/**
* Convert to Ebsco individual record syntax
*
* @param string $id
* @return Url
*/
public function getRecordUrl($id)
{
if (!strstr($id, "-")) {
throw new \Exception("could not find record");
}
// database and id come in on same value, so split 'em
$database = Parser::removeRight($id, "-");
$id = Parser::removeLeft($id, "-");
// get results
$query = "AN {$id}";
// construct url
$url = $this->host . '/Search?' . 'prof=' . $this->username . '&pwd=' . $this->password . '&authType=&ipprof=' . '&query=' . urlencode($query) . "&db={$database}" . '&startrec=1&numrec=1&format=detailed';
return new Url($url);
}
/**
* Convert to EDS individual record syntax
*
* @param string $id
* @return Url
*/
public function getRecordUrl($id)
{
if ($id == "") {
throw new \DomainException('No record ID supplied');
}
$database = Parser::removeRight($id, "-");
$id = Parser::removeLeft($id, "-");
// build request
$url = $this->server . 'retrieve?';
$url .= 'dbid=' . $database;
$url .= '&an=' . urlencode($id);
$url .= '&includefacets=n';
return new Url($url, $this->headers);
}
/**
* Return an individual record
*
* @param string record identifier
* @return ResultSet
*/
protected function doGetRecord($id)
{
$results = new Search\ResultSet($this->config);
// get the record from the database
$record = $this->datamap->getRecordByID($id);
// no record found?
if ($record == null) {
$results->total = 0;
return $results;
}
// got one
$results->total = 1;
$result = $this->createSearchResult($record);
// corrupted record, look out
if ($result->corrupted == true) {
$fixed = false;
$data = $record->marc;
// go back to the original search engine and fetch it again
$class_name = 'Application\\Model\\' . ucfirst($result->source) . '\\Engine';
if (class_exists($class_name)) {
try {
$engine = new $class_name();
$new_results = $engine->getRecord($result->original_id);
if ($new_results->total > 0) {
$result = $new_results->getRecord(0);
$fixed = true;
}
} catch (NotFoundException $e) {
if (strstr($data, 'Xerxes_TransRecord')) {
$data = '<?xml version="1.0"?>' . Parser::removeLeft($data, '<?xml version="1.0"?>');
$data = Parser::removeRight($data, '</xerxes_record>') . '</xerxes_record>';
$xerxes_record = new Xerxes\Record();
$xerxes_record->loadXML($data);
$record->xerxes_record = $xerxes_record;
// recreate the result, since we need the openurl and such
$result = $this->createSearchResult($record);
$fixed = true;
}
}
} elseif (strstr($data, 'Xerxes_MetalibRecord')) {
$data = '<?xml' . Parser::removeLeft($data, '<?xml');
$data = Parser::removeRight($data, '</x_server_response>') . '</x_server_response>';
$xerxes_record = new \Xerxes_MetalibRecord();
$xerxes_record->loadXML($data);
$record->xerxes_record = $xerxes_record;
// recreate the result, since we need the openurl and such
$result = $this->createSearchResult($record);
$fixed = true;
}
if ($fixed == false) {
throw new \Exception('Sorry, this record has been corrupted');
}
}
// if a catalog record, fetch holdings
if ($record->xerxes_record instanceof Solr\Record) {
try {
$engine = new Solr\Engine();
$solr_results = $engine->getRecord($result->original_id);
$holdings = $solr_results->getRecord(0)->getHoldings();
$result->setHoldings($holdings);
} catch (\Exception $e) {
trigger_error('saved records holdings lookup: ' . $e->getMessage(), E_USER_WARNING);
}
}
$results->addResult($result);
return $results;
}
public function map()
{
// score
$this->score = (string) $this->document->documentElement->getAttribute("RANK");
// record data
$record = $this->document->documentElement->getElementsByTagName("record")->item(0);
// print $this->document->saveXML();
$control = $this->getElement($record, "control");
$display = $this->getElement($record, "display");
$links = $this->getElement($record, "links");
$search = $this->getElement($record, "search");
$sort = $this->getElement($record, "sort");
$addata = $this->getElement($record, "addata");
$facets = $this->getElement($record, "facets");
$sourceid = "";
if ($control != null) {
$sourceid = $this->getElementValue($control, "sourceid");
}
if ($display != null) {
// database name
$this->database_name = $this->getElementValue($display, "source");
$this->database_name = strip_tags($this->database_name);
// journal
$this->journal = $this->getElementValue($display, "ispartof");
// snippet
$this->snippet = $this->getElementValue($display, "snippet");
$this->snippet = strip_tags($this->snippet);
// description
$this->abstract = $this->getElementValue($display, "description");
$this->abstract = strip_tags($this->abstract);
// language
$this->language = $this->getElementValue($display, "language");
