function unapiExplainResponse($unapiID)
{
global $contentTypeCharset;
// these variables are specified in 'ini.inc.php'
$unapiCollectionDoc = new XMLDocument();
$unapiCollectionDoc->setEncoding($contentTypeCharset);
$unapiCollection = new XML("formats");
if (!empty($unapiID)) {
$unapiCollection->setTagAttribute("id", $unapiID);
}
// Recommended format names are given at <http://unapi.stikipad.com/unapi/show/existing+formats>
// TODO: add 'ISI', 'ODF XML' and 'Word XML'
addNewBranch($unapiCollection, "format", array("name" => "bibtex", "type" => "text/plain", "docs" => "http://en.wikipedia.org/wiki/BibTeX"), "");
// function 'addNewBranch()' is defined in 'webservice.inc.php'
addNewBranch($unapiCollection, "format", array("name" => "endnote", "type" => "text/plain", "docs" => "http://www.ecst.csuchico.edu/~jacobsd/bib/formats/endnote.html"), "");
addNewBranch($unapiCollection, "format", array("name" => "ris", "type" => "text/plain", "docs" => "http://www.adeptscience.co.uk/kb/article/A626"), "");
addNewBranch($unapiCollection, "format", array("name" => "atom", "type" => "application/atom+xml", "docs" => "http://www.atomenabled.org/developers/syndication/"), "");
addNewBranch($unapiCollection, "format", array("name" => "mods", "type" => "application/xml", "docs" => "http://www.loc.gov/standards/mods/"), "");
addNewBranch($unapiCollection, "format", array("name" => "oai_dc", "type" => "application/xml", "docs" => "http://www.openarchives.org/OAI/openarchivesprotocol.html#dublincore"), "");
addNewBranch($unapiCollection, "format", array("name" => "srw_dc", "type" => "application/xml", "docs" => "http://www.loc.gov/standards/sru/"), "");
addNewBranch($unapiCollection, "format", array("name" => "srw_mods", "type" => "application/xml", "docs" => "http://www.loc.gov/standards/sru/"), "");
addNewBranch($unapiCollection, "format", array("name" => "html", "type" => "text/html", "docs" => "http://www.w3.org/MarkUp/"), "");
addNewBranch($unapiCollection, "format", array("name" => "rtf", "type" => "application/rtf", "docs" => "http://en.wikipedia.org/wiki/Rich_Text_Format"), "");
addNewBranch($unapiCollection, "format", array("name" => "pdf", "type" => "application/pdf", "docs" => "http://partners.adobe.com/public/developer/pdf/index_reference.html"), "");
addNewBranch($unapiCollection, "format", array("name" => "latex", "type" => "application/x-latex", "docs" => "http://en.wikipedia.org/wiki/LaTeX"), "");
addNewBranch($unapiCollection, "format", array("name" => "markdown", "type" => "text/plain", "docs" => "http://daringfireball.net/projects/markdown/"), "");
addNewBranch($unapiCollection, "format", array("name" => "text", "type" => "text/plain"), "");
$unapiCollectionDoc->setXML($unapiCollection);
$unapiCollectionString = $unapiCollectionDoc->getXMLString();
return $unapiCollectionString;
}
function srwGenerateBaseTags($srwOperation)
{
global $exportFormat;
// this is needed so that we can distinguish between "SRW_DC XML" and "SRW_MODS XML" record formats
$srwCollection = new XML("srw:" . $srwOperation);
$srwCollection->setTagAttribute("xmlns:srw", "http://www.loc.gov/zing/srw/");
if ($srwOperation == "searchRetrieveResponse") {
$srwCollection->setTagAttribute("xmlns:diag", "http://www.loc.gov/zing/srw/diagnostic/");
$srwCollection->setTagAttribute("xmlns:xcql", "http://www.loc.gov/zing/cql/xcql/");
if (preg_match("/DC/i", $exportFormat)) {
$srwCollection->setTagAttribute("xmlns:srw_dc", "info:srw/schema/1/dc-v1.1");
$srwCollection->setTagAttribute("xmlns:dc", "http://purl.org/dc/elements/1.1/");
$srwCollection->setTagAttribute("xmlns:prism", "http://prismstandard.org/namespaces/1.2/basic/");
} else {
// add namespace declarations for "SRW_MODS XML":
$srwCollection->setTagAttribute("xmlns:mods", "http://www.loc.gov/mods/v3");
}
}
// elseif ($srwOperation == "explainResponse")
// {
// $srwCollection->setTagAttribute("xmlns:zr", "http://explain.z3950.org/dtd/2.0/");
// }
addNewBranch($srwCollection, "srw:version", array(), "1.1");
// function 'addNewBranch()' is defined in 'webservice.inc.php'
return $srwCollection;
}
function modsRecord($row)
{
global $databaseBaseURL;
// these variables are defined in 'ini.inc.php'
global $contentTypeCharset;
global $fileVisibility;
global $fileVisibilityException;
global $filesBaseURL;
global $convertExportDataToUTF8;
// defined in 'transtab_unicode_charset.inc.php' and 'transtab_latin1_charset.inc.php'
global $alnum, $alpha, $cntrl, $dash, $digit, $graph, $lower, $print, $punct, $space, $upper, $word, $patternModifiers;
$exportPrivate = True;
// This will be a global variable or will be used
// when modsRow is called and will determine if we
// export user-specific data
$exportRecordURL = True;
// Specifies whether an attribution string containing
// the URL to the refbase database record (and the last
// modification date) shall be written to the notes branch.
// Note that this string is required by the "-A|--append"
// feature of the 'refbase' command line client
// convert this record's modified date/time info to UNIX time stamp format:
// => "date('D, j M Y H:i:s O')", e.g. "Sat, 15 Jul 2006 22:24:16 +0200"
// function 'generateRFC2822TimeStamp()' is defined in 'include.inc.php'
$currentDateTimeStamp = generateRFC2822TimeStamp($row['modified_date'], $row['modified_time']);
// --- BEGIN TYPE * ---
// |
// | These apply to everything
// this is a stupid hack that maps the names of the '$row' array keys to those used
// by the '$formVars' array (which is required by function 'generateCiteKey()')
// (eventually, the '$formVars' array should use the MySQL field names as names for its array keys)
$formVars = buildFormVarsArray($row);
// function 'buildFormVarsArray()' is defined in 'include.inc.php'
// generate or extract the cite key for this record
// (note that charset conversion can only be done *after* the cite key has been generated,
// otherwise cite key generation will produce garbled text!)
$citeKey = generateCiteKey($formVars);
// function 'generateCiteKey()' is defined in 'include.inc.php'
// Create an XML object for a single record.
