【SPL標(biāo)準(zhǔn)庫專題(2)】Iterator

focusj 發(fā)布于2019-07-01 12:47 / 2440人閱讀

摘要：由于沒有方法，所以就不能用重置游標(biāo)，這個(gè)時(shí)候已經(jīng)到最后了，所以為空一種占位符形式的迭代器，不執(zhí)行任何操作。檢查是否含有子節(jié)點(diǎn)輸出所有字節(jié)點(diǎn)將一個(gè)樹形結(jié)構(gòu)的迭代器展開為一維結(jié)構(gòu)。

Iterator界面

本段內(nèi)容來自阮一峰老師再加自己的部分注解

SPL規(guī)定，所有部署了Iterator界面的class，都可以用在foreach Loop中。Iterator界面中包含5個(gè)必須部署的方法：

* current()

This method returns the current index"s value. You are solely
responsible for tracking what the current index is as the 
interface does not do this for you. (返回當(dāng)前索引值)

* key()

This method returns the value of the current index"s key. For 
foreach loops this is extremely important so that the key 
value can be populated. (返回當(dāng)前的索引key)

* next()

This method moves the internal index forward one entry. (迭代中的內(nèi)部指針往前進(jìn)一步)

* rewind()

This method should reset the internal index to the first element. (重置迭代中的內(nèi)部指針)

* valid()

This method should return true or false if there is a current 
element. It is called after rewind() or next(). (驗(yàn)證內(nèi)部指針是否到最后一行)

Example

class ArrayReloaded implements Iterator {
  /**
   * 如前一篇文章所說,該類實(shí)現(xiàn)了Iterator接口,所以該類對(duì)象就是ZEND_ITER_OBJECT,對(duì)于ZEND_ITER_OBJECT的類對(duì)象,會(huì)通過調(diào)用對(duì)象實(shí)現(xiàn)的Iterator接口相關(guān)函數(shù)來進(jìn)行foreach。
   */
  private $array = array();
  private $valid = FALSE;

  function __construct($array) {
    $this->array = $array;
  }

  function rewind(){
    /**
     *  reset: 將數(shù)組的內(nèi)部指針指向第一個(gè)單元,如果數(shù)組為空則返回false;
     *  所以下述語句表示: 數(shù)組不為空并且已重置到第一個(gè)單元;
     */
    $this->valid = (FALSE !== reset($this->array));
  }

  function current(){
    return current($this->array);
  }

  function key(){
    return key($this->array);
  }

  function next(){
    /**
     * next: 將數(shù)組中的內(nèi)部指針向前移動(dòng)一位
     * 返回?cái)?shù)組內(nèi)部指針指向的下一個(gè)單元的值，或當(dāng)沒有更多單元時(shí)返回 FALSE。
     * 所以下述語句表示: 如果還有下一個(gè)單元的話,指針移動(dòng)到下個(gè)單元并返回true;
     */
    $this->valid = (FALSE !== next($this->array));
  }

  function valid(){
    return $this->valid;
  }

  #↑↑ 以上5個(gè)方法是必須實(shí)現(xiàn)的接口方法,也可以再擴(kuò)展prev和end等方法,后續(xù)會(huì)介紹一些SPL內(nèi)置的實(shí)現(xiàn)了itertor接口的類,這些類可以拿來直接使用
}

$arr = array(
  "color1" => "red",
  "color2" => "blue",
  "color3" => "green",
  "color4" => "plack",
  "color5" => "purple"
);

$colors = new ArrayReloaded($arr);

# 通過foreach來遍歷
foreach($colors as $k => $v){
  echo $k.":".$v."
";
}

# 通過while來遍歷
/**
 * 1: foreach的內(nèi)部實(shí)現(xiàn)方式其實(shí)也是如此,事實(shí)上直接用while來遍歷性能更高
 * 2: 在使用迭代器來遍歷的時(shí)候,一定要記住要rewind和next,而PHP的foreach遍歷早已把rewind和next給集成了;
 */
echo "";
$colors -> rewind();
while($colors -> valid()){
  echo $colors -> key().":".$colors -> current()."
";
  $colors -> next();
}

ArrayIterator

ArrayIterator implements ArrayAccess , SeekableIterator , Countable , Serializable

這是一個(gè)非常有用的迭代器,里面實(shí)現(xiàn)了排序,添加,篩選等foreach不能直接實(shí)現(xiàn)的方法(都是要全部遍歷出來再進(jìn)行判斷處理,代碼不優(yōu)雅維護(hù)性差)

