250) {
/* apply OTS */
// deriving ratio from word count - doesn't quite work
$ratio = 20000/$count;
$ratiofullstring = (string)$ratio;
$dotpos = strpos($ratiofullstring, ".");
$ratiostring = substr($ratiofullstring, 0, $dotpos);
exec("cat ".$cleanfilename." | ots -r ".$ratiostring, $output);
// exec("cat ".$cleanfilename." | ots -r 20", $output);
} else {
/* pass through text, it's only short */
exec("cat ".$cleanfilename, $output);
}
$abstract = implode(" ", $output);
// quick & dirty
$abstract = htmlspecialchars($abstract);
echo '';
echo '';
echo ''.$abstract.'';
unset($output);
exec("cat ".$cleanfilename." | ots -a ", $output);
$keywords = implode($output);
$keywords = substr($keywords, 21);
$keywords = substr($keywords, 0, strlen($keywords)-1);
$keywordsarray = explode(",", $keywords);
foreach ($keywordsarray as $keyword) {
$keyword = trim($keyword);
if($keyword != '') {
echo '';
// echo '';
// echo ' ';
// echo ' ';
// echo ' ';
// echo ' ';
// echo '';
}
}
echo '';
echo '';
/*
# create RDF -
# http://purl.org/dc/terms/abstract
# dcterms:subject - URIs for keywords
*/
function get_and_save($url, $filename) {
$err_msg = '';
$out = fopen($filename, 'wb');
if ($out == FALSE){
print "File " . $filename . "not opened
";
exit;
}
$ch = curl_init();
curl_setopt($ch, CURLOPT_FILE, $out);
curl_setopt($ch, CURLOPT_HEADER, 0);
curl_setopt($ch, CURLOPT_URL, $url);
curl_exec($ch);
/* echo "
Error is : ".curl_error ( $ch); */
curl_close($ch);
}
function strip($oldfile, $newfile) {
$content = file_get_contents($oldfile);
$content = strip_html_tags($content);
file_put_contents($newfile, $content);
/*
$newpointer = fopen($newfile, 'w');
$handle = @fopen($oldfile, 'r');
if ($handle) {
while (!feof($handle)) {
$buffer = fgetss($handle, 4096);
fwrite($newpointer, $buffer);
}
fclose($handle);
fclose($newpointer);
}
*/
}
/**
* Remove HTML tags, including invisible text such as style and
* script code, and embedded objects. Add line breaks around
* block-level tags to prevent word joining after tag removal.
from http://nadeausoftware.com/articles/2007/09/php_tip_how_strip_html_tags_web_page
*/
function strip_html_tags( $text )
{
$text = preg_replace(
array(
// Remove invisible content
'@]*?>.*?@siu',
'@]*?>.*?@siu',
'@]*?.*?@siu',
'@]*?.*?@siu',
'@]*?.*?@siu',
'@]*?.*?@siu',
'@]*?.*?@siu',
'@]*?.*?@siu',
'@]*?.*?@siu',
/* Add line breaks before and after blocks
'@