使用方式 ~~~ $keyword="自媒体"; $keyword = urlencode($keyword); $catchUrl = "https://www.toutiao.com/search_content/?offset=0&format=json&keyword={$keyword}&autoload=true&count=20&cur_tab=1"; $h=new JqHttp(); $htmlcode = $h->ihttp_get($catchUrl); $htmlcode = $htmlcode['content']; $htmlcode = base64_encode($htmlcode); $htmlcode = $h->ihttp_post('http://we7cc.csdn123.net/toutiao_news/now.catch.php', array('htmlcode' => $htmlcode)); $htmlcode = $htmlcode['content']; $htmlcode = preg_replace('/^\s+|\s+$/', '', $htmlcode); $htmlcode = base64_decode($htmlcode); $linkArr = unserialize($htmlcode); print_r($linkArr); ~~~ 类实现 ~~~ <?php namespace jqstu; class JqHttp { var $curlconf=[ 'TIMEOUT'=>60, 'useCert'=>false, 'SSLCERT_PATH'=>'', 'SSLKEY_PATH'=>'', ]; public function __construct($conf=[]) { $this->curlconf=array_merge($this->curlconf, $conf); } /** * @param $url * @return array * @throws \Exception */ public function ihttp_get($url) { return self::ihttp_request($url); } /** * @param $url * @param $data * @return array * @throws \Exception */ public function ihttp_post($url, $data) { $headers = array('Content-Type' => 'application/x-www-form-urlencoded'); return self::ihttp_request($url, $data, $headers); } /** * @param $url * @param string $post * @param array $extra * @param int $timeout * @return array * @throws \Exception */ public function ihttp_request($url, $post = '', $extra = array()) { $timeout=$this->curlconf['TIMEOUT']; if (function_exists('curl_init') && function_exists('curl_exec') && $timeout > 0) { $ch = self::ihttp_build_curl($url, $post, $extra, $timeout); $data = curl_exec($ch); $status = curl_getinfo($ch); $errno = curl_errno($ch); $error = curl_error($ch); curl_close($ch); if ($errno || empty($data)) { Jqutil::JqException($error,$errno); } else { return self::ihttp_response_parse($data); } } $urlset = self::ihttp_parse_url($url, true); if (!empty($urlset['ip'])) { $urlset['host'] = $urlset['ip']; } $body = self::ihttp_build_httpbody($url, $post, $extra); if ($urlset['scheme'] == 'https') { $fp = self::ihttp_socketopen('ssl://' . $urlset['host'], $urlset['port'], $errno, $error); } else { $fp = self::ihttp_socketopen($urlset['host'], $urlset['port'], $errno, $error); } stream_set_blocking($fp, $timeout > 0 ? true : false); stream_set_timeout($fp, ini_get('default_socket_timeout')); if (!$fp) { Jqutil::JqException($error,1); } else { fwrite($fp, $body); $content = ''; if($timeout > 0) { while (!feof($fp)) { $content .= fgets($fp, 512); } } fclose($fp); return self::ihttp_response_parse($content, true); } } private function ihttp_socketopen($hostname, $port = 80, &$errno, &$errstr, $timeout = 15) { $fp = ''; if(function_exists('fsockopen')) { $fp = @fsockopen($hostname, $port, $errno, $errstr, $timeout); } elseif(function_exists('pfsockopen')) { $fp = @pfsockopen($hostname, $port, $errno, $errstr, $timeout); } elseif(function_exists('stream_socket_client')) { $fp = @stream_socket_client($hostname.':'.$port, $errno, $errstr, $timeout); } return $fp; } private function ihttp_build_httpbody($url, $post, $extra) { $urlset = self::ihttp_parse_url($url, true); if ($urlset) { return $urlset; } if (!empty($urlset['ip'])) { $extra['ip'] = $urlset['ip']; } $body = ''; if (!empty($post) && is_array($post)) { $filepost = false; $boundary = Jqutil::random(40); foreach ($post as $name => &$value) { if ((is_string($value) && substr($value, 0, 1) == '@') && file_exists(ltrim($value, '@'))) { $filepost = true; $file = ltrim($value, '@'); $body .= "--$boundary\r\n"; $body .= 'Content-Disposition: form-data; name="'.$name.'"; filename="'.basename($file).'"; Content-Type: application/octet-stream'."\r\n\r\n"; $body .= file_get_contents($file)."\r\n"; } else { $body .= "--$boundary\r\n"; $body .= 'Content-Disposition: form-data; name="'.$name.'"'."\r\n\r\n"; $body .= $value."\r\n"; } } if (!$filepost) { $body = http_build_query($post, '', '&'); } else { $body .