$record = new XML("mods");
$record->setTagAttribute("version", "3.2");
if (!empty($citeKey)) {
$record->setTagAttribute("ID", $citeKey);
}
// titleInfo
// Regular Title
if (!empty($row['title'])) {
$record->setTagContent(encodeXMLField('title', $row['title']), "mods/titleInfo/title");
}
// Translated Title
// NOTE: This field is excluded by the default cite SELECT method
if (!empty($row['orig_title'])) {
$orig_title = new XMLBranch("titleInfo");
$orig_title->setTagAttribute("type", "translated");
$orig_title->setTagContent(encodeXMLField('orig_title', $row['orig_title']), "titleInfo/title");
$record->addXMLBranch($orig_title);
}
// name
// author
if (!empty($row['author'])) {
if (preg_match("/ *\\(eds?\\)\$/", $row['author'])) {
$author = preg_replace("/[ \r\n]*\\(eds?\\)/i", "", $row['author']);
$nameArray = separateNames("author", "/\\s*;\\s*/", "/\\s*,\\s*/", "/(?<=^|[{$word}])[^-{$word}]+|(?<=^|[{$upper}])(?=\$|[{$upper}])/{$patternModifiers}", $author, "personal", "editor");
} else {
if ($row['type'] == "Map") {
$nameArray = separateNames("author", "/\\s*;\\s*/", "/\\s*,\\s*/", "/(?<=^|[{$word}])[^-{$word}]+|(?<=^|[{$upper}])(?=\$|[{$upper}])/{$patternModifiers}", $row['author'], "personal", "cartographer");
} else {
$nameArray = separateNames("author", "/\\s*;\\s*/", "/\\s*,\\s*/", "/(?<=^|[{$word}])[^-{$word}]+|(?<=^|[{$upper}])(?=\$|[{$upper}])/{$patternModifiers}", $row['author'], "personal", "author");
}
}
foreach ($nameArray as $singleName) {
$record->addXMLBranch($singleName);
}
}
// originInfo
if (!empty($row['year']) || !empty($row['publisher']) || !empty($row['place'])) {
$origin = new XMLBranch("originInfo");
// dateIssued
if (!empty($row['year'])) {
$origin->setTagContent(encodeXMLField('year', $row['year']), "originInfo/dateIssued");
}
// Book Chapters and Journal Articles only have a dateIssued
// (editions, places, and publishers are associated with the host)
if (!preg_match("/^(Book Chapter|Journal Article)\$/", $row['type'])) {
// publisher
if (!empty($row['publisher'])) {
$origin->setTagContent(encodeXMLField('publisher', $row['publisher']), "originInfo/publisher");
}
// place
if (!empty($row['place'])) {
$origin->setTagContent(encodeXMLField('place', $row['place']), "originInfo/place/placeTerm");
$origin->setTagAttribute("type", "text", "originInfo/place/placeTerm");
}
// edition
if (!empty($row['edition'])) {
$origin->setTagContent(encodeXMLField('edition', $row['edition']), "originInfo/edition");
}
}
if ($origin->hasBranch()) {
$record->addXMLBranch($origin);
}
}
// language
if (!empty($row['language'])) {
$record->setTagContent(encodeXMLField('language', $row['language']), "mods/language");
}
// abstract
// NOTE: This field is excluded by the default cite SELECT method
if (!empty($row['abstract'])) {
$abstract = new XMLBranch("abstract");
$abstract->setTagContent(encodeXMLField('abstract', $row['abstract']));
if (!empty($row['summary_language'])) {
$abstract->setTagAttribute("lang", encodeXMLField('summary_language', $row['summary_language']));
}
$record->addXMLBranch($abstract);
}
// subject
// keywords
if (!empty($row['keywords'])) {
$subjectArray = array();
$subjectArray = preg_split("/\\s*;\\s*/", $row['keywords']);
// "unrelated" keywords
foreach ($subjectArray as $singleSubject) {
$subjectBranch = new XMLBranch("subject");
$topicArray = array();
$topicArray = preg_split("/\\s*,\\s*/", $singleSubject);
// "related" keywords
foreach ($topicArray as $singleTopic) {
$topicBranch = new XMLBranch("topic");
$topicBranch->setTagContent(encodeXMLField('keywords', $singleTopic));
$subjectBranch->addXMLBranch($topicBranch);
}
$record->addXMLBranch($subjectBranch);
}
}
// user_keys
// NOTE: a copy of the above. Needs to be a separate function later.
if (!empty($row['user_keys']) && $exportPrivate) {
$subjectArray = array();
$subjectArray = preg_split("/\\s*;\\s*/", $row['user_keys']);
// "unrelated" user_keys
foreach ($subjectArray as $singleSubject) {
$subjectBranch = new XMLBranch("subject");
$topicArray = array();
$topicArray = preg_split("/\\s*,\\s*/", $singleSubject);
// "related" user_keys
foreach ($topicArray as $singleTopic) {
$topicBranch = new XMLBranch("topic");
$topicBranch->setTagContent(encodeXMLField('user_keys', $singleTopic));
$subjectBranch->addXMLBranch($topicBranch);
}
$record->addXMLBranch($subjectBranch);
}
}
// user_groups
// NOTE: a copy of the above. Needs to be a separate function later.
if (!empty($row['user_groups']) && $exportPrivate) {
$subjectArray = array();
$subjectArray = preg_split("/\\s*;\\s*/", $row['user_groups']);
// "unrelated" user_groups
foreach ($subjectArray as $singleSubject) {
$subjectBranch = new XMLBranch("subject");
$topicArray = array();
$topicArray = preg_split("/\\s*,\\s*/", $singleSubject);
// "related" user_groups
foreach ($topicArray as $singleTopic) {
$topicBranch = new XMLBranch("topic");
$topicBranch->setTagContent(encodeXMLField('user_groups', $singleTopic));
$subjectBranch->addXMLBranch($topicBranch);
}
$record->addXMLBranch($subjectBranch);
}
}
// notes
if (!empty($row['notes'])) {
$record->setTagContent(encodeXMLField('notes', $row['notes']), "mods/note");
}
// user_notes
if (!empty($row['user_notes']) && $exportPrivate) {
// replaces any generic notes
$record->setTagContent(encodeXMLField('user_notes', $row['user_notes']), "mods/note");
}
// refbase attribution string
if ($exportRecordURL) {
$attributionBranch = new XMLBranch("note");
$attributionBranch->setTagContent("exported from refbase (" . $databaseBaseURL . "show.php?record=" . $row['serial'] . "), last updated on " . $currentDateTimeStamp);
$record->addXMLBranch($attributionBranch);
}
// typeOfResource
// maps are 'cartographic', software is 'software, multimedia',
// and everything else is 'text'
$type = new XMLBranch("typeOfResource");
if ($row['type'] == "Map") {
$type->setTagContent("cartographic");
} else {
if ($row['type'] == "Software") {
$type->setTagContent("software, multimedia");
} else {
$type->setTagContent("text");
}
}
if ($row['type'] == "Manuscript") {
$type->setTagAttribute("manuscript", "yes");
}
$record->addXMLBranch($type);
// location
// Physical Location
// NOTE: This field is excluded by the default cite SELECT method
// This should also be parsed later
if (!empty($row['location'])) {
$location = new XMLBranch("location");
$locationArray = array();
$locationArray = preg_split("/\\s*;\\s*/", $row['location']);
foreach ($locationArray as $singleLocation) {
$locationBranch = new XMLBranch("physicalLocation");
$locationBranch->setTagContent(encodeXMLField('location', $singleLocation));
$location->addXMLBranch($locationBranch);
}
$record->addXMLBranch($location);
}
// URL (also an identifier, see below)
// NOTE: This field is excluded by the default cite SELECT method
if (!empty($row['url'])) {
$location = new XMLBranch("location");
$location->setTagContent(encodeXMLField('url', $row['url']), "location/url");
$record->addXMLBranch($location);
}
// Include a link to any corresponding FILE if one of the following conditions is met:
// - the variable '$fileVisibility' (defined in 'ini.inc.php') is set to 'everyone'
// - the variable '$fileVisibility' is set to 'login' AND the user is logged in
// - the variable '$fileVisibility' is set to 'user-specific' AND the 'user_permissions' session variable contains 'allow_download'
// - the array variable '$fileVisibilityException' (defined in 'ini.inc.php') contains a pattern (in array element 1) that matches the contents of the field given (in array element 0)
if ($fileVisibility == "everyone" or $fileVisibility == "login" and isset($_SESSION['loginEmail']) or $fileVisibility == "user-specific" and (isset($_SESSION['user_permissions']) and preg_match("/allow_download/", $_SESSION['user_permissions'])) or !empty($fileVisibilityException) and preg_match($fileVisibilityException[1], $row[$fileVisibilityException[0]])) {
// file
// Note that when converting MODS to Endnote or RIS, Bibutils will include the above
// URL (if given), otherwise it'll take the URL from the 'file' field. I.e. for
// Endnote or RIS, the URL to the PDF is only included if no regular URL is available.
