/**
* @brief Parse OpenURL parameters and return referent
*
* @param params Array of OpenURL parameters
* @param referent Referent object to populate
*
*/
function parse_openurl($params, &$referent)
{
global $debug;
$referent->authors = array();
foreach ($params as $key => $value) {
switch ($key) {
case 'rft_val_fmt':
switch ($value) {
case 'info:ofi/fmt:kev:mtx:journal':
$referent->genre = 'article';
break;
case 'info:ofi/fmt:kev:mtx:book':
$referent->genre = 'book';
break;
default:
if (!isset($referent->genre)) {
$referent->genre = 'unknown';
}
break;
}
break;
// Article title
// Article title
case 'rft.atitle':
case 'atitle':
$title = $value[0];
$title = preg_replace('/\\.$/', '', $title);
$title = strip_tags($title);
$title = html_entity_decode($title, ENT_NOQUOTES, 'UTF-8');
$referent->title = $title;
$referent->genre = 'article';
break;
// Book title
// Book title
case 'rft.btitle':
case 'btitle':
$referent->title = $value[0];
$referent->genre = 'book';
break;
// Journal title
// Journal title
case 'rft.jtitle':
case 'rft.title':
case 'title':
$secondary_title = trim($value[0]);
$secondary_title = preg_replace('/^\\[\\[/', '', $secondary_title);
$secondary_title = preg_replace('/\\]\\]$/', '', $secondary_title);
$referent->secondary_title = $secondary_title;
$referent->genre = 'article';
break;
case 'rft.issn':
case 'issn':
$ISSN_proto = $value[0];
$clean = ISN_clean($ISSN_proto);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$referent->issn = canonical_ISSN($ISSN_proto);
$referent->genre = 'article';
}
break;
// Identifiers
// Identifiers
case 'rft_id':
case 'id':
foreach ($value as $v) {
// DOI
if (preg_match('/^(info:doi\\/|doi:)(?<doi>.*)/', $v, $match)) {
$referent->doi = $match['doi'];
}
// URL
if (preg_match('/^http:\\/\\//', $v, $match)) {
$referent->url = $v;
}
// LSID
if (preg_match('/^urn:lsid:/', $v, $match)) {
$referent->lsid = $v;
}
}
break;
// Authors
// Authors
case 'rft.au':
case 'au':
foreach ($value as $v) {
$parts = parse_name($v);
$author = new stdClass();
if (isset($parts['last'])) {
$author->lastname = $parts['last'];
}
if (isset($parts['suffix'])) {
$author->suffix = $parts['suffix'];
}
if (isset($parts['first'])) {
$author->forename = $parts['first'];
if (array_key_exists('middle', $parts)) {
$author->forename .= ' ' . $parts['middle'];
}
}
$referent->authors[] = $author;
}
break;
default:
$k = str_replace("rft.", '', $key);
$referent->{$k} = $value[0];
break;
}
}
// Clean
// Dates
if (isset($referent->date)) {
if (preg_match('/^[0-9]{4}$/', $referent->date)) {
$referent->year = $referent->date;
$referent->date = $referent->date . '-00-00';
}
if (preg_match('/^(?<year>[0-9]{4})-(?<month>[0-9]{2})-(?<day>[0-9]{2})$/', $referent->date, $match)) {
$referent->year = $match['year'];
$referent->date = $match['year'] . '-' . $match['month'] . '-' . $match['day'];
}
}
// Zotero
if (isset($referent->pages)) {
// Note "u" option in regular expression, so that we match UTF-8 characters such as –
if (preg_match('/(?<spage>[0-9]+)[\\-|–](?<epage>[0-9]+)/u', $referent->pages, $match)) {
$referent->spage = $match['spage'];
$referent->epage = $match['epage'];
unset($referent->pages);
}
}
// Endnote epage may have leading "-" as it splits spage-epage to generate OpenURL
if (isset($referent->epage)) {
$referent->epage = preg_replace('/^\\-/', '', $referent->epage);
}
// Single page
if (isset($referent->pages)) {
if (is_numeric($referent->pages)) {
$referent->spage = $referent->pages;
$referent->epage = $referent->pages;
unset($referent->pages);
}
}
// Journal titles with series numbers are split into title,series fields
if (preg_match('/(?<title>.*),?\\s+series\\s+(?<series>[0-9]+)$/i', $referent->secondary_title, $match)) {
$referent->secondary_title = $match['title'];
$referent->series = $match['series'];
}
// Volume might have series information
if (preg_match('/^series\\s+(?<series>[0-9]+),\\s*(?<volume>[0-9]+)$/i', $referent->volume, $match)) {
