Im searching a safe and fast way to use a shared object.
I asked the question already here: https://github.com/krakjoe/pthreads/issues/470
but obviuously this wasnt the right place.
Trying to share an object (Threaded) with many other contextes (Thread).
All threads are updating this shard object -- they can set own requests and have to respond to requests from others also.
Now that krakjoe responded that lock/unlock wont be available in 7 i got a problem.
I know about :.synchronized but have no idea how to use it to get it working for my needs.
How can i use ::synchronized to write methods like
lock()
unlock()
is_locked() -- check if locked, and if so dont try - just go on and try later
EDIT:
i wrote a (imo) very easy test script.
this script includes no syc/lock/... methods atm.
it should just show what im trying to do.
im still searching a way to use :: to make this shared safe.
code:
<?php
/*
TEST:
create n threads
each will
- Shared::set() its own ref
- check if Shared::exists() its own ref
- Shared::get() its ref back
- call method ::isRunning() at returned val to easily check if is ref or got overwritten by another context
TODO:
using ::synchronized to handle multi-context-access
NOTES:
every method as public to prevent pthreads v2 "Method Modifiers - Special Behaviour"
see: "Method Modifiers - Special Behaviour"
at http://blog.krakjoe.ninja/2015/08/a-letter-from-future.html
*/
class Shared extends Threaded
{
public $data;
public function exists($ident)
{
return isset($this->data[$ident]);
}
public function set($ident, $ref)
{
$return = false;
if(!isset($this->data[$ident])){
$data = $this->data;
$data[$ident] = $ref;
$this->data = $data;
$return = $this->data[$ident];
}
#echo __METHOD__ . '(' . $ident . ') => ' . gettype($return) . PHP_EOL;
return $return;
}
public function get($ident)
{
$return = false;
if($this->exists($ident) === true){
$data = $this->data;
$return = $data[$ident];
unset($data[$ident]);
$this->data = $data;
}
#echo __METHOD__ . '(' . $ident . ') => ' . gettype($return) . PHP_EOL;
return $return;
}
}
class T extends Thread
{
public $count;
public function __construct(Shared $Shared, $ident)
{
$this->Shared = $Shared;
$this->ident = $ident;
}
public function run()
{
$slowdown = true;
$this->count = 0;
while(true){
if($slowdown){
// "don't allow usleep or sleep" : https://github.com/krakjoe/pthreads/commit/a157b34057b0f584b4db326f30961b5c760dead8
// loop a bit to simulate work:
$start = microtime(true);
$until = rand(1, 100000)/1000000;
while(microtime(true)-$start < $until){
// ...
}
}
if($this->Shared->exists($this->ident) === true){
$ref = $this->Shared->get($this->ident);
}
else{
$ref = $this->Shared->set($this->ident, $this);
}
// calling a method on $ref -- if not a ref we crash
$ref->isRunning();
unset($ref);
$this->count++;
}
}
}
echo 'start ...' . PHP_EOL;
$n = 8;
$Shared = new Shared();
for($i = 0, $refs = array(); $i < $n; $i++){
$refs[$i] = new T($Shared, $i);
$refs[$i]->start();
}
while(!empty($refs)){
// print status:
if(!isset($t)or microtime(true)-$t > 1){
$t = microtime(true);
echo 'status: ' . count($refs) . ' running atm ...' . PHP_EOL;
}
// join crashed threads:
foreach($refs as $i => $thread){
if($thread->isRunning() === false){
echo 'T-' . $i . ' stopped after ' . $thread->count . PHP_EOL;
if($thread->isJoined() === false){
$thread->join();
}
unset($refs[$i]);
}
}
}
echo 'no thread running anymore.' . PHP_EOL;
/* output
start ...
status: 8 running atm ...
Notice: Undefined offset: 6 in ...\shared_test.php on line 33
Fatal error: Call to a member function isRunning() on null in ...\shared_test.php on line 82
T-6 stopped after 10
status: 7 running atm ...
