From 4bfd8094e7a2c85ee951945c3d193596d10fb3f0 Mon Sep 17 00:00:00 2001
From: Demian Katz <demian.katz@villanova.edu>
Date: Mon, 2 Jul 2012 14:41:25 -0400
Subject: [PATCH] Began porting WorldCat functionality (untested).
---
module/VuFind/src/VuFind/Connection/SRU.php | 333 +++++++++++++
.../VuFind/src/VuFind/Connection/WorldCat.php | 215 ++++++++
.../src/VuFind/Connection/WorldCatUtils.php | 466 ++++++++++++++++++
module/VuFind/src/VuFind/XSLT/Processor.php | 62 +++
4 files changed, 1076 insertions(+)
create mode 100644 module/VuFind/src/VuFind/Connection/SRU.php
create mode 100644 module/VuFind/src/VuFind/Connection/WorldCat.php
create mode 100644 module/VuFind/src/VuFind/Connection/WorldCatUtils.php
create mode 100644 module/VuFind/src/VuFind/XSLT/Processor.php
diff --git a/module/VuFind/src/VuFind/Connection/SRU.php b/module/VuFind/src/VuFind/Connection/SRU.php
new file mode 100644
index 00000000000..c8030f2f6e8
--- /dev/null
+++ b/module/VuFind/src/VuFind/Connection/SRU.php
@@ -0,0 +1,333 @@
+<?php
+/**
+ * SRU Search Interface
+ *
+ * PHP version 5
+ *
+ * Copyright (C) Andrew Nagy 2008.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2,
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Andrew S. Nagy <vufind-tech@lists.sourceforge.net>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/system_classes#searching Wiki
+ */
+namespace VuFind\Connection;
+use VuFind\Http\Client as HttpClient, VuFind\Log\Logger,
+ VuFind\XSLT\Processor as XSLTProcessor;
+
+/**
+ * SRU Search Interface
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Andrew S. Nagy <vufind-tech@lists.sourceforge.net>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/system_classes#searching Wiki
+ */
+class SRU
+{
+ /**
+ * Logger object for debug info (or false for no debugging).
+ */
+ protected $logger;
+
+ /**
+ * Whether to Serialize to a PHP Array or not.
+ * @var bool
+ */
+ protected $raw = false;
+
+ /**
+ * The HTTP_Request object used for REST transactions
+ * @var object HTTP_Request
+ */
+ protected $client;
+
+ /**
+ * The host to connect to
+ * @var string
+ */
+ protected $host;
+
+ /**
+ * The version to specify in the URL
+ * @var string
+ */
+ protected $sruVersion = '1.1';
+
+ /**
+ * Constructor
+ *
+ * Sets up the SOAP Client
+ *
+ * @param string $host The URL of the eXist Server
+ */
+ public function __construct($host)
+ {
+ // Initialize properties needed for HTTP connection:
+ $this->host = $host;
+ $this->client = new HttpClient();
+
+ // Don't waste time generating debug messages if nobody is listening:
+ $this->logger = Logger::debugNeeded() ? Logger::getInstance() : false;
+ }
+
+ /**
+ * Build Query string from search parameters
+ *
+ * @param array $search An array of search parameters
+ *
+ * @throws Exception
+ * @return array An array of query results
+ */
+ public function buildQuery($search)
+ {
+ foreach ($search as $params) {
+ if ($params['lookfor'] != '') {
+ $query = (isset($query)) ? $query . ' ' . $params['bool'] . ' ' : '';
+ switch ($params['field']) {
+ case 'title':
+ $query .= 'dc.title="' . $params['lookfor'] . '" OR ';
+ $query .= 'dc.title=' . $params['lookfor'];
+ break;
+ case 'id':
+ $query .= 'rec.id=' . $params['lookfor'];
+ break;
+ case 'author':
+ preg_match_all('/"[^"]*"|[^ ]+/', $params['lookfor'], $wordList);
+ $author = array();
+ foreach ($wordList[0] as $phrase) {
+ if (substr($phrase, 0, 1) == '"') {
+ $arr = explode(
+ ' ', substr($phrase, 1, strlen($phrase) - 2)
+ );
+ $author[] = implode(' AND ', $arr);
+ } else {
+ $author[] = $phrase;
+ }
+ }
+ $author = implode(' ', $author);
+ $query .= 'dc.creator any "' . $author . '" OR';
+ $query .= 'dc.creator any ' . $author;
+ break;
+ case 'callnumber':
+ break;
+ case 'publisher':
+ break;
+ case 'year':
+ $query = 'dc.date=' . $params['lookfor'];
+ break;
+ case 'series':
+ break;
+ case 'language':
+ break;
+ case 'toc':
+ break;
+ case 'topic':
+ break;
+ case 'geo':
+ break;
+ case 'era':
+ break;
+ case 'genre':
+ break;
+ case 'subject':
+ break;
+ case 'isn':
+ break;
+ case 'all':
+ default:
+ $query = 'dc.title="' . $params['lookfor'] . '" OR dc.title=' .