$referent->volume = $match['volume'];
$referent->series = $match['series'];
}
// Roman to Arabic volume
if (!is_numeric($referent->volume)) {
if (preg_match('/^[ivxicl]+$/', $referent->volume)) {
$referent->volume = arabic($referent->volume);
}
}
// Author array might not be populated, in which case add author from aulast and aufirst fields
if (count($referent->authors) == 0 && (isset($referent->aulast) && isset($referent->aufirst))) {
$author = new stdClass();
$author->lastname = $referent->aulast;
$author->forename = $referent->aufirst;
$referent->authors[] = $author;
}
// Use aulast and aufirst to ensure first author name properly parsed
if (isset($referent->aulast) && isset($referent->aufirst)) {
$author = new stdClass();
$author->lastname = $referent->aulast;
$author->forename = $referent->aufirst;
$referent->authors[0] = $author;
}
// EndNote encodes accented characters, which break journal names
if (isset($referent->secondary_title)) {
$referent->secondary_title = preg_replace('/%9F/', 'ü', $referent->secondary_title);
}
}
/**
* @brief Classify an identifier string
*
* Use regular expressions to match (and clean) the identifier.
*
* @return Array
*/
function IdentifierKind($id)
{
$result = array();
$identifierType = IDENTIFIER_UNKNOWN;
$identifierString = $id;
$matches = array();
// DOI
if (preg_match('/^(http:\\/\\/dx.doi.org\\/|doi:|info:doi\\/|info:doi\\/http:\\/\\/dx.doi.org\\/)?(10.[0-9]*\\/(.*))/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_DOI;
}
if (IDENTIFIER_DOI != $identifierType) {
// Handle
if (preg_match('/^(http:\\/\\/hdl.handle.net\\/|hdl:|info:hdl\\/)?(([0-9][0-9]*(.[0-9]*)?)\\/(.*))/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_HANDLE;
}
}
// SICI
if (preg_match('/^(http:\\/\\/links.jstor.org\\/sici\\?sici=|info:sici\\/|sici:)(.*)/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_SICI;
}
// PubMed
if (preg_match('/http:\\/\\/www.ncbi.nlm.nih.gov/', $identifierString)) {
preg_match('/list_uids=([0-9]+)/', $identifierString, $matches);
//print_r($matches);
if (isset($matches[1])) {
$identifierString = $matches[1];
$identifierType = IDENTIFIER_PUBMED;
}
if (preg_match('/http:\\/\\/www.ncbi.nlm.nih.gov\\/pubmed\\/([0-9]+)/', $identifierString, $matches)) {
$identifierString = $matches[1];
$identifierType = IDENTIFIER_PUBMED;
}
}
if (preg_match('/^(pmid:|info:pmid\\/)([0-9]*)/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_PUBMED;
}
if (preg_match('/^(genbank:|info:ddbj-embl-genbank\\/)(.*)/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_GENBANK;
}
if (preg_match('/^(gi:)([0-9]*)/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_GI;
}
// LSIDs
if (preg_match("/^(lsidres:)?([uU][rR][nN]:[lL][sS][iI][dD]:([A-Za-z0-9][\\w\\(\\)\\+\\,\\-\\.\\=\\@\\;\$\"\\!\\*\\']*):([A-Za-z0-9][\\w\\(\\)\\+\\,\\-\\.\\=\\@\\;\$\"\\!\\*\\']*):[A-Za-z0-9][\\w\\(\\)\\+\\,\\-\\.\\=\\@\\;\$\"\\!\\*\\']*(:[A-Za-z0-9][\\w\\(\\)\\+\\,\\-\\.\\=\\@\\;\$\"\\!\\*\\']*)?)\$/", $identifierString, $matches)) {
// 3 is the authority, 4 is the namespace
$identifierString = $matches[2];
$identifierType = IDENTIFIER_LSID;
}
// ISSN
if (IDENTIFIER_UNKNOWN == $identifierType) {
$ISSN_proto = $id;
// strip any prefix
$ISSN_proto = preg_replace("/^issn:/i", '', $ISSN_proto);
$clean = ISN_clean($ISSN_proto);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$identifierString = canonical_ISSN($ISSN_proto);
$identifierType = IDENTIFIER_ISSN;
}
}
// URL
if (IDENTIFIER_UNKNOWN == $identifierType) {
if (preg_match('/^(http:\\/\\/(.*))/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_URL;
}
}
// OCLC
if (IDENTIFIER_UNKNOWN == $identifierType) {
if (preg_match('/^(info:oclcnum\\/)(.*)/i', $identifierString, $matches)) {
$identifierString = $matches[2];
$identifierType = IDENTIFIER_OCLC;
}
}
$result['identifier_type'] = $identifierType;
$result['identifier_string'] = $identifierString;
//print_r($result);
return $result;
}
