This script has the feature that extract English wikipedia's category has only alphabets, digits, and spaces. %php category_enals.php<database> <output_file>

category_enals.php

<?php
  $db = "127.0.0.1";
  $db_name = $argv[1];
  $table = "category";

  $username = "mysql";
  $password = "";

  $output_file = $argv[2];

  $con = mysql_connect($db, $username, $password);
  if (!$con) {
  exit('Database connection error');
  }

  $result = mysql_select_db($db_name, $con);
  if (!$result) {
  exit('Database select error');
  }

  $fw = fopen($output_file, 'a+');

  $startTime = microtime(true);

  $query = "SELECT cat_title FROM $table;";
  $result = mysql_query($query, $con);
  if(!$result){
    echo "SELECT ERROR aborted" . "\n";
    $con = mysql_close($con);
    if (!$con) {
      echo 'Database close error';
    }
    fclose($fw);
    exit;
  }

  while ($item = mysql_fetch_array($result)){
    $title = $item['cat_title'];
    mb_regex_encoding("UTF-8");
    if (preg_match("/^[a-zA-Z0-9 ]+$/u",$title)) {
      $title = $title;
      fwrite($fw,$title . "\n");
    }
  }
  $endTime = microtime(true);
  $elapsedTime = $endTime - $startTime;


  $con = mysql_close($con);
  if (!$con) {
  exit('Database close error');
  }

  fclose($fw);
  echo "Done.\n";

?>