if (!empty($row['file'])) {
$location = new XMLBranch("location");
if (preg_match('#^(https?|ftp|file)://#i', $row['file'])) {
// if the 'file' field contains a full URL (starting with "http://", "https://", "ftp://", or "file://")
$URLprefix = "";
// we don't alter the URL given in the 'file' field
} else {
// if the 'file' field contains only a partial path (like 'polarbiol/10240001.pdf') or just a file name (like '10240001.pdf')
// use the base URL of the standard files directory as prefix:
if (preg_match('#^/#', $filesBaseURL)) {
// absolute path -> file dir is located outside of refbase root dir
$URLprefix = 'http://' . $_SERVER['HTTP_HOST'] . $filesBaseURL;
} else {
// relative path -> file dir is located within refbase root dir
$URLprefix = $databaseBaseURL . $filesBaseURL;
}
}
$location->setTagContent(encodeXMLField('file', $URLprefix . $row['file']), "location/url");
$location->setTagAttribute("displayLabel", "Electronic full text", "location/url");
// the 'access' attribute requires MODS v3.2 or greater:
$location->setTagAttribute("access", "raw object", "location/url");
$record->addXMLBranch($location);
}
}
// identifier
// url
if (!empty($row['url'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('url', $row['url']));
$identifier->setTagAttribute("type", "uri");
$record->addXMLBranch($identifier);
}
// doi
if (!empty($row['doi'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('doi', $row['doi']));
$identifier->setTagAttribute("type", "doi");
$record->addXMLBranch($identifier);
}
// pubmed
// NOTE: Until refbase stores PubMed & arXiv IDs in a better way,
// we extract them from the 'notes' field
if (preg_match("/PMID *: *\\d+/i", $row['notes'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(preg_replace("/.*?PMID *: *(\\d+).*/i", "\\1", $row['notes']));
$identifier->setTagAttribute("type", "pubmed");
$record->addXMLBranch($identifier);
}
// arxiv
// NOTE: see note for pubmed
if (preg_match("/arXiv *: *[^ ;]+/i", $row['notes'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(preg_replace("/.*?arXiv *: *([^ ;]+).*/i", "\\1", $row['notes']));
$identifier->setTagAttribute("type", "arxiv");
$record->addXMLBranch($identifier);
}
// cite_key
if (!empty($citeKey)) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('cite_key', $citeKey));
$identifier->setTagAttribute("type", "citekey");
$record->addXMLBranch($identifier);
}
// local--CALL NUMBER
// NOTE: This should really be parsed!
if (!empty($row['call_number'])) {
$identifierArray = array();
$identifierArray = preg_split("/\\s*;\\s*/", $row['call_number']);
foreach ($identifierArray as $singleIdentifier) {
if (!preg_match("/@\\s*\$/", $singleIdentifier)) {
$identifierBranch = new XMLBranch("identifier");
$identifierBranch->setTagContent(encodeXMLField('call_number', $singleIdentifier));
$identifierBranch->setTagAttribute("type", "local");
$record->addXMLBranch($identifierBranch);
}
}
}
// --- END TYPE * ---
// -----------------------------------------
// --- BEGIN TYPE != ABSTRACT || BOOK CHAPTER || CONFERENCE ARTICLE || JOURNAL ARTICLE || MAGAZINE ARTICLE || NEWSPAPER ARTICLE ---
// |
// | BOOK WHOLE, CONFERENCE VOLUME, JOURNAL, MANUAL, MANUSCRIPT, MAP, MISCELLANEOUS, PATENT,
// | REPORT, and SOFTWARE have some info as a branch off the root, whereas ABSTRACT, BOOK CHAPTER,
// | CONFERENCE ARTICLE, JOURNAL ARTICLE, MAGAZINE ARTICLE and NEWSPAPER ARTICLE place it in the relatedItem branch.
if (!preg_match("/^(Abstract|Book Chapter|Conference Article|Journal Article|Magazine Article|Newspaper Article)\$/", $row['type'])) {
// name
// editor
if (!empty($row['editor'])) {
$editor = $row['editor'];
$author = $row['author'];
if (preg_match("/ *\\(eds?\\)\$/", $editor)) {
$editor = preg_replace("/[ \r\n]*\\(eds?\\)/i", "", $editor);
}
if (preg_match("/ *\\(eds?\\)\$/", $author)) {
$author = preg_replace("/[ \r\n]*\\(eds?\\)/i", "", $author);
}
if ($editor != $author) {
$nameArray = separateNames("editor", "/\\s*;\\s*/", "/\\s*,\\s*/", "/(?<=^|[{$word}])[^-{$word}]+|(?<=^|[{$upper}])(?=\$|[{$upper}])/{$patternModifiers}", $editor, "personal", "editor");
foreach ($nameArray as $singleName) {
$record->addXMLBranch($singleName);
}
}
}
// corporate
// (we treat a 'corporate_author' similar to how Bibutils converts the BibTeX
// 'organization' field to MODS XML, i.e., we add a separate name element with
// a 'type="corporate"' attribute and an 'author' role (or a 'degree grantor'
// role in case of theses))
if (!empty($row['corporate_author'])) {
$nameBranch = new XMLBranch("name");
$nameBranch->setTagAttribute("type", "corporate");
$nameBranch->setTagContent(encodeXMLField('corporate_author', $row['corporate_author']), "name/namePart");
if (empty($row['thesis'])) {
$nameBranch->setTagContent("author", "name/role/roleTerm");
} else {
// thesis
$nameBranch->setTagContent("degree grantor", "name/role/roleTerm");
}
$nameBranch->setTagAttribute("authority", "marcrelator", "name/role/roleTerm");
$nameBranch->setTagAttribute("type", "text", "name/role/roleTerm");
$record->addXMLBranch($nameBranch);
}
// conference
if (!empty($row['conference'])) {
$nameBranch = new XMLBranch("name");
$nameBranch->setTagAttribute("type", "conference");
$nameBranch->setTagContent(encodeXMLField('conference', $row['conference']), "name/namePart");
$record->addXMLBranch($nameBranch);
}
// genre
// type
// NOTE: Is there a better MARC genre[1] for 'manuscript?'
// [1]<http://www.loc.gov/marc/sourcecode/genre/genrelist.html>
$genremarc = new XMLBranch("genre");
$genre = new XMLBranch("genre");
// NOTE: According to the MARC "Source Codes for Genre"[1]
// the MARC authority should be 'marcgt', not 'marc'.
// [1]<http://www.loc.gov/marc/sourcecode/genre/genresource.html>
$genremarc->setTagAttribute("authority", "marcgt");
if (empty($row['thesis'])) {
// theses will get their own genre (see below)
if ($row['type'] == "Book Whole") {
$record->setTagContent("monographic", "mods/originInfo/issuance");
$genremarc->setTagContent("book");
} else {
if ($row['type'] == "Conference Volume") {
$genremarc->setTagContent("conference publication");
} else {
if ($row['type'] == "Journal") {
$genremarc->setTagContent("periodical");
$genre->setTagContent("academic journal");
} else {
if ($row['type'] == "Manual") {
// should we set '<issuance>monographic' here (and for the ones below)?