Notice: Undefined offset: 4 in ...\shared_test.php on line 33
Fatal error: Call to a member function isRunning() on null in ...\shared_test.php on line 82
T-4 stopped after 35
status: 6 running atm ...
Notice: Undefined offset: 7 in ...\shared_test.php on line 33
Fatal error: Call to a member function isRunning() on null in ...\shared_test.php on line 82
T-7 stopped after 43
status: 5 running atm ...
status: 5 running atm ...
status: 5 running atm ...
[...]
*/
?>
Threaded objects are already thread safe, that is to say that, any time you read, write, check for the existence of, or delete (unset) a member, the operation is atomic - no other context can perform any of the aforementioned operations while the first operation takes place. The same is true for engine handlers that the user is unaware of, everything down to the lowest level is implicitly safe.
Quite re-assuring, however ... This has obvious limits when the logic gets more complex, such as checking the existence of a member before setting or doing something else with it, as you are doing: While the operations on the object are atomic, there is nothing to stop another context unseting a member between the call to isset and the call to read the property/dimension.
This applies to PHP7 (pthreads v3+)
Safety, and integrity are two different things here. When integrity is important you can use Threaded::synchronized in PHP7 to preserve it correctly. In PHP5 you could preserve it also, but the code would be more complicated, as would the explanation.
Your second example should run indefinitely, if I understand it's logic. So I'm using that assumption to construct the correct code, I'm going to make further assumptions about what you might want to do in this endless loop and provide some insight where it seems required.
<?php
class Referee extends Threaded {
public function find(string $ident, Threaded $reference) {
return $this->synchronized(function () use($ident, $reference) {
if (isset($this[$ident])) {
return $this[$ident];
} else return ($this[$ident] = $reference);
});
}
public function foreach(Closure $closure) {
$this->synchronized(function() use($closure) {
foreach ($this as $ident => $reference) {
$closure($ident, $reference);
}
});
}
}
class Test extends Thread {
public function __construct(Referee $referee, string $ident, bool $delay) {
$this->referee = $referee;
$this->ident = $ident;
$this->delay = $delay;
}
public function run() {
while (1) {
if ($this->delay) {
$this->synchronized(function(){
$this->wait(1000000);
});
}
$reference =
$this->referee->find($this->ident, $this);
/* do something with reference here, I guess */
/* do something with all references here */
$this->referee->foreach(function($ident, $reference){
var_dump(Thread::getCurrentThreadId(),
$reference->getIdent(),
$reference->isRunning());
});
}
}
public function getIdent() {
return $this->ident;
}
private $referee;
private $ident;
private $delay;
}
$referee = new Referee();
$threads = [];
$thread = 0;
$idents = [
"smelly",
"dopey",
"bashful",
"grumpy",
"sneezy",
"sleepy",
"happy",
"naughty"
];
while ($thread < 8) {
$threads[$thread] = new Test($referee, $idents[$thread], rand(0, 1));
$threads[$thread]->start();
$thread++;
}
foreach ($threads as $thread)
$thread->join();
?>
So we'll look at the differences, I'll tell you why they are as they are and how else you might write them, you already know that we're not talking about safety now, but integrity, you are afforded the (quite remarkable) assumption that anything you write is "safe", as explained.
The first major difference, is this:
if ($this->delay) {
$this->synchronized(function(){
$this->wait(1000000);
});
}
This is simply a suitable way to make a Thread wait, you wouldn't have to use the Thread itself to synchronize, you could use any Threaded object. The benefit of doing things properly, in case not clear is that, sleep and usleep do not leave threads in a receptive state, using ::wait does.
In the real world, where you really should only ever wait for something, that would be a more complex block, it might (and should) look more like:
if ($this->delay) {
$this->synchronized(function(){
while ($this->condition) {
$this->wait(1000000);
}
});
}
Note: waiting for a timeout is technically waiting for something, however, you might be awoken by something other than the timeout having been reached, and code should be prepared for that.