+ $params['lookfor'] . ' OR dc.creator="' .
+ $params['lookfor'] . '" OR dc.creator=' .
+ $params['lookfor'] . ' OR dc.subject="' .
+ $params['lookfor'] . '" OR dc.subject=' .
+ $params['lookfor'] . ' OR dc.description=' .
+ $params['lookfor'] . ' OR dc.date=' . $params['lookfor'];
+ break;
+ }
+ }
+ }
+
+ return $query;
+ }
+
+ /**
+ * Get records similiar to one record
+ *
+ * @param array $record An associative array of the record data
+ * @param string $id The record id
+ * @param int $max The maximum records to return; Default is 5
+ *
+ * @throws Exception
+ * @return array An array of query results
+ */
+ public function getMoreLikeThis($record, $id, $max = 5)
+ {
+ // More Like This Query
+ $query = 'title="' . $record['245']['a'] . '" ' .
+ "NOT rec.id=$id";
+
+ // Query String Parameters
+ $options = array('operation' => 'searchRetrieve',
+ 'query' => $query,
+ 'maximumRecords' => $max,
+ 'startRecord' => 1,
+ 'recordSchema' => 'marcxml');
+
+ if ($this->logger) {
+ $this->logger->debug('More Like This Query: ' . print_r($query, true));
+ }
+
+ return $this->call('GET', $options);
+ }
+
+ /**
+ * Scan
+ *
+ * @param string $clause The CQL clause specifying the start point
+ * @param int $pos The position of the start point in the response
+ * @param int $maxTerms The maximum number of terms to return
+ *
+ * @return string XML response
+ */
+ public function scan($clause, $pos = null, $maxTerms = null)
+ {
+ $options = array('operation' => 'scan',
+ 'scanClause' => $clause);
+ if (!is_null($pos)) {
+ $options['responsePosition'] = $pos;
+ }
+ if (!is_null($maxTerms)) {
+ $options['maximumTerms'] = $maxTerms;
+ }
+
+ return $this->call('GET', $options, false);
+ }
+
+ /**
+ * Search
+ *
+ * @param string $query The search query
+ * @param string $start The record to start with
+ * @param string $limit The amount of records to return
+ * @param string $sortBy The value to be used by for sorting
+ * @param string $schema Record schema to use in results list
+ * @param bool $process Process into array (true) or return raw (false)
+ *
+ * @throws Exception
+ * @return array An array of query results
+ */
+ public function search($query, $start = 1, $limit = null, $sortBy = null,
+ $schema = 'marcxml', $process = true
+ ) {
+ if ($this->logger) {
+ $this->logger->debug('Query: ' . print_r($query, true));
+ }
+
+ // Query String Parameters
+ $options = array('operation' => 'searchRetrieve',
+ 'query' => $query,
+ 'startRecord' => ($start) ? $start : 1,
+ 'recordSchema' => $schema);
+ if (!is_null($limit)) {
+ $options['maximumRecords'] = $limit;
+ }
+ if (!is_null($sortBy)) {
+ $options['sortKeys'] = $sortBy;
+ }
+
+ return $this->call('GET', $options, $process);
+ }
+
+ /**
+ * Check for HTTP errors in a response.
+ *
+ * @param Zend_Http_Response $result The response to check.
+ *
+ * @throws Exception
+ * @return void
+ */
+ public function checkForHttpError($result)
+ {
+ if (!$result->isSuccess()) {
+ throw new Exception('HTTP error ' . $result->getStatus());
+ }
+ }
+
+ /**
+ * Submit REST Request
+ *
+ * @param string $method HTTP Method to use: GET or POST
+ * @param array $params An array of parameters for the request
+ * @param bool $process Should we convert the MARCXML?