$genremarc->setTagContent("instruction");
$genre->setTagContent("manual");
} else {
if ($row['type'] == "Manuscript") {
$genremarc->setTagContent("loose-leaf");
$genre->setTagContent("manuscript");
} else {
if ($row['type'] == "Map") {
$genremarc->setTagContent("map");
} else {
if ($row['type'] == "Miscellaneous") {
$genre->setTagContent("miscellaneous");
} else {
if ($row['type'] == "Patent") {
$genremarc->setTagContent("patent");
} else {
if ($row['type'] == "Report") {
$genremarc->setTagContent("technical report");
$genre->setTagContent("report");
} else {
if ($row['type'] == "Software") {
// $genremarc->setTagContent("programmed text"); // would this be correct?
$genre->setTagContent("software");
} else {
if (!empty($row['type'])) {
// catch-all: don't use a MARC genre
$genre->setTagContent(encodeXMLField('type', $row['type']));
}
}
}
}
}
}
}
}
}
}
}
if ($genremarc->hasLeaf()) {
$record->addXMLBranch($genremarc);
}
if ($genre->hasLeaf()) {
$record->addXMLBranch($genre);
}
} else {
// if (!empty($row['thesis']))
$record->setTagContent("monographic", "mods/originInfo/issuance");
$thesismarc = new XMLBranch("genre");
$thesis = new XMLBranch("genre");
$thesismarc->setTagContent("thesis");
$thesismarc->setTagAttribute("authority", "marcgt");
// tweak thesis names so that Bibutils will recognize them:
if ($row['thesis'] == "Master's thesis") {
$row['thesis'] = "Masters thesis";
}
$thesis->setTagContent(encodeXMLField('thesis', $row['thesis']));
$record->addXMLBranch($thesismarc);
$record->addXMLBranch($thesis);
}
// physicalDescription
// pages
if (!empty($row['pages'])) {
$description = new XMLBranch("physicalDescription");
$pages = new XMLBranch("extent");
$pages->setTagAttribute("unit", "pages");
if (preg_match("/[0-9] *- *[0-9]/", $row['pages'])) {
// if a page range
// split the page range into start and end pages
list($pagestart, $pageend) = preg_split('/\\s*[-]\\s*/', $row['pages']);
if ($pagestart < $pageend) {
// extents MUST span multiple pages
$pages->setTagContent(encodeXMLField('pages', $pagestart), "extent/start");
$pages->setTagContent(encodeXMLField('pages', $pageend), "extent/end");
} else {
$pages->setTagContent(encodeXMLField('pages', $row['pages']));
}
} else {
if (preg_match("/^\\d\\d*\\s*pp?.?\$/", $row['pages'])) {
list($pagetotal) = preg_split('/\\s*pp?/', $row['pages']);
$pages->setTagContent(encodeXMLField('pages', $pagetotal), "extent/total");
} else {
$pages->setTagContent(encodeXMLField('pages', $row['pages']));
}
}
$description->addXMLBranch($pages);
$record->addXMLBranch($description);
}
// identifier
// isbn
if (!empty($row['isbn'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('isbn', $row['isbn']));
$identifier->setTagAttribute("type", "isbn");
$record->addXMLBranch($identifier);
}
// issn
if (!empty($row['issn'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('issn', $row['issn']));
$identifier->setTagAttribute("type", "issn");
$record->addXMLBranch($identifier);
}
// series
if (!empty($row['series_editor']) || !empty($row['series_title']) || !empty($row['abbrev_series_title']) || !empty($row['series_volume']) || !empty($row['series_issue'])) {
$record->addXMLBranch(serialBranch($row['series_editor'], $row['series_title'], $row['abbrev_series_title'], $row['series_volume'], $row['series_issue']));
}
} else {
// if (preg_match("/^(Abstract|Book Chapter|Conference Article|Journal Article|Magazine Article|Newspaper Article)$/", $row['type']))
// relatedItem
$related = new XMLBranch("relatedItem");
$related->setTagAttribute("type", "host");
// title (Publication)
if (!empty($row['publication'])) {
$related->setTagContent(encodeXMLField('publication', $row['publication']), "relatedItem/titleInfo/title");
}
// title (Abbreviated Journal)
if (!empty($row['abbrev_journal'])) {
$titleabbrev = new XMLBranch("titleInfo");
$titleabbrev->setTagAttribute("type", "abbreviated");
$titleabbrev->setTagContent(encodeXMLField('abbrev_journal', $row['abbrev_journal']), "titleInfo/title");
$related->addXMLBranch($titleabbrev);
}
// name
// editor
if (!empty($row['editor'])) {
$editor = $row['editor'];
if (preg_match("/ *\\(eds?\\)\$/", $editor)) {
$editor = preg_replace("/[ \r\n]*\\(eds?\\)/i", "", $editor);
}
$nameArray = separateNames("editor", "/\\s*;\\s*/", "/\\s*,\\s*/", "/(?<=^|[{$word}])[^-{$word}]+|(?<=^|[{$upper}])(?=\$|[{$upper}])/{$patternModifiers}", $editor, "personal", "editor");
foreach ($nameArray as $singleName) {
$related->addXMLBranch($singleName);
}
}
// corporate
// NOTE: a copy of the code for 'corporate_author' above.
// Needs to be a separate function later.
if (!empty($row['corporate_author'])) {
$nameBranch = new XMLBranch("name");
$nameBranch->setTagAttribute("type", "corporate");
$nameBranch->setTagContent(encodeXMLField('corporate_author', $row['corporate_author']), "name/namePart");
if (empty($row['thesis'])) {
$nameBranch->setTagContent("author", "name/role/roleTerm");
} else {
// thesis
$nameBranch->setTagContent("degree grantor", "name/role/roleTerm");
}
$nameBranch->setTagAttribute("authority", "marcrelator", "name/role/roleTerm");
$nameBranch->setTagAttribute("type", "text", "name/role/roleTerm");
$related->addXMLBranch($nameBranch);
}
// conference
// NOTE: a copy of the code for 'conference' above.
// Needs to be a separate function later.
