/**
* Check if the uploaded file has a similar name to an existing
* file which would then be a candidate for a revised file.
* @param $uploadedFile SubmissionFile
* @param $submissionFiles array a list of submission files to
* check the uploaded file against.
* @return integer the if of the possibly revised file or null
* if no matches were found.
*/
function &_checkForRevision(&$uploadedFile, &$submissionFiles)
{
// Get the file name.
$uploadedFileName = $uploadedFile->getOriginalFileName();
// Start with the minimal required similarity.
$minPercentage = SUBMISSION_MIN_SIMILARITY_OF_REVISION;
// Find out whether one of the files belonging to the current
// file stage matches the given file name.
$possibleRevisedFileId = null;
$matchedPercentage = 0;
foreach ($submissionFiles as $submissionFile) {
/* @var $submissionFile SubmissionFile */
// Do not consider the uploaded file itself.
if ($uploadedFile->getFileId() == $submissionFile->getFileId()) {
continue;
}
// Do not consider files from different publication formats.
if ($uploadedFile->getAssocType() == ASSOC_TYPE_REPRESENTATION && $submissionFile->getAssocType() == ASSOC_TYPE_REPRESENTATION && $uploadedFile->getAssocId() != $submissionFile->getAssocId()) {
continue;
}
// Test whether the current submission file is similar
// to the uploaded file. (Transliterate to ASCII -- the
// similar_text function can't handle UTF-8.)
import('lib.pkp.classes.core.Transcoder');
$transcoder = new Transcoder('UTF-8', 'ASCII', true);
similar_text($a = $transcoder->trans($uploadedFileName), $b = $transcoder->trans($submissionFile->getOriginalFileName()), $matchedPercentage);
if ($matchedPercentage > $minPercentage && !$this->_onlyNumbersDiffer($a, $b)) {
// We found a file that might be a possible revision.
$possibleRevisedFileId = $submissionFile->getFileId();
// Reset the min percentage to this comparison's precentage
// so that only better matches will be considered from now on.
$minPercentage = $matchedPercentage;
}
}
// Return the id of the file that we found similar.
return $possibleRevisedFileId;
}
/**
* Normalize a string in an unknown (non-UTF8) encoding into a valid UTF-8 sequence
* @param $str string input string
* @return string
*/
static function utf8_normalize($str)
{
import('lib.pkp.classes.core.Transcoder');
if (String::hasMBString()) {
// NB: CP-1252 often segfaults; we've left it out here but it will detect as 'ISO-8859-1'
$mb_encoding_order = 'UTF-8, UTF-7, ASCII, ISO-8859-1, EUC-JP, SJIS, eucJP-win, SJIS-win, JIS, ISO-2022-JP';
$detected_encoding = mb_detect_encoding($str, $mb_encoding_order, false);
} elseif (function_exists('iconv') && strlen(iconv('CP1252', 'UTF-8', $str)) != strlen(iconv('ISO-8859-1', 'UTF-8', $str))) {
// use iconv to detect CP-1252, assuming default ISO-8859-1
$detected_encoding = 'CP1252';
} else {
// assume ISO-8859-1, PHP default
$detected_encoding = 'ISO-8859-1';
}
// transcode CP-1252/ISO-8859-1 into HTML entities; this works because CP-1252 is mapped onto ISO-8859-1
if ('ISO-8859-1' == $detected_encoding || 'CP1252' == $detected_encoding) {
$trans = new Transcoder('CP1252', 'HTML-ENTITIES');
$str = $trans->trans($str);
}
// transcode from detected encoding to to UTF-8
$trans = new Transcoder($detected_encoding, 'UTF-8');
$str = $trans->trans($str);
return $str;
}