if (preg_match('/^http:\\/\\/dx.doi.org\\//', $url)) {
$obj->doi = $url;
$obj->doi = str_replace('http://dx.doi.org/', '', $obj->doi);
$done = true;
}
if (!$done) {
$obj->url = $url;
}
}
// issn
$nc = $xpath->query("atom:content/xhtml:div/xhtml:table/xhtml:tr[@class='ISSN']/xhtml:td", $node);
foreach ($nc as $n) {
$issn = $n->firstChild->nodeValue;
// Format ISSN
$clean = ISN_clean($issn);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$obj->issn = canonical_ISSN($issn);
}
}
// volume
$nc = $xpath->query("atom:content/xhtml:div/xhtml:table/xhtml:tr[@class='volume']/xhtml:td", $node);
foreach ($nc as $n) {
$obj->volume = $n->firstChild->nodeValue;
}
// issue
$nc = $xpath->query("atom:content/xhtml:div/xhtml:table/xhtml:tr[@class='issue']/xhtml:td", $node);
foreach ($nc as $n) {
$obj->issue = $n->firstChild->nodeValue;
}
// pages
function journal_title_from_issn($issn, $language_code = 'en')
{
global $db;
$title = '';
// Format ISSN
$clean = ISN_clean($issn);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$issn = canonical_ISSN($issn);
$sql = 'SELECT * FROM issn WHERE (issn = ' . $db->Quote($issn) . ')
AND (language_code=' . $db->Quote($language_code) . ') ORDER BY LENGTH(title) DESC LIMIT 1';
$result = $db->Execute($sql);
if ($result == false) {
die("failed: " . $sql);
}
if ($result->NumRows() == 1) {
$title = $result->fields['title'];
}
}
return $title;
}
function store_in_cache($item)
{
global $db;
// sanity check (to do)
$sql = 'INSERT INTO article_cache(';
$columns = '';
$values = ') VALUES (';
$first = true;
// ISSN
if (isset($item->issn)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
// Format ISSN
$clean = ISN_clean($item->issn);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$columns .= 'issn';
$values .= $db->qstr(canonical_ISSN($item->issn));
}
$first = false;
}
// eISSN
if (isset($item->eissn)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
// Format ISSN
$clean = ISN_clean($item->eissn);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$columns .= 'eissn';
$values .= $db->qstr(canonical_ISSN($item->eissn));
}
$first = false;
}
// Volume
if (isset($item->volume)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'volume';
$values .= $db->qstr($item->volume);
}
// Issue
if (isset($item->issue)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'issue';
$values .= $db->qstr($item->issue);
}
// Spage
if (isset($item->spage)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'spage';
$values .= $db->qstr($item->spage);
}
// EPage
if (isset($item->epage)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'epage';
$values .= $db->qstr($item->epage);
$columns .= ',hard';
$values .= ',1';
}
// Year
if (isset($item->year)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'year';
$values .= $db->qstr($item->year);
}
// Date
if (isset($item->date)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'date';
$values .= $db->qstr($item->date);
}
// Article title
if (isset($item->atitle)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
// Clean up and make UTF-8
$atitle = html_entity_decode($atitle, ENT_QUOTES, "utf-8");
$atitle = strip_tags($item->atitle);
$columns .= 'atitle';
$values .= $db->qstr(trim($atitle));
}
// Journal title
if (isset($item->title)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'title';
$values .= $db->qstr(trim($item->title));
}
// URL
if (isset($item->url)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'url';
$values .= $db->qstr($item->url);
}
// doi
if (isset($item->doi)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'doi';
$values .= $db->qstr($item->doi);
}
// handle
if (isset($item->hdl)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'hdl';
$values .= $db->qstr($item->hdl);
}
// sici
if (isset($item->sici)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'sici';
$values .= $db->qstr($item->sici);
}
// pmid
if (isset($item->pmid)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'pmid';