= "--$boundary\r\n"; } } $method = empty($post) ? 'GET' : 'POST'; $fdata = "{$method} {$urlset['path']}{$urlset['query']} HTTP/1.1\r\n"; $fdata .= "Accept: */*\r\n"; $fdata .= "Accept-Language: zh-cn\r\n"; if ($method == 'POST') { $fdata .= empty($filepost) ? "Content-Type: application/x-www-form-urlencoded\r\n" : "Content-Type: multipart/form-data; boundary=$boundary\r\n"; } $fdata .= "Host: {$urlset['host']}\r\n"; $fdata .= "User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0.1) Gecko/20100101 Firefox/9.0.1\r\n"; if (function_exists('gzdecode')) { $fdata .= "Accept-Encoding: gzip, deflate\r\n"; } $fdata .= "Connection: close\r\n"; if (!empty($extra) && is_array($extra)) { foreach ($extra as $opt => $value) { if (!Jqutil::strexists($opt, 'CURLOPT_')) { $fdata .= "{$opt}: {$value}\r\n"; } } } if ($body) { $fdata .= 'Content-Length: ' . strlen($body) . "\r\n\r\n{$body}"; } else { $fdata .= "\r\n"; } return $fdata; } /** * @param $url * @param $post * @param $extra * @param $timeout * @return resource * @throws \Exception */ private function ihttp_build_curl($url, $post, $extra, $timeout) { if (!function_exists('curl_init') || !function_exists('curl_exec')) { Jqutil::JqException("curl扩展未开启"); } $urlset = self::ihttp_parse_url($url); // print_r($urlset); if (!empty($urlset['ip'])) { $extra['ip'] = $urlset['ip']; } $ch = curl_init(); if (!empty($extra['ip'])) { $extra['Host'] = $urlset['host']; $urlset['host'] = $extra['ip']; unset($extra['ip']); } if ($urlset['port']=80){ $urlset['port']=''; } if(empty($urlset['query'])){ $urlset['query']=''; } curl_setopt($ch, CURLOPT_URL, $urlset['scheme'] . '://' . $urlset['host'] . $urlset['port'] . $urlset['path'] . $urlset['query']); curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); @curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1); curl_setopt($ch, CURLOPT_HEADER, 1); @curl_setopt($ch, CURLOPT_HTTP_VERSION, CURL_HTTP_VERSION_1_0); if ($post) { if (is_array($post)) { $filepost = false; foreach ($post as $name => &$value) { if (version_compare(phpversion(), '5.5') >= 0 && is_string($value) && substr($value, 0, 1) == '@') { $post[$name] = new \CURLFile(ltrim($value, '@')); } if ((is_string($value) && substr($value, 0, 1) == '@') || (class_exists('CURLFile') && $value instanceof CURLFile)) { $filepost = true; } } if (!$filepost) { $post = http_build_query($post); } curl_setopt($ch, CURLOPT_POST, 1); curl_setopt($ch, CURLOPT_POSTFIELDS, $post); } } $proxy=array('host'=>'127.0.0.1','auth'=>''); if (!empty($proxy['host']) || !empty($proxy['auth'])){ $urls = parse_url($proxy['host']); if (!empty($urls['host'])){ curl_setopt($ch, CURLOPT_PROXY, "{$urls['host']}:{$urls['port']}"); $proxytype = 'CURLPROXY_' . strtoupper($urls['scheme']); if (!empty($urls['scheme']) && defined($proxytype)) { curl_setopt($ch, CURLOPT_PROXYTYPE, constant($proxytype)); } else { curl_setopt($ch, CURLOPT_PROXYTYPE, CURLPROXY_HTTP); curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1); } if (!empty($proxy['auth'])) { curl_setopt($ch, CURLOPT_PROXYUSERPWD, $proxy['auth']); } } } if (defined('CURL_SSLVERSION_TLSv1')) { curl_setopt($ch, CURLOPT_SSLVERSION, CURL_SSLVERSION_TLSv1); } curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, $timeout); curl_setopt($ch, CURLOPT_TIMEOUT, $timeout); //安全证书检查 $curlconf=$this->curlconf; if ($curlconf['useCert'] && !empty($curlconf['SSLCERT_PATH']) || !empty($curlconf['SSLKEY_PATH'])){ curl_setopt($ch,CURLOPT_SSL_VERIFYPEER,TRUE); curl_setopt($ch,CURLOPT_SSL_VERIFYHOST,2);//严格校验 curl_setopt($ch,CURLOPT_SSLCERTTYPE,'PEM'); curl_setopt($ch,CURLOPT_SSLCERT, $curlconf['SSLCERT_PATH']); curl_setopt($ch,CURLOPT_SSLKEYTYPE,'PEM'); curl_setopt($ch,CURLOPT_SSLKEY,$curlconf['SSLKEY_PATH']); }else{ curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false); curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 0); curl_setopt($ch, CURLOPT_SSLVERSION, 