向东博客 专注WEB应用 构架之美 --- 构架之美，在于尽态极妍 | 应用之美，在于药到病除|赢在IT，Playin' with IT,Focus on Killer Application,Marketing Meets Technology.

背景：用try catch去捕获php的curl错误，是不行的，
The PHP cURL functions do not throw exceptions, so there will never be anything to catch. You'll just have to check the return value for a boolean false.
As to why the cURL timeout does not seem to be occurring after 30 seconds, I'm not sure.

<?php     $url="12122222.com";     $agents = "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1";   $ch = curl_init();   curl_setopt($ch, CURLOPT_HEADER, 0);   curl_setopt($ch, CURLOPT_VERBOSE, FALSE);   curl_setopt($ch, CURLOPT_RETURNTRANSFER, TRUE);   curl_setopt($ch, CURLOPT_FOLLOWLOCATION, TRUE);   curl_setopt($ch, CURLOPT_USERAGENT, $agents);   curl_setopt($ch, CURLOPT_URL, $url);   $curlResp = curl_exec($ch);     if(curl_errno($ch))     {         echo 'error:' . curl_error($ch);     }     /*   if ($curlResp === FALSE) {     throw new Exception("Curl Has Some Wrong ,Please Check It。");   }     */   curl_close($ch);
上面代码返回：
error:Could not resolve host: 12122222.com; Host not found

打开注释代码返回：
Fatal error: Uncaught exception 'Exception' with message 'Curl Has Some Wrong ,Please Check It。' in D:\wamp\www\curl_try_catch.php on line 14
Exception: Curl Has Some Wrong ,Please Check It。 in D:\wamp\www\curl_try_catch.php on line 14
---
返回一些状态码，URl如下：curl.haxx.se/libcurl/c/libcurl-errors.html
再进行捕获。
来自：
http://stackoverflow.com/questions/3987006/how-to-catch-curl-errors-in-php

You can use the curl_error() function to know if there was some error, example:

if(curl_errno($c))
{
    echo 'error:' . curl_error($c);
}
===================================================

I'm trying to debug this cURL-operation. The var_dump() is returning bool(false)
How can I make it exit the try in that case?
function parse($url, $headonly = TRUE ){     $agents = 'Mozilla/5.0 (X11; U; Linux i686; en-US) AppleWebKit/534.16 (KHTML, like Gecko) Chrome/10.0.648.204 Safari/534.16';     try     {         $ch = curl_init();         curl_setopt($ch, CURLOPT_HEADER, 0);         curl_setopt($ch, CURLOPT_VERBOSE, FALSE);         curl_setopt($ch, CURLOPT_RETURNTRANSFER, TRUE);         curl_setopt($ch, CURLOPT_FOLLOWLOCATION, TRUE);         curl_setopt($ch, CURLOPT_USERAGENT, $agents);         curl_setopt($ch, CURLOPT_URL, $url);         $curlResp = curl_exec($ch);         curl_close($ch);         var_dump($curlResp); //RETURNS bool(false)         die;         $resp = str_replace("class=l","class='l'",$curlResp);         return $resp;     }     catch( Exception $e)     {         $strResponse = "";         $strErrorCode = $e->getCode();         $strErrorMessage = $e->getMessage();         print_r($arrCurlInfo, $strErrorCode, $strErrorMessage);         die;     }  //end catch }

Why not try throwing an Exception when false is returned?

if ($curlResp === FALSE) {     throw new Exception(); }

Look here: php.net/manual/en/function.curl-exec.php; basically, false means the operation failed.
Found it: echo curl_error($ch); "Couldn't connect to host"

http://stackoverflow.com/questions/7069822/curl-try-catch-problem


群讨论：
回忆未来-向东-Jàck(372647***)  上午 10:49:40
PHP try catch 能抓到Curl的timeout吗
一周工作五天(243801***)  上午 10:50:58
不能
回忆未来-向东-Jàck(372647***)  上午 10:52:15
常规处理办法是？
三德子(277940***)  上午 10:52:31
curl没有抛异常 除非你去做个errorhandler
回忆未来-向东-Jàck(372647***)  上午 10:52:39
前一段时间我写过博文，但没有根治：http://www.jackxiang.com/post/6445/
errorhandler  ？
有示例demo么。
三德子(277940***)  上午 10:53:02
在errorhandler中 抛errorexception
看手册 errorexception这一章
回忆未来-向东-Jàck(372647***)  上午 10:53:48
嗯，超时捕获有code值没？
三德子(277940***)  上午 10:53:57
<?php function exception_error_handler($errno, $errstr, $errfile, $errline ) { throw new ErrorException($errstr, 0, $errno, $errfile, $errline); } set_error_handler("exception_error_handler"); /* Trigger exception */ strpos(); ?>
一周工作五天(243801***)  上午 10:55:41
搞那么复杂干嘛  直接把 超时时间设置长些不就ok了

