I have used a curl single init to issue an HTTP Get and all worked fine.
Now I tried to use a multi init (as I need to get multiple URLs) and I get a 401 message with "This request requires HTTP authentication" on the response to the Get.
Same Curl options where used on both cases.
Here is the code for th multi init and below it the single init function.
protected function _multiQueryRunkeeper($uri, $subscribersInfo,$acceptHeader) {
$curlOptions = array(
CURLOPT_URL => 'https://api.runkeeper.com' . $uri,
CURLOPT_SSL_VERIFYPEER => false,
CURLOPT_RETURNTRANSFER => 1,
CURLOPT_TIMEOUT => 8,
CURLOPT_HTTPAUTH => CURLAUTH_ANY,
CURLOPT_HTTPGET => true
);
$curl_array = array();
$mh = curl_multi_init();
foreach ($subscribersInfo as $i => $subscriber) {
$curl_array[$i] = curl_init();
curl_setopt_array($curl_array[$i],$curlOptions);
curl_setopt($curl_array[$i], CURLOPT_HEADER,
array('Authorization: Bearer '.$subscriber['token'],
'Accept: application/vnd.com.runkeeper.' . $acceptHeader));
curl_multi_add_handle($mh,$curl_array[$i]);
}
$running = NULL;
do {
usleep(10000);
curl_multi_exec($mh,$running);
} while($running > 0);
$subscribersWorkoutFeed = array();
foreach($subscribersInfo as $i => $subscriber)
{
$subscribersWorkoutFeed[$i] = curl_multi_getcontent($curl_array[$i]);
curl_multi_remove_handle($mh, $curl_array[$i]);
}
curl_multi_close($mh);
return $subscribersWorkoutFeed;
}
protected function _singleQueryRunkeeper($uri, $subscriberToken,$acceptHeader) {
try{
// get fitness user's fitness activities from Runkeeper
$this->_curl = isset($this->_curl)? $this->_curl : curl_init();
$curlOptions = array(
CURLOPT_URL => 'https://api.runkeeper.com' . $uri,
CURLOPT_HTTPHEADER => array('Authorization: Bearer '.$subscriberToken,
'Accept: application/vnd.com.runkeeper.' . $acceptHeader),
CURLOPT_SSL_VERIFYPEER => false,
CURLOPT_RETURNTRANSFER => 1,
CURLOPT_TIMEOUT => 8,
CURLOPT_HTTPGET => true
);
curl_setopt_array($this->_curl,$curlOptions);
$response = curl_exec($this->_curl);
if($response == false) {
if (Zend_Registry::isRegistered('logger')) {
$logger = Zend_Registry::get('logger');
$logger->log('Curl error on _singleQueryRunkeeper: '
. curl_error($this->_curl), Zend_Log::INFO);
}
return null;
}
$data = Zend_Json::decode($response);
return($data);
} catch(Exception $e){
if (Zend_Registry::isRegistered('logger')) {
$logger = Zend_Registry::get('logger');
$logger->log('exception occured on getUsersLatestWorkoutsFromRK. Curl error'
. curl_error($this->_curl), Zend_Log::INFO);
}
}
}
Related
I am trying to perform multiple POST REST Call. The catch: doing multiple POST calls at the same time. I am fully aware and have worked with the library guzzle but I haven't figured away to do this properly. I can perform GET calls asynchronously but nothing at the same level for POST calls. Then I came across pthreads and I read through the documentation and was a bit confused on how to even start it off. I have compiled php with the pthreads extension.
Could someone advise how to perform multiple POST calls at the same time and be able to gather the responses for later manipulation?
The below is a basic implementation that loops and waits. Very slow overall.