+ *
+ * @return string|SimpleXMLElement The response from the XServer
+ */
+ protected function call($method = 'GET', $params = null, $process = true)
+ {
+ if ($params) {
+ $query = array('version='.$this->sruVersion);
+ foreach ($params as $function => $value) {
+ if (is_array($value)) {
+ foreach ($value as $additional) {
+ $additional = urlencode($additional);
+ $query[] = "$function=$additional";
+ }
+ } else {
+ $value = urlencode($value);
+ $query[] = "$function=$value";
+ }
+ }
+ $queryString = implode('&', $query);
+ }
+
+ if ($this->logger) {
+ $this->logger->debug(
+ 'Connect: ' . print_r($this->host . '?' . $queryString, true)
+ );
+ }
+
+ // Send Request
+ $this->client->resetParameters();
+ $this->client->setUri($this->host . '?' . $queryString);
+ $result = $this->client->setMethod($method)->send();
+ $this->checkForHttpError($result);
+
+ // Return processed or unprocessed response, as appropriate:
+ return $process ? $this->process($result->getBody()) : $result->getBody();
+ }
+
+ /**
+ * Process an SRU response. Returns either the raw XML string or a
+ * SimpleXMLElement based on the contents of the class' raw property.
+ *
+ * @param string $result SRU response
+ *
+ * @return string|SimpleXMLElement
+ */
+ protected function process($result)
+ {
+ if (substr($result, 0, 5) != '<?xml') {
+ throw new Exception('Cannot Load Results');
+ }
+
+ // Send back either the raw XML or a SimpleXML object, as requested:
+ $result = XSLTProcessor::process('sru-convert.xsl', $result);
+ return $this->raw ? $result : simplexml_load_string($result);
+ }
+}
diff --git a/module/VuFind/src/VuFind/Connection/WorldCat.php b/module/VuFind/src/VuFind/Connection/WorldCat.php
new file mode 100644
index 00000000000..80c3b6a0639
--- /dev/null
+++ b/module/VuFind/src/VuFind/Connection/WorldCat.php
@@ -0,0 +1,215 @@
+<?php
+/**
+ * Class for accessing OCLC WorldCat search API
+ *
+ * PHP version 5
+ *
+ * Copyright (C) Andrew Nagy 2008.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2,
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Andrew S. Nagy <vufind-tech@lists.sourceforge.net>
+ * @author Demian Katz <demian.katz@villanova.edu>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/system_classes Wiki
+ */
+namespace VuFind\Connection;
+use VuFind\Config\Reader as ConfigReader;
+
+/**
+ * WorldCat SRU Search Interface
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Andrew S. Nagy <vufind-tech@lists.sourceforge.net>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/system_classes Wiki
+ */
+class WorldCat extends SRU
+{
+ protected $wskey;
+ protected $limitCodes;
+
+ /**
+ * Constructor
+ */
+ public function __construct()
+ {
+ parent::__construct(
+ 'http://www.worldcat.org/webservices/catalog/search/sru'
+ );
+ $config = ConfigReader::getConfig();
+ $this->wskey = isset($config->WorldCat->apiKey)
+ ? $config->WorldCat->apiKey : null;
+ $this->limitCodes = isset($config->WorldCat->LimitCodes)
+ ? $config->WorldCat->LimitCodes : null;
+ }
+
+ /**
+ * Get holdings information for the specified record.
+ *
+ * @param string $id Record to obtain holdings for.
+ *
+ * @throws Exception
+ * @return SimpleXMLElement
+ */
+ public function getHoldings($id)
+ {
+ $this->client->resetParameters();
+ $uri = "http://www.worldcat.org/webservices/catalog/content/libraries/{$id}";
+ $uri .= "?wskey={$this->wskey}&servicelevel=full";
+ $this->client->setUri($uri);
+ if ($this->logger) {
+ $this->logger->debug('Connect: ' . $uri);
+ }
+ $result = $this->client->request('POST');
+ $this->checkForHttpError($result);
+
+ return simplexml_load_string($result->getBody());
+ }
+
+ /**
+ * Retrieve a specific record.