Such that, another context is able to notify the Thread that it should stop waiting and shutdown gracefully, or carry out some other important action immediately, simply by synchronizing, changing a condition and notifying the Thread.
For predictable code, it's extremely important to get comfortable with how synchronized, wait and notify work.
Next we have our logic for setting and or getting the reference:
$reference =
$this->referee->find($this->ident, $this);
Which invokes this:
public function find(string $ident, Threaded $reference) {
return $this->synchronized(function () use($ident, $reference) {
if (isset($this[$ident])) {
return $this[$ident];
} else return ($this[$ident] = $reference);
});
}
This is badly named, naming things is hard, but you can see that integrity is preserved by synchronization while these grouped operations take place. The same method could also be used to fetch a reference to another object, with a bit of tweaking.
I guess you do something with that particular reference (which is always going to be $this currently). I can't guess what. Moving on ...
I've made the assumption that you'll want to do something with each of these Threads, and you want to preserve the integrity of the data while the entire iteration takes place:
$this->referee->foreach(function($ident, $reference){
var_dump(Thread::getCurrentThreadId(),
$reference->getIdent(),
$reference->isRunning());
});
Which invokes:
public function foreach(Closure $closure) {
$this->synchronized(function() use($closure) {
foreach ($this as $ident => $reference) {
$closure($ident, $reference);
}
});
}
This is how you would do such a thing.
It's worthy of mention that synchronized is not necessarily required here; just as nothing bad will happen if you remove a member from an array you are iterating over, nothing bad will happen if you unset or set or do anything else to an object while iteration occurs.
Related
We have PHP code in production that sometimes fails with "Call to member function on null", although the same code path executes fine several times before that in one invocation. We have a test that reproduces the error consistently at the same run of the loop.
I already proved that the object gets created correctly in the factory even if it gets returned as null. The factory method must not return null in any case, as indicated in the DocBlock. This question is not related to nullable return types or something like that.
The process does not exceed memory or runtime limitations and I already tried turning off the garbage collector, but no luck. The error happens both in PHP 7.0 and 7.3 on Debian, did not try on other versions or operating systems.
I am not allowed to paste the real code here, but I wrote a simple mockup to explain in more detail. Please keep in mind that this demo code will not result in the error, it is just meant to show the general structure of the program that runs into this fault.
// Three placeholder classes with common methods
class Bender
{
public function common()
{
echo "Bend, bend!" . PHP_EOL;
}
}
class Clamper
{
public function common()
{
echo "Clamp, clamp!" . PHP_EOL;
}
}
class Worker
{
public function common()
{
echo "Work, work!" . PHP_EOL;
}
}
// abstract class with static factory to produce objects
abstract class MomCorp
{
/**
* Factory to create one of several objects
*
* #param string $name
* #return Bender|Clamper|Worker
*/
public static function factory($name)
{
$type = self::managementDecision($name);
switch ($type)
{
case "bender":
$robot = new Bender();
break;
case "clamper":
$robot = new Clamper();
break;
default:
$robot = new Worker();
}
// optional QA works flawlessly here, object is fine all the time!
// $robot->common();
return $robot;
}
public static function managementDecision($name)
{
// irrelevant magic happens on $name here
return "bender";
}
}
foreach (['Rodriguez', 'Angle-ine', 'Flexo', 'Billie'] as $newName)
{
echo "$newName: ";
// The next two lines break after some loops - why?
// The perfectly functional object in factory gets returned as null
$bot = MomCorp::factory($newName);
$bot->common();
}
// SAMPLE OUTPUT
// Rodriguez: Bend, bend!
// Angle-ine: Bend, bend!
// Flexo: Bend, bend!