----------------------------------------------------------------------------------------------
强制用try catch作下实践如下：
    public function testTimeOut(){       $user_mobile_gsidlogin_url = "http://jackxiang.com/test/testMaxTimeUrl";       $data = array();//Post Data       try{         $vericode = Remote::get($user_mobile_gsidlogin_url,         array           (             CURLOPT_USERAGENT      => 'Mozilla/5.0 (compatible; KoPHP v3.0 +http://kophp.com/)',             CURLOPT_POST => TRUE,             CURLOPT_POSTFIELDS => http_build_query($data),                                   CURLOPT_CONNECTTIMEOUT => 5,             CURLOPT_TIMEOUT        => 5,           )         );//发送GET请求         echo $vericode;       }       catch (Exception $e){         echo "Max Time Out...time out。<hr>";         echo  $e->getmessage();       }     }          public function testMaxTimeUrl(){     sleep(10);     echo "10 sec pass!";     }
访问：http://jackxiang.com/test/testTimeOut
Max Time Out...time out。
Error fetching remote http://jackxiang.com/test/testMaxTimeUrl [ status 0 ]
Error fetching remote http://jackxiang.com/test/testMaxTimeUrl [ status 0 ]
Operation timed out after 5007 milliseconds with 0 bytes received
其实上述说明，不在curl里捕获，在外面一层也是能捕捉到的，只是错误的message提示是：Error：Operation timed out after 5007 milliseconds with 0 bytes received！
________________________________________________________________________________________________
在生产中的代码，用try catch实现的Remote:get，来自KO框架，作了下简单修改（有时Url为空，不知是谁调的，Add Time:2014/03/26）：
    * @param   string   remote URL      * @param   array    curl options      * @return  string      * @throws  KoException      */     public static function get($url, array $options = NULL)     {         if ($options === NULL) {             // Use default options             $options = self::$default_options;         } else {             // Add default options             $options = $options + self::$default_options;         }         $log = Log::instance(__CLASS__)->attach(new Log_File(DATA_PATH . 'logs'.DIRECTORY_SEPARATOR . date('Ymd').DIRECTORY_SEPARATOR, __CLASS__.'-' . date('Ymd') . '.log'));         // The transfer must always be returned         $options[CURLOPT_RETURNTRANSFER] = TRUE;         $response = false;         try {             // Open a new remote connection             $remote = curl_init($url);             if(KO::$threshold >= 3){                 $ldata = array('uri' => $url, 'options' => $options);                 $log->add('remote_log', $ldata);             }                              // Set connection options             curl_setopt_array($remote, $options);             if (array_key_exists(CURLOPT_POST, $options)) {                 curl_setopt($remote, CURLOPT_HTTPHEADER, array('Expect:'));             }             // Get the response             $response = curl_exec($remote);             // Get the response information             $code = curl_getinfo($remote, CURLINFO_HTTP_CODE);             if ($code AND $code < 200 OR $code > 299) {                 $error = $response;             } elseif ($response === FALSE) {                 $error = curl_error($remote);             }             // Close the connection             curl_close($remote);             if (isset($error)) {                 $data = array(':url' => $url, ':code' => $code, ':error' => $error);                 $log->add('remote-error', $data);                 throw new KoException('Error fetching remote :url [ status :code ] :error', $data);             }         } catch (Exception $e) {             foreach (Ko::trace($trace) as $i => $step){                 if($step['function'] == "Remote::get"){                     $traceInfoStr = "Remote::get called by: ".$step['file']." line:".$step['line']." param:".var_export($step['args'],TRUE);                 }             }             $data = array(':url' => $url, ':code' => $code, ':error' => $e->getMessage(),":traceinfo" => $traceInfoStr);             $log->add('remote-error', $data);             throw new KoException('Error fetching remote :url [ status :code ] :error', $data);         }         return $response;     }

作者：jackxiang@向东博客 专注WEB应用 构架之美 --- 构架之美，在于尽态极妍 | 应用之美，在于药到病除
地址：http://jackxiang.com/post/6445/
版权所有。转载时必须以链接形式注明作者和原始出处及本声明！