data && strpos($response->data, "Malformed DOI") !== FALSE) { //try again without parsing the doi as sometimes parsing breaks it $response = upei_roblib_ill_doi_load($doi, FALSE); } $headers = array_change_key_case($response->headers); if ($response->data && strpos($headers['content-type'], 'text/html') === FALSE && strpos($response->data, "Malformed DOI") === FALSE ) { $crossref_xml = new DOMDocument('1.0'); if (!$crossref_xml->loadXML($response->data)) { drupal_set_message(t('Error parseing DOI response, @msg')); return $form; } else { return upei_roblib_ill_populate_form_doi($crossref_xml, $form); } } else { drupal_set_message(t('Error retrieve data using DOI of @doi', array('@doi' => $doi)), 'error'); return $form; } } /** * A convenience function for retrieving individual nodes from the Crossref XML without throwing notices. * * @param DomDocument $crossref_xml * The Crossref XML * @param string $node_name * The node we want a value for, we are assuming the xml only has one node with the given name. * @param int $index * If there are more than one element we can ask for the value of each element * @return string * Returns the value of the element */ function upei_roblib_ill_get_xml_node($crossref_xml, $node_name, $index = 0) { // TODO rewrite this function to use full xpath return empty($crossref_xml->getElementsbyTagName($node_name)->item($index)) ? '' : $crossref_xml->getElementsbyTagName($node_name)->item($index)->nodeValue; } /** * Parse the response from Crossref and populate the request form * @param DomDocument $crossref_xml * The crossref xml * @param array $form * Drupal form * @return mixed * An updated Drupal form */ function upei_roblib_ill_populate_form_doi($crossref_xml, $form) { $full_title = upei_roblib_ill_get_xml_node($crossref_xml, 'full_title'); if (empty($full_title)) { $full_title = upei_roblib_ill_get_xml_node($crossref_xml, 'conference_name'); } $form['ArticleTitle']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'title'); if (empty($full_title)) { $form['Genre']['#value'] = 'book'; $form['ArticleTitle']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'title',1); if(isset($form['ArticleTitle']['#value'])) { $form['Genre']['#value'] = 'chapter'; } $full_title = upei_roblib_ill_get_xml_node($crossref_xml, 'title' , 0); } $form['Title']['#value'] = $full_title; $form['ISSN']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'issn'); $form['Date']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'year'); $form['ImageOrPageNumber']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'first_page'); $form['PagesRequested']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'first_page') . '-' . upei_roblib_ill_get_xml_node($crossref_xml, 'last_page'); $form['Volume']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'volume'); $form['Issue']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'issue'); $form['ISBN']['#value'] = upei_roblib_ill_get_xml_node($crossref_xml, 'isbn'); foreach ($crossref_xml->getElementsbyTagName('person_name') as $person) { if ($person->getAttribute('contributor_role') == 'author' && $person->getAttribute('sequence') == 'first' ) { $article_author = upei_roblib_ill_get_xml_node($person, 'surname') . ',' . upei_roblib_ill_get_xml_node($person, 'given_name'); } if ($person->getAttribute('contributor_role') == 'editor' && $person->getAttribute('sequence') == 'first' ) { $author = upei_roblib_ill_get_xml_node($person, 'surname') . ',' . upei_roblib_ill_get_xml_node($person, 'given_name'); } } $form['ArticleAuthor']['#value'] = $article_author; $form['Author']['#value'] = $author; return $form; } /** * Get the DOI name for the given url. * * @param string $doi_url * A DOI url, or a DOI name with or without the "doi:" prefix. If the $url * is a CrossRef ID passed in as doi:10.1016/j.tiv.2011.10.017 or * 10.1111/eve.12339 it will not get changed by running it through parse_url. * * @return string * A string containing the DOI name. */ function upei_roblib_ill_doi_name_from_url($doi_url) { // Allows for DOI to be entered in the following formats: 10.1111/eva.12339, // http://dx.doi.org/10.1111/eva.12339, http://doi.org/10.1111/eva.12339, // or doi:10.1111/eva.12339. Removing whitespace before parsing the url and // then removing any trailing "/" from the returned path. $doi_name = trim(parse_url(trim($doi_url), PHP_URL_PATH), '/'); return $doi_name; } /** * Perform a request to CrossRef for the given ID. * * @param string $doi * A DOI to lookop. May be provided with or without the "doi:" prefix. Can * also pass a full DOI url (i.e, http://doi.org/10.1111/eva.12339 ). * @param bool parse * parse the doi to remove prefix and whitespace etc. * * @return object * An object as provided by drupal_http_request(). */ function upei_roblib_ill_doi_load($doi, $parse = TRUE) { // Allows for $id to pass a DOI url string or the DOI name. $id = $parse ? upei_roblib_ill_doi_name_from_url($doi) : $doi; $openurl = variable_get('upei_roblib_ill_doi_openurl', 'http://www.crossref.org/openurl'); $openurl_pid = variable_get('upei_roblib_ill_doi_openurl_pid', 'user@example.com'); $url = url($openurl, array( 'query' => array( 'noredirect' => 'true', 'pid' => $openurl_pid, 'format' => 'unixref', 'id' => ((strpos($id, 'doi:') === 0) ? $id : 'doi:' . $id), ), )); return drupal_http_request($url); }