$values .= $db->qstr($item->pmid);
}
// pdf
if (isset($item->pdf)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'pdf';
$values .= $db->qstr($item->pdf);
}
// publisher id, such as OAI urn
if (isset($item->publisher_id)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'publisher_id';
$values .= $db->qstr($item->publisher_id);
}
// XML url (e.g., from Scielo
if (isset($item->xml_url)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'xml_url';
$values .= $db->qstr($item->xml_url);
}
// abstract
if (isset($item->abstract)) {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
// Clean up and make UTF-8
$abstract = html_entity_decode($abstract, ENT_QUOTES, "utf-8");
$abstract = strip_tags($item->abstract);
$columns .= 'abstract';
$values .= $db->qstr($abstract);
}
// Set open_access flag to 1 if item is Open access
// availability
if (isset($item->availability)) {
if ($item->availability == 'Open access') {
if (!$first) {
$columns .= ',';
}
if (!$first) {
$values .= ',';
}
if ($first) {
$first = false;
}
$columns .= 'open_access';
$values .= $db->qstr('Y');
}
}
$sql .= $columns . $values . ');';
//echo $sql;
// Store
$result = $db->Execute($sql);
if ($result == false) {
die("failed [" . __LINE__ . "]: " . $sql);
}
$id = $db->Insert_ID();
// Authors (don't actually need them, but seems a shame to through this information away)
if (isset($item->authors)) {
store_authors($id, $item->authors);
}
return $id;
}
/**
* @brief Parse OpenURL parameters and return context object
*
* @param params Array of OpenURL parameters
* @param context_object Context object to populate
*
*/
function parse_openurl($params, &$context_object)
{
global $debug;
$context_object->referring_entity = new stdClass();
$context_object->referent = new stdClass();
$context_object->referent->type = 'unknown';
foreach ($params as $key => $value) {
switch ($key) {
case 'ctx_ver':
$context_object->version = $value[0];
break;
case 'rfe_id':
$context_object->referring_entity->id = $value[0];
break;
case 'rft_val_fmt':
switch ($value) {
case 'info:ofi/fmt:kev:mtx:journal':
$context_object->referent->type = 'article';
break;
case 'info:ofi/fmt:kev:mtx:book':
$context_object->referent->type = 'book';
break;
default:
if (!isset($context_object->referent->type)) {
$context_object->referent->type = 'Unknown';
}
break;
}
break;
// Article title
// Article title
case 'rft.atitle':
case 'atitle':
$title = $value[0];
$title = preg_replace('/\\.$/', '', $title);
$title = strip_tags($title);
$title = html_entity_decode($title, ENT_NOQUOTES, 'UTF-8');
$context_object->referent->title = $title;
$context_object->referent->type = 'article';
break;
// Book title
// Book title
case 'rft.btitle':
case 'btitle':
$context_object->referent->title = $value[0];
$context_object->referent->type = 'book';
break;
// Journal title
// Journal title
case 'rft.jtitle':
case 'rft.title':
case 'title':
$publication_outlet = trim($value[0]);
$publication_outlet = preg_replace('/^\\[\\[/', '', $publication_outlet);
$publication_outlet = preg_replace('/\\]\\]$/', '', $publication_outlet);
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
$context_object->referent->journal->name = $publication_outlet;
$context_object->referent->type = 'article';
break;
case 'rft.issn':
case 'issn':
$ISSN_proto = $value[0];
$clean = ISN_clean($ISSN_proto);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$identifier = new stdclass();
$identifier->type = 'issn';
$identifier->id = canonical_ISSN($ISSN_proto);
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
$context_object->referent->journal->identifier[] = $identifier;
}
break;
// Identifiers
// Identifiers
case 'rft_id':
case 'id':
foreach ($value as $v) {
// DOI
if (preg_match('/^(info:doi\\/|doi:)(?<doi>.*)/', $v, $match)) {
$identifier = new stdclass();
$identifier->type = 'doi';
$identifier->id = $match['doi'];
$context_object->referent->identifier[] = $identifier;
}
// Handle
