $this->getDescription() . '.' . $filterKey;
} else {
$formDescriptionKey = $this->getDescription() . 'Confirm';
}
} else {
$templateMgr->assign('filterDisplayName', '');
$formDescriptionKey = $this->getDescription() . 'Template';
}
$templateMgr->assign('formTitle', $this->getTitle());
$templateMgr->assign('formDescription', $formDescriptionKey);
return parent::fetch($request);
}
/**
* @see PubIdPlugin::getPubId()
*/
function getPubId($pubObject, $preview = false)
{
$urn = $pubObject->getStoredPubId($this->getPubIdType());
if (!$urn) {
// Determine the type of the publishing object
$pubObjectType = $this->getPubObjectType($pubObject);
// Initialize variables for publication objects
$issue = $pubObjectType == 'Issue' ? $pubObject : null;
$article = $pubObjectType == 'Article' ? $pubObject : null;
$galley = $pubObjectType == 'Galley' ? $pubObject : null;
// Get the journal id of the object
if (in_array($pubObjectType, array('Issue', 'Article'))) {
$journalId = $pubObject->getJournalId();
} else {
// Retrieve the published article
assert(is_a($pubObject, 'SubmissionFile'));
$articleDao = DAORegistry::getDAO('ArticleDAO');
$article = $articleDao->getById($pubObject->getArticleId(), null, true);
if (!$article) {
return null;
}
// Now we can identify the journal
$journalId = $article->getJournalId();
}
// get the journal
$journal = $this->getJournal($journalId);
if (!$journal) {
return null;
}
$journalId = $journal->getId();
// Check whether URNs are enabled for the given object type
$urnEnabled = $this->getSetting($journalId, "enable{$pubObjectType}URN") == '1';
if (!$urnEnabled) {
return null;
}
// Retrieve the issue
if (!is_a($pubObject, 'Issue')) {
assert(!is_null($article));
$issueDao = DAORegistry::getDAO('IssueDAO');
$issue = $issueDao->getIssueByArticleId($article->getId(), $journal->getId(), true);
}
// Retrieve the URN prefix
$urnPrefix = $this->getSetting($journal->getId(), 'urnPrefix');
if (empty($urnPrefix)) {
return null;
}
// Generate the URN suffix
$urnSuffixSetting = $this->getSetting($journal->getId(), 'urnSuffix');
switch ($urnSuffixSetting) {
case 'publisherId':
$urnSuffix = (string) call_user_func_array(array($pubObject, "getBest{$pubObjectType}Id"), array(&$journal));
// When the suffix equals the object's ID then
// require an object-specific prefix to be sure that the suffix is unique
if ($pubObjectType != 'Article' && $urnSuffix === (string) $pubObject->getId()) {
$urnSuffix = strtolower_codesafe($pubObjectType[0]) . $urnSuffix;
}
if (!empty($urnSuffix)) {
$urn = $urnPrefix . $urnSuffix;
if ($this->getSetting($journal->getId(), 'checkNo')) {
$urn .= $this->_calculateCheckNo($urn);
}
}
break;
case 'customIdentifier':
$urnSuffix = $pubObject->getData('urnSuffix');
if (!empty($urnSuffix)) {
$urn = $urnPrefix . $urnSuffix;
}
break;
case 'pattern':
$urnSuffix = $this->getSetting($journal->getId(), "urn{$pubObjectType}SuffixPattern");
// %j - journal initials
$urnSuffix = String::regexp_replace('/%j/', String::strtolower($journal->getAcronym($journal->getPrimaryLocale())), $urnSuffix);
// %x - custom identifier
if ($pubObject->getStoredPubId('publisher-id')) {
$urnSuffix = String::regexp_replace('/%x/', $pubObject->getStoredPubId('publisher-id'), $urnSuffix);
}
if ($issue) {
// %v - volume number
$urnSuffix = String::regexp_replace('/%v/', $issue->getVolume(), $urnSuffix);
// %i - issue number
$urnSuffix = String::regexp_replace('/%i/', $issue->getNumber(), $urnSuffix);
// %Y - year
$urnSuffix = String::regexp_replace('/%Y/', $issue->getYear(), $urnSuffix);
}
if ($article) {
// %a - article id