$postDatas = [
['field' => 'test'],
['field' => 'test1'],
['field' => 'test2'],
];
foreach ($postDatas as $postData) {
$curl = curl_init();
curl_setopt_array($curl, array(
CURLOPT_URL => "https://www.apisite.com",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode($postData),
CURLOPT_HTTPHEADER => [
"cache-control: no-cache",
"connection: keep-alive",
"content-type: application/json",
"host: some.apisite.com",
],
));
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}
}
That if the task is reduced to working with the API then you probably need to use http://php.net/manual/ru/function.curl-multi-exec.php
public function getMultiUrl() {
//If the connections are very much split the queue into parts
$parts = array_chunk($this->urlStack, self::URL_ITERATION_SIZE , TRUE);
//base options
$options = [
CURLOPT_USERAGENT => 'MyAPP',
CURLOPT_HEADER => false,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_POST => true,
];
foreach ($parts as $urls) {
$mh = curl_multi_init();
$active = null;
$connects = [];
foreach ($urls as $i => $url) {
$options[CURLOPT_POSTFIELDS] = $url['postData'];
$connects[$i] = curl_init($url['queryUrl']);
curl_setopt_array($connects[$i], $options);
curl_multi_add_handle($mh, $connects[$i]);
}
do {
$status = curl_multi_exec($mh, $active);
$info = curl_multi_info_read($mh);
if (false !== $info) {
var_dump($info);
}
} while ($status === CURLM_CALL_MULTI_PERFORM || $active);
foreach ($connects as $i => $conn) {
$content = curl_multi_getcontent($conn);
file_put_contents($this->dir . $i, $content);
curl_close($conn);
}
}
}
I have used
http://requests.ryanmccue.info/ and https://github.com/rmccue/Requests
I am using Request library, but any other library can also be suggested.
My code for CodeIgniter
class Home extends CI_Controller{
public function index(){
$this->load->library('PHPRequest');
$this->rest_client();
}
function rest_client(){
$user = 'myusername';
$pass = 'mypass';
$BaseApiUrl = 'myurl';
$headers = array('Accept' => 'application/json');
$options = array('auth' => new Requests_Auth_Basic(array($user, $pass)));
$request = Requests::get($BaseApiUrl, $headers, $options);
var_dump($request->status_code);
var_dump($request->body);
}
}
But I am getting the following error:
int(401) string(28) "HTTP Digest: Access denied. "
Using Curl PHP Now
$options = array(
CURLOPT_URL => $url,
CURLOPT_HEADER => false,
CURLOPT_VERBOSE => true,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_FOLLOWLOCATION => true,
CURLOPT_SSL_VERIFYPEER => false, // for https
CURLOPT_USERPWD => $username . ":" . $password,
CURLOPT_HTTPAUTH => CURLAUTH_DIGEST
);
$ch = curl_init();
curl_setopt_array( $ch, $options );
try {
$raw_response = curl_exec( $ch );
// validate CURL status
if(curl_errno($ch))
throw new Exception(curl_error($ch), 500);
// validate HTTP status code (user/password credential issues)
$status_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
if ($status_code != 200)
throw new Exception("Response with Status Code [" . $status_code . "].", 500);
} catch(Exception $ex) {
if ($ch != null) curl_close($ch);
throw new Exception($ex);
}
if ($ch != null) curl_close($ch);
return json_decode($raw_response);
I am trying to fetch the header info from multiple webpages. I tried to do so using single cURL requests using the code shown below :
<?php
$arr = array(
"John", "Mary",
"William", " Peter",
"James", "Emma",
"George", "Elizabeth",
"Charles", "Margaret",
);
$ch = curl_init();
for($i=0; $i<sizeOf($arr); $i++){
$url = "https://example.com/".$arr[$i];
$options = array(
CURLOPT_URL => $url,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_HEADER => true,
CURLOPT_FOLLOWLOCATION => true,
CURLOPT_ENCODING => "",
CURLOPT_SSL_VERIFYPEER => FALSE,
CURLOPT_AUTOREFERER => true,
CURLOPT_CONNECTTIMEOUT => 120,
CURLOPT_TIMEOUT => 120,
CURLOPT_MAXREDIRS => 10,