// peer reviewed
$peer_reviewed = $this->getElementValue($display, 'lds50');
if ($peer_reviewed == 'peer_reviewed') {
$this->refereed = true;
}
}
if ($search != null) {
// record id
$this->record_id = $this->getElementValue($search, "recordid");
// year
$this->year = $this->getElementValue($search, "creationdate");
// issn
$issn = $this->getElementValue($search, "issn");
$issn = preg_replace('/\\D/', "", $issn);
array_push($this->issns, $issn);
// authors
$authors = $this->getElementValues($search, "creatorcontrib");
foreach ($authors as $author) {
array_push($this->authors, new Xerxes\Record\Author($author, null, "personal"));
}
// format
$format = $this->getElementValue($search, "rsrctype");
$this->format()->setInternalFormat($format);
// create a readable display
$format_display = self::createReadableLabel($format);
$this->format()->setPublicFormat($format_display);
}
// article data
if ($addata != null) {
$this->journal_title = $this->start_page = $this->getElementValue($addata, "jtitle");
$this->volume = $this->getElementValue($addata, "volume");
$this->issue = $this->getElementValue($addata, "issue");
$this->start_page = $this->getElementValue($addata, "spage");
$this->end_page = $this->getElementValue($addata, "epage");
// primo's own ris type
$ris_type = $this->getElementValue($addata, 'ristype');
if ($ris_type != "") {
$this->format()->setNormalizedFormat($ris_type);
}
// abstract
$abstract = $this->getElementValue($addata, "abstract");
if ($this->abstract == "") {
$this->abstract = strip_tags($abstract);
}
}
// subjects
if ($facets != null) {
$topics = $this->getElementValues($facets, "topic");
foreach ($topics as $topic) {
$subject_object = new Xerxes\Record\Subject();
$subject_object->value = $topic;
$subject_object->display = $topic;
array_push($this->subjects, $subject_object);
}
}
// title
if ($sort != null) {
$this->title = $this->getElementValue($sort, "title");
}
// direct link
$backlink = $this->getElementValue($links, "backlink");
if ($backlink != "") {
$backlink = Parser::removeLeft($backlink, '$$U');
$url = Parser::removeRight($backlink, '$$E');
$message = Parser::removeLeft($backlink, '$$E');
$link = new Link($url);
$link->setType(Link::ONLINE);
$this->links[] = $link;
}
// Gale title clean-up, because for some reason unknown to man they put weird
// notes and junk at the end of the title. so remove them here and add them to notes.
if (stristr($sourceid, "gale") || stristr($sourceid, "muse")) {
$gale_regex = '/\\(([^)]*)\\)/';
$matches = array();
if (preg_match_all($gale_regex, $this->title, $matches) != 0) {
$this->title = preg_replace($gale_regex, "", $this->title);
foreach ($matches[1] as $match) {
array_push($this->notes, $match);
}
}
if (strpos($this->abstract, 'the full-text of this article') !== false) {
$this->abstract = Parser::removeLeft($this->abstract, 'Abstract:');
}
}
}
/**
* Extracts the MARC data from the HTML response and converts it to MARC-XML
*
* @param string $marc marc data as string
* @return DOMDocument marc-xml document
*/
protected function extractMarc($response)
{
$xml = Parser::convertToDOMDocument("<record xmlns=\"http://www.loc.gov/MARC21/slim\" />");
$marc = "";
// marc data as text
$arrTags = array();
// array to hold each MARC tag
if (!stristr($response, "<pre>")) {
// didn't find a record
return $xml;
}
// parse out MARC data
$marc = Parser::removeLeft($response, "<pre>");
$marc = Parser::removeRight($marc, "</pre>");
// remove break-tabs for easier parsing
$marc = str_replace(" \n ", " ", $marc);
$marc = str_replace("\n ", " ", $marc);
$marc = trim($marc);
// assign the marc values to the array based on Unix LF as delimiter
$arrTags = explode("\n", $marc);
foreach ($arrTags as $strTag) {
// assign tag # and identifiers
$strTagNumber = substr($strTag, 0, 3);
$strId1 = substr($strTag, 4, 1);
$strId2 = substr($strTag, 5, 1);
// assign data and clean it up
$data = substr($strTag, 7);
// only convert all data to utf8 if told to do so,
// but always do it to the leader, since it has mangled chars
if ($this->convert_to_utf8 == true || $strTagNumber == "LEA") {
if (function_exists("mb_convert_encoding")) {
$data = mb_convert_encoding($data, "UTF-8");
} else {
$data = utf8_encode($data);
}
}
$data = Parser::escapeXml($data);
$data = trim($data);
if ($strTagNumber == "LEA") {
// leader
$objLeader = $xml->createElementNS($this->marc_ns, "leader", $data);
$xml->documentElement->appendChild($objLeader);
} elseif ($strTagNumber == "REC") {
// Pseudo-MARC "REC" data field to store the INNOPAC
// bibliographic record number in subfield a.