+ *
+ * @param string $id Record ID to retrieve
+ *
+ * @throws Exception
+ * @return string MARC XML
+ */
+ public function getRecord($id)
+ {
+ $this->client->resetParameters();
+ $uri = 'http://www.worldcat.org/webservices/catalog/content/' . $id;
+ $uri .= "?wskey={$this->wskey}&servicelevel=full";
+ $this->client->setUri($uri);
+ if ($this->logger) {
+ $this->logger->debug('Connect: ' . $uri);
+ }
+ $result = $this->client->request('POST');
+ $this->checkForHttpError($result);
+
+ return $result->getBody();
+ }
+
+ /**
+ * Search
+ *
+ * @param string $query The search query
+ * @param string $oclcCode An OCLC code to exclude from results
+ * @param int $page The page of records to start with
+ * @param int $limit The number of records to return per page
+ * @param string $sort The value to be used by for sorting
+ *
+ * @throws Exception
+ * @return array An array of query results
+ */
+ public function search($query, $oclcCode = null, $page = 1, $limit = 10,
+ $sort = null
+ ) {
+ // Exclude current library from results
+ if ($oclcCode) {
+ $query .= ' not srw.li all "' . $oclcCode . '"';
+ }
+
+ // Submit query
+ $start = ($page-1) * $limit;
+ $params = array('query' => $query,
+ 'startRecord' => $start,
+ 'maximumRecords' => $limit,
+ 'sortKeys' => empty($sort) ? 'relevance' : $sort,
+ 'servicelevel' => 'full',
+ 'wskey' => $this->wskey);
+
+ // Establish a limitation on searching by OCLC Codes
+ if (!empty($this->limitCodes)) {
+ $params['oclcsymbol'] = $this->limitCodes;
+ }
+
+ return simplexml_load_string($this->call('POST', $params, false));
+ }
+
+ /**
+ * Build Query string from search parameters
+ *
+ * @param array $search An array of search parameters
+ *
+ * @throws Exception
+ * @return string The query
+ */
+ public function buildQuery($search)
+ {
+ $groups = array();
+ $excludes = array();
+ if (is_array($search)) {
+ $query = '';
+
+ foreach ($search as $params) {
+ // Advanced Search
+ if (isset($params['group'])) {
+ $thisGroup = array();
+ // Process each search group
+ foreach ($params['group'] as $group) {
+ // Build this group individually as a basic search
+ $thisGroup[] = $this->buildQuery(array($group));
+ }
+ // Is this an exclusion (NOT) group or a normal group?
+ if ($params['group'][0]['bool'] == 'NOT') {
+ $excludes[] = join(" OR ", $thisGroup);
+ } else {
+ $groups[]
+ = join(" ".$params['group'][0]['bool']." ", $thisGroup);
+ }
+ }
+
+ // Basic Search
+ if (isset($params['lookfor']) && $params['lookfor'] != '') {
+ // Clean and validate input -- note that index may be in a
+ // different field depending on whether this is a basic or
+ // advanced search.
+ $lookfor = str_replace('"', '', $params['lookfor']);
+ if (isset($params['field'])) {
+ $index = $params['field'];
+ } else if (isset($params['index'])) {
+ $index = $params['index'];
+ } else {
+ $index = 'srw.kw';
+ }
+
+ // The index may contain multiple parts -- we want to search all
+ // listed index fields:
+ $index = explode(':', $index);
+ $clauses = array();
+ foreach ($index as $currentIndex) {
+ $clauses[] = "{$currentIndex} all \"{$lookfor}\"";
+ }
+ $query .= '(' . implode(' OR ', $clauses) . ')';
+ }
+ }
+ }
+
+ // Put our advanced search together
+ if (count($groups) > 0) {
+ $query = "(" . join(") " . $search[0]['join'] . " (", $groups) . ")";
+ }
+ // and concatenate exclusion after that
+ if (count($excludes) > 0) {
+ $query .= " NOT ((" . join(") OR (", $excludes) . "))";
+ }
+
+ // Ensure we have a valid query to this point
+ return isset($query) ? $query : '';
+ }
+}
diff --git a/module/VuFind/src/VuFind/Connection/WorldCatUtils.php b/module/VuFind/src/VuFind/Connection/WorldCatUtils.php
new file mode 100644
index 00000000000..d0e9142627e
--- /dev/null
+++ b/module/VuFind/src/VuFind/Connection/WorldCatUtils.php
@@ -0,0 +1,466 @@
+<?php
+/**
+ * World Cat Utilities
+ *
+ * PHP version 5
+ *
+ * Copyright (C) Villanova University 2010.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2,
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Demian Katz <demian.katz@villanova.edu>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/system_classes Wiki
+ */
+namespace VuFind\Connection;
+use VuFind\Config\Reader as ConfigReader, VuFind\Log\Logger,
+ VuFind\XSLT\Processor as XSLTProcessor;
+
+/**
+ * World Cat Utilities
+ *
+ * Class for accessing helpful WorldCat APIs.