if (!empty($row['conference'])) {
$nameBranch = new XMLBranch("name");
$nameBranch->setTagAttribute("type", "conference");
$nameBranch->setTagContent(encodeXMLField('conference', $row['conference']), "name/namePart");
$related->addXMLBranch($nameBranch);
}
// originInfo
$relorigin = new XMLBranch("originInfo");
// dateIssued
if (!empty($row['year'])) {
$relorigin->setTagContent(encodeXMLField('year', $row['year']), "originInfo/dateIssued");
}
// publisher
if (!empty($row['publisher'])) {
$relorigin->setTagContent(encodeXMLField('publisher', $row['publisher']), "originInfo/publisher");
}
// place
if (!empty($row['place'])) {
$relorigin->setTagContent(encodeXMLField('place', $row['place']), "originInfo/place/placeTerm");
$relorigin->setTagAttribute("type", "text", "originInfo/place/placeTerm");
}
// edition
if (!empty($row['edition'])) {
$relorigin->setTagContent(encodeXMLField('edition', $row['edition']), "originInfo/edition");
}
if ($relorigin->hasBranch()) {
$related->addXMLBranch($relorigin);
}
// genre (and originInfo/issuance)
if (empty($row['thesis'])) {
// theses will get their own genre (see below)
if (preg_match("/^(Journal Article|Magazine Article)\$/", $row['type'])) {
$related->setTagContent("continuing", "relatedItem/originInfo/issuance");
$genremarc = new XMLBranch("genre");
$genre = new XMLBranch("genre");
$genremarc->setTagContent("periodical");
$genremarc->setTagAttribute("authority", "marcgt");
if ($row['type'] == "Magazine Article") {
$genre->setTagContent("magazine");
} else {
$genre->setTagContent("academic journal");
}
$related->addXMLBranch($genremarc);
$related->addXMLBranch($genre);
} else {
if ($row['type'] == "Abstract") {
$record->setTagContent("abstract or summary", "mods/genre");
$record->setTagAttribute("authority", "marcgt", "mods/genre");
} else {
if ($row['type'] == "Conference Article") {
$related->setTagContent("conference publication", "relatedItem/genre");
$related->setTagAttribute("authority", "marcgt", "relatedItem/genre");
} else {
if ($row['type'] == "Newspaper Article") {
$related->setTagContent("continuing", "relatedItem/originInfo/issuance");
$related->setTagContent("newspaper", "relatedItem/genre");
$related->setTagAttribute("authority", "marcgt", "relatedItem/genre");
} else {
// if ($row['type'] == "Book Chapter")
$related->setTagContent("monographic", "relatedItem/originInfo/issuance");
$related->setTagContent("book", "relatedItem/genre");
$related->setTagAttribute("authority", "marcgt", "relatedItem/genre");
}
}
}
}
} else {
// if (!empty($row['thesis']))
$thesismarc = new XMLBranch("genre");
$thesis = new XMLBranch("genre");
$thesismarc->setTagContent("thesis");
$thesismarc->setTagAttribute("authority", "marcgt");
// tweak thesis names so that Bibutils will recognize them:
if ($row['thesis'] == "Master's thesis") {
$row['thesis'] = "Masters thesis";
}
$thesis->setTagContent(encodeXMLField('thesis', $row['thesis']));
$related->addXMLBranch($thesismarc);
$related->addXMLBranch($thesis);
}
if (!empty($row['year']) || !empty($row['volume']) || !empty($row['issue']) || !empty($row['pages'])) {
$part = new XMLBranch("part");
if (!empty($row['year'])) {
$part->setTagContent(encodeXMLField('year', $row['year']), "date");
}
if (!empty($row['volume'])) {
$detailvolume = new XMLBranch("detail");
$detailvolume->setTagContent(encodeXMLField('volume', $row['volume']), "detail/number");
$detailvolume->setTagAttribute("type", "volume");
$part->addXMLBranch($detailvolume);
}
if (!empty($row['issue'])) {
$detailnumber = new XMLBranch("detail");
$detailnumber->setTagContent(encodeXMLField('issue', $row['issue']), "detail/number");
$detailnumber->setTagAttribute("type", "issue");
$part->addXMLBranch($detailnumber);
}
if (!empty($row['pages'])) {
if (preg_match("/[0-9] *- *[0-9]/", $row['pages'])) {
// if a page range
// split the page range into start and end pages
list($pagestart, $pageend) = preg_split('/\\s*[-]\\s*/', $row['pages']);
if ($pagestart < $pageend) {
// extents MUST span multiple pages
$pages = new XMLBranch("extent");
$pages->setTagContent(encodeXMLField('pages', $pagestart), "extent/start");
$pages->setTagContent(encodeXMLField('pages', $pageend), "extent/end");
$pages->setTagAttribute("unit", "page");
} else {
$pages = new XMLBranch("detail");
if ($pagestart == $pageend) {
// single-page item
$pages->setTagContent(encodeXMLField('pages', $pagestart), "detail/number");
} else {
$pages->setTagContent(encodeXMLField('pages', $row['pages']), "detail/number");
}
$pages->setTagAttribute("type", "page");
}
} else {
$pages = new XMLBranch("detail");
$pages->setTagContent(encodeXMLField('pages', $row['pages']), "detail/number");
$pages->setTagAttribute("type", "page");
}
$part->addXMLBranch($pages);
}
$related->addXMLBranch($part);
}
// identifier
// isbn
if (!empty($row['isbn'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('isbn', $row['isbn']));
$identifier->setTagAttribute("type", "isbn");
$related->addXMLBranch($identifier);
}
// issn
if (!empty($row['issn'])) {
$identifier = new XMLBranch("identifier");
$identifier->setTagContent(encodeXMLField('issn', $row['issn']));
$identifier->setTagAttribute("type", "issn");
$related->addXMLBranch($identifier);
}
// series
if (!empty($row['series_editor']) || !empty($row['series_title']) || !empty($row['abbrev_series_title']) || !empty($row['series_volume']) || !empty($row['series_issue'])) {
$related->addXMLBranch(serialBranch($row['series_editor'], $row['series_title'], $row['abbrev_series_title'], $row['series_volume'], $row['series_issue']));
}
$record->addXMLBranch($related);
}
// --- END TYPE == ABSTRACT || BOOK CHAPTER || CONFERENCE ARTICLE || JOURNAL ARTICLE || MAGAZINE ARTICLE || NEWSPAPER ARTICLE ---
return $record;
}
function openSearchGenerateBaseTags($openSearchOperation)
{
if ($openSearchOperation == "Error") {
// OpenSearch Atom XML is used for diagnostics
$atomCollection = atomGenerateBaseTags($openSearchOperation);
} elseif ($openSearchOperation == "Description") {
$atomCollection = new XML("OpenSearchDescription");
$atomCollection->setTagAttribute("xmlns", "http://a9.com/-/spec/opensearch/1.1/");
$atomCollection->setTagAttribute("xmlns:opensearch", "http://a9.com/-/spec/opensearch/1.1/");
$atomCollection->setTagAttribute("xmlns:mozilla", "http://www.mozilla.org/2006/browser/search/");
}
return $atomCollection;
}
function odfSpreadsheetTableRow($recordExportArray, $rowType)
{
// create an XML object for a single record
$record = new XML("table:table-row");
if ($rowType == "heading") {
$record->setTagAttribute("table:style-name", "ro1");
foreach ($recordExportArray as $odfIndex => $indexValue) {
$tableCell = new XMLBranch("table:table-cell");
$tableCell->setTagAttribute("office:value-type", "string");
$tableCell->setTagContent($odfIndex, "table:table-cell/text:p");
$record->addXMLBranch($tableCell);
}
} else {
$record->setTagAttribute("table:style-name", "ro2");
foreach ($recordExportArray as $odfIndex => $indexValue) {
$tableCell = new XMLBranch("table:table-cell");
if (!empty($indexValue)) {
$tableCell->setTagAttribute("office:value-type", "string");
$tableCell->setTagContent($indexValue, "table:table-cell/text:p");
}
$record->addXMLBranch($tableCell);
}
}
return $record;
}
function atomGenerateBaseTags($atomOperation)
{
global $officialDatabaseName;
// these variables are specified in 'ini.inc.php'
global $databaseBaseURL;
global $feedbackEmail;
global $contentTypeCharset;
global $convertExportDataToUTF8;
global $logoImageURL;
global $faviconImageURL;
global $query;
$atomCollection = new XML("feed");
$atomCollection->setTagAttribute("xmlns", "http://www.w3.org/2005/Atom");
$atomCollection->setTagAttribute("xmlns:opensearch", "http://a9.com/-/spec/opensearch/1.1/");
$atomCollection->setTagAttribute("xmlns:unapi", "http://unapi.info/");
// NOTE: is the unAPI namespace ok? Or should we use "http://unapi.info/specs/", or maybe something like "http://purl.org/unapi/ns/" ?
