这篇笔记用来整顿Collection 在Laravel 的理论利用场景。
求和
需要：遍历$orders 数组，求price 的和。
<?php// 引入packagerequire __DIR__ . '/vendor/autoload.php';$orders = [[ 'id' => 1, 'user_id' => 1, 'number' => '13908080808', 'status' => 0, 'fee' => 10, 'discount' => 44, 'order_products'=> [ ['order_id'=>1,'product_id'=>1,'param'=>'6寸','price'=>555.00,'product'=>['id'=>1,'name'=>'蛋糕名称','images'=>[]]], ['order_id'=>1,'product_id'=>1,'param'=>'7寸','price'=>333.00,'product'=>['id'=>1,'name'=>'蛋糕名称','images'=>[]]], ],]];- 应用传统的foreach 形式进行遍历：
$sum = 0;foreach ($orders as $order) { foreach ($order['order_products'] as $item) { $sum += $item['price']; }}echo $sum;- 应用汇合的map、flatten、sum：
$sum = collect($orders)->map(function($order){ return $order['order_products'];})->flatten(1)->map(function($order){ return $order['price'];})->sum();echo $sum;map：遍历汇合，返回一个新的汇合。
flatten：将多维数组转换为一维。
sum：返回数组的和。
- 应用汇合的flatMap、pluck、sum：
$sum = collect($orders)->flatMap(function($order){ return $order['order_products'];})->pluck('price')->sum();echo $sum;flatMap：和map 相似，不过区别在于flatMap 能够间接应用返回的新汇合。
- 应用汇合的flatMap、sum：
$sum = collect($orders)->flatMap(function($order){ return $order['order_products'];})->sum('price');sum：能够接管一个列名作为参数进行求和。
格式化数据
需要：将如下构造的数组，格式化成上面的新数组。
// 带格式化数组$gates = [ 'BaiYun_A_A17', 'BeiJing_J7', 'ShuangLiu_K203', 'HongQiao_A157', 'A2', 'BaiYun_B_B230'];// 新数组$boards = [ 'A17', 'J7', 'K203', 'A157', 'A2', 'B230'];- 应用foreach 进行遍历：
$res = [];foreach($gates as $key => $gate) { if(strpos($gate, '_') === false) { $res[$key] = $gate; }else{ $offset = strrpos($gate, '_') + 1; $res[$key] = mb_substr($gate , $offset); }}var_dump($res);- 应用汇合的map以及php 的explode、end：
$res = collect($gates)->map(function($gate) { $parts = explode('_', $gate); return end($parts);});- 应用汇合的map、explode、last、toArray：
$res = collect($gates)->map(function($gate) { return collect(explode('_', $gate))->last();})->toArray();explode：将字符串进行宰割成数组
last：获取最初一个元素
统计GitHub Event
首先，通过此链接获取到集体事件json。
一个 PushEvent计 5 分，一个 CreateEvent 计 4 分，一个 IssueCommentEvent计 3 分，一个 IssueCommentEvent 计 2 分，除此之外的其它类型的事件计 1 分，计算以后用户的工夫得分总和。
$opts = [ 'http' => [ 'method' => 'GET', 'header' => [ 'User-Agent: PHP' ] ]];$context = stream_context_create($opts);$events = json_decode(file_get_contents('http://api.github.com/users/0xAiKang/events', false, $context), true);- 传统foreach 形式：
$eventTypes = []; // 事件类型$score = 0; // 总得分foreach ($events as $event) { $eventTypes[] = $event['type'];}foreach($eventTypes as $eventType) { switch ($eventType) { case 'PushEvent': $score += 5; break; case 'CreateEvent': $score += 4; break; case 'IssueEvent': $score += 3; break; case 'IssueCommentEvent': $score += 2; break; default: $score += 1; break; }}- 应用汇合的map、pluck、sum 办法：
$score = $events->pluck('type')->map(function($eventType) { switch ($eventType) { case 'PushEvent': return 5; case 'CreateEvent': return 4; case 'IssueEvent': return 3; case 'IssueCommentEvent': return 2; default: return 1; }})->sum();应用汇合的链式编程，能够很好地解决下面那种屡次遍历的问题。