$urnSuffix = String::regexp_replace('/%a/', $article->getId(), $urnSuffix);
// %p - page number
if ($article->getPages()) {
$urnSuffix = String::regexp_replace('/%p/', $article->getPages(), $urnSuffix);
}
}
if ($galley) {
// %g - galley id
$urnSuffix = String::regexp_replace('/%g/', $galley->getId(), $urnSuffix);
}
if (!empty($urnSuffix)) {
$urn = $urnPrefix . $urnSuffix;
if ($this->getSetting($journal->getId(), 'checkNo')) {
$urn .= $this->_calculateCheckNo($urn);
}
}
break;
default:
$urnSuffix = String::strtolower($journal->getAcronym($journal->getPrimaryLocale()));
if ($issue) {
$urnSuffix .= '.v' . $issue->getVolume() . 'i' . $issue->getNumber();
} else {
$urnSuffix .= '.v%vi%i';
}
if ($article) {
$urnSuffix .= '.' . $article->getId();
}
if ($galley) {
$urnSuffix .= '.g' . $galley->getId();
}
$urn = $urnPrefix . $urnSuffix;
if ($this->getSetting($journal->getId(), 'checkNo')) {
$urn .= $this->_calculateCheckNo($urn);
}
}
if ($urn && !$preview) {
$this->setStoredPubId($pubObject, $pubObjectType, $urn);
}
}
return $urn;
}
/**
* @see PubIdPlugin::getPubId()
*/
function getPubId(&$pubObject, $preview = false)
{
$doi = null;
if (!$this->isExcluded($pubObject)) {
// Determine the type of the publishing object.
$pubObjectType = $this->getPubObjectType($pubObject);
// Initialize variables for publication objects.
$issue = $pubObjectType == 'Issue' ? $pubObject : null;
$article = $pubObjectType == 'Article' ? $pubObject : null;
$galley = $pubObjectType == 'Galley' ? $pubObject : null;
$suppFile = $pubObjectType == 'SuppFile' ? $pubObject : null;
// Get the journal id of the object.
if (in_array($pubObjectType, array('Issue', 'Article'))) {
$journalId = $pubObject->getJournalId();
} else {
// Retrieve the published article.
assert(is_a($pubObject, 'ArticleFile'));
$articleDao =& DAORegistry::getDAO('PublishedArticleDAO');
/* @var $articleDao PublishedArticleDAO */
$article =& $articleDao->getPublishedArticleByArticleId($pubObject->getArticleId(), null, true);
if (!$article) {
return null;
}
// Now we can identify the journal.
$journalId = $article->getJournalId();
}
$journal =& $this->getJournal($journalId);
if (!$journal) {
return null;
}
$journalId = $journal->getId();
// Check whether DOIs are enabled for the given object type.
$doiEnabled = $this->getSetting($journalId, "enable{$pubObjectType}Doi") == '1';
if (!$doiEnabled) {
return null;
}
// If we already have an assigned DOI, use it.
$storedDOI = $pubObject->getStoredPubId('doi');
if ($storedDOI) {
return $storedDOI;
}
// Retrieve the issue.
if (!is_a($pubObject, 'Issue')) {
assert(!is_null($article));
$issueDao =& DAORegistry::getDAO('IssueDAO');
/* @var $issueDao IssueDAO */
$issue =& $issueDao->getIssueByArticleId($article->getId(), $journal->getId(), true);
}
if ($issue && $journalId != $issue->getJournalId()) {
return null;
}
// Retrieve the DOI prefix.
$doiPrefix = $this->getSetting($journalId, 'doiPrefix');
if (empty($doiPrefix)) {
return null;
}
// Generate the DOI suffix.
$doiSuffixGenerationStrategy = $this->getSetting($journalId, 'doiSuffix');
switch ($doiSuffixGenerationStrategy) {
case 'publisherId':
switch ($pubObjectType) {
case 'Issue':
$doiSuffix = (string) $pubObject->getBestIssueId($journal);
break;
case 'Article':
$doiSuffix = (string) $pubObject->getBestArticleId($journal);
break;
case 'Galley':
$doiSuffix = (string) $pubObject->getBestGalleyId($journal);
break;
case 'SuppFile':
$doiSuffix = (string) $pubObject->getBestSuppFileId($journal);
break;
default:
assert(false);
}
// When the suffix equals the object's ID then
// require an object-specific prefix to be sure that
// the suffix is unique.
