split.php 14 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465
  1. <?php
  2. //强力拆分复合词
  3. require_once '../public/casesuf.inc';
  4. require_once '../studio/dict_find_un.inc';
  5. require_once '../studio/sandhi.php';
  6. require_once "../path.php";
  7. require_once "../public/_pdo.php";
  8. if(isset($_POST["word"])){
  9. $input_word=mb_strtolower(trim($_POST["word"]),'UTF-8');
  10. if(trim($input_word)==""){
  11. echo "Empty";
  12. return;
  13. }
  14. $arrWords = str_getcsv($input_word,"\n");//支持批量拆分
  15. }
  16. else{
  17. ?>
  18. <!--debug only-->
  19. <form action="split.php" method="post">
  20. Words: <textarea type="text" name="word"></textarea>
  21. <input name="debug" type="hidden" />批量查询,单词之间用英文逗号分隔。
  22. <div>
  23. <input type="checkbox" name = "express" checked /> 快速搜索(遇到第一个连音规则成功就返回)
  24. </div>
  25. <input type="submit">
  26. </form>
  27. <?php
  28. return;
  29. }
  30. if(isset($_POST["express"])){
  31. if($_POST["express"]==="on"){
  32. $_express = true;
  33. }
  34. else{
  35. $_express = false;
  36. }
  37. }
  38. else{
  39. $_express = false;
  40. }
  41. global $dbh;
  42. $dns = "sqlite:"._FILE_DB_PART_;
  43. $dbh = new PDO($dns, "", "",array(PDO::ATTR_PERSISTENT=>true));
  44. $dbh->setAttribute(PDO::ATTR_ERRMODE, PDO::ERRMODE_WARNING);
  45. global $path;
  46. global $confidence;
  47. global $result;
  48. global $part ;
  49. $part= array();
  50. $path[]=array("",0);
  51. $path[]=array("",0);
  52. $path[]=array("",0);
  53. $path[]=array("",0);
  54. $path[]=array("",0);
  55. $path[]=array("",0);
  56. $path[]=array("",0);
  57. $path[]=array("",0);
  58. $path[]=array("",0);
  59. $path[]=array("",0);
  60. $path[]=array("",0);
  61. $path[]=array("",0);
  62. $path[]=array("",0);
  63. $path[]=array("",0);
  64. $path[]=array("",0);
  65. $path[]=array("",0);
  66. $path[]=array("",0);
  67. global $sandhi ;
  68. //sandhi 语尾表
  69. $sandhi[]=array("a"=>"","b"=>"","c"=>"","len"=>0,"adj_len"=>0,"advance"=>false);
  70. $sandhi[]=array("a"=>"a","b"=>"a","c"=>"ā","len"=>1,"adj_len"=>0,"advance"=>false);
  71. $sandhi[]=array("a"=>"ā","b"=>"ā","c"=>"ā","len"=>1,"adj_len"=>0,"advance"=>false);
  72. $sandhi[]=array("a"=>"a","b"=>"ā","c"=>"ā","len"=>1,"adj_len"=>0,"advance"=>false);
  73. $sandhi[]=array("a"=>"ā","b"=>"a","c"=>"ā","len"=>1,"adj_len"=>0,"advance"=>false);
  74. $sandhi[]=array("a"=>"a","b"=>"e","c"=>"e","len"=>1,"adj_len"=>0,"advance"=>false);
  75. $sandhi[]=array("a"=>"a","b"=>"i","c"=>"i","len"=>1,"adj_len"=>0,"advance"=>false);
  76. $sandhi[]=array("a"=>"a","b"=>"o","c"=>"o","len"=>1,"adj_len"=>0,"advance"=>false);
  77. $sandhi[]=array("a"=>"a","b"=>"u","c"=>"o","len"=>1,"adj_len"=>0,"advance"=>false);