if (preg_match('/^(info:hdl\\/|hdl:)(?<hdl>.*)/', $v, $match)) {
$identifier = new stdclass();
$identifier->type = 'handle';
$identifier->id = $match['hdl'];
$context_object->referent->identifier[] = $identifier;
}
// PMID
if (preg_match('/^(info:pmid\\/|pmid:)(?<pmid>.*)/', $v, $match)) {
$identifier = new stdclass();
$identifier->type = 'pmid';
$identifier->id = $match['pmid'];
$context_object->referent->identifier[] = $identifier;
}
// PMC
if (preg_match('/^(pmc:)(?<pmc>.*)/', $v, $match)) {
$identifier = new stdclass();
$identifier->type = 'pmc';
$identifier->id = $match['pmc'];
$context_object->referent->identifier[] = $identifier;
}
// Without INFO-URI prefix
// LSID
if (preg_match('/^urn:lsid:/', $v)) {
$identifier = new stdclass();
$identifier->type = 'lsid';
$identifier->id = $v;
$context_object->referent->identifier[] = $identifier;
}
// URL (including PDFs)
if (preg_match('/^http:\\/\\//', $v)) {
$matched = false;
// PDF
if (!$matched) {
if (preg_match('/\\.pdf/', $v)) {
$matched = true;
$context_object->referent->pdf = $v;
}
}
// BioStor
if (!$matched) {
if (preg_match('/http:\\/\\/biostor.org\\/reference\\/(?<id>\\d+)$/', $v, $match)) {
$matched = true;
$identifier = new stdclass();
$identifier->type = 'biostor';
$identifier->id = $match['id'];
$context_object->referent->identifier[] = $identifier;
}
}
if (!$matched) {
$context_object->referent->link = new stdclass();
$context_object->referent->link->url = $v;
}
}
}
break;
// Authors
// Authors
case 'rft.au':
case 'au':
foreach ($value as $v) {
$parts = parse_name($v);
$author = new stdClass();
if (isset($parts['last'])) {
$author->lastname = $parts['last'];
}
if (isset($parts['suffix'])) {
$author->suffix = $parts['suffix'];
}
if (isset($parts['first'])) {
$author->forename = $parts['first'];
if (array_key_exists('middle', $parts)) {
$author->forename .= ' ' . $parts['middle'];
}
}
$context_object->referent->author[] = $author;
}
break;
// article details
// article details
case 'rft.volume':
case 'volume':
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
$context_object->referent->journal->volume = $value[0];
break;
case 'rft.issue':
case 'issue':
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
$context_object->referent->journal->issue = $value[0];
break;
case 'rft.spage':
case 'spage':
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
$context_object->referent->journal->pages = $value[0];
break;
case 'rft.epage':
case 'epage':
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
if (isset($context_object->referent->journal->pages)) {
$context_object->referent->journal->pages .= '--' . $value[0];
} else {
$context_object->referent->journal->pages = $value[0];
}
break;
case 'rft.pages':
case 'pages':
if (!isset($context_object->referent->journal)) {
$context_object->referent->journal = new stdclass();
}
$context_object->referent->journal->pages = $value[0];
break;
default:
$k = str_replace("rft.", '', $key);
$context_object->referent->{$k} = $value[0];
break;
}
}
// Clean
// Dates
if (isset($context_object->referent->date)) {
if (preg_match('/^[0-9]{4}$/', $context_object->referent->date)) {
$context_object->referent->year = $context_object->referent->date;
$context_object->referent->date = $context_object->referent->date . '-00-00';
}
if (preg_match('/^(?<year>[0-9]{4})-(?<month>[0-9]{2})-(?<day>[0-9]{2})$/', $context_object->referent->date, $match)) {
$context_object->referent->year = $match['year'];
$context_object->referent->date = $match['year'] . '-' . $match['month'] . '-' . $match['day'];
}
}
// Zotero
if (isset($context_object->referent->pages)) {
// Note "u" option in regular expression, so that we match UTF-8 characters such as –
if (preg_match('/(?<spage>[0-9]+)[\\-|–](?<epage>[0-9]+)/u', $context_object->referent->pages, $match)) {
$context_object->referent->spage = $match['spage'];
$context_object->referent->epage = $match['epage'];
unset($context_object->referent->pages);
}
}
