Initial commit

lotgd-web/lotgd/lib/spell.php

@@ -0,0 +1,108 @@
+<?php
+// translator ready
+// addnews ready
+// mail ready
+$spell_dictionary = array();
+function spell($input,$words=false,$prefix="<span style='border: 1px dotted #FF0000;'>",$postfix="</span>"){
+	global $spell_dictionary;
+	if ($words===false)
+		$words = getsetting("dictionary","/usr/share/dict/words");
+	if (file_exists($words)){
+		if (!is_array($spell_dictionary) || count($spell_dictionary)==0){
+			//retrieve dictionary
+			$dict = file($words);
+			//sanitize the keys to drop linefeeds from the words
+			$dict = join("",$dict);
+			$dict = explode("\n",$dict);
+
+			$dict = array_flip($dict);
+			//words not typically found in a dict file
+			$dict['a']=1;
+			$dict['I']=1;
+			$spell_dictionary =& $dict;
+		}else{
+			$dict = &$spell_dictionary;
+		}
+		//Common Contractions
+		$contractions = array(
+			"n't"=>"n't", //haven't
+			"'s"=>"'s", //Joe's going to, also possessive noun
+			"'ll"=>"'ll", //we'll
+			"'re"=>"'re", //they're
+			"'ve"=>"'ve", //Where've you been all day?
+			"'m"=>"'m", //What'm I supposed to say?
+			"'d"=>"'d", //He'd
+		);
+		$input = preg_split("/([<>])/",$input,-1,PREG_SPLIT_DELIM_CAPTURE);
+		$intag = false;
+		$output = "";
+		while (list($key,$val)=each($input)){
+			if ($val=="<"){
+				$intag = true;
+			}elseif ($val==">"){
+				$intag = false;
+			}elseif (!$intag){
+				//spellcheck data not found within tags.
+				$line =
+					preg_split("/([\t\n\r[:space:]-])/",
+						$val,-1,PREG_SPLIT_DELIM_CAPTURE);
+				$val = "";
+				while (list($k,$v)=each($line)){
+					$lookups = array();
+					$i=0;
+					//look for common variations on words
+					$v1 = trim($v);
+					if ($v1>"") {
+						$lookups[$v1]=$i++;
+						$lookups[strtolower($v1)]=$i++;
+					}
+					//search for contraction endings
+					reset($contractions);
+					//strip trailing punctuation
+					$v2 = preg_replace("/[.?!\"']+$/","",$v);
+					while (list($cont,$throwaway)=each($contractions)){
+						if (substr($v2,strlen($v2)-strlen($cont)) == $cont){
+							$v1 = substr($v2,0,strlen($v2)-strlen($cont));
+							if ($v1>"") {
+								$lookups[$v1]=$i++;
+								$lookups[strtolower($v1)]=$i++;
+							}
+						}
+					}
+					$v1 = preg_replace("/[^a-zA-Z]/","",trim($v));
+					if ($v1>"") {
+						$lookups[$v1]=$i++;
+						$lookups[strtolower($v1)]=$i++;
+					} else {
+						 //if there's no alpha chars, we have no lookups to do
+						$lookups = array();
+					}
+					if (count($lookups)>0){
+						$found = false;
+						while (list($k1,$v1)=each($lookups)){
+							if (isset($dict[$k1])){
+								$found = true;
+								break;
+							}
+						}
+					}else{
+						$found = true;
+					}
+					if (!$found){
+						if (preg_match("/[[:digit:]]/",$v)) $found=true;
+					}
+					if (!$found){
+						$val = $val.$prefix.$v.$postfix;
+					}else{
+						$val.=$v;
+					}
+				}//end while
+			}//end if
+			$output.=$val;
+		}//end while
+	}else{
+		$output = $input;
+	}
+	return $output;
+}
+?>