$objRecNum = $xml->createElementNS($this->marc_ns, "datafield");
$objRecNum->setAttribute("tag", "REC");
$objRecNum->setAttribute("ind1", ' ');
$objRecNum->setAttribute("ind2", ' ');
$objRecNumSub = $xml->createElementNS($this->marc_ns, "subfield", strtolower($data));
$objRecNumSub->setAttribute("code", 'a');
$objRecNum->appendChild($objRecNumSub);
$xml->documentElement->appendChild($objRecNum);
} elseif ((int) $strTagNumber <= 8) {
// control fields
$objControlField = $xml->createElementNS($this->marc_ns, "controlfield", $data);
$objControlField->setAttribute("tag", $strTagNumber);
$xml->documentElement->appendChild($objControlField);
} else {
// data fields
$objDataField = $xml->createElementNS($this->marc_ns, "datafield");
$objDataField->setAttribute("tag", $strTagNumber);
$objDataField->setAttribute("ind1", $strId1);
$objDataField->setAttribute("ind2", $strId2);
// if first character is not a pipe symbol, then this is the default |a subfield
// so make that explicit for the array
if (substr($data, 0, 1) != "|") {
$data = "|a " . $data;
}
// split the subfield data on the pipe and add them in using the first
// character after the delimiter as the subfield code
$arrSubFields = explode("|", $data);
foreach ($arrSubFields as $strSubField) {
if ($strSubField != "") {
$code = substr($strSubField, 0, 1);
$data = trim(substr($strSubField, 1));
// check for a url, in which case we need to ensure there are no spaces;
// which can happen on the wrap of the data in the marc display
if (strlen($data) > 4) {
if (substr($data, 0, 4) == "http") {
$data = str_replace(" ", "", $data);
}
}
$objSubField = $xml->createElementNS($this->marc_ns, "subfield", $data);
$objSubField->setAttribute("code", $code);
$objDataField->appendChild($objSubField);
}
}
$xml->documentElement->appendChild($objDataField);
}
}
return $xml;
}
/**
* Do the actual fetch of an individual record
*
* @param string record identifier
* @return Results
*/
protected function doGetRecord($id)
{
if (!strstr($id, "-")) {
throw new \Exception("could not find record");
}
// database and id come in on same value, so split 'em
$database = Parser::removeRight($id, "-");
$id = Parser::removeLeft($id, "-");
// get results
$results = $this->doSearch("AN {$id}", $database, 1, 1);
return $results;
}
/**
* Add links to the query object limits
*
* @param Query $query
*/
public function addQueryLinks(Query $query)
{
// we have to pass in the query object here rather than take
// the property above because adding the links doesn't seem
// to reflect back in the main object, even though they should
// be references, maybe because limit objects are in an array?
// add current query to query object itself
$params = $query->extractSearchParams();
$params['controller'] = $this->request->getParam('controller');
$params['action'] = 'search';
$params['source'] = $this->request->getParam('source');
$params['sort'] = $this->request->getParam('sort');
// url
$query->url = $this->request->url_for($params);
// query only
$query->url_query = Parser::removeLeft($query->url, '?');
// advanced search
$params['action'] = 'advanced';
$query->url_advanced = $this->request->url_for($params);
// search option links
$search = $this->registry->getConfig('search');
if ($search instanceof \SimpleXMLElement) {
$controller_map = $this->request->getControllerMap();
// combined results
$combined = $controller_map->getUrlAlias('combined');
$combined_id = $combined . '_' . $query->getHash();
if ($this->request->getSessionData($combined_id) != null) {
$params = $query->extractSearchParams();
$params['controller'] = $combined;
$params['action'] = "results";
$search->combined_url = $this->request->url_for($params);
}
// individual search options
foreach ($search->xpath("//option") as $option) {
$id = (string) $option["id"];
if ((string) $option["source"] != '') {
$id .= '_' . (string) $option["source"];
}
$id .= '_' . $this->query->getHash();
// format the number
// is this the current tab?
if ($this->request->getControllerName() == (string) $option["id"] && ($this->request->getParam('source') == (string) $option["source"] || (string) $option["source"] == '')) {
// mark as current
$option->addAttribute('current', "1");
}
// create url based on the search terms only!