Example

 "red",
  "color3" => "green",
  "color4" => "plack",
  "color2" => "blue",
  "color5" => "purple"
);

// $colors = new ArrayIterator($arr); //可以直接通過實(shí)例一個(gè)數(shù)組迭代器對(duì)象,然后while這個(gè)迭代器;

//但以下的方式要更易于擴(kuò)展
//先實(shí)例一個(gè)array對(duì)象
$colorsObj = new ArrayObject($arr);
$it = $colorsObj -> getIterator(); //獲得當(dāng)前的ArrayIterator

# 通過while來遍歷
$it -> rewind();
while($it -> valid()){
  echo $it -> key().":".$it -> current()."
";
  $it -> next();
}

#通過iterator迭代器來遍歷就變得很靈活
echo $colorsObj -> count(); //元素?cái)?shù)量統(tǒng)計(jì)

//從第三個(gè)開始遍歷
$it -> rewind(); //凡是要使用迭代器之前先重置;
if($it -> valid()){
  $it -> seek(2); //從0開始的,第二個(gè)位置
  while($it -> valid()){
    echo $it -> key().":".$it -> current()."
";
    $it -> next();
  }
}

//對(duì)索引名進(jìn)行升序排列
$it -> ksort();
foreach($it as $k => $v){
  echo $k ."-->". $v."
";
}

//對(duì)索引值進(jìn)行排序
$it -> asort();
foreach($it as $k => $v){
  echo $k ."-->". $v."
";
}

//這些對(duì)象方法是否很熟悉? 這就是上一篇文章中說到的 "SPL是一種使object（物體）模仿array（數(shù)組）行為的interfaces和classes"

AppendIterator

按順序迭代訪問幾個(gè)不同的迭代器。例如，希望在一次循環(huán)中迭代訪問兩個(gè)或者更多的組合。這個(gè)迭代器的append方法類似于array_merge()函數(shù)來合并數(shù)組。

Example

$arr1 = array(
  "color1" => "red",
  "color3" => "green",
  "color4" => "plack",
  "color2" => "blue",
  "color5" => "purple"
);

$arr2 = array(
  "fruit1" => "apple",
  "fruit2" => "orange",
  "fruit3" => "banana",
  "fruit4" => "grape",
  "fruit5" => "strawberry",
);

$ao1 = new ArrayIterator($arr1);
$ao2 = new ArrayIterator($arr2);
$iterator = new AppendIterator();
$iterator -> append($ao1);
$iterator -> append($ao2);
foreach($iterator as $k => $v){
  echo $k.":".$v,"
";
}

MultipleIterator

迭代器的鏈接器，更多參考連接 http://php.net/manual/en/clas...

$person_id = new ArrayIterator(array("001", "002", "003")); 
$person_name = new ArrayIterator(array("張三", "李四", "王五")); 
$person_age = new ArrayIterator(array(22, 23, 11)); 
$mit = new MultipleIterator(MultipleIterator::MIT_KEYS_ASSOC); 
$mit->attachIterator($person_id, "ID"); 
$mit->attachIterator($person_name, "NAME"); 
$mit->attachIterator($person_age, "AGE"); 
echo"連接的迭代器個(gè)數(shù):".$mit->countIterators() . "
"; //3 
foreach ($mit as $person) { 
    print_r($person); 
} 
/**output
Array
(
    [ID] => 001
    [NAME] => 張三
    [AGE] => 22
)
Array
(
    [ID] => 002
    [NAME] => 李四
    [AGE] => 23
)
Array
(
    [ID] => 003
    [NAME] => 王五
    [AGE] => 11
)
**/

LimitIterator

返回給定數(shù)量的結(jié)果以及從集合中取出結(jié)果的起始索引點(diǎn)


FilterIterator
基于OuterIterator接口，用于過濾數(shù)據(jù)，返回符合條件的元素。必須實(shí)現(xiàn)一個(gè)抽象方法accept()，此方法必須為迭代器的當(dāng)前項(xiàng)返回true或false
class UserFilter extends FilterIterator{
  private $userFilter;

  public function  __construct(Iterator $iterator , $filter){
    parent::__construct($iterator);
    //要過濾的參數(shù)
    $this->userFilter = $filter;
  }

  public function accept(){
    /*
     * getInnerIterator(): 獲得內(nèi)部的迭代器
     * current(): 然后獲取當(dāng)前的元素
     * in strcmp(string str1,string str2) 區(qū)分字符串中字母大小寫地比較,返回0就相同
     * int strcasecmp(string str1,string str2) 忽略字符串中字母大小寫地比較，返回0就相同
     * 如果accept返回false的話就過濾掉
     */
    $user = $this->getInnerIterator()->current();
    if (strcasecmp($user["name"] , $this->userFilter) == 0) {
      return false;
    }
    return true;
  }
}