// Endnote epage may have leading "-" as it splits spage-epage to generate OpenURL
if (isset($context_object->referent->epage)) {
$context_object->referent->epage = preg_replace('/^\\-/', '', $context_object->referent->epage);
}
// Journal titles with series numbers are split into title,series fields
if (preg_match('/(?<title>.*),?\\s+series\\s+(?<series>[0-9]+)$/i', $context_object->referent->journal->name, $match)) {
$context_object->referent->journal->name = $match['title'];
$context_object->referent->journal->series = $match['series'];
}
// Volume might have series information
if (preg_match('/^series\\s+(?<series>[0-9]+),\\s*(?<volume>[0-9]+)$/i', $context_object->referent->journal->volume, $match)) {
$context_object->referent->journal->volume = $match['volume'];
$context_object->referent->journal->series = $match['series'];
}
// Author array might not be populated, in which case add author from aulast and aufirst fields
if (isset($context_object->referent->author)) {
if (count($context_object->referent->author) == 0 && (isset($context_object->referent->aulast) && isset($context_object->referent->aufirst))) {
$author = new stdClass();
$author->surname = $context_object->referent->aulast;
$author->forename = $context_object->referent->aufirst;
$context_object->referent->author[] = $author;
}
}
// Use aulast and aufirst to ensure first author name properly parsed
if (isset($context_object->referent->aulast) && isset($context_object->referent->aufirst)) {
$author = new stdClass();
$author->surname = $context_object->referent->aulast;
$author->forename = $context_object->referent->aufirst;
$context_object->referent->author[0] = $author;
}
// EndNote encodes accented characters, which break journal names
if (isset($context_object->referent->publication_outlet)) {
$context_object->referent->publication_outlet = preg_replace('/%9F/', 'ü', $context_object->referent->publication_outlet);
}
}
function GetParameters($parameters)
{
//print_r($parameters);
foreach ($parameters as $k => $v) {
switch ($k) {
case 'submit':
break;
case 'rft_id':
$this->StoreIdentifier($v);
break;
default:
if (preg_match('/^rft_/', $k)) {
switch ($k) {
case 'rft_val_fmt':
switch ($v) {
case 'info:ofi/fmt:kev:mtx:journal':
$this->values['genre'] = 'article';
break;
default:
$this->values['genre'] = 'article';
break;
}
break;
case 'rft_jtitle':
$this->values['title'] = trim($v);
break;
default:
$key = $k;
$key = str_replace("rft_", '', $key);
$this->values[$key] = trim($v);
break;
}
} else {
if (preg_match('/^rfr/', $k)) {
// eat referrer info
} else {
if ($this->version == 0.1) {
$key = '';
switch ($k) {
case 'title':
if (isset($parameters['genre'])) {
if ($parameters['genre'] == 'book') {
$key = 'btitle';
}
if ($parameters['genre'] == 'article') {
$key = 'title';
}
if ($parameters['genre'] == 'journal') {
$key = 'title';
}
} else {
// assume it's an article
$key = 'title';
$this->values['genre'] = 'article';
}
break;
case 'id':
$this->StoreIdentifier(urldecode($v));
break;
case 'vol':
$key = 'volume';
// liberal input (catch case of vol=)
break;
default:
$key = $k;
break;
}
if ($key != '') {
$this->values[$key] = trim($v);
}
}
}
}
break;
}
}
// Clean
// Endnote may have leading "-" as it splits spage-epage to generate OpenURL
if (isset($this->values['epage'])) {
$this->values['epage'] = str_replace("-", "", $this->values['epage']);
}
// If an article has an issue but not a volume, we will use the issue as the volume
if (!isset($this->values['volume'])) {
if (isset($this->values['issue'])) {
$this->values['volume'] = $this->values['issue'];
unset($this->values['issue']);
}
}
// Ensure ISSN is formatted correctly (may lack hyphen if coming from Wiki)
if (isset($this->values['issn'])) {
$ISSN_proto = $this->values['issn'];
$clean = ISN_clean($ISSN_proto);
$class = ISSN_classifier($clean);
if ($class == "checksumOK") {
$this->values['issn'] = canonical_ISSN($ISSN_proto);
} else {
unset($this->values['issn']);
}
}
}