);
curl_setopt_array( $ch, $options );
$response = curl_exec($ch);
$httpCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
if ( $httpCode != 200 ){
echo $arr[$i]." Error<br>";
} else {
echo $arr[$i]." Success<br>";
}
}
curl_close($ch);
?>
But this code seems to take a very long execution time. I searched the internet & found curl_multi_exec which could be used to run multiple cURL requests at a time. So now I use this code :
<?php
ini_set('max_execution_time', 0);
$arr = array(
"John", "Mary",
"William", " Peter",
"James", "Emma",
"George", "Elizabeth",
"Charles", "Margaret",
);
function multiRequest($data) {
// array of curl handles
$curly = array();
// data to be returned
$result = array();
// multi handle
$mh = curl_multi_init();
// loop through $data and create curl handles
// then add them to the multi-handle
foreach ($data as $id => $d) {
$curly[$id] = curl_init();
$url = "https://example.com/".$data[$id];
$options = array(
CURLOPT_URL => $url,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_HEADER => true,
CURLOPT_FOLLOWLOCATION => true,
CURLOPT_ENCODING => "",
CURLOPT_SSL_VERIFYPEER => FALSE,
CURLOPT_AUTOREFERER => true,
CURLOPT_CONNECTTIMEOUT => 120,
CURLOPT_TIMEOUT => 120,
CURLOPT_MAXREDIRS => 10,
);
// extra options?
if (!empty($options)) {
curl_setopt_array($curly[$id], $options);
}
curl_multi_add_handle($mh, $curly[$id]);
}
// execute the handles
$running = null;
do {
curl_multi_exec($mh, $running);
} while($running > 0);
// get content and remove handles
foreach($curly as $id => $c) {
$result[$id] = curl_multi_getcontent($c);
//Code to fetch header info
curl_multi_remove_handle($mh, $c);
}
// all done
curl_multi_close($mh);
return $result;
}
multiRequest($arr);
?>
How to fetch multiple header_info from curl_multi_init HTTP request?
This code from your first example:
$httpCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
if ( $httpCode != 200 ){
echo $arr[$i]." Error<br>";
} else {
echo $arr[$i]." Success<br>";
}
will work even if the curl handle was executed by curl_multi_exec().
In your second example, replace this code:
// get content and remove handles
foreach($curly as $id => $c) {
$result[$id] = curl_multi_getcontent($c);
//Code to fetch header info
curl_multi_remove_handle($mh, $c);
}
with this:
// get content and remove handles
foreach($curly as $id => $c) {
$result[$id] = curl_multi_getcontent($c);
$httpCode = curl_getinfo($c, CURLINFO_HTTP_CODE);
$url = curl_getinfo($c, CURLINFO_EFFECTIVE_URL);
if ( $httpCode != 200 ){
echo $url." Error<br>";
} else {
echo $url." Success<br>";
}
curl_multi_remove_handle($mh, $c);
}
I am trying to ingest an image into a Fedora Commons repository using PHP. Here is the code I'm using:
$queryArgs = array(
"label" => "label goes here",
"format" => "info:fedora/fedora-system:METSFedoraExt-1.1",
"namespace" => $prefix,
"ownerID" => $fedoraUsername,
"logMessage" => "log message goes here"
);
$url = sprintf(
"%s://%s:%d/%s/objects/%s?%s",
"http",
$host,
$port,
"fedora",
$prefix . ":" . $identifier,
http_build_query($queryArgs)
);
$headers = array("Accept: text/xml", "Content-Type: image/jpg");
$userPassword = $fedoraUsername . ":" . $fedoraPassword;
$verifyPeer = false;
$fileContents = file_get_contents("http://localhost/path/to/image.jpg");
$curlOptions = array(
CURLOPT_URL => $url,
CURLOPT_HTTPHEADER => $headers,
CURLOPT_USERPWD => $userPassword,
CURLOPT_HTTPAUTH => CURLAUTH_BASIC,
CURLOPT_SSL_VERIFYPEER => $verifyPeer,
CURLOPT_FOLLOWLOCATION => true,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_POST => true,
CURLOPT_POSTFIELDS => $fileContents
);
$curlHandle = curl_init();
curl_setopt_array($curlHandle, $curlOptions);
curl_exec($curlHandle);
error_log(print_r(curl_getinfo($curlHandle), true));
At the end I print out whatever curl_getinfo has to say, notably [http_code] => 415.