// Billie: Call to a member function common() on null
Has anyone experienced the same and has any hints on what might cause such an error and how to fix it?
here is what im doing
im using
pthreads from - pthreads.org
php Simple Html DOM parser from - simplehtmldom.sourceforge.net
now the process of what i (will) do is:
I am reading a bulk of links that is from a text file.
I initialize a thread to have a separate process
I am creating a log file for this thread so that i will know, what happens later on.
now, this is my code for my thread class.
class ReadLinks extends Thread {
private $conn;
private $links;
private $fileObj;
public function __construct($conn, $links, $fileObj) {
//.. well just asign this to the global variables
}
public function run() {
try {
$this->logMsg("Start Reading Reviews");
$this->readLinks();
} catch (Exception $ex) {
$this->logMsg($ex);
}
$this->closeLog();
}
private function readLinks() {
$this->logMsg("Links");
foreach ($this->links as $link) {
$link = trim(preg_replace('/\s\s+/', ' ', $link));
$this->logMsg("Link: " . $link);
$html = html_readLink($link);
break;
}
}
private function logMsg($msg) {//something to write on the text file
}
private function closeLog() {//closes the textfile
}}
$conn - is my mysqli link to have db actions in the future
$links - is an array of links to be read.
$fileObj- is a resource return from fopen(). ( well to write into a file)
now who is that html_readlink,
its an outer function that is like this:
function html_readLink($link) {
return file_get_html($link);}
basically it is the resource returned by a function from simple html dom parser
now, i have as well a function that reads a link alone to do the other (different business requirement) and im using the simple html dom parser with ease.
with the pthreads, i tried writing the file(logs first) so that i can ensure that everything as well works fine.
about contacting db is not yet sure., well ill try to figure it out if it works fine, but logically it should work.
now when i called this class: its like this:
try {
$thread = new readLinks($conn, $Links, createlog());
if ($thread->start()) {
$thread->join();
} else {
echo "something i need to research if this happens";
}
} catch (Exception $err) {
echo $err; //something i need to research as well if this happens
}
i got this error
Warning: Invalid argument supplied for foreach() in C:\my\path\to\simplehtmldom_1_5\simple_html_dom.php on line 1119
that simplehtmldom code is:
function clear()
{
foreach ($this->nodes as $n) {$n->clear(); $n = null;}
// This add next line is documented in the sourceforge repository. 2977248 as a fix for ongoing memory leaks that occur even with the use of clear.
if (isset($this->children)) foreach ($this->children as $n) {$n->clear(); $n = null;}
if (isset($this->parent)) {$this->parent->clear(); unset($this->parent);}
if (isset($this->root)) {$this->root->clear(); unset($this->root);}
unset($this->doc);
unset($this->noise);
}
now that is the source code coming from simple html dom. that foreach is the one that is returning the error. now my other code using simple html dom doesn't have a problem with simple html dom. but with pthreads i got this error.
also, when i change my codes and didn't use pthreads, (had some revisions like this:
on pthreads class:
class ReadLinks {// extends Thread {
//insert other codes
public function readLinks() {
$this->logMsg("Links");
foreach ($this->links as $link) {
$link = trim(preg_replace('/\s\s+/', ' ', $link));
$this->logMsg("Link: " . $link);
$html = html_readLink($link);
$this->logMsg(getTitle($html));
//
break;
}
}
and change the way this is called like this:
try {
$thread = new ReadLinks($conn, $revLinks, createlog());
$thread->readLinks();
// if ($thread->start()) {
// $thread->join();
// } else {
// echo "something i need to research if this happens";
// }
} catch (Exception $err) {
echo $err; //something i need to debug and research if this happens
}
everything works fine, i get the desired results.
pthreads is something i need to use since loading bulk links and reading each of them is quite a time consuming process. and i need it to be on a separate thread. now i dont know whats wrong with these pthreads, or simple html dom parser. have i done something unnecessary/wrong? is there other way to do this?
anyone??