$array = array (array ("name" => "Jonathan" , "id" => "5") , array ("name" => "Abdul" , "id" => "22"),array ("name" => "zhouzhou" , "id" => "9"));
$object = new ArrayObject($array);
//去除掉名為abdul的人員
$iterator = new UserFilter($object->getIterator() , "abdul");
foreach ($iterator as $result) {
  echo $result["name"];
}

/**output
 * Jonathan
 **/
RegexIterator
繼承FilterIterator，支持使用正則表達(dá)式模式匹配和修改迭代器中的元素。經(jīng)常用于將字符串匹配。
更多參考： http://cn2.php.net/manual/zh/...
//可以實(shí)現(xiàn):  preg_match_all(), preg_match(), preg_replace(),preg_split()等函數(shù)的功能

$a = new ArrayIterator(array("test1", "test2", "test3"));
$i = new RegexIterator($a, "/^(test)(d+)/", RegexIterator::REPLACE);
$i->replacement = "$2:$1";
print_r(iterator_to_array($i));

/**output
Array
(
[0] => 1:test
[1] => 2:test
[2] => 3:test
)
 **/
IteratorIterator
一種通用類型的迭代器，所有實(shí)現(xiàn)了Traversable接口的類都可以被它迭代訪問。
CachingIterator
用來執(zhí)行提前讀取一個(gè)元素的迭代操作，例如可以用于確定當(dāng)前元素是否為最后一個(gè)元素。
$array = array ("koala" , "kangaroo" , "wombat" , "wallaby" , "emu" , "kiwi" , "kookaburra" , "platypus");
$object = new CachingIterator(new ArrayIterator($array));
foreach ($object as $value) {
  echo $value;
  if ($object->hasNext()) {
    echo ","; //如果有下一項(xiàng)的話才輸出 突出不了該迭代器的作用啊，其他迭代器也可以搞定的
  }
}
/**output
 * koala,kangaroo,wombat,wallaby,emu,kiwi,kookaburra,platypus
 **/
SeekableIterator
用于創(chuàng)建非順序訪問的迭代器，允許跳轉(zhuǎn)到迭代器中的任何一點(diǎn)上。
$array = array("apple", "banana", "cherry", "damson", "elderberry");
$iterator = new ArrayIterator($array);
$iterator->seek(3); //起始0從第3個(gè)開始取；
echo $iterator->current()."
";
/**output
damson
 **/
NoRewindIterator
用于不能多次迭代的集合，適用于在迭代過程中執(zhí)行一次性操作。
$fruit = array("apple", "banana", "cranberry");
$arr = new ArrayObject($fruit);
$it = new NoRewindIterator($arr->getIterator());
echo "Fruit A:
";
foreach ($it as $item) {
  echo $item . "
";
}

echo "Fruit B:
"; 

// ↓↓ 由于NoRewindIterator沒有rewind方法，所以foreach就不能用rewind重置游標(biāo)，這個(gè)時(shí)候$it已經(jīng)到最后了，所以為空；
foreach ($it as $item) {
  echo $item . "
";
}

/**output
Fruit A:
apple
banana
cranberry
Fruit B:
 **/
EmptyIterator
一種占位符形式的迭代器，不執(zhí)行任何操作。當(dāng)要實(shí)現(xiàn)某個(gè)抽象類的方法并且這個(gè)方法需要返回一個(gè)迭代器時(shí)，可以使用這種迭代器。
InfiniteIterator
用于持續(xù)地訪問數(shù)據(jù)，當(dāng)?shù)阶詈笠粋€(gè)元素時(shí)，會(huì)再次從第一個(gè)元素開始迭代訪問。
$arrayit = new ArrayIterator(array("cat", "dog"));
$infinite = new InfiniteIterator($arrayit);