if ($pubObjectType != 'Article' && $doiSuffix === (string) $pubObject->getId()) {
$doiSuffix = strtolower_codesafe($pubObjectType[0]) . $doiSuffix;
}
break;
case 'customId':
$doiSuffix = $pubObject->getData('doiSuffix');
break;
case 'pattern':
$doiSuffix = $this->getSetting($journalId, "doi{$pubObjectType}SuffixPattern");
// %j - journal initials
$doiSuffix = String::regexp_replace('/%j/', String::strtolower($journal->getLocalizedSetting('initials', $journal->getPrimaryLocale())), $doiSuffix);
// %x - custom identifier
if ($pubObject->getStoredPubId('publisher-id')) {
$doiSuffix = String::regexp_replace('/%x/', $pubObject->getStoredPubId('publisher-id'), $doiSuffix);
}
if ($issue) {
// %v - volume number
$doiSuffix = String::regexp_replace('/%v/', $issue->getVolume(), $doiSuffix);
// %i - issue number
$doiSuffix = String::regexp_replace('/%i/', $issue->getNumber(), $doiSuffix);
// %Y - year
$doiSuffix = String::regexp_replace('/%Y/', $issue->getYear(), $doiSuffix);
}
if ($article) {
// %a - article id
$doiSuffix = String::regexp_replace('/%a/', $article->getId(), $doiSuffix);
// %p - page number
if ($article->getPages()) {
$doiSuffix = String::regexp_replace('/%p/', $article->getPages(), $doiSuffix);
}
}
if ($galley) {
// %g - galley id
$doiSuffix = String::regexp_replace('/%g/', $galley->getId(), $doiSuffix);
}
if ($suppFile) {
// %s - supp file id
$doiSuffix = String::regexp_replace('/%s/', $suppFile->getId(), $doiSuffix);
}
break;
default:
$doiSuffix = String::strtolower($journal->getLocalizedSetting('initials', $journal->getPrimaryLocale()));
if ($issue) {
$doiSuffix .= '.v' . $issue->getVolume() . 'i' . $issue->getNumber();
} else {
$doiSuffix .= '.v%vi%i';
}
if ($article) {
$doiSuffix .= '.' . $article->getId();
}
if ($galley) {
$doiSuffix .= '.g' . $galley->getId();
}
if ($suppFile) {
$doiSuffix .= '.s' . $suppFile->getId();
}
}
if (empty($doiSuffix)) {
return null;
}
// Join prefix and suffix.
$doi = $doiPrefix . '/' . $doiSuffix;
if (!$preview) {
// Save the generated DOI.
$this->setStoredPubId($pubObject, $pubObjectType, $doi);
}
}
return $doi;
}
/**
* Convert limited HTML into a string.
* @param $html string
* @return string
*/
static function html2text($html)
{
$html = String::regexp_replace('/<[\\/]?p>/', "\n", $html);
$html = String::regexp_replace('/<li>/', '• ', $html);
$html = String::regexp_replace('/<\\/li>/', "\n", $html);
$html = String::regexp_replace('/<br[ ]?[\\/]?>/', "\n", $html);
$html = html_entity_decode(strip_tags($html), ENT_COMPAT, 'UTF-8');
return $html;
}
/**
* Suggest a username given the first and last names.
* @return string
*/
function suggestUsername($firstName, $lastName)
{
$initial = String::substr($firstName, 0, 1);
$suggestion = String::regexp_replace('/[^a-zA-Z0-9_-]/', '', String::strtolower($initial . $lastName));
$userDao =& DAORegistry::getDAO('UserDAO');
for ($i = ''; $userDao->userExistsByUsername($suggestion . $i); $i++) {
}
return $suggestion . $i;
}
function cleanFileName($var)
{
$var = String::regexp_replace('/[^\\w\\-\\.]/', '', $var);
if (!FilesHandler::fileNameFilter($var)) {
$var = time() . '';
}
return $var;
}
/**
* Create an identifier element with the object's DOI.
* @param $object Issue|PublishedArticle|ArticleGalley|SuppFile
* @param $relationType string One of the DATACITE_RELTYPE_* constants.
* @return XMLNode|DOMImplementation|null Can be null if the given ID Type
* has not been assigned to the given object.