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Demian Katz <demian.katz@villanova.edu>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/system_classes Wiki
+ */
+class WorldCatUtils
+{
+ /**
+ * Get the WorldCat ID from the config file.
+ *
+ * @return string
+ */
+ protected function getWorldCatId()
+ {
+ static $wcId = null;
+ if (is_null($wcId)) {
+ $config = ConfigReader::getConfig();
+ $wcId = isset($config->WorldCat->id)
+ ? $config->WorldCat->id : false;
+ }
+ return $wcId;
+ }
+
+ /**
+ * Retrieve results from the index using the XISBN service.
+ *
+ * @param string $isbn ISBN of main record
+ *
+ * @return array ISBNs for related items (may be empty).
+ */
+ public function getXISBN($isbn)
+ {
+ // Build URL
+ $url = 'http://xisbn.worldcat.org/webservices/xid/isbn/' .
+ urlencode(is_array($isbn) ? $isbn[0] : $isbn) .
+ '?method=getEditions&format=csv';
+ if ($wcId = $this->getWorldCatId()) {
+ $url .= '&ai=' . urlencode($wcId);
+ }
+
+ // Print Debug code
+ Logger::getInstance()->debug("XISBN: $url");
+
+ // Fetch results
+ $isbns = array();
+ if ($fp = @fopen($url, "r")) {
+ while (($data = fgetcsv($fp, 1000, ",")) !== false) {
+ // Filter out non-ISBN characters and validate the length of
+ // whatever is left behind; this will prevent us from treating
+ // error messages like "invalidId" or "overlimit" as ISBNs.
+ $isbn = preg_replace('/[^0-9xX]/', '', $data[0]);
+ if (strlen($isbn) < 10) {
+ continue;
+ }
+ $isbns[] = $isbn;
+ }
+ }
+
+ return $isbns;
+ }
+
+ /**
+ * Retrieve results from the index using the XOCLCNUM service.
+ *
+ * @param string $oclc OCLC number of main record
+ *
+ * @return array ISBNs for related items (may be empty).
+ */
+ public function getXOCLCNUM($oclc)
+ {
+ // Build URL
+ $url = 'http://xisbn.worldcat.org/webservices/xid/oclcnum/' .
+ urlencode(is_array($oclc) ? $oclc[0] : $oclc) .
+ '?method=getEditions&format=csv';
+ if ($wcId = $this->getWorldCatId()) {
+ $url .= '&ai=' . urlencode($wcId);
+ }
+
+ // Print Debug code
+ Logger::getInstance()->debug("XOCLCNUM: $url");
+
+ // Fetch results
+ $results = array();
+ if ($fp = @fopen($url, "r")) {
+ while (($data = fgetcsv($fp, 1000, ",")) !== false) {
+ // Filter out non-numeric characters and validate the length of
+ // whatever is left behind; this will prevent us from treating
+ // error messages like "invalidId" or "overlimit" as ISBNs.
+ $current = preg_replace('/[^0-9]/', '', $data[0]);
+ if (empty($current)) {
+ continue;
+ }
+ $results[] = $current;
+ }
+ }
+
+ return $results;
+ }
+
+ /**
+ * Retrieve results from the index using the XISSN service.
+ *
+ * @param string $issn ISSN of main record
+ *
+ * @return array ISSNs for related items (may be empty).
+ */
+ public function getXISSN($issn)
+ {
+ // Build URL
+ $url = 'http://xissn.worldcat.org/webservices/xid/issn/' .
+ urlencode(is_array($issn) ? $issn[0] : $issn) .