- 应用汇合中的map、pluck、get 办法：
$score = $events->pluck('type')->map(function($eventType) { return collect([ 'PushEvent'=> 5, 'CreateEvent'=> 4, 'IssueEvent'=> 3, 'IssueCommentEvent'=> 2 ])->get($eventType, 1); // 如果不存在则默认等于1})->sum();- 尝试将该需要，封装成一个类：
class GithubScore { private $events; private function __construct($events){ $this->events = $events; } public static function score($events) { return (new static($events))->scoreEvents(); } private function scoreEvents() { return $this->events->pluck('type')->map(function($eventType){ return $this->lookupEventScore($eventType, 1); })->sum(); } public function lookupEventScore($eventType, $default_value) { return collect([ 'PushEvent'=> 5, 'CreateEvent'=> 4, 'IssueEvent'=> 3, 'IssueCommentEvent'=> 2 ])->get($eventType, $default_value); // 如果不存在则默认等于1 }}var_dump(GithubScore::score($events));格式化数据
需要：将以下数据格式化成新的构造。
$messages = [ 'Should be working now for all Providers.', 'If you see one where spaces are in the title let me know.', 'But there should not have blank in the key of config or .env file.'];// 格式化之后的后果- Should be working now for all Providers. \n- If you see one where spaces are in the title let me know. \n- But there should not have blank in the key of config or .env file.- 传统的foreach 形式：
$comment = '- ' . array_shift($messages);foreach ($messages as $message) { $comment .= "\n - ${message}";}var_dump($comment);- 应用汇合的map、implode办法：
$comment = collect($messages)->map(function($message){ return '- ' . $message;})->implode("\n");var_dump($comment);多个数组求差
需要：两组数据别离代表去年的营收和往年的营收，求每个月的盈亏状况。
$lastYear = [ 6345.75, 9839.45, 7134.60, 9479.50, 9928.0, 8652.00, 7658.40, 10245.40, 7889.40, 3892.40, 3638.40, 2339.40];$thisYear = [ 6145.75, 6895.00, 3434.00, 9349350, 9478.60, 7652.80, 4758.40, 10945.40, 3689.40, 8992.40, 7588.40, 2239.40];- 传统的foreach 形式：
$profit = [];foreach($thisYear as $key => $monthly){ $profit[$key] = $monthly - $lastYear[$key];}var_dump($profit);- 应用汇合的zip、first、last：
$profit = collect($thisYear)->zip($lastYear)->map(function($monthly){ return $monthly->first() - $monthly->last();});zip：将给定数组的值与相应索引处的原汇合的值合并在一起。
创立lookup 数组
需要：将如下数组格式化成上面的后果：
$employees = [ [ 'name' => 'example', 'email' => 'example@exmaple.com', 'company' => 'example Inc.' ], [ 'name' => 'Lucy', 'email' => 'lucy@example.com', 'company' => 'ibm Inc.' ], [ 'name' => 'Taylor', 'email' => 'toylor@laravel.com', 'company'=>'Laravel Inc.' ]];// 格式化之后的后果$lookup = [ 'example' => 'example@example.com', 'Lucy' => ‘lucy@example.com’, 'Taylor'=> 'toylor@laravel.com'];- 传统的foreach 形式：
$emails = [];foreach ($employees as $key => $value) { $emails[$value['name']] = $value['email'];}- 应用汇合的reduce 办法：
$emails = collect($employees)->reduce(function($emailLookup, $employee){ $emailLookup[$employee['name']] = $employee['email']; return $emailLookup;},[]);reduce：将每次迭代的后果传递给下一次迭代直到汇合缩小为单个值。
- 应用汇合的pluck 办法：
$emails = collect($employees)->pluck('name', 'email');参考链接
- collection在理论开发中的应用