EDIT
i followed the answer of Prafulla Kumar Sahu:
the new code for the function clear() of simple html dom is:
function clear() {
if (is_array($this->nodes) || $this->nodes instanceof Traversable) {
foreach ($this->nodes as $n) {
$n->clear();
$n = null;
}
}
// This add next line is documented in the sourceforge repository. 2977248 as a fix for ongoing memory leaks that occur even with the use of clear.
if (isset($this->children))
foreach ($this->children as $n) {
$n->clear();
$n = null;
}
if (isset($this->parent)) {
$this->parent->clear();
unset($this->parent);
}
if (isset($this->root)) {
$this->root->clear();
unset($this->root);
}
unset($this->doc);
unset($this->noise);
}
the result is: it eliminated the error
but it is not the desired result
when using the function
$x=$resource->find($selector,0);
//resource is the return obj of file_gets_content, selector is my css selector string
it returns null/empty where in fact it should have a value.
ive checked a separate function that uses the simple html dom after i updated their code, seems it wasn't affected, and it is working properly. but with my pthread class, it is not working correctly.
The code I have doesn't have a foreach on line 1119; maybe you have an older version. You're getting a warning only, what problem(s) do you see in the results?
1117 // save dom as string
1118 function save($filepath='')
1119 {
1120 $ret = $this->root->innertext();
1121 if ($filepath!=='') file_put_contents($filepath, $ret, LOCK_EX);
1122 return $ret;
1123 }
It happens if the variable you are trying to traverse using foreach is not irritable so please check if your variable is either an array or instanceof Traversable class .
*It may be because you are not getting any value for that variable you want to traverse.
so, I would suggest you to use is_array( $whatever ) || $whatever instanceof Traversable just before foreach.
ie.
if( is_array( $whatever ) || $whatever instanceof Traversable ){
foreach( $whatever as $what ){
//some code
}
}
In your case it is
function clear()
{
foreach ($this->nodes as $n) {$n->clear(); $n = null;}
// This add next line is documented in the sourceforge repository. 2977248 as a fix for ongoing memory leaks that occur even with the use of clear.
if (isset($this->children)) foreach ($this->children as $n) {$n->clear(); $n = null;}
if (isset($this->parent)) {$this->parent->clear(); unset($this->parent);}
if (isset($this->root)) {$this->root->clear(); unset($this->root);}
unset($this->doc);
unset($this->noise);
}
source:- https://github.com/jalbertbowden/simplehtmldom/blob/master/simplehtmldom_1_5/simple_html_dom.php#L1119
this means you are unable to get $this->nodes correctly, so please var_dump it before you are calling function clear or before the foreach .
How should I multithread some php-cli code that needs a timeout?
I'm using PHP 5.6 on Centos 6.6 from the command line.
I'm not very familiar with multithreading terminology or code. I'll simplify the code here but it is 100% representative of what I want to do.
The non-threaded code currently looks something like this:
$datasets = MyLibrary::getAllRawDataFromDBasArrays();
foreach ($datasets as $dataset) {
MyLibrary::processRawDataAndStoreResultInDB($dataset);
}
exit; // just for clarity
I need to prefetch all my datasets, and each processRawDataAndStoreResultInDB() cannot fetch it's own dataset. Sometimes processRawDataAndStoreResultInDB() takes too long to process a dataset, so I want to limit the amount of time it has to process it.
So you can see that making it multithreaded would
Speed it up by allowing multiple processRawDataAndStoreResultInDB() to execute at the same time
Use set_time_limit() to limit the amount of time each one has to process each dataset
Notice that I don't need to come back to my main program. Since this is a simplification, you can trust that I don't want to collect all the processed datasets and do a single save into the DB after they are all done.