*/
function &_relatedIdentifierElement(&$object, $relationType)
{
$id = $object->getPubId('doi');
if (empty($id)) {
return $nullVar;
$nullVar = null;
}
if ($this->getTestMode()) {
$id = String::regexp_replace('#^[^/]+/#', DATACITE_API_TESTPREFIX . '/', $id);
}
return $this->createElementWithText('relatedIdentifier', $id, array('relatedIdentifierType' => DATACITE_IDTYPE_DOI, 'relationType' => $relationType));
}
/**
* Generate a unique username for a user based on the user's name.
* @param $user ImportedUser the user to be modified by this function
*/
function generateUsername(&$user)
{
$userDao =& DAORegistry::getDAO('UserDAO');
$baseUsername = String::regexp_replace('/[^A-Z0-9]/i', '', $user->getLastName());
if (empty($baseUsername)) {
$baseUsername = String::regexp_replace('/[^A-Z0-9]/i', '', $user->getFirstName());
}
if (empty($username)) {
// Default username if we can't use the user's last or first name
$baseUsername = '******';
}
for ($username = $baseUsername, $i = 1; $userDao->userExistsByUsername($username, true); $i++) {
$username = $baseUsername . $i;
}
$user->setUsername($username);
}
/**
* @see FunctionalDoiExportTest::cleanXml()
*/
protected function cleanXml($xml)
{
// Fix URLs.
$xml = String::regexp_replace('#http://[^\\s]+/index.php/(test|index)#', 'http://example.com/index.php/test', $xml);
// Fix sent date.
$xml = String::regexp_replace('/<SentDate>[0-9]{12}<\\/SentDate>/', '<SentDate>201111082218</SentDate>', $xml);
// Fix version.
$xml = String::regexp_replace('/(<MessageNote>[^<]*)([0-9]\\.){4}(<\\/MessageNote>)/', '\\1x.x.x.x.\\3', $xml);
return parent::cleanXml($xml);
}
/**
* Translate query keywords.
* @param $searchPhrase string
* @return The translated search phrase.
*/
function _translateSearchPhrase($searchPhrase, $backwards = false)
{
static $queryKeywords;
if (is_null($queryKeywords)) {
// Query keywords.
$queryKeywords = array(String::strtoupper(__('search.operator.not')) => 'NOT', String::strtoupper(__('search.operator.and')) => 'AND', String::strtoupper(__('search.operator.or')) => 'OR');
}
if ($backwards) {
$translationTable = array_flip($queryKeywords);
} else {
$translationTable = $queryKeywords;
}
// Translate the search phrase.
foreach ($translationTable as $translateFrom => $translateTo) {
$searchPhrase = String::regexp_replace("/(^|\\s){$translateFrom}(\\s|\$)/i", "\\1{$translateTo}\\2", $searchPhrase);
}
return $searchPhrase;
}
/**
* Build the test file UI as understood by Selenium.
* @param string $fileFormat The test file format, e.g. "pdf".
* @return string
*/
private function _getTestFileUri($fileFormat)
{
$testFile = 'tests/functional/pages/editor/test-files/test-article.' . $fileFormat;
self::assertTrue(file_exists($testFile));
$testFile = realpath($testFile);
if (Core::isWindows()) {
$testFile = str_replace(DIRECTORY_SEPARATOR, '/', $testFile);
$testFile = String::regexp_replace('%^[A-Z]:/%', '/', $testFile);
}
$testFile = 'file://' . $testFile;
return $testFile;
}
/**
* Mark an object as "registered"
* by saving it's DOI to the object's
* "registeredDoi" setting.
* We prefix the setting with the plug-in's
* id so that we do not get name clashes
* when several DOI registration plug-ins
* are active at the same time.
* @parem $request Request
* @param $object Issue|PublishedArticle|ArticleGalley
* @parem $testPrefix string
*/
function markRegistered($request, $object, $testPrefix)
{
$registeredDoi = $object->getPubId('doi');
assert(!empty($registeredDoi));
if ($this->isTestMode($request)) {
$registeredDoi = String::regexp_replace('#^[^/]+/#', $testPrefix . '/', $registeredDoi);
}
$this->saveRegisteredDoi($object, $registeredDoi);
}
/**
* Retrieve the DOI of an object. The DOI will be
* patched if we are in test mode.