HTTP Error 415 Unsupported media type
What am I doing wrong here?
I finally figured it out. I ended up making two different requests: one to create a new empty object in Fedora, the other to attach a datastream to that object. Here is the code (I put all this in a class named Fedora, but this is not necessary and might not fit your needs):
<?php
class Fedora {
// Use cURL with the provided functions and return the result if the HTTP Code recieved matches the expected HTTP Code
private function curlThis($curlOptions, $expectedHttpCode) {
$returnValue = false;
try {
$curlHandle = curl_init();
if ($curlHandle === false) {
throw new Exception(
"`curl_init()` returned `false`"
);
}
$settingOptionsSucceeded = curl_setopt_array($curlHandle, $curlOptions);
if ($settingOptionsSucceeded === false) {
throw new Exception(
sprintf(
"`curl_setopt_array(...)` returned false. Error: %s. Info: %s",
curl_error($curlHandle),
print_r(curl_getinfo($curlHandle), true)
),
curl_errno($curlHandle)
);
}
$curlReturn = curl_exec($curlHandle);
if ($curlReturn === false) {
throw new Exception(
sprintf(
"`curl_exec(...)` returned false. Error: %s. Info: %s",
curl_error($curlHandle),
print_r(curl_getinfo($curlHandle), true)
),
curl_errno($curlHandle)
);
}
$httpCode = curl_getinfo($curlHandle, CURLINFO_HTTP_CODE);
if ($httpCode === false) {
throw new Exception(
sprintf(
"`curl_getinfo(...)` returned false. Error: %s.",
curl_error($curlHandle)
),
curl_errno($curlHandle)
);
}
if ($httpCode !== $expectedHttpCode) {
throw new Exception(
sprintf(
"`curl_getinfo(...)` returned an unexpected http code (expected %s, but got %s). Error: %s. Complete info: %s",
$expectedHttpCode,
$httpCode,
curl_error($curlHandle),
print_r(curl_getinfo($curlHandle), true)
),
curl_errno($curlHandle)
);
}
$returnValue = $curlReturn;
} catch (Exception $e) {
trigger_error(
sprintf(
"(%d) %s",
$e->getCode(),
$e->getMessage()
),
E_USER_ERROR
);
}
return $returnValue;
}
// Create a new empty object in Fedora Commons and return its pid
private function createNewEmptyObject($prefix, $id) {
$returnValue = false;
// Build URL
$protocol = variable_get("fedora_protocol"); // 'http'
$host = variable_get("fedora_host");
$port = variable_get("fedora_port"); // '8082'
$context = variable_get("fedora_context"); // 'fedora'
$pid = $prefix . ":" . $id;
$url = sprintf(
"%s://%s:%d/%s/objects/%s",
$protocol,
$host,
$port,
$context,
$pid
);
// Build cURL options
$userPassword = variable_get("fedora_username") . ":" . variable_get("fedora_password"); // username:password
$verifyPeer = false; // false for ignoring self signed certificates
$headers = array("Accept: text/xml", "Content-Type: text/xml");
$curlOptions = array(
CURLOPT_URL => $url,
CURLOPT_HTTPHEADER => $headers,
CURLOPT_USERPWD => $userPassword,
CURLOPT_HTTPAUTH => CURLAUTH_BASIC,
CURLOPT_SSL_VERIFYPEER => $verifyPeer,
CURLOPT_FOLLOWLOCATION => true,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_POST => true
);
// Try `cURL`ing
$result = $this->curlThis($curlOptions, 201);
if ($result === $pid) {
$returnValue = $result;
}
return $returnValue;
}