//必須限制否則就是死循環(huán)
$limit = new LimitIterator($infinite, 0, 7);
foreach ($limit as $value) {
  echo "$value
";
}
RecursiveArrayIterator
創(chuàng)建一個(gè)用于遞歸形式數(shù)組結(jié)構(gòu)的迭代器，類似于多維數(shù)組.它為許多更復(fù)雜的迭代器提供了所需的操作，如RecursiveTreeIterator和RecursiveIteratorIterator迭代器。
$fruits = array("a" => "lemon", "b" => "orange", array("a" => "apple", "p" => "pear"));
$iterator = new RecursiveArrayIterator($fruits);
while ($iterator->valid()) {
  //檢查是否含有子節(jié)點(diǎn)
  if ($iterator->hasChildren()) {
    //輸出所有字節(jié)點(diǎn)
    foreach ($iterator->getChildren() as $key => $value) {
      echo $key . " : " . $value . "
";
    }
  } else {
    echo "No children.
";
  }
  $iterator->next();
}

/**output
No children.
No children.
a : apple
p : pear
 **/
RecursiveIteratorIterator
將一個(gè)樹形結(jié)構(gòu)的迭代器展開為一維結(jié)構(gòu)。
$fruits = array("a" => "lemon", "b" => "orange", array("a" => "apple", "p" => "pear","c" => ["a","b"]));
$arrayiter = new RecursiveArrayIterator($fruits);
$iteriter = new RecursiveIteratorIterator($arrayiter);
foreach ($iteriter as $key => $value) {
  $d = $iteriter->getDepth();
  echo "depth=$d k=$key v=$value
";
}
/**output
depth=0 k=a v=lemon
depth=0 k=b v=orange
depth=1 k=a v=apple
depth=1 k=p v=pear
depth=2 k=0 v=a
depth=2 k=1 v=b
 **/
RecursiveTreeIterator
以可視在方式顯示一個(gè)樹形結(jié)構(gòu)。
$hey = array("a" => "lemon", "b" => "orange", array("a" => "apple", "p" => "pear"));
$awesome = new RecursiveTreeIterator(
  new RecursiveArrayIterator($hey),
  null, null, RecursiveIteratorIterator::LEAVES_ONLY
);
foreach ($awesome as $line)
  echo $line . PHP_EOL;

/**output
|-lemon
|-orange
  |-apple
  -pear
 **/
ParentIterator
是一個(gè)擴(kuò)展的FilterIterator迭代器，它可以過濾掉來自于RecursiveIterator迭代器的非父元素，只找出子節(jié)點(diǎn)的鍵值。通俗來說，就是去枝留葉。
$hey = array ("a" => "lemon" , "b" => "orange" , array ("a" => "apple" , "p" => "pear"));
$arrayIterator = new RecursiveArrayIterator($hey);
$it = new ParentIterator($arrayIterator);
print_r(iterator_to_array($it));
/**output
 * Array
 * (
 * [0] => Array
 * (
 * [a] => apple
 * [p] => pear
 * )
 * )
 **/
RecursiveFilterIterator
是FilterIterator迭代器的遞歸形式，也要求實(shí)現(xiàn)抽象的accept()方法，但在這個(gè)方法中應(yīng)該使用$this->getInnerIterator()方法訪問當(dāng)前正在迭代的迭代器。
class TestsOnlyFilter extends RecursiveFilterIterator{
  public function accept(){
    // 找出含有“葉”的元素
    return $this->hasChildren() || (mb_strpos($this->current() , "葉") !== false);
  }
}

$array = array ("葉1" , array ("力2" , "葉3" , "葉4") , "葉5");
$iterator = new RecursiveArrayIterator($array);
$filter = new TestsOnlyFilter($iterator);
$filter = new RecursiveIteratorIterator($filter);
print_r(iterator_to_array($filter));
/**output
 * Array
 * (
 * [0] => 葉1
 * [1] => 葉3 //只會(huì)找出含葉的元素，不會(huì)把元素成員全部顯示出來
 * [2] => 葉5
 * )
 **/
RecursiveRegexIterator
是RegexIterator迭代器的遞歸形式，只接受RecursiveIterator迭代器作為迭代對(duì)象。
$rArrayIterator = new RecursiveArrayIterator(array ("葉1" , array ("tet3" , "葉4" , "葉5")));
$rRegexIterator = new RecursiveRegexIterator(
  $rArrayIterator , "/^葉/" , RecursiveRegexIterator::ALL_MATCHES
);
foreach ($rRegexIterator as $key1 => $value1) {
  if ($rRegexIterator->hasChildren()) {
    // print all children
    echo "Children: ";
    foreach ($rRegexIterator->getChildren() as $key => $value) {
      echo $value . " ";
    }
    echo "
";
  } else {
    echo "No children
";
  }
}
/**output
 * No children
 * Children: 葉4 葉5
 **/
RecursiveCachingIterator
在RecursiveIterator迭代器上執(zhí)行提前讀取一個(gè)元素的遞歸操作。
CallbackFilterIterator(PHP5.4)
同時(shí)執(zhí)行過濾和回調(diào)操作，在找到一個(gè)匹配的元素之后會(huì)調(diào)用回調(diào)函數(shù)。
$hey = array ("李1" , "葉2" , "葉3" , "葉4" , "葉5" , "葉6" ,);
$arrayIterator = new RecursiveArrayIterator($hey);