  78. $sandhi[]=array("a"=>"u","b"=>"a","c"=>"o","len"=>1,"adj_len"=>0,"advance"=>false);
  79. $sandhi[]=array("a"=>"u","b"=>"u","c"=>"ū","len"=>1,"adj_len"=>0,"advance"=>false);
  80. $sandhi[]=array("a"=>"a","b"=>"u","c"=>"u","len"=>1,"adj_len"=>0,"advance"=>false);
  81. $sandhi[]=array("a"=>"a","b"=>"ī","c"=>"ī","len"=>1,"adj_len"=>0,"advance"=>false);
  82. $sandhi[]=array("a"=>"a","b"=>"ū","c"=>"ū","len"=>1,"adj_len"=>0,"advance"=>false);
  83. $sandhi[]=array("a"=>"a","b"=>"i","c"=>"e","len"=>1,"adj_len"=>0,"advance"=>false);
  84. $sandhi[]=array("a"=>"i","b"=>"i","c"=>"ī","len"=>1,"adj_len"=>0,"advance"=>false);
  85. $sandhi[]=array("a"=>"i","b"=>"e","c"=>"e","len"=>1,"adj_len"=>0,"advance"=>false);
  86. $sandhi[]=array("a"=>"i","b"=>"a","c"=>"ya","len"=>2,"adj_len"=>0,"advance"=>false);
  87. $sandhi[]=array("a"=>"a","b"=>"atth","c"=>"atth","len"=>4,"adj_len"=>0,"advance"=>false);
  88. $sandhi[]=array("a"=>"taṃ","b"=>"n","c"=>"tann","len"=>4,"adj_len"=>0,"advance"=>false);
  89. $sandhi[]=array("a"=>"[ṃ]","b"=>"api","c"=>"mpi","len"=>3,"adj_len"=>0,"advance"=>false);
  90. $sandhi[]=array("a"=>"[ṃ]","b"=>"eva","c"=>"meva","len"=>4,"adj_len"=>0,"advance"=>false);
  91. $sandhi[]=array("a"=>"[o]","b"=>"iva","c"=>"ova","len"=>3,"adj_len"=>0,"advance"=>false);
  92. $sandhi[]=array("a"=>"a","b"=>"ādi","c"=>"ādi","len"=>3,"adj_len"=>0,"advance"=>false);
  93. $sandhi[]=array("a"=>"a[ānaṃ]","b"=>"a","c"=>"ānama","len"=>5,"adj_len"=>0,"advance"=>false);
  94. $sandhi[]=array("a"=>"a","b"=>"iti","c"=>"āti","len"=>3,"adj_len"=>0,"advance"=>false);
  95. $sandhi[]=array("a"=>"[ṃ]","b"=>"ca","c"=>"ñca","len"=>3,"adj_len"=>0,"advance"=>false);
  96. $sandhi[]=array("a"=>"[ṃ]","b"=>"iti","c"=>"nti","len"=>3,"adj_len"=>0,"advance"=>false);
  97. $sandhi[]=array("a"=>"[ṃ]","b"=>"a","c"=>"ma","len"=>2,"adj_len"=>0,"advance"=>false);
  98. $sandhi[]=array("a"=>"[ṃ]","b"=>"ā","c"=>"mā","len"=>2,"adj_len"=>0,"advance"=>false);
  99. $sandhi[]=array("a"=>"[ṃ]","b"=>"u","c"=>"mu","len"=>2,"adj_len"=>0,"advance"=>false);
  100. $sandhi[]=array("a"=>"[ṃ]","b"=>"h","c"=>"ñh","len"=>2,"adj_len"=>0,"advance"=>false);
  101. $sandhi[]=array("a"=>"ā","b"=>"[ṃ]","c"=>"am","len"=>2,"adj_len"=>0,"advance"=>false);
  102. $sandhi[]=array("a"=>"ī","b"=>"[ṃ]","c"=>"im","len"=>2,"adj_len"=>0,"advance"=>false);
  103. $sandhi[]=array("a"=>"ati","b"=>"tabba","c"=>"atabba","len"=>6,"adj_len"=>0,"advance"=>false);
  104. $sandhi[]=array("a"=>"ati","b"=>"tabba","c"=>"itabba","len"=>6,"adj_len"=>0,"advance"=>false);
  105. $sandhi[]=array("a"=>"iti","b"=>"a","c"=>"icca","len"=>4,"adj_len"=>0,"advance"=>false);
  106. /*
  107. $sandhi[]=array("a"=>"u[ūnaṃ]","b"=>"a","c"=>"ūnama","len"=>5,"adj_len"=>0,"advance"=>false);