private function attachDatastream ($pid, $file, $datastreamID) {
$returnValue = false;
// Build URL
$protocol = variable_get("fedora_protocol"); // "http"
$host = variable_get("fedora_host");
$port = variable_get("fedora_port"); // 8082
$context = variable_get("fedora_context"); // fedora
$url = sprintf(
"%s://%s:%d/%s/objects/%s/datastreams/%s?controlGroup=M", // M stands for 'Managed Content'
$protocol,
$host,
$port,
$context,
$pid,
$datastreamID
);
// Build cURL options
$userPassword = variable_get("fedora_username") . ":" . variable_get("fedora_password"); // username:password
$verifyPeer = false; // false for ignoring self signed certificates
$headers = array("Accept: text/xml", "Content-Type: " . $file->filemime);
$fileContents = file_get_contents("sites/default/files/images/" . $file->filename);
$curlOptions = array(
CURLOPT_URL => $url,
CURLOPT_HTTPHEADER => $headers,
CURLOPT_USERPWD => $userPassword,
CURLOPT_HTTPAUTH => CURLAUTH_BASIC,
CURLOPT_SSL_VERIFYPEER => $verifyPeer,
CURLOPT_FOLLOWLOCATION => true,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_POST => true,
CURLOPT_POSTFIELDS => $fileContents
);
// Try `cURL`ing
$result = $this->curlThis($curlOptions, 201);
if ($result === $pid) {
$returnValue = $result;
}
return $returnValue;
}
public function ingest($namespace, $identifier, $file) {
$pid = $this->createNewEmptyObject($namespace, $identifier);
if ($pid) {
$result = $this->attachDatastream($pid, $file, "IMAGE");
if ($result) {
error_log("Successfully ingested a file!");
} else {
error_log("FAILED ATTACHING DATASTREAM TO NEW OBJECT");
}
} else {
error_log("FAILED CREATING NEW EMPTY OBJECT");
}
}
}
$fedora = new Fedora();
/*
* $file is an object obtained by uploading a file into a drupal file system (at /drupal/sites/default/files/images/singe.jpg). It has the following attributes:
* $file->filemime === "image/jpeg"
* $file->filename === "singe.jpg"
*/
$fedora->ingest('namespace', 'identifier', $file);
?>
Can you give me some idea how to improve this function so it handles unexpected reply when server returns output that is not in xml, eg a simple server error message in html and then retry fetching the xml?
function fetch_xml($url, $timeout=15)
{
$ch = curl_init();
curl_setopt_array($ch, array(
CURLOPT_HEADER => 0,
CURLOPT_RETURNTRANSFER => 1,
CURLOPT_CONNECTTIMEOUT => (int)$timeout,
CURLOPT_FOLLOWLOCATION => 1,
CURLOPT_URL => $url)
);
$xml_data = curl_exec($ch);
curl_close($ch);
if (!empty($xml_data)) {
return new SimpleXmlElement($xml_data);
}
else {
return null;
}
}
You can give this a try. I haven't tested it out.
function fetch_xml($url, $timeout = 15, $max_attempts = 5, $attempts = 0)
{
$ch = curl_init();
curl_setopt_array($ch, array(
CURLOPT_HEADER => 0,
CURLOPT_RETURNTRANSFER => 1,
CURLOPT_CONNECTTIMEOUT => (int)$timeout,
CURLOPT_FOLLOWLOCATION => 1,
CURLOPT_URL => $url)
);
$xml_data = curl_exec($ch);
curl_close($ch);
if ($attempts <= $max_attempts && !empty($xml_data)) // don't infinite loop
{
try
{
return new SimpleXmlElement($xml_data);
}
catch (Exception $e)
{
return fetch_xml($url, (int)$timeout, $max_attempts, $attempts++);
}
}
return NULL;
}