1); } curl_setopt($ch, CURLOPT_USERAGENT, 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0.1) Gecko/20100101 Firefox/9.0.1'); if (!empty($extra) && is_array($extra)) { $headers = array(); foreach ($extra as $opt => $value) { if (Jqutil::strexists($opt, 'CURLOPT_')) { curl_setopt($ch, constant($opt), $value); } elseif (is_numeric($opt)) { curl_setopt($ch, $opt, $value); } else { $headers[] = "{$opt}: {$value}"; } } if (!empty($headers)) { curl_setopt($ch, CURLOPT_HTTPHEADER, $headers); } } return $ch; } /** * @param $url * @param bool $set_default_port * @return mixed * @throws \Exception */ private function ihttp_parse_url($url, $set_default_port = false) { if (empty($url)) { Jqutil::JqException("url参数没有填写"); } $urlset = parse_url($url); if (!empty($urlset['scheme']) && !in_array($urlset['scheme'], array('http', 'https'))) { Jqutil::JqException("只能使用 http 及 https 协议"); } if (!empty($urlset['query'])) { $urlset['query'] = "?{$urlset['query']}"; } if (Jqutil::strexists($url, 'https://') && !extension_loaded('openssl')) { if (!extension_loaded("openssl")) { Jqutil::JqException("请开启您PHP环境的openssl"); } } if ($set_default_port && empty($urlset['port'])) { $urlset['port'] = $urlset['scheme'] == 'https' ? '443' : '80'; } if (empty($urlset['host'])) { $siteroot = $_SERVER['SERVER_NAME']; $current_url = parse_url($siteroot); $urlset['host'] = $current_url['host']; $urlset['scheme'] = $current_url['scheme']; $urlset['path'] = $current_url['path'] . '/' . str_replace('./', '', $urlset['path']); $urlset['ip'] = '127.0.0.1'; } elseif (!self::ihttp_allow_host($urlset['host'])) { Jqutil::JqException('host 非法'); } if (array_key_exists("port",$urlset)) { $urlset['port']=$urlset['port']; } else { $urlset['port']=80; } return $urlset; } /** * @param $data * @param bool $chunked * @return array */ private function ihttp_response_parse($data, $chunked = false){ $rlt = array(); $headermeta = explode('HTTP/', $data); if (count($headermeta) > 2) { $data = 'HTTP/' . array_pop($headermeta); } $pos = strpos($data, "\r\n\r\n"); $split1[0] = substr($data, 0, $pos); $split1[1] = substr($data, $pos + 4, strlen($data)); $split2 = explode("\r\n", $split1[0], 2); preg_match('/^(\S+) (\S+) (.*)$/', $split2[0], $matches); $rlt['code'] = $matches[2]; $rlt['status'] = $matches[3]; $rlt['responseline'] = $split2[0]; $isgzip = false; $ischunk = false; $header = explode("\r\n", $split2[1]); foreach ($header as $v){ $pos = strpos($v, ':'); $key = substr($v, 0, $pos); $value = trim(substr($v, $pos + 1)); $rlt['headers'][$key] = $value; if(!$isgzip && strtolower($key) == 'content-encoding' && strtolower($value) == 'gzip') { $isgzip = true; } if(!$ischunk && strtolower($key) == 'transfer-encoding' && strtolower($value) == 'chunked') { $ischunk = true; } } if($chunked && $ischunk) { $rlt['content'] = self::ihttp_response_parse_unchunk($split1[1]); } else { $rlt['content'] = $split1[1]; } $rlt['content'] = $split1[1]; if($isgzip && function_exists('gzdecode')) { $rlt['content'] = gzdecode($rlt['content']); } if($rlt['code'] == '100') { return self::ihttp_response_parse($rlt['content']); } return $rlt; } /** * @param null $str * @return bool|null|string */ private function ihttp_response_parse_unchunk($str = null) { if(!is_string($str) or strlen($str) < 1) { return false; } $eol = "\r\n"; $add = strlen($eol); $tmp = $str; $str = ''; do { $tmp = ltrim($tmp); $pos = strpos($tmp, $eol); if($pos === false) { return false; } $len = hexdec(substr($tmp, 0, $pos)); if(!is_numeric($len) or $len < 0) { return false; } $str .= substr($tmp, ($pos + $add), $len); $tmp = substr($tmp, ($len + $pos + $add)); $check = trim($tmp); } while(!empty($check)); unset($tmp); return $str; } /** * @param $host * @return bool */ private function ihttp_allow_host($host) { if (Jqutil::strexists($host, '@')) { return false; } $pattern = "/^(10|172|192|127)/"; if (preg_match($pattern, $host)) { return false; } return true; } } ~~~