  108. $sandhi[]=array("a"=>"ī[īnaṃ]","b"=>"a","c"=>"īnama","len"=>5,"adj_len"=>0,"advance"=>false);
  109. $sandhi[]=array("a"=>"ā","b"=>"iti","c"=>"āti","len"=>3,"adj_len"=>0);
  110. $sandhi[]=array("a"=>"a","b"=>"iti","c"=>"āti","len"=>3,"adj_len"=>0);
  111. $sandhi[]=array("a"=>"e","b"=>"iti","c"=>"eti","len"=>3,"adj_len"=>0);
  112. $sandhi[]=array("a"=>"ī","b"=>"iti","c"=>"īti","len"=>3,"adj_len"=>0);
  113. $sandhi[]=array("a"=>"i","b"=>"iti","c"=>"īti","len"=>3,"adj_len"=>0);
  114. $sandhi[]=array("a"=>"o","b"=>"iti","c"=>"oti","len"=>3,"adj_len"=>0);
  115. $sandhi[]=array("a"=>"ū","b"=>"iti","c"=>"ūti","len"=>3,"adj_len"=>0);
  116. $sandhi[]=array("a"=>"u","b"=>"iti","c"=>"ūti","len"=>3,"adj_len"=>0);
  117. $sandhi[]=array("a"=>"ṃ","b"=>"iti","c"=>"nti","len"=>3,"adj_len"=>0);
  118. $sandhi[]=array("a"=>"ṃ","b"=>"ca","c"=>"ñca","len"=>3,"adj_len"=>0);
  119. $sandhi[]=array("a"=>"a","b"=>"eva","c"=>"eva","len"=>3,"adj_len"=>0);
  120. $sandhi[]=array("a"=>"ā","b"=>"eva","c"=>"āyeva","len"=>5,"adj_len"=>0);
  121. $sandhi[]=array("a"=>"e","b"=>"eva","c"=>"eva","len"=>3,"adj_len"=>0);
  122. $sandhi[]=array("a"=>"i","b"=>"eva","c"=>"yeva","len"=>4,"adj_len"=>0);
  123. $sandhi[]=array("a"=>"ī","b"=>"eva","c"=>"iyeva","len"=>5,"adj_len"=>0);
  124. $sandhi[]=array("a"=>"ī","b"=>"eva","c"=>"īyeva","len"=>5,"adj_len"=>0);
  125. $sandhi[]=array("a"=>"o","b"=>"eva","c"=>"ova","len"=>3,"adj_len"=>0);
  126. $sandhi[]=array("a"=>"u","b"=>"eva","c"=>"veva","len"=>3,"adj_len"=>0);
  127. $sandhi[]=array("a"=>"a","b"=>"eva","c"=>"evā","len"=>3,"adj_len"=>0);
  128. $sandhi[]=array("a"=>"e","b"=>"eva","c"=>"evā","len"=>3,"adj_len"=>0);
  129. $sandhi[]=array("a"=>"i","b"=>"eva","c"=>"yevā","len"=>4,"adj_len"=>0);
  130. $sandhi[]=array("a"=>"ī","b"=>"eva","c"=>"yevā","len"=>4,"adj_len"=>0);
  131. $sandhi[]=array("a"=>"ī","b"=>"eva","c"=>"iyevā","len"=>4,"adj_len"=>0);
  132. $sandhi[]=array("a"=>"ī","b"=>"eva","c"=>"īyevā","len"=>4,"adj_len"=>0);
  133. $sandhi[]=array("a"=>"o","b"=>"eva","c"=>"ovā","len"=>4,"adj_len"=>0);
  134. $sandhi[]=array("a"=>"ā","b"=>"api","c"=>"āpi","len"=>3,"adj_len"=>0);
  135. $sandhi[]=array("a"=>"a","b"=>"api","c"=>"āpi","len"=>3,"adj_len"=>0);
  136. $sandhi[]=array("a"=>"e","b"=>"api","c"=>"epi","len"=>3,"adj_len"=>0);
  137. $sandhi[]=array("a"=>"ī","b"=>"api","c"=>"īpi","len"=>3,"adj_len"=>0);
  138. $sandhi[]=array("a"=>"i","b"=>"api","c"=>"īpi","len"=>3,"adj_len"=>0);
  139. $sandhi[]=array("a"=>"o","b"=>"api","c"=>"opi","len"=>3,"adj_len"=>0);
  140. $sandhi[]=array("a"=>"ū","b"=>"api","c"=>"ūpi","len"=>3,"adj_len"=>0);
  141. $sandhi[]=array("a"=>"u","b"=>"api","c"=>"ūpi","len"=>3,"adj_len"=>0);