+ //'?method=getEditions&format=csv';
+ '?method=getEditions&format=xml';
+ if ($wcId = $this->getWorldCatId()) {
+ $url .= '&ai=' . urlencode($wcId);
+ }
+
+ // Print Debug code
+ Logger::getInstance()->debug("XISSN: $url");
+
+ // Fetch results
+ $issns = array();
+ $xml = @file_get_contents($url);
+ if (!empty($xml)) {
+ $data = simplexml_load_string($xml);
+ if (!empty($data) && isset($data->group->issn)
+ && count($data->group->issn) > 0
+ ) {
+ foreach ($data->group->issn as $issn) {
+ $issns[] = (string)$issn;
+ }
+ }
+ }
+
+ return $issns;
+ }
+
+ /**
+ * Support function for getIdentitiesQuery(); is the provided name component
+ * worth considering as a first or last name?
+ *
+ * @param string $current Name chunk to examine.
+ *
+ * @return boolean Should we use this as a name?
+ */
+ protected function isUsefulNameChunk($current)
+ {
+ // Some common prefixes and suffixes that we do not want to treat as first
+ // or last names:
+ static $badChunks = array('jr', 'sr', 'ii', 'iii', 'iv', 'v', 'vi', 'vii',
+ 'viii', 'ix', 'x', 'junior', 'senior', 'esq', 'mr', 'mrs', 'miss', 'dr');
+
+ // Clean up the input string:
+ $current = str_replace('.', '', strtolower($current));
+
+ // We don't want to use empty, numeric or known bad strings!
+ if (empty($current) || is_numeric($current)
+ || in_array($current, $badChunks)
+ ) {
+ return false;
+ }
+ return true;
+ }
+
+ /**
+ * Support function for getRelatedIdentities() -- parse a name into a query
+ * for WorldCat Identities.
+ *
+ * @param string $name Name to parse.
+ *
+ * @return mixed False if useless string; Identities query otherwise.
+ */
+ protected function getIdentitiesQuery($name)
+ {
+ // Clean up user query and try to find name components within it:
+ $name = trim(str_replace(array('"', ',', '-'), ' ', $name));
+ $parts = explode(' ', $name);
+ $first = $last = '';
+ foreach ($parts as $current) {
+ $current = trim($current);
+ // Do we want to store this chunk?
+ if ($this->isUsefulNameChunk($current)) {
+ // Is the first name empty? If so, save this there.
+ if (empty($first)) {
+ $first = $current;
+ } else if (strlen($current) > 2 || empty($last)) {
+ // If this isn't the first name, we always want to save it as the
+ // last name UNLESS it's an initial, in which case we'll only
+ // save it if we don't already have something better!
+ $last = $current;
+ }
+ }
+ }
+
+ // Fail if we found no useful name components; otherwise, build up the query
+ // based on whether we found a first name only or both first and last names:
+ if (empty($first) && empty($last)) {
+ return false;
+ } else if (empty($last)) {
+ return "local.Name=\"{$first}\"";
+ } else {
+ return "local.Name=\"{$last}\" and local.Name=\"{$first}\"";
+ }
+ }
+
+ /**
+ * Support method for getRelatedIdentities() -- extract subject headings from
+ * the current node of the Identities API response.
+ *
+ * @param array $current Current response node.
+ *
+ * @return array Extracted subject headings.
+ */
+ protected function processIdentitiesSubjects($current)
+ {
+ // Normalize subjects array if it has only a single entry:
+ $subjects = isset($current->fastHeadings->fast) ?
+ $current->fastHeadings->fast : null;
+ if (isset($subjects->tag)) {
+ $subjects = array($subjects);
+ }
+
+ // Collect subjects for current name:
+ $retVal = array();
+ if (!is_null($subjects) && count($subjects) > 0) {
+ foreach ($subjects as $currentSubject) {
+ if ($currentSubject['tag'] == '650') {
+ $text = (string)$currentSubject;
+ if (!empty($text)) {
+ // Double dash will cause problems with Solr searches, so
+ // represent subject heading subdivisions differently:
+ $retVal[] = str_replace('--', ': ', $text);
+ }
+ }
+ }
+ }
+
+ return $retVal;
+ }
+
+ /**
+ * Given a name string, get related identities. Inspired by Eric Lease
+ * Morgan's Name Finder demo (http://zoia.library.nd.edu/sandbox/name-finder/).
