/**
* Converts a string to a normalized (no-spaces, non-letters) string
*
* @param string $subject original string
* @return string normalized string
*/
public static function normalize($region)
{
// this is influenced by the setlocale() call with category LC_CTYPE; see PopulateDatabases.php
$normalized = iconv('UTF-8', 'ASCII//TRANSLIT', $region);
$normalized = Parser::strtolower($normalized);
$normalized = str_replace("&", "", $normalized);
$normalized = str_replace("'", "", $normalized);
$normalized = str_replace("+", "-", $normalized);
$normalized = str_replace(" ", "-", $normalized);
$normalized = Parser::preg_replace('/\\W/', "-", $normalized);
while (strstr($normalized, "--")) {
$normalized = str_replace("--", "-", $normalized);
}
return $normalized;
}
/**
* Create a normalize category name (lowercase, just alpha and dashes)
* from supplied name
*
* @param string $name [optional] will otherwise use name property
*/
public function setNormalizedFromName($name = null)
{
if ($name == null) {
$name = $this->name;
}
// convert accented character and the like to just ascii equivalent
// this is influenced by the setlocale() call with category LC_CTYPE
$this->normalized = iconv('UTF-8', 'ASCII//TRANSLIT', $name);
$this->normalized = Parser::strtolower($this->normalized);
// strip out weird characters
$this->normalized = str_replace("&", "", $this->normalized);
$this->normalized = str_replace("'", "", $this->normalized);
// convert these to dashes
$this->normalized = str_replace("+", "-", $this->normalized);
$this->normalized = str_replace(" ", "-", $this->normalized);
// now any other non-word character to a dash
$this->normalized = Parser::preg_replace('/\\W/', "-", $this->normalized);
// pair multiple dashes down to one
while (strstr($this->normalized, "--")) {
$this->normalized = str_replace("--", "-", $this->normalized);
}
return $this;
}
/**
* Lower-case the phrase
*/
public function toLower()
{
$this->phrase = Parser::strtolower($this->phrase);
return $this;
}
public function extractFormat($data_fields)
{
if (is_array($data_fields)) {
$data_fields = implode(" ", $data_fields);
// combine them into a string
}
$data_fields = Parser::strtolower($data_fields);
if (strstr($data_fields, 'dissertation')) {
return self::Thesis;
}
if (strstr($data_fields, 'proceeding')) {
return self::ConferenceProceeding;
}
if (strstr($data_fields, 'conference')) {
return self::ConferencePaper;
}
if (strstr($data_fields, 'hearing')) {
return self::Hearing;
}
if (strstr($data_fields, 'working')) {
return self::UnpublishedWork;
}
if (strstr($data_fields, 'book review') || strstr($data_fields, 'review-book')) {
return self::BookReview;
}
if (strstr($data_fields, 'film review') || strstr($data_fields, 'film-book')) {
return self::Review;
}
if (strstr("{$data_fields} ", 'review ')) {
return self::Review;
}
if (strstr($data_fields, 'book art') || strstr($data_fields, 'book ch') || strstr($data_fields, 'chapter')) {
return self::BookSection;
}
if (strstr($data_fields, 'journal')) {
return self::Article;
}
if (strstr($data_fields, 'periodical') || strstr($data_fields, 'serial')) {
return self::Article;
}
if (strstr($data_fields, 'book')) {
return self::Book;
}
if (strstr($data_fields, 'pamphlet')) {
return self::Pamphlet;
}
if (strstr($data_fields, 'essay')) {
return self::Article;
}
if (strstr($data_fields, 'article')) {
return self::Article;
}
// if we got this far, just return unknown
return self::Unknown;
}
/**
* Add global array as xml to request xml document
*
* @param DOMDocument $xml [by reference] request xml document
* @param DOMNode $objAppend [by reference] node to append values to
* @param array $arrValues global array
*/
private function addElement(&$xml, &$objAppend, $arrValues)
{
foreach ($arrValues as $key => $value) {
// @todo: change this to 'data' element and fix xslt
// need to make sure the xml element has a valid name
// and not something crazy with spaces or commas, etc.