  142. $sandhi[]=array("a"=>"u","b"=>"api","c"=>"upi","len"=>3,"adj_len"=>0);
  143. $sandhi[]=array("a"=>"ṃ","b"=>"api","c"=>"mpi","len"=>3,"adj_len"=>0);
  144. */
  145. //$sandhi[]=array("a"=>"a","b"=>"a","c"=>"a","len"=>1,"adj_len"=>-1,"advance"=>true);
  146. //$sandhi[]=array("a"=>"ī","b"=>"","c"=>"i","len"=>1,"adj_len"=>0,"advance"=>true);
  147. $search = array('aa', 'ae', 'ai', 'ao', 'au', 'aā', 'aī', 'aū', 'ea', 'ee', 'ei', 'eo', 'eu', 'eā', 'eī', 'eū', 'ia', 'ie', 'ii', 'io', 'iu', 'iā', 'iī', 'iū', 'oa', 'oe', 'oi', 'oo', 'ou', 'oā', 'oī', 'oū', 'ua', 'ue', 'ui', 'uo', 'uu', 'uā', 'uī', 'uū', 'āa', 'āe', 'āi', 'āo', 'āu', 'āā', 'āī', 'āū', 'īa', 'īe', 'īi', 'īo', 'īu', 'īā', 'īī', 'īū', 'ūa', 'ūe', 'ūi', 'ūo', 'ūu', 'ūā', 'ūī', 'ūū');
  148. $replace = array('a-a', 'a-e', 'a-i', 'a-o', 'a-u', 'a-ā', 'a-ī', 'a-ū', 'e-a', 'e-e', 'e-i', 'e-o', 'e-u', 'e-ā', 'e-ī', 'e-ū', 'i-a', 'i-e', 'i-i', 'i-o', 'i-u', 'i-ā', 'i-ī', 'i-ū', 'o-a', 'o-e', 'o-i', 'o-o', 'o-u', 'o-ā', 'o-ī', 'o-ū', 'u-a', 'u-e', 'u-i', 'u-o', 'u-u', 'u-ā', 'u-ī', 'u-ū', 'ā-a', 'ā-e', 'ā-i', 'ā-o', 'ā-u', 'ā-ā', 'ā-ī', 'ā-ū', 'ī-a', 'ī-e', 'ī-i', 'ī-o', 'ī-u', 'ī-ā', 'ī-ī', 'ī-ū', 'ū-a', 'ū-e', 'ū-i', 'ū-o', 'ū-u', 'ū-ā', 'ū-ī', 'ū-ū');
  149. $allword = array();
  150. foreach($arrWords as $oneword){
  151. //预处理
  152. //将双元音拆开
  153. $word = str_replace($search, $replace, $oneword);
  154. if(isset($_POST["debug"])){
  155. echo "Look up:{$word}<br>";
  156. }
  157. //按连字符拆开处理
  158. $arrword = str_getcsv($word,"-");
  159. $t1=microtime_float();
  160. $output = array();
  161. foreach($arrword as $oneword){
  162. $result = array();//全局变量,递归程序的输出容器
  163. if(mb_strlen($oneword,"UTF-8")<30){
  164. mySplit2($oneword,0,$_express);
  165. }
  166. else{
  167. mySplit2($oneword,0,$_express);
  168. }
  169. arsort($result);//按信心指数排序
  170. $wordlist = array();
  171. $iMax = 5;
  172. $iCount = 0;
  173. foreach($result as $row=>$value){
  174. $iCount++;
  175. $word_part = array();
  176. $word_part["word"] = $row;
  177. $word_part["confidence"] = $value;
  178. $wordlist[] = $word_part;
  179. if($iCount>=$iMax){
  180. break;
  181. }
  182. }
  183. $output[] = $wordlist;
  184. if(isset($_POST["debug"])){
  185. echo "<h2>{$oneword}</h2>";
  186. echo "<h4>".count($result)."</h4>";
  187. }
  188. $iCount=0;
  189. foreach($result as $row=>$value){
  190. if($iCount>10){
  191. break;
  192. }
  193. $iCount++;
  194. $level=$value*90;
  195. if(isset($_POST["debug"])){
  196. echo $row."-[".$value."]<br>";
  197. }
  198. }
  199. /*
  200. 后处理
  201. -ssāpi=-[ssa]-api
  202. */
  203. }
  204. $t2 = microtime_float();
  205. $one_split["data"]=$output;
  206. $one_split["time"]= $auto_split_times;