$isYe = function($current){
  return mb_strpos($current , "葉") !== false;
};

$rs = new CallbackFilterIterator($arrayIterator , $isYe);
print_r(iterator_to_array($rs));

/**output
 * Array
 * (
 * [0] => 葉2
 * [1] => 葉3
 * [2] => 葉4
 * [3] => 葉5
 * [4] => 葉6
 * )
 **/
RecursiveCallbackFilterIterator(PHP5.4)
在RecursiveIterator迭代器上進(jìn)行遞歸操作，同時(shí)執(zhí)行過濾和回調(diào)操作，在找到一個(gè)匹配的元素之后會(huì)調(diào)用回調(diào)函數(shù)。
function doesntStartWithLetterT($current){
  $rs = $current->getFileName();
  return $rs[0] !== "T";
}

$rdi = new RecursiveDirectoryIterator(__DIR__);
$files = new RecursiveCallbackFilterIterator($rdi , "doesntStartWithLetterT");
foreach (new RecursiveIteratorIterator($files) as $file) {
  echo $file->getPathname() . PHP_EOL;
}

DirectoryIterator
目錄文件遍歷器，提供了查詢當(dāng)前文件的所有信息的方法（是否可讀可寫，所屬，權(quán)限等等），具體參考 http://cn2.php.net/manual/zh/...
$it = new DirectoryIterator("../");
foreach ($it as $file) {
  //用isDot ()方法分別過濾掉“.”和“..”目錄
  if (!$it->isDot()) {
    echo $file . "
";
  }
}

RecursiveDirectoryIterator
遞歸目錄文件遍歷器，可實(shí)現(xiàn)列出所有目錄層次結(jié)構(gòu)，而不是只操作一個(gè)目錄。具體看：http://cn2.php.net/manual/zh/...
//列出指定目錄中所有文件
$path = realpath("../");
$objects = new RecursiveIteratorIterator(new RecursiveDirectoryIterator($path) , RecursiveIteratorIterator::SELF_FIRST);
foreach ($objects as $name => $object) {
  echo "$name
";
}
FilesystemIterator
是DirectoryIterator的遍歷器
$it = new FilesystemIterator("../");
foreach ($it as $fileinfo) {
  echo $fileinfo->getFilename() . "
";
}

GlobIterator
帶匹配模式的文件遍歷器
$iterator = new GlobIterator("*.php");
if (!$iterator->count()) {
  echo "無php文件";
} else {
  $n = 0;
  printf("總計(jì) %d 個(gè)php文件
" , $iterator->count());
  foreach ($iterator as $item) {
    printf("[%d] %s
" , ++ $n , $iterator->key());
  }
}
SimpleXMLIterator
XMl文檔訪問迭代器，可實(shí)現(xiàn)訪問xml中所有節(jié)點(diǎn)
$xml = <<
        
            PHP Basics
            Jim Smith
        
        XML basics

XML;
// SimpleXML轉(zhuǎn)換為數(shù)組
function sxiToArray($sxi){
  $a = array ();
  for ($sxi->rewind();$sxi->valid();$sxi->next()) {
    if (!array_key_exists($sxi->key() , $a)) {
      $a[$sxi->key()] = array ();
    }
    if ($sxi->hasChildren()) {
      $a[$sxi->key()][] = sxiToArray($sxi->current());
    } else {
      $a[$sxi->key()][] = strval($sxi->current());
    }
  }
  return $a;
}

$xmlIterator = new SimpleXMLIterator($xml);
$rs = sxiToArray($xmlIterator);
print_r($rs);
/**output
 * Array
 * (
 * [book] => Array
 * (
 * [0] => Array
 * (
 * [title] => Array
 * (
 * [0] => PHP Basics
 * )
 * [author] => Array
 * (
 * [0] => Jim Smith
 * )
 * )
 * [1] => XML basics
 * )
 * )
 **/
參考鏈接：
http://www.ruanyifeng.com/blo...
http://www.cnblogs.com/Script...