I'd like to do something like:
class MyWorkerThread extends SomeThreadType {
public function __construct($timeout, $dataset) {
$this->timeout = $timeout;
$this->dataset = $dataset;
}
public function run() {
set_time_limit($this->timeout);
MyLibrary::processRawDataAndStoreResultInDB($this->dataset);
}
}
$numberOfThreads = 4;
$pool = somePoolClass($numberOfThreads);
$pool->start();
$datasets = MyLibrary::getAllRawDataFromDBasArrays();
$timeoutForEachThread = 5; // seconds
foreach ($datasets as $dataset) {
$thread = new MyWorkerThread($timeoutForEachThread, $dataset);
$thread->addCallbackOnTerminated(function() {
if ($this->isTimeout()) {
MyLibrary::saveBadDatasetToDb($dataset);
}
}
$pool->addToQueue($thread);
}
$pool->waitUntilAllWorkersAreFinished();
exit; // for clarity
From my research online I've found the PHP extension pthreads which I can use with my thread-safe php CLI, or I could use the PCNTL extension or a wrapper library around it (say, Arara/Process)
https://github.com/krakjoe/pthreads (and the example directory)
https://github.com/Arara/Process (pcntl wrapper)
When I look at them and their examples though (especially the pthreads pool example) I get confused quickly by the terminology and which classes I should use to achieve the kind of multithreading I'm looking for.
I even wouldn't mind creating the pool class myself, if I had a isRunning(), isTerminated(), getTerminationStatus() and execute() function on a thread class, as it would be a simple queue.
Can someone with more experience please direct me to which library, classes and functions I should be using to map to my example above? Am I taking the wrong approach completely?
Thanks in advance.
Here comes an example using worker processes. I'm using the pcntl extension.
/**
* Spawns a worker process and returns it pid or -1
* if something goes wrong.
*
* #param callback function, closure or method to call
* #return integer
*/
function worker($callback) {
$pid = pcntl_fork();
if($pid === 0) {
// Child process
exit($callback());
} else {
// Main process or an error
return $pid;
}
}
$datasets = array(
array('test', '123'),
array('foo', 'bar')
);
$maxWorkers = 1;
$numWorkers = 0;
foreach($datasets as $dataset) {
$pid = worker(function () use ($dataset) {
// Do DB stuff here
var_dump($dataset);
return 0;
});
if($pid !== -1) {
$numWorkers++;
} else {
// Handle fork errors here
echo 'Failed to spawn worker';
}
// If $maxWorkers is reached we need to wait
// for at least one child to return
if($numWorkers === $maxWorkers) {
// $status is passed by reference
$pid = pcntl_wait($status);
echo "child process $pid returned $status\n";
$numWorkers--;
}
}
// (Non blocking) wait for the remaining childs
while(true) {
// $status is passed by reference
$pid = pcntl_wait($status, WNOHANG);
if(is_null($pid) || $pid === -1) {
break;
}
if($pid === 0) {
// Be patient ...
usleep(50000);
continue;
}
echo "child process $pid returned $status\n";
}
I'm trying to make an error logging class, I have some functions to set up various output methods like DB, file, return and screen. I want all errors to be stored into an array and when __destruct() is called, I want to stop the client from waiting for data and the log details about errors the user experienced. This way they don't have to report errors to me.
I have 2 modes, a simple GUI to test the functionality and the actual script generates responses in JSON, machine to machine. For GUI the final dump is fine but for JSON it destroys the response. So all error reporting is off and I have to handle any errors that would be dumped on screen myself, hence the $return in function flush_log($return) which make the function return a string if set to true.
After reporting flushing the errors I want to:unset($this->log_arrays)
or empty: $this->log_arrays=Array();, but it is out of scope - I understand why, my function uses a local copy - but how do I reset the array?
[EDIT]:
I tried:
$this->log_arrays = Array();
$this->log_arrays = null;
array popping:
for ($i = 1; count($this->log_arrays); $i++)
{
array_pop($this->log_arrays);
}
But I think none of it can work because within a class function you work with copies of variables, so they're basically out of scope.