$params = $query->extractSearchParams();
$params['controller'] = $controller_map->getUrlAlias((string) $option["id"]);
$params['action'] = "results";
$params['source'] = (string) $option["source"];
$params['sort'] = $this->request->getParam('sort');
// results url
$url = $this->request->url_for($params);
$option->addAttribute('url', $url);
// hits url
$params['action'] = 'hits';
$url = $this->request->url_for($params);
$option->addAttribute('url_hits', $url);
// cached search hit count?
foreach ($this->request->getAllSessionData() as $session_id => $session_value) {
// does this value in the cache have the save id as our tab?
$id = str_replace("_" . $query->getHash(), "", $session_id);
if ($id == (string) $option["id"]) {
// yup, so add it
$option->addAttribute('hits', Parser::number_format($session_value));
}
}
}
// header('Content-type: text/xml'); echo $search->asXML(); exit;
$this->registry->setConfig('search', $search);
}
// links to remove facets
foreach ($query->getLimits() as $limit) {
$params = $this->currentParams();
$value = $limit->value;
if ($limit->display != "") {
$value = $limit->display;
}
// urlencode here necessary to support the urlencode above on 'key' urls
$params = Parser::removeFromArray($params, urlencode($limit->param), $value);
$limit->remove_url = $this->request->url_for($params);
}
}
public function map()
{
// score
$this->score = (string) $this->document->documentElement->getAttribute("RANK");
// record data
$record = $this->document->documentElement->getElementsByTagName("record")->item(0);
// print $this->document->saveXML();
$display = $this->getElement($record, "display");
$search = $this->getElement($record, "search");
$sort = $this->getElement($record, "sort");
$addata = $this->getElement($record, "addata");
$facets = $this->getElement($record, "facets");
if ($display != null) {
// database name
$this->database_name = $this->getElementValue($display, "source");
$this->database_name = strip_tags($this->database_name);
// journal
$this->journal = $this->getElementValue($display, "ispartof");
// snippet
$this->snippet = $this->getElementValue($display, "snippet");
$this->snippet = strip_tags($this->snippet);
// description
$this->abstract = $this->getElementValue($display, "description");
$this->abstract = strip_tags($this->abstract);
// language
$this->language = $this->getElementValue($display, "language");
}
if ($search != null) {
// record id
$this->record_id = $this->getElementValue($search, "recordid");
// year
$this->year = $this->getElementValue($search, "creationdate");
// issn
$issn = $this->getElementValue($search, "issn");
$issn = preg_replace('/\\D/', "", $issn);
array_push($this->issns, $issn);
// authors
$authors = $this->getElementValues($search, "creatorcontrib");
foreach ($authors as $author) {
array_push($this->authors, new Xerxes\Record\Author($author, null, "personal"));
}
// format
// @todo: create map for internal
$this->format->setFormat($this->getElementValue($search, "rsrctype"));
}
// article data
if ($addata != null) {
$this->journal_title = $this->start_page = $this->getElementValue($addata, "jtitle");
$this->volume = $this->getElementValue($addata, "volume");
$this->issue = $this->getElementValue($addata, "issue");
$this->start_page = $this->getElementValue($addata, "spage");
$this->end_page = $this->getElementValue($addata, "epage");
// abstract
$abstract = $this->getElementValue($addata, "abstract");
if ($this->abstract == "") {
$this->abstract = strip_tags($abstract);
}
// gale madness
if (stristr($this->database_name, "gale")) {
if (strpos($this->abstract, 'the full-text of this article') !== false) {
$this->abstract = Parser::removeLeft($this->abstract, 'Abstract:');
}
}
}
// subjects
if ($facets != null) {
$topics = $this->getElementValues($facets, "topic");
foreach ($topics as $topic) {
$subject_object = new Xerxes\Record\Subject();
$subject_object->value = $topic;
$subject_object->display = $topic;
array_push($this->subjects, $subject_object);
}
}
// title
if ($sort != null) {
$this->title = $this->getElementValue($sort, "title");
}
// Gale title clean-up, because for some reason unknown to man they put weird
// notes and junk at the end of the title. so remove them here and add them to notes.
// @todo factor this out somehow? since this is very similar to Gale MetalibRecord code
if (stristr($this->database_name, "gale") || stristr($this->database_name, "muse")) {
$strGaleRegExp = '/\\(([^)]*)\\)/';
$arrMatches = array();
if (preg_match_all($strGaleRegExp, $this->title, $arrMatches) != 0) {
$this->title = preg_replace($strGaleRegExp, "", $this->title);
foreach ($arrMatches[1] as $strMatch) {
array_push($this->notes, $strMatch);
}
}
}
}