从内部 PHP 函数保存数据

Question

我正在使用Rolling cUrl PHP 库通过 JSON 从 API 中获取数据。

在 Rolling CURL 库文档中，我尝试了以下示例，效果很好：

$rollingCurl = new \RollingCurl\RollingCurl();
$rollingCurl
    ->get('https://www.google.com')
    ->get('http://www.msn.com')
    ->setCallback(function(\RollingCurl\Request $request, \RollingCurl\RollingCurl $rollingCurl) {
        // parsing html with regex is evil (http://bit.ly/3x9sQX), but this is just a demo
        if (preg_match("#<title>(.*)</title>#i", $request->getResponseText(), $out)) {
            $title = $out[1];
        }
        else {
            $title = '[No Title Tag Found]';
        }
        echo "Fetch complete for (" . $request->getUrl() . ") $title " . PHP_EOL;
    })
    ->setSimultaneousLimit(3)
    ->execute();

我尝试根据需要对其进行更改，并将每个“抓取”的数据保存在一个数组中，但我找不到从回调或外部保存数据的方法。

这是我尝试过的：

$rollingCurl = new \RollingCurl\RollingCurl();
$rollingCurl
    ->get('https://www.myapi.com')
    ->get('https://www.myapi2.com')
    ->setCallback(function(\RollingCurl\Request $request, \RollingCurl\RollingCurl $rollingCurl) {
        $output = json_decode($request->getResponseText());
        $data[$x++] = $output; //How do I save it outside this function?
    })
    ->setSimultaneousLimit(3)
    ->execute();
//Can I get the content of each API call without the callback?
print_r($data); // returns nothing :(

所以我的问题是：

1. 如何在此函数之外保存数据数组？

2. 可以不通过回调（执行后）获取每个API调用的内容吗？

Answer 1

我认为在回调闭包中初始化 $data 和 use 应该可以解决您的问题。否则，您的 $data 数组仅设置在您的闭包范围内（在声明回调函数时请注意 use (&$data)）：

/** IMPORTANT **/
$data = array();
/***************/

$rollingCurl = new \RollingCurl\RollingCurl();
$rollingCurl
    ->get('https://www.myapi.com')
    ->get('https://www.myapi2.com')
    ->setCallback(function(\RollingCurl\Request $request, \RollingCurl\RollingCurl $rollingCurl) use (&$data) {
        $output = json_decode($request->getResponseText());
        $data[$x++] = $output;
    })
    ->setSimultaneousLimit(3)
    ->execute();
// Yes, you can!
print_r($data);

我喜欢的另一个不错的功能是 Closure 范围绑定，它只是一个快速且未经测试的示例（当您需要更多信息时请告诉我:)）：

class CurlHandler {
    $data = array();
    public function processOutput($data) {
        // Do something with the data retrieved here
        $this->data[] = $data;
    }
    public function getData() { return $this->data; }
}

$curlResponseHandler = new CurlHandler();

// Your callback to be passed into \RollingCurl\RollingCurl::setCallback()
$closure = function(\RollingCurl\Request $request, \RollingCurl\RollingCurl $rollingCurl) {
    $output = json_decode($request->getResponseText());
    $this->processOutput($output);
}

// Bind your callback Closure to your CurlHandler scope
// => the scope binding makes $this inside your callback Closure refer to your CurlHandler instance
$closure->bindTo($curlResponseHandler);

$rollingCurl = new \RollingCurl\RollingCurl();
$rollingCurl
    ->get('https://www.myapi.com')
    ->get('https://www.myapi2.com')
    ->setCallback($closure)
    ->setSimultaneousLimit(3)
    ->execute();

// Get all your outputs from your CurlHandler class now:
print_r($curlResponseHandler->getData());

或者最后但并非最不重要的一点是直接在您的 CurlHandler 类上构造一个回调，在这种情况下这可能是我最喜欢的：

class CurlHandler {
    $data = array();
    public function processOutput(\RollingCurl\Request $request, \RollingCurl\RollingCurl $rollingCurl) {
        $output = json_decode($request->getResponseText());
        $this->data[] = $output;
    }
    public function getData() { return $this->data; }
}

$curlResponseHandler = new CurlHandler();

$rollingCurl = new \RollingCurl\RollingCurl();
$rollingCurl
    ->get('https://www.myapi.com')
    ->get('https://www.myapi2.com')
    ->setCallback(array($curlResponseHandler, 'processOutput'))
    ->setSimultaneousLimit(3)
    ->execute();

// Get all your outputs from your CurlHandler class now:
print_r($curlResponseHandler->getData());

最后一个选项假定setCallback() 通常接受可调用对象，而不仅仅是Closures。

希望对您有所帮助，干杯！