  207. $one_split["second"]= $t2-$t1;
  208. $allword[] = $one_split;
  209. if(isset($_POST["debug"])){
  210. echo "<div>";
  211. echo "<br>查询【{$auto_split_times}】次";
  212. echo "time:".($t2-$t1);
  213. echo "</div>";
  214. }
  215. }
  216. if(isset($_POST["debug"])){
  217. echo "<pre style='margin:2em;padding:1em;background-color:#e9e9e9;'>";
  218. print_r($allword);
  219. echo "</pre>";
  220. }
  221. echo json_encode($allword,JSON_UNESCAPED_UNICODE);
  222. /*
  223. 用于数组连接字符串
  224. */
  225. function myfunction($v1,$v2)
  226. {
  227. return $v1 . "+" . $v2;
  228. }
  229. function microtime_float()
  230. {
  231. list($usec, $sec) = explode(" ", microtime());
  232. return ((float)$usec + (float)$sec);
  233. }
  234. function dict_lookup($word){
  235. global $dbh;
  236. $query = "select weight from part where \"word\" = ? ";
  237. $stmt = $dbh->prepare($query);
  238. $stmt->execute(array($word));
  239. $row = $stmt->fetch(PDO::FETCH_NUM);
  240. if ($row) {
  241. return $row[0];
  242. } else {
  243. return 0;
  244. }
  245. }
  246. /*
  247. 查找某个单词是否在现有词典出现
  248. 返回信心指数
  249. */
  250. function isExsit($word,$adj_len=0){
  251. global $auto_split_times;
  252. global $result;
  253. global $part;
  254. global $confidence;
  255. $auto_split_times++;
  256. if(isset($_POST["debug"])){
  257. echo "<div>正在查询:{$word}</div>";
  258. }
  259. $isFound=false;
  260. if(isset($part["{$word}"]))
  261. {
  262. if($part["{$word}"]>0){
  263. $isFound=true;
  264. $count=$part["{$word}"]+1;
  265. }
  266. }
  267. else{
  268. $db=dict_lookup($word);
  269. //加入查询缓存
  270. $part["{$word}"] = $db;
  271. if($db>0){
  272. $isFound=true;
  273. $count=$db+1;
  274. }
  275. else{
  276. }
  277. }
  278. if($isFound)
  279. {
  280. if(isset($confidence["{$word}"])){
  281. $cf=$confidence["{$word}"];
  282. }
  283. else{
  284. $len=mb_strlen($word,"UTF-8")+$adj_len;
  285. $len_correct=1.2;
  286. $count2=1.1+pow($count,1.18);
  287. $conf_num=pow(1/$count2,pow(($len-1),$len_correct));
  288. $cf=round(1/(1+640*$conf_num),9);
  289. $confidence["{$word}"]=$cf;
  290. }
  291. return($cf);
  292. }
  293. else{
  294. return(-1);
  295. }
  296. }
  297. /*
  298. 核心拆分函数
  299. $strWord, 要查询的词
  300. $deep, 当前递归深度
  301. $express=true, 快速查询
  302. $adj_len=0 长度校正系数
  303. $c_threshhold 信心指数阈值
  304. */
  305. function mySplit2($strWord,$deep,$express=false,$adj_len=0,$c_threshhold=0.8){
  306. global $path;
  307. global $result;
  308. global $sandhi ;
  309. $output = array();
  310. //达到最大搜索深度,返回
  311. if($deep>=16){
  312. $word = "";
  313. $cf=1.0;
  314. for($i=0;$i<$deep;$i++){
  315. $word .= $path[$i][0];
  316. if(isset($_POST["debug"])){