+ * Return value is an associative array where key = author name and value =
+ * subjects used in that author's works.
+ *
+ * @param string $name Name to search for (any format).
+ * @param int $maxRecords Max # of records to read from API (more = slower).
+ *
+ * @return mixed False on error, otherwise array of related names.
+ */
+ public function getRelatedIdentities($name, $maxRecords = 10)
+ {
+ // Build the WorldCat Identities API query:
+ $query = $this->getIdentitiesQuery($name);
+ if (!$query) {
+ return false;
+ }
+
+ // Get the API response:
+ $url = "http://worldcat.org/identities/search/PersonalIdentities" .
+ "?query=" . urlencode($query) .
+ "&version=1.1" .
+ "&operation=searchRetrieve" .
+ "&recordSchema=info%3Asrw%2Fschema%2F1%2FIdentities" .
+ "&maximumRecords=" . intval($maxRecords) .
+ "&startRecord=1" .
+ "&resultSetTTL=300" .
+ "&recordPacking=xml" .
+ "&recordXPath=" .
+ "&sortKeys=holdingscount";
+ $xml = @file_get_contents($url);
+
+ // Translate XML to object:
+ $data = simplexml_load_string($xml);
+
+ // Give up if expected data is missing:
+ if (!isset($data->records->record)) {
+ return false;
+ }
+
+ // Loop through data and collect names and related subjects:
+ $output = array();
+ foreach ($data->records->record as $current) {
+ // Build current name string:
+ $current = isset($current->recordData->Identity->nameInfo) ?
+ $current->recordData->Identity->nameInfo : null;
+ if (isset($current['type']) && $current['type'] == 'personal'
+ && !empty($current->rawName->suba)
+ ) {
+ $currentName = $current->rawName->suba .
+ (isset($current->rawName->subd) ?
+ ', ' . $current->rawName->subd : '');
+
+ // Get subject list for current identity; if the current name is a
+ // duplicate of a previous name, merge the subjects together:
+ $subjects = $this->processIdentitiesSubjects($current);
+ $output[$currentName] = isset($output[$currentName])
+ ? array_unique(array_merge($output[$currentName], $subjects))
+ : $subjects;
+ }
+ }
+
+ return $output;
+ }
+
+ /**
+ * Given a subject term, get related (broader/narrower/alternate) terms.
+ * Loosely adapted from Eric Lease Morgan's Term Finder demo (see
+ * http://zoia.library.nd.edu/sandbox/term-finder/). Note that this is
+ * intended as a fairly fuzzy search -- $term need not be an exact subject
+ * heading; this function will return best guess matches in the 'exact'
+ * key, possible broader terms in the 'broader' key and possible narrower
+ * terms in the 'narrower' key of the return array.
+ *
+ * @param string $term Term to get related terms for.
+ * @param string $vocabulary Vocabulary to search (default = LCSH; see OCLC docs
+ * for other options).
+ * @param int $maxRecords Max # of records to read from API (more = slower).
+ *
+ * @return mixed False on error, otherwise array of related terms,
+ * keyed by category.
+ */
+ public function getRelatedTerms($term, $vocabulary = 'lcsh', $maxRecords = 10)
+ {
+ // Strip quotes from incoming term:
+ $term = str_replace('"', '', $term);
+
+ // Build the request URL:
+ $url = "http://tspilot.oclc.org/" . urlencode($vocabulary) . "/?" .
+ // Search for the user-supplied term in both preferred and alternative
+ // fields!
+ "query=oclcts.preferredTerm+%3D+%22" . urlencode($term) .
+ "%22+OR+oclcts.alternativeTerms+%3D+%22" . urlencode($term) . "%22" .
+ "&version=1.1" .
+ "&operation=searchRetrieve" .
+ "&recordSchema=info%3Asrw%2Fschema%2F1%2Fmarcxml-v1.1" .
+ "&maximumRecords=" . intval($maxRecords) .
+ "&startRecord=1" .
+ "&resultSetTTL=300" .
+ "&recordPacking=xml" .
+ "&recordXPath=" .
+ "&sortKeys=recordcount";
+
+ // Get the API response:
+ $data = @file_get_contents($url);
+
+ // Extract plain MARCXML from the WorldCat response:
+ $marcxml = XSLTProcessor::process('wcterms-marcxml.xsl', $data);
+
+ // Try to parse the MARCXML into a File_MARC object; if this fails,
+ // we probably have bad MARCXML, which may indicate an API failure
+ // or an empty record set. Just give up if this happens!