[/EDIT]:
This is an already simplified class.. :
<?php
class log_strings extends mysqli
{
private $log_arrays = Array();
public function __construct($output_to_file=false, $output_to_db=true, $fall_back_file=true, $arguments, $ip=null)
{
// Setup mysqli connection, file handle or report error if one or all have failed.
// Also check wich outputs should be used and keep that info for later.
}
public function log($level, $string)
{
$log_arrays[] = Array('level' => $level, 'string' => $string);
}
public function __destruct()
{
$this->flush_log();
}
public function flush_log($return=false)
{
if (!isset($log_arrays) && count($log_arrays) == 0)
{
return true;
}
if ($return)
{
return $this->return_output();
}
else
{
$success = false;
// if enabled, output to db
if ($this->output_to_db)
{
$success = $success || $this->mysqli_output();
}
// if enabled or if db failed and fallback is enabled, output to file
if ($this->output_to_file || ($this->fall_back_file && !$success))
{
$success = $success || $this->file_output();
}
// if neither file or db succeeded, dump on screen
if ($success = false)
{
$this->screen_dump();
}
return true;
}
unset($this->log_arrays); // <= This is what it is all about!
}
private function screen_dump()
{
foreach($this->log_arrays as $array)
{
echo "<strong>{$array['level']}</strong>{$array['string']}<br/>\n";
}
}
private function mysqli_output()
{
// Output to db functionally equal to $this->screen_dump()
}
private function file_output()
{
// Output to file functionally equal to $this->screen_dump()
}
private function return_output()
{
// Return output functionally equal to $this->screen_dump()
}
}
?>
Resetting the array should work
$this->log_arrays = array();
Unset a class property is a very bad idea. Because it may be used in other methods or other classes using a potentionally getter of your class.
I am trying to implement the singleton pattern in php like described here in Example #2:
http://www.php.net/singleton
When I run the example code
$singleton = Example::singleton(); // prints "Creating new instance."
echo $singleton->increment(); // 0
echo $singleton->increment(); // 1
$singleton = Example::singleton(); // reuses existing instance now
echo $singleton->increment(); // 2
echo $singleton->increment(); // 3
it allways ends with Fatal Error 'Clone is not allowed.' directly after 'Creating new instance.'
I would expect that there is no reason for php to call the __clone-method.
In another real-life project of mine I want to have a singleton PlayerManager that holds Player-objects in an array (loaded only once in __construct) and has functions like GetPlayers() or GetPlayersByID($id).
In my script I write something like
$pm = PlayerManager::GetInstance();
$p1 = $pm->GetPlayerByID(0);
echo $p1->SomeNumber; //100
$p1->SomeNumber = 200;
$p2 = $pm->GetPlayerByID(0);
echo $p2->SomeNumber; //100 and not 200, as I would expect
Can someome give me some hints how to implement the PlayerManager using the Singleton pattern correct? I'm not sure if it is only a problem with the singleton or also a problem with returning object references...
I'm not quiet sure why you're getting that error (post your singleton class if you want help with that). Though I always preferred this version to the one you're using, it's a bit simpler: http://www.talkphp.com/advanced-php-programming/1304-how-use-singleton-design-pattern.html
So with the above, your code would look like:
class Counter
{
$CurrentValue = 0;
// Store the single instance of Database
private static $m_pInstance;
private function __construct() { }
public static function getInstance()
{
if (!self::$m_pInstance)
{
self::$m_pInstance = new Counter();
}
return self::$m_pInstance;
}
public function increment ($by)
{
$this->CurrentValue += $by;
return $this->CurrentValue;
}
public function getValue ()
{
return $this->CurrentValue;
}
}
And to use:
$counter = Counter::getInstance();
echo $counter->increment(); // 0
echo $counter->increment(); // 1
$counter = null;
$counter = Counter::getInstance(); // reuses existing instance now
echo $counter->increment(); // 2
echo $counter->increment(); // 3
Tell me how that works out for you.