  317. $word .="(".$path[$i][1].")-";
  318. }
  319. else{
  320. $word .= "-";
  321. }
  322. $cf=$cf*$path[$i][1];
  323. }
  324. $len=pow(mb_strlen($strWord,"UTF-8"),3);
  325. $cf+=(0-$len)/($len+150);
  326. $word .= "{$strWord}";
  327. $result[$word]=$cf;
  328. return;
  329. }
  330. //直接找到
  331. $confidence=isExsit($strWord,$adj_len);
  332. if($confidence>=0){
  333. $output[] = array($strWord,"",$confidence);
  334. }
  335. else{
  336. $confidence=isExsit("[".$strWord."]");
  337. if($confidence>=0){
  338. $output[] = array("[".$strWord."]","",$confidence);
  339. }
  340. }
  341. //如果开头有双辅音,去掉第一个辅音。因为巴利语中没有以双辅音开头的单词。
  342. $doubleword="kkggccjjṭṭḍḍttddppbb";
  343. if(mb_strlen($strWord,"UTF-8")>2){
  344. $left2=mb_substr($strWord,0,2,"UTF-8");
  345. if(mb_strpos($doubleword,$left2,0,"UTF-8")!==FALSE){
  346. $strWord=mb_substr($strWord,1,NULL,"UTF-8");
  347. }
  348. }
  349. $len=mb_strlen($strWord,"UTF-8");
  350. if($len>2){
  351. for($i=$len;$i>1;$i--){
  352. foreach($sandhi as $row){
  353. if(mb_substr($strWord,$i-$row["len"],$row["len"],"UTF-8")==$row["c"]){
  354. $str1=mb_substr($strWord,0,$i-$row["len"],"UTF-8").$row["a"];
  355. $str2=$row["b"].mb_substr($strWord,$i,NULL,"UTF-8");
  356. $confidence=isExsit($str1,$adj_len);
  357. if($confidence >= $c_threshhold){
  358. $output[] = array($str1,$str2,$confidence,$row["adj_len"]);
  359. if($express){
  360. break;
  361. }
  362. }
  363. }
  364. }
  365. }
  366. }
  367. if(count($output)>0){
  368. foreach($output as $part){
  369. $path[$deep][0]=$part[0];
  370. $path[$deep][1]=$part[2];
  371. if($part[1]!=""){
  372. mySplit2($part[1],($deep+1),$express,$part[3],$c_threshhold);
  373. }
  374. else{
  375. $word = "";
  376. $cf=1.0;
  377. for($i=0;$i<$deep;$i++){
  378. $word .= $path[$i][0]."+";
  379. if(isset($_POST["debug"])){
  380. $word .= "(".$path[$i][1].")-";
  381. }
  382. $cf=$cf*$path[$i][1];
  383. }
  384. $word .= $part[0];
  385. if(isset($_POST["debug"])){
  386. $word .= "({$part[2]})";
  387. }
  388. $cf=$cf+$part[2]*0.1;
  389. if($cf >= $c_threshhold){
  390. $result[$word]=$cf;
  391. }
  392. }
  393. }
  394. }
  395. else{
  396. $word = "";
  397. $cf=1.0;
  398. for($i=0;$i<$deep;$i++){
  399. $word .= $path[$i][0]."+";
  400. if(isset($_POST["debug"])){
  401. $word .= "(".$path[$i][1].")-";
  402. }
  403. $cf=$cf*$path[$i][1];
  404. }
  405. $len=pow(mb_strlen($strWord,"UTF-8"),3);
  406. $cf+=(0-$len)/($len+150);
  407. if(isset($_POST["debug"])){
  408. $word .= $strWord."(0)";
  409. }
  410. else{
  411. $word .= $strWord;
  412. }
  413. if($cf >= $c_threshhold){
  414. $result[$word]=$cf;
  415. }
  416. }
  417. }
  418. ?>