$atomCollection->setTagAttribute("xmlns:dc", "http://purl.org/dc/elements/1.1/");
$atomCollection->setTagAttribute("xmlns:dcterms", "http://purl.org/dc/terms/");
$atomCollection->setTagAttribute("xmlns:prism", "http://prismstandard.org/namespaces/1.2/basic/");
$officialDatabaseNameConv = encodeHTMLspecialchars($officialDatabaseName);
// function 'encodeHTMLspecialchars()' is defined in 'include.inc.php'
if ($atomOperation != "Error") {
// convert database name to UTF-8:
// (if '$convertExportDataToUTF8' is set to "yes" in 'ini.inc.php' and character encoding is not UTF-8 already)
if ($convertExportDataToUTF8 == "yes" and $contentTypeCharset != "UTF-8") {
$officialDatabaseNameConv = convertToCharacterEncoding("UTF-8", "IGNORE", $officialDatabaseNameConv);
}
// function 'convertToCharacterEncoding()' is defined in 'include.inc.php'
}
// ----------------------------------------------------------
// Add feed-level tags:
// (not yet used: category, contributor, rights)
// - 'title':
addNewBranch($atomCollection, "title", array("type" => "text"), $officialDatabaseNameConv);
// - 'subtitle':
if ($atomOperation == "Error") {
addNewBranch($atomCollection, "subtitle", array(), "Search error!");
} else {
// ...extract the 'WHERE' clause from the SQL query to include a natural-language version (well, sort of) within the 'subtitle' element:
$queryWhereClause = extractWHEREclause($query);
// function 'extractWHEREclause()' is defined in 'include.inc.php'
// construct a meaningful feed description based on the actual 'WHERE' clause:
// TODO: For Atom XML, the query string should not get HTML encoded!
$subTitle = "Displays records where " . encodeHTML(explainSQLQuery($queryWhereClause));
// functions 'encodeHTML()' and 'explainSQLQuery()' are defined in 'include.inc.php'
addNewBranch($atomCollection, "subtitle", array(), $subTitle);
}
// - 'updated':
// (TODO: the timestamp in the 'updated' element should really only get updated if any of the matching records was updated, right?)
addNewBranch($atomCollection, "updated", array(), generateISO8601TimeStamp());
// function 'generateISO8601TimeStamp()' is defined in 'include.inc.php'
// - 'author':
$authorBranch = new XMLBranch("author");
$authorBranch->setTagContent($officialDatabaseNameConv, "author/name");
$authorBranch->setTagContent($feedbackEmail, "author/email");
$authorBranch->setTagContent($databaseBaseURL, "author/uri");
$atomCollection->addXMLBranch($authorBranch);
// - 'generator', 'icon', 'logo':
addNewBranch($atomCollection, "generator", array("uri" => "http://www.refbase.net/", "version" => "0.9.5"), "Web Reference Database (http://refbase.sourceforge.net)");
addNewBranch($atomCollection, "icon", array(), $databaseBaseURL . $faviconImageURL);
addNewBranch($atomCollection, "logo", array(), $databaseBaseURL . $logoImageURL);
// - 'link' (more links will be added in function 'atomCollection()'):
// - link to OpenSearch Description file:
atomLink($atomCollection, $databaseBaseURL . "opensearch.php?operation=explain", "search", "OpenSearch", $officialDatabaseNameConv);
// - link to unAPI server:
atomLink($atomCollection, $databaseBaseURL . "unapi.php", "unapi:unapi-server", "unAPI", "unAPI");
return $atomCollection;
}
function oaidcRecord($row, $metadataPrefix = "oai_dc", $addNameSpaceInfo = true)
{
global $databaseBaseURL;
// these variables are defined in 'ini.inc.php'
global $contentTypeCharset;
global $fileVisibility;
global $fileVisibilityException;
global $filesBaseURL;
global $convertExportDataToUTF8;
global $defaultCiteStyle;
global $citeStyle;
global $alnum, $alpha, $cntrl, $dash, $digit, $graph, $lower, $print, $punct, $space, $upper, $word, $patternModifiers;
// defined in 'transtab_unicode_charset.inc.php' and 'transtab_latin1_charset.inc.php'
// The array '$transtab_refbase_unicode' contains search & replace patterns for conversion from refbase markup to Unicode entities.
global $transtab_refbase_unicode;
// defined in 'transtab_refbase_unicode.inc.php'
// The array '$transtab_refbase_ascii' contains search & replace patterns for conversion from refbase markup to plain text.
global $transtab_refbase_ascii;
// defined in 'transtab_refbase_ascii.inc.php'
// Define inline text markup to generate a plain text citation string:
// (to be included within a 'dcterms:bibliographicCitation' element)
$markupPatternsArrayPlain = array("bold-prefix" => "", "bold-suffix" => "", "italic-prefix" => "", "italic-suffix" => "", "underline-prefix" => "", "underline-suffix" => "", "endash" => "-", "emdash" => "-", "ampersand" => "&", "double-quote" => '"', "double-quote-left" => '"', "double-quote-right" => '"', "single-quote" => "'", "single-quote-left" => "'", "single-quote-right" => "'", "less-than" => "<", "greater-than" => ">", "newline" => "\n");
// This is a stupid hack that maps the names of the '$row' array keys to those used
// by the '$formVars' array (which is required by function 'generateCiteKey()')
// (eventually, the '$formVars' array should use the MySQL field names as names for its array keys)
$formVars = buildFormVarsArray($row);
// function 'buildFormVarsArray()' is defined in 'include.inc.php'
// Generate or extract the cite key for this record:
// (to be included within a 'dc:identifier' element)
$citeKey = generateCiteKey($formVars);
// function 'generateCiteKey()' is defined in 'include.inc.php'
// Generate OpenURL data:
// (to be included within a 'dc:identifier' element)
$openURL = openURL($row, "openurl:");
// function 'openURL()' is defined in 'openurl.inc.php'
// Encode special chars and perform charset conversions:
foreach ($row as $rowFieldName => $rowFieldValue) {
// We only convert those special chars to entities which are supported by XML:
// function 'encodeHTMLspecialchars()' is defined in 'include.inc.php'
$row[$rowFieldName] = encodeHTMLspecialchars($row[$rowFieldName]);
// Convert field data to UTF-8:
// (if '$convertExportDataToUTF8' is set to "yes" in 'ini.inc.php' and character encoding is not UTF-8 already)
// (Note that charset conversion can only be done *after* the cite key has been generated, otherwise cite key
// generation will produce garbled text!)
// function 'convertToCharacterEncoding()' is defined in 'include.inc.php'
if ($convertExportDataToUTF8 == "yes" and $contentTypeCharset != "UTF-8") {
$row[$rowFieldName] = convertToCharacterEncoding("UTF-8", "IGNORE", $row[$rowFieldName]);
}
}
// Defines field-specific search & replace 'actions' that will be applied to all those refbase fields that are listed in the corresponding 'fields' element:
// (If you don't want to perform any search and replace actions, specify an empty array, like: '$fieldSpecificSearchReplaceActionsArray = array();'.
// Note that the search patterns MUST include the leading & trailing slashes -- which is done to allow for mode modifiers such as 'imsxU'.)