$strSafeKey = Parser::strtolower(preg_replace('/\\W/', '_', $key));
if (is_array($value)) {
foreach ($value as $strKey => $strValue) {
$objElement = $xml->createElement($strSafeKey);
$objElement->setAttribute('original_key', $key);
$objElement->setAttribute("key", $strKey);
$objAppend->appendChild($objElement);
if (is_array($strValue)) {
// multi-dimensional arrays will be recursively added
$this->addElement($xml, $objElement, $strValue);
} else {
$objElement->nodeValue = Parser::escapeXml($strValue);
}
}
} else {
$objElement = $xml->createElement($strSafeKey, Parser::escapeXml($value));
$objElement->setAttribute('original_key', $key);
$objAppend->appendChild($objElement);
}
}
}
/**
* Best-guess regular expression for extracting volume, issue, pagination,
* broken out here for clarity
*
* @param string $strJournalInfo any journal info, usually from 773
* @return array
*/
private function extractJournalData($strJournalInfo)
{
$arrFinal = array();
$arrCapture = array();
// we'll drop the whole thing to lower case and padd it
// with spaces to make parsing easier
$strJournalInfo = " " . Parser::strtolower($strJournalInfo) . " ";
// volume
if (preg_match('/ v[a-z]{0,5}[\\.]{0,1}[ ]{0,3}([0-9]{1,})/', $strJournalInfo, $arrCapture) != 0) {
$arrFinal["volume"] = $arrCapture[1];
$strJournalInfo = str_replace($arrCapture[0], "", $strJournalInfo);
}
// issue
if (preg_match('/ i[a-z]{0,4}[\\.]{0,1}[ ]{0,3}([0-9]{1,})/', $strJournalInfo, $arrCapture) != 0) {
$arrFinal["issue"] = $arrCapture[1];
$strJournalInfo = str_replace($arrCapture[0], "", $strJournalInfo);
} elseif (preg_match('/ n[a-z]{0,5}[\\.]{0,1}[ ]{0,3}([0-9]{1,})/', $strJournalInfo, $arrCapture) != 0) {
$arrFinal["issue"] = $arrCapture[1];
$strJournalInfo = str_replace($arrCapture[0], "", $strJournalInfo);
}
// pages
if (preg_match("/([0-9]{1,})-([0-9]{1,})/", $strJournalInfo, $arrCapture) != 0) {
$arrFinal["spage"] = $arrCapture[1];
$arrFinal["epage"] = $arrCapture[2];
$strJournalInfo = str_replace($arrCapture[0], "", $strJournalInfo);
} elseif (preg_match('/ p[a-z]{0,3}[\\.]{0,1}[ ]{0,3}([0-9]{1,})/', $strJournalInfo, $arrCapture) != 0) {
$arrFinal["spage"] = $arrCapture[1];
$strJournalInfo = str_replace($arrCapture[0], "", $strJournalInfo);
}
return $arrFinal;
}
protected function toTitleCase($strInput)
{
$arrMatches = "";
// matches from regular expression
$arrSmallWords = "";
// words that shouldn't be capitalized if they aren't the first word.
$arrWords = "";
// individual words in input
$strFinal = "";
// final string to return
$strLetter = "";
// first letter of subtitle, if any
// if there are no lowercase letters (and its sufficiently long a title to
// not just be an aconym or something) then this is likely a title stupdily
// entered into a database in ALL CAPS, so drop it entirely to
// lower-case first
$iMatch = preg_match("/[a-z]/", $strInput);
if ($iMatch == 0 && strlen($strInput) > 10) {
$strInput = Parser::strtolower($strInput);
}
// array of small words
$arrSmallWords = array('of', 'a', 'the', 'and', 'an', 'or', 'nor', 'but', 'is', 'if', 'then', 'else', 'when', 'at', 'from', 'by', 'on', 'off', 'for', 'in', 'out', 'over', 'to', 'into', 'with', 'as');
// split the string into separate words
$arrWords = explode(' ', $strInput);
foreach ($arrWords as $key => $word) {
// if this word is the first, or it's not one of our small words, capitalise it
if ($key == 0 || !in_array(Parser::strtolower($word), $arrSmallWords)) {
// make sure first character is not a quote or something
if (preg_match("/^[^a-zA-Z0-9]/", $word)) {
$first = substr($word, 0, 1);
$rest = substr($word, 1);
$arrWords[$key] = $first . ucwords($rest);
} else {
$arrWords[$key] = ucwords($word);
}
} elseif (in_array(Parser::strtolower($word), $arrSmallWords)) {
$arrWords[$key] = Parser::strtolower($word);
}
}
// join the words back into a string
$strFinal = implode(' ', $arrWords);
// catch subtitles
if (preg_match("/: ([a-z])/", $strFinal, $arrMatches)) {
$strLetter = ucwords($arrMatches[1]);
$strFinal = preg_replace("/: ([a-z])/", ": " . $strLetter, $strFinal);
}
// catch words that start with double quotes
if (preg_match("/\"([a-z])/", $strFinal, $arrMatches)) {
$strLetter = ucwords($arrMatches[1]);
$strFinal = preg_replace("/\"[a-z]/", "\"" . $strLetter, $strFinal);
}
// catch words that start with a single quote
// need to be a little more cautious here and make sure there is a space before the quote when
// inside the title to ensure this isn't a quote for a contraction or for possisive; separate
// case to handle when the quote is the first word
if (preg_match("/ '([a-z])/", $strFinal, $arrMatches)) {
$strLetter = ucwords($arrMatches[1]);
$strFinal = preg_replace("/ '[a-z]/", " '" . $strLetter, $strFinal);
}
if (preg_match("/^'([a-z])/", $strFinal, $arrMatches)) {
$strLetter = ucwords($arrMatches[1]);
$strFinal = preg_replace("/^'[a-z]/", "'" . $strLetter, $strFinal);
}
return $strFinal;
}