+ try {
+ $marc = new File_MARCXML($marcxml, File_MARCXML::SOURCE_STRING);
+ } catch (File_MARC_Exception $e) {
+ return false;
+ }
+
+ // Initialize arrays:
+ $exact = array();
+ $broader = array();
+ $narrower = array();
+
+ while ($record = $marc->next()) {
+ // Get exact terms:
+ $actual = $record->getField('150');
+ if ($actual) {
+ $main = $actual->getSubfield('a');
+ if ($main) {
+ // Some versions of File_MARCXML seem to have trouble returning
+ // strings properly (giving back XML objects instead); let's
+ // cast to string to be sure we get what we expect!
+ $main = (string)$main->getData();
+
+ // Add subdivisions:
+ $subdivisions = $actual->getSubfields('x');
+ if ($subdivisions) {
+ foreach ($subdivisions as $current) {
+ $main .= ', ' . (string)$current->getData();
+ }
+ }
+
+ // Only save the actual term if it is not a subset of the
+ // requested term.
+ if (!stristr($term, $main)) {
+ $exact[] = $main;
+ }
+ }
+ }
+
+ // Get broader/narrower terms:
+ $related = $record->getFields('550');
+ foreach ($related as $current) {
+ $type = $current->getSubfield('w');
+ $value = $current->getSubfield('a');
+ if ($type && $value) {
+ $type = (string)$type->getData();
+ $value = (string)$value->getData();
+ if ($type == 'g') {
+ // Don't save exact matches to the user-entered term:
+ if (strcasecmp($term, $value) != 0) {
+ $broader[] = $value;
+ }
+ } else if ($type == 'h') {
+ // Don't save exact matches to the user-entered term:
+ if (strcasecmp($term, $value) != 0) {
+ $narrower[] = $value;
+ }
+ }
+ }
+ }
+ }
+
+ // Send back everything we found, sorted and filtered for uniqueness; note
+ // that we do NOT sort FAST results since they support relevance ranking.
+ // As of this writing, other vocabularies do not support relevance.
+ if ($vocabulary !== 'fast') {
+ natcasesort($exact);
+ natcasesort($broader);
+ natcasesort($narrower);
+ }
+ return array(
+ 'exact' => array_unique($exact),
+ 'broader' => array_unique($broader),
+ 'narrower' => array_unique($narrower)
+ );
+ }
+}
\ No newline at end of file
diff --git a/module/VuFind/src/VuFind/XSLT/Processor.php b/module/VuFind/src/VuFind/XSLT/Processor.php
new file mode 100644
index 00000000000..3023e29c434
--- /dev/null
+++ b/module/VuFind/src/VuFind/XSLT/Processor.php
@@ -0,0 +1,62 @@
+<?php
+/**
+ * VuFind XSLT wrapper
+ *
+ * PHP version 5
+ *
+ * Copyright (C) Villanova University 2010.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2,
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Demian Katz <demian.katz@villanova.edu>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/ Wiki
+ */
+namespace VuFind\XSLT;
+
+/**
+ * VuFind XSLT wrapper
+ *
+ * @category VuFind2
+ * @package Support_Classes
+ * @author Demian Katz <demian.katz@villanova.edu>
+ * @license http://opensource.org/licenses/gpl-2.0.php GNU General Public License
+ * @link http://vufind.org/wiki/ Wiki
+ */
+class Processor
+{
+ /**
+ * Perform an XSLT transformation and return the results.
+ *
+ * @param string $xslt Name of stylesheet (in application/xsl directory)
+ * @param string $xml XML to transform with stylesheet
+ *
+ * @return string Transformed XML
+ */
+ public static function process($xslt, $xml)
+ {
+ $style = new DOMDocument();
+ // TODO: support local overrides
+ $style->load(APPLICATION_PATH . '/xsl/' . $xslt);
+ $xsl = new XSLTProcessor();
+ $xsl->importStyleSheet($style);
+ $doc = new DOMDocument();
+ if ($doc->loadXML($xml)) {
+ return $xsl->transformToXML($doc);
+ }
+ return '';
+ }
+}
--
GitLab