// "/Search Pattern/" => "Replace Pattern"
$fieldSpecificSearchReplaceActionsArray = array();
if ($convertExportDataToUTF8 == "yes") {
$fieldSpecificSearchReplaceActionsArray[] = array('fields' => array("title", "publication", "abbrev_journal", "address", "keywords", "abstract", "orig_title", "series_title", "abbrev_series_title", "notes"), 'actions' => $transtab_refbase_unicode);
}
// Apply field-specific search & replace 'actions' to all fields that are listed in the 'fields' element of the arrays contained in '$fieldSpecificSearchReplaceActionsArray':
foreach ($fieldSpecificSearchReplaceActionsArray as $fieldActionsArray) {
foreach ($row as $rowFieldName => $rowFieldValue) {
if (in_array($rowFieldName, $fieldActionsArray['fields'])) {
$row[$rowFieldName] = searchReplaceText($fieldActionsArray['actions'], $rowFieldValue, true);
}
}
}
// function 'searchReplaceText()' is defined in 'include.inc.php'
// Fetch the name of the citation style file that's associated with the style given in '$citeStyle':
$citeStyleFile = getStyleFile($citeStyle);
// function 'getStyleFile()' is defined in 'include.inc.php'
if (empty($citeStyleFile)) {
$citeStyle = $defaultCiteStyle;
// if the given cite style could not be found, we'll use the default cite style which is defined by the '$defaultCiteStyle' variable in 'ini.inc.php'
$citeStyleFile = getStyleFile($citeStyle);
}
// Include the found citation style file *once*:
include_once "cite/" . $citeStyleFile;
// Generate a proper citation for this record, ordering attributes according to the chosen output style & record type:
// - Plain text version of citation string:
$recordCitationPlain = citeRecord($row, $citeStyle, "", $markupPatternsArrayPlain, false);
// function 'citeRecord()' is defined in the citation style file given in '$citeStyleFile' (which, in turn, must reside in the 'styles' directory of the refbase root directory)
// Convert any refbase markup that remains in the citation string (such as _italic_ or **bold**) to plain text:
$recordCitationPlain = searchReplaceText($transtab_refbase_ascii, $recordCitationPlain, true);
// Convert any remaining refbase markup in the 'title', 'keywords' & 'abstract' fields to plain text:
$row['title'] = searchReplaceText($transtab_refbase_ascii, $row['title'], true);
$row['keywords'] = searchReplaceText($transtab_refbase_ascii, $row['keywords'], true);
$row['abstract'] = searchReplaceText($transtab_refbase_ascii, $row['abstract'], true);
// Strip any " (ed)" or " (eds)" suffix from author/editor string:
if (preg_match("/ *\\(eds?\\)\$/", $row['author'])) {
$row['author'] = preg_replace("/[ \r\n]*\\(eds?\\)/i", "", $row['author']);
}
if (preg_match("/ *\\(eds?\\)\$/", $row['editor'])) {
$row['editor'] = preg_replace("/[ \r\n]*\\(eds?\\)/i", "", $row['editor']);
}
// Include a link to any corresponding file if one of the following conditions is met:
// - the variable '$fileVisibility' (defined in 'ini.inc.php') is set to 'everyone'
// - the variable '$fileVisibility' is set to 'login' AND the user is logged in
// - the variable '$fileVisibility' is set to 'user-specific' AND the 'user_permissions' session variable contains 'allow_download'
// - the array variable '$fileVisibilityException' (defined in 'ini.inc.php') contains a pattern (in array element 1) that matches the contents of the field given (in array element 0)
//
// TODO: - the URL-generating code should be made into a dedicated function (since it's shared with 'modsxml.inc.php' and 'atomxml.inc.php')
$printURL = false;
if ($fileVisibility == "everyone" or $fileVisibility == "login" and isset($_SESSION['loginEmail']) or $fileVisibility == "user-specific" and (isset($_SESSION['user_permissions']) and preg_match("/allow_download/", $_SESSION['user_permissions'])) or !empty($fileVisibilityException) and preg_match($fileVisibilityException[1], $row[$fileVisibilityException[0]])) {
if (!empty($row['file'])) {
if (preg_match('#^(https?|ftp|file)://#i', $row['file'])) {
$URLprefix = "";
// we don't alter the URL given in the 'file' field
} else {
// use the base URL of the standard files directory as prefix:
if (preg_match('#^/#', $filesBaseURL)) {
// absolute path -> file dir is located outside of refbase root dir
$URLprefix = 'http://' . $_SERVER['HTTP_HOST'] . $filesBaseURL;
} else {
// relative path -> file dir is located within refbase root dir
$URLprefix = $databaseBaseURL . $filesBaseURL;
}
}
$printURL = true;
}
}
// ----------------------------------------------------------
// Start OAI_DC XML record:
if (!empty($metadataPrefix)) {
$recordPrefix = $metadataPrefix . ":";
}
$record = new XML($recordPrefix . "dc");
// create an XML object for a single record
if ($addNameSpaceInfo) {
if ($metadataPrefix == "oai_dc") {
$record->setTagAttribute("xmlns:oai_dc", "http://www.openarchives.org/OAI/2.0/oai_dc/");
} elseif ($metadataPrefix == "srw_dc") {
$record->setTagAttribute("xmlns:srw_dc", "info:srw/schema/1/dc-v1.1");
}
$record->setTagAttribute("xmlns:dc", "http://purl.org/dc/elements/1.1/");
if ($metadataPrefix == "oai_dc") {
$record->setTagAttribute("xmlns:xsi", "http://www.w3.org/2001/XMLSchema-instance");
$record->setTagAttribute("xsi:schemaLocation", "http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd");
} elseif ($metadataPrefix == "srw_dc") {
$record->setTagAttribute("xmlns:prism", "http://prismstandard.org/namespaces/1.2/basic/");
}
}
// Add Dublin Core elements:
// NOTE: With a few exceptions, we try to adhere to the guidelines given at
// "Using simple Dublin Core to describe eprints" by Andy Powell et al.
// See: <http://eprints-uk.rdn.ac.uk/project/docs/simpledc-guidelines/>
// - 'dc:title':
if (!empty($row['title'])) {
addMetaElement($record, "dc", "title", array(), $row['title']);
}
// function 'addMetaElement()' is defined in 'webservice.inc.php'
// - 'dc:creator':
if (!empty($row['author']) and $row['author'] != $row['editor']) {
addMetaElement($record, "dc", "creator", array(), $row['author']);
}
// - 'dc:creator':
// TODO: add refbase corporate author(s) as 'dc:creator'
// - 'dc:contributor':
if (!empty($row['editor'])) {
addMetaElement($record, "dc", "contributor", array(), $row['editor']);
}
// - 'dc:description':
if (!empty($row['abstract'])) {
addMetaElement($record, "dc", "description", array(), $row['abstract']);
}
// - 'dc:identifier':
// - DOI:
if (!empty($row['doi'])) {
addMetaElement($record, "dc", "identifier", array(), $row['doi'], "doi");
}
// - PMID:
if (!empty($row['notes']) and preg_match("/PMID *: *\\d+/i", $row['notes'])) {
addMetaElement($record, "dc", "identifier", array(), $row['notes'], "pmid");
}
// - arXiv:
if (!empty($row['notes']) and preg_match("/arXiv *: *[^ ;]+/i", $row['notes'])) {
addMetaElement($record, "dc", "identifier", array(), $row['notes'], "arxiv");
}
// - ISBN:
if (!empty($row['isbn'])) {
addMetaElement($record, "dc", "identifier", array(), $row['isbn'], "isbn");
}
// - OpenURL:
addMetaElement($record, "dc", "identifier", array(), $openURL, "openurl");
// - refbase ID:
addMetaElement($record, "dc", "identifier", array(), $databaseBaseURL . generateURL("show.php", "html", array("record" => $row['serial']), true), "url");
// - Cite key:
addMetaElement($record, "dc", "identifier", array(), $citeKey, "citekey");
// - Bibliographic citation:
// NOTE: In 'atomxml.inc.php', the bibliographic citation is put into a
// 'dcterms:bibliographicCitation' element so that it can be uniquely
// identified and extracted easily. However, in case of simple Dublin
// Core output, we just put it into a 'dc:identifier' element and
// use a "citation:" prefix.
addMetaElement($record, "dc", "identifier", array(), encodeHTMLspecialchars($recordCitationPlain), "citation");
// - 'dc:source':
// NOTE: - In <http://eprints-uk.rdn.ac.uk/project/docs/simpledc-guidelines/>,
// Andy Powell et al. recommend that this element should NOT be used!