* @param $object Issue|PublishedArticle|ArticleGalley
* @return string
*/
function _getDoi(&$object)
{
$doi = $object->getPubId('doi');
if (!empty($doi) && $this->getTestMode()) {
$doi = String::regexp_replace('#^[^/]+/#', MEDRA_WS_TESTPREFIX . '/', $doi);
}
return $doi;
}
function _cleanFileName($var)
{
$var = String::regexp_replace('/[^\\w\\-\\.]/', '', $var);
if (!$this->_fileNameFilter($var)) {
$var = time() . '';
}
return $var;
}
/**
* Override the built-in smarty truncate modifier to support mbstring and HTML tags
* text properly, if possible.
*/
function smartyTruncate($string, $length = 80, $etc = '...', $break_words = false, $middle = false, $skip_tags = true)
{
if ($length == 0) {
return '';
}
if (String::strlen($string) > $length) {
$originalLength = String::strlen($string);
if ($skip_tags) {
if ($middle) {
$tagsReverse = array();
$this->_removeTags($string, $tagsReverse, true, $length);
}
$tags = array();
$string = $this->_removeTags($string, $tags, false, $length);
}
$length -= min($length, String::strlen($etc));
if (!$middle) {
if (!$break_words) {
$string = String::regexp_replace('/\\s+?(\\S+)?$/', '', String::substr($string, 0, $length + 1));
} else {
$string = String::substr($string, 0, $length + 1);
}
if ($skip_tags) {
$string = $this->_reinsertTags($string, $tags);
}
return $this->_closeTags($string) . $etc;
} else {
$firstHalf = String::substr($string, 0, $length / 2);
$secondHalf = String::substr($string, -$length / 2);
if ($break_words) {
if ($skip_tags) {
$firstHalf = $this->_reinsertTags($firstHalf, $tags);
$secondHalf = $this->reinsertTags($secondHalf, $tagsReverse, true);
return $this->_closeTags($firstHalf) . $etc . $this->_closeTags($secondHalf, true);
} else {
return $firstHalf . $etc . $secondHalf;
}
} else {
for ($i = $length / 2; $string[$i] != ' '; $i++) {
$firstHalf = String::substr($string, 0, $i + 1);
}
for ($i = $length / 2; String::substr($string, -$i, 1) != ' '; $i++) {
$secondHalf = String::substr($string, -$i - 1);
}
if ($skip_tags) {
$firstHalf = $this->_reinsertTags($firstHalf, $tags);
$secondHalf = $this->reinsertTags($secondHalf, $tagsReverse, strlen($string));
return $this->_closeTags($firstHalf) . $etc . $this->_closeTags($secondHalf, true);
} else {
return $firstHalf . $etc . $secondHalf;
}
}
}
} else {
return $string;
}
}
/**
* Check whether the given DOI appears on the object's page.
* @param $objectType string
* @param $expectedDoi string/boolean the expected DOI or false if
* no DOI should be present.
*/
private function checkDoiDisplay($objectType, $expectedDoi)
{
$url = $this->getUrl($objectType, 1);
$this->verifyAndOpen($url);
try {
if ($expectedDoi === false) {
$visibleElement = $this->pages[$objectType]['visible'];
if (strpos($objectType, 'citations') !== false) {
$this->assertNotText($visibleElement, 'doi');
} else {
$this->assertElementNotPresent($visibleElement);
}
foreach (array('DC-meta', 'Google-meta') as $doiMetaAttribute) {
if (isset($this->pages[$objectType][$doiMetaAttribute])) {
$doiMetaElement = String::regexp_replace('/@[^@]+$/', '', $this->pages[$objectType][$doiMetaAttribute]);
$this->assertElementNotPresent($doiMetaElement);
}
}
} else {
$expectedDoiPattern = "(^|.* ){$expectedDoi}(\$| .*)";
$doiText = $this->getText($this->pages[$objectType]['visible']);
if ($expectedDoi == '10.1234/t.v1i1.1') {
$fata = 'morgana';
}
$this->assertText($this->pages[$objectType]['visible'], $expectedDoiPattern);
foreach (array('DC-meta', 'Google-meta') as $doiMetaAttribute) {
if (isset($this->pages[$objectType][$doiMetaAttribute])) {
$this->assertAttribute($this->pages[$objectType][$doiMetaAttribute], $expectedDoiPattern);
}
}
}
} catch (Exception $e) {
throw $this->improveException($e, $objectType);
}
}
