php遍历树的常用方法汇总_php技巧

一、递归的深度优先的算法：

<?php
define('DS', DIRECTORY_SEPARATOR);
function rec_list_files($from = '.')
{
  if(!is_dir($from)) {
    return array();
  }
  $files = array();
  if($dh = opendir($from))
  {
    while(false !== ($file = readdir($dh))) {
      if($file == '.' || $file == '..') {
        continue;
      }
      $path = $from . DS . $file;
       
      if (is_file($path)) {
        $files[] = $path;
      }
      $files = array_merge($files, rec_list_files($path));
    }
    closedir($dh);
  }
  return $files;
}
function profile($func, $trydir)
{
  $mem1 = memory_get_usage();
  echo '<pre>----------------------- Test run for '.$func.'() ';
  flush();
  $time_start = microtime(true);
  $list = $func($trydir);
  //print_r($list);
  $time = microtime(true) - $time_start;
  echo 'Finished : '.count($list).' files</pre>';
  $mem2 = memory_get_peak_usage();
  printf('<pre>Max memory for '.$func.'() : %0.2f kbytes Running time for '.$func.'() : %0.f s</pre>',
  ($mem2-$mem1)/1024.0, $time);
  return $list;
}
profile('rec_list_files', "D:wwwserver");
?>

二、递归的深度优先的算法（用了一个栈来实现）

<?php
define('DS', DIRECTORY_SEPARATOR);
function deep_first_list_files($from = '.')
{
  if(!is_dir($from)) {
    return false;
  }
  $files = array();
  $dirs = array($from);
  while(NULL !== ($dir = array_pop($dirs))) {
    if( $dh = opendir($dir)) {
      while( false !== ($file = readdir($dh))) {
        if($file == '.' || $file == '..') {
          continue;
        }
        $path = $dir . DS . $file;
        if(is_dir($path)) {
          $dirs[] = $path;
        } else {
          $files[] = $path;
        }
      }
      closedir($dh);
    }
  }
  return $files;
}
function profile($func, $trydir)
{
  $mem1 = memory_get_usage();
  echo '<pre>----------------------- Test run for '.$func.'() ';
  flush();
  $time_start = microtime(true);
  $list = $func($trydir);
  //print_r($list);
  $time = microtime(true) - $time_start;
  echo 'Finished : '.count($list).' files</pre>';
  $mem2 = memory_get_peak_usage();
  printf('<pre>Max memory for '.$func.'() : %0.2f kbytes Running time for '.$func.'() : %0.f s</pre>',
  ($mem2-$mem1)/1024.0, $time);
  return $list;
}
profile('deep_first_list_files', "D:wwwserver");
?>

三、非递归的广度优先算法（用了一个队列来实现）

<?php
define('DS', DIRECTORY_SEPARATOR);
function breadth_first_files($from = '.') {
  $queue = array(rtrim($from, DS).DS);// normalize all paths
  $files = array();
  while($base = array_shift($queue )) {
    if (($handle = opendir($base))) {
      while (($child = readdir($handle)) !== false) {
        if( $child == '.' || $child == '..') {
          continue;
        }
        if (is_dir($base.$child)) {
          $combined_path = $base.$child.DS;
          array_push($queue, $combined_path);
        } else {
          $files[] = $base.$child;
        }
      }
      closedir($handle);
    } // else unable to open directory => NEXT CHILD
  }
  return $files; // end of tree, file not found
}
function profile($func, $trydir)
{
  $mem1 = memory_get_usage();
  echo '<pre>----------------------- Test run for '.$func.'() ';
  flush();
  $time_start = microtime(true);
  $list = $func($trydir);
  //print_r($list);
  $time = microtime(true) - $time_start;
  echo 'Finished : '.count($list).' files</pre>';
  $mem2 = memory_get_peak_usage();
  printf('<pre>Max memory for '.$func.'() : %0.2f kbytes Running time for '.$func.'() : %0.f s</pre>',
  ($mem2-$mem1)/1024.0, $time);
  return $list;
}
profile('breadth_first_files', "D:wwwserver");
?>

希望本文所述对大家的php程序设计有所帮助。