// However, we use 'dc:source' elements for publication & series info
// (publication/series title plus volume & issue) to provide a dedicated
// source string that's easily readable and parsable.
// Example: <dc:source>Polar Biology, Vol. 25, No. 10</dc:source>
// - While we could also append the page info to the publication
// 'dc:source' element, this info is more pertinent to the article
// itself and is thus not included. For 'srw_dc:dc' output, page info is
// included in PRISM elements (see below).
// - All metadata (including the page info) are also provided as a machine
// parsable citation in form of an OpenURL ContextObject (see above).
// - Publication info:
// NOTE: We only include the 'dc:source' element for 'oai_dc:dc' output. In case of 'srw_dc:dc'
// output, we use the more fine-grained PRISM elements instead (see below)
if ($metadataPrefix == "oai_dc" and (!empty($row['publication']) or !empty($row['abbrev_journal']))) {
if (!empty($row['publication'])) {
$source = $row['publication'];
} elseif (!empty($row['abbrev_journal'])) {
$source = $row['abbrev_journal'];
}
if (!empty($row['volume'])) {
$source .= ", Vol. " . $row['volume'];
}
if (!empty($row['issue'])) {
$source .= ", No. " . $row['issue'];
}
if (!empty($source)) {
addMetaElement($record, "dc", "source", array(), $source);
}
}
// - Series info:
if (!empty($row['series_title']) or !empty($row['abbrev_series_title'])) {
if (!empty($row['series_title'])) {
$series = $row['series_title'];
} elseif (!empty($row['abbrev_series_title'])) {
$series = $row['abbrev_series_title'];
}
if (!empty($row['series_volume'])) {
$series .= ", Vol. " . $row['series_volume'];
}
if (!empty($row['series_issue'])) {
$series .= ", No. " . $row['series_issue'];
}
if (!empty($series)) {
addMetaElement($record, "dc", "source", array(), $series);
}
// NOTE: To distinguish between regular publication & series info,
// should we better use a "series:" prefix here? If so, use:
// addMetaElement($record, "dc", "source", array(), $series, "series");
}
// - ISSN:
// NOTE: for 'srw_dc:dc' output, we put the ISSN into the 'prism:issn' element
if ($metadataPrefix == "oai_dc" and !empty($row['issn'])) {
addMetaElement($record, "dc", "source", array(), $row['issn'], "issn");
}
// - 'dc:date':
if (!empty($row['year'])) {
addMetaElement($record, "dc", "date", array(), $row['year']);
}
// - 'dc:type':
if (!empty($row['type'])) {
addMetaElement($record, "dc", "type", array(), $row['type'], $row['thesis']);
}
// In case of a thesis, we add another 'dc:type' element with the actual thesis type:
if (!empty($row['thesis'])) {
addMetaElement($record, "dc", "type", array(), $row['thesis']);
}
// - 'dc:format':
// TODO: ideally, we should parse the content of the refbase 'medium' field and map it
// to a media-type term from <http://www.iana.org/assignments/media-types/>
if (!empty($row['medium'])) {
$mediaType = $row['medium'];
} else {
$mediaType = "text";
}
addMetaElement($record, "dc", "format", array(), $mediaType);
// - 'dc:subject':
// TODO: add user-specific keywords (from field 'user_keys') if the user is logged in
if (!empty($row['keywords'])) {
addMetaElement($record, "dc", "subject", array(), $row['keywords']);
}
// - 'dc:coverage':
// TODO: should we add contents from the refbase 'area' field as 'dc:coverage' element(s)?
// - 'dc:relation':
// - Related URL:
if (!empty($row['url'])) {
addMetaElement($record, "dc", "relation", array(), $row['url'], "url");
}
// - Related FILE:
if ($printURL) {
addMetaElement($record, "dc", "relation", array(), $URLprefix . $row['file'], "file");
}
// - 'dc:publisher':
if (!empty($row['publisher'])) {
addMetaElement($record, "dc", "publisher", array(), $row['publisher']);
}
// - 'dc:language':
// TODO: convert to ISO notation (i.e. "en" instead of "English", etc)
if (!empty($row['language'])) {
addMetaElement($record, "dc", "language", array(), $row['language']);
}
// ----------------------------------------------------------
// Add PRISM elements:
// NOTE: When using the 'srw_dc' namespace (i.e. 'info:srw/schema/1/dc-v1.1' as detailed at
// <http://www.loc.gov/standards/sru/resources/dc-schema.html>), I don't think it's allowed
// to include anything but the fifteen elements from simple Dublin Core. Is this correct?
// If so, then:
//
// TODO: Do we need to put the PRISM elements in <extraRecordData> instead? Or can we put them within
// a separate branch outside of (and next to) the '<srw_dc:dc>' element? Or shall we better omit
// them entirely?
// More info on SRU Extra Data>: <http://www.loc.gov/standards/sru/specs/extra-data.html>
//
// See also "Mixing DC metadata with other metadata schemas" in "Guidelines for implementing
// Dublin Core in XML" <http://dublincore.org/documents/dc-xml-guidelines/>
if ($metadataPrefix == "srw_dc") {
// - 'prism:issn':
if (!empty($row['issn'])) {
addMetaElement($record, "prism", "issn", array(), $row['issn']);
}
// - 'prism:publicationName':
if (!empty($row['publication'])) {
addMetaElement($record, "prism", "publicationName", array(), $row['publication']);
} elseif (!empty($row['abbrev_journal'])) {
addMetaElement($record, "prism", "publicationName", array(), $row['abbrev_journal']);
}
// - 'prism:publicationDate':
if (!empty($row['year'])) {
addMetaElement($record, "prism", "publicationDate", array(), $row['year']);
}
// - 'prism:volume':
if (!empty($row['volume'])) {
addMetaElement($record, "prism", "volume", array(), $row['volume']);
}
// - 'prism:number':
if (!empty($row['issue'])) {
addMetaElement($record, "prism", "number", array(), $row['issue']);
}
// - 'prism:startingPage', 'prism:endingPage':
// TODO: Similar code is used in 'include.in.php', 'modsxml.inc.php' and 'openurl.inc.php',
// so this should be made into a dedicated function!
if (!empty($row['pages']) and preg_match("/\\d+/i", $row['pages'])) {
$pages = preg_replace("/^\\D*(\\d+)( *[{$dash}]+ *\\d+)?.*/i{$patternModifiers}", "\\1\\2", $row['pages']);
// extract page range (if there's any), otherwise just the first number
$startPage = preg_replace("/^\\D*(\\d+).*/i", "\\1", $row['pages']);
// extract starting page
$endPage = extractDetailsFromField("pages", $pages, "/\\D+/", "[-1]");
// extract ending page (function 'extractDetailsFromField()' is defined in 'include.inc.php')
// NOTE: To extract the ending page, we'll use function 'extractDetailsFromField()'
// instead of just grabbing a matched regex pattern since it'll also work
// when just a number but no range is given (e.g. when startPage = endPage)
// - 'prism:startingPage':
if (preg_match("/\\d+ *[{$dash}]+ *\\d+/i{$patternModifiers}", $row['pages'])) {
// if there's a page range
addMetaElement($record, "prism", "startingPage", array(), $startPage);
}
// - 'prism:endingPage':
addMetaElement($record, "prism", "endingPage", array(), $endPage);
}
}
return $record;
}
