====== PHP True Async: Scope and Structured Concurrency ======

  * Version: 1.0
  * Date: 2025-11-03
  * Author: Edmond [HT], edmondifthen@proton.me
  * Status: Draft
  * First Published at: http://wiki.php.net/rfc/true_async_scope
  * Git: https://github.com/true-async
  * Related RFC: [[rfc:true_async|PHP True Async (Base RFC)]]

===== Introduction =====

This RFC extends the [[rfc:true_async|PHP True Async (Base RFC)]] with **Scope** and **Structured Concurrency** functionality.

**Note:** This RFC builds upon the base async functionality. Please read the [[rfc:true_async|Base RFC]] first.

Structured concurrency is a programming paradigm that ties the lifetime of concurrent operations to the lexical scope in which they are created. This provides several benefits:

  * **Automatic lifetime management**: Coroutines are automatically cancelled when their scope ends
  * **Predictable resource cleanup**: Resources are released in a deterministic order
  * **Error propagation**: Exceptions in child coroutines are propagated to parent scope
  * **Prevention of "fire-and-forget"**: Eliminates zombie coroutines that outlive their intended lifetime

The ''Scope'' class provides a container for managing groups of coroutines with shared lifetime and cancellation semantics.

===== Goals =====

The **Scope and Structured Concurrency** implementation pursues the following goals:

  * **Prevent resource leaks**: Ensure all coroutines complete or are properly cancelled before scope exits
  * **Predictable lifetime management**: Make coroutine lifetimes explicit and manageable
  * **Error handling**: Provide clear error propagation from child coroutines to parent scope
  * **Zombie prevention**: Eliminate coroutines that continue executing after their intended lifetime
  * **Flexible patterns**: Support both fire-and-forget and structured concurrency patterns
  * **Critical sections**: Provide mechanisms to protect code sections from cancellation

===== Proposal =====

==== Overview ====

=== Short glossary ===

| Term                       | Description                                                                    | Section                                     |
| **Scope**                  | A container for managing the lifecycle of coroutines                           | Scope                                       |
| **Zombie coroutine**       | A coroutine that continues execution after its Scope has been destroyed        | Scope disposal                              |
| **Structured concurrency** | A paradigm where coroutine lifetimes are bound to lexical scope                | Structured concurrency                      |

This **RFC** describes structured concurrency functionality for PHP, which includes:


=== Scope ===
A container that manages coroutine lifetimes.
Example
<code php>
use function Async\spawn;

$scope = new Async\Scope();
$scope->spawn(function() {
    // Coroutine bound to $scope
    spawn(function() {
        // Coroutine bound to $scope
    });
});

// Dispose of the scope after 5 seconds
sleep(5);
$scope->disposeSafely();
</code>

=== Structured concurrency ===

<code php>
use Async\Scope;

function getUserProfile(int $userId): array {
    $scope = new Scope();
    $profile = [];

    $scope->spawn(function() use ($userId, &$profile) {
        $profile['data'] = fetchUserData($userId);
    });

    $scope->spawn(function() use ($userId, &$profile) {
        $profile['orders'] = fetchUserOrders($userId);
    });

    $scope->spawn(function() use ($userId, &$profile) {
        $profile['settings'] = fetchUserSettings($userId);
    });

    $scope->awaitCompletion(Async\timeout(30000));
    return $profile;
}
</code>

=== Spawn with scope ===

To create a coroutine in a specific ''Scope'',
you need to use the ''Scope::spawn'' method.:

<code php>
$scope = new Async\Scope();

// Launch a coroutine in the $scope
$coroutine = $scope->spawn(function():string {
    return gethostbyname('php.net');
});

function defineTargetIpV4(string $host): string {
    return gethostbyname($host);
}

$coroutine = $scope->spawn(defineTargetIpV4(...), $host);
</code>

==== Coroutine Scope ====

> **Coroutine Scope** — the space associated with coroutines created using the ''spawn'' expression.

=== Coroutine lifetime ===

> A coroutine launched without an **explicitly defined** Scope has an **unknown** lifetime.

A coroutine created with spawn inherits the current Scope and its lifetime.
If code calls a third-party library function (or another module) that uses ''spawn()'' internally,
the library ends up creating a coroutine whose lifetime is unknown to it.
The programmer **should carefully** consider such situations
and determine whether an undefined coroutine lifetime is what they expect in that context.

If the coroutine lifetime matters, it is recommended to explicitly define a ''Scope''
to clearly manage coroutine lifetimes.

=== Motivation ===

Sometimes it is necessary to gain control not only over a currently running coroutine,
but also over all coroutines that will be launched within a new one — without having direct access to them.
This could be the case for web server code that handles requests in separate coroutines
and does not know how many additional coroutines will be launched,
or a ''JobExecutor'' that wants to manage the lifecycle of running jobs.

Without such control, the application code loses the ability to resist runtime errors,
which increases the risk of a complete service failure.

This is why the **Coroutine Scope** pattern is of critical importance in the context of ensuring reliability.

The main use cases for ''Scope'' are:

  - Controlling the lifetime of coroutines created within a single scope (**Point of responsibility**)
  - Handling errors from all coroutines within the scope
  - Binding the lifetime of the scope's coroutines to the lifetime of a **PHP object**
  - Creating a hierarchy of scopes to manage coroutines in a structured way

Binding ''Scope'' to objects is a good practice that has proven effective in **Kotlin**.
By allowing coroutines to be tied to an object (this could be a ''Screen'' or a ''ViewModel''),
it is possible to avoid the error where coroutines outlive the object that manages them.

For frameworks, it can be useful to be able to control all coroutines created within a ''Scope'',
to apply context-dependent constraints to them.

=== Scope propagation ===

By default, all coroutines are associated with the **Global Coroutine Scope**:

<code php>
use function Async\spawn;

spawn('file_get_contents', 'file1.txt'); // <- global scope

function readFile(string $file): void {
    return file_get_contents($file); // <- global scope
}

function mainTask(): void { // <- global scope
    spawn('readFile', 'file1.txt'); // <- global scope
}

spawn('mainTask'); // <- global scope
</code>

If an application never creates **custom Scopes**, its behavior is similar to coroutines in ''Go'':
  * Coroutines are not explicitly linked to each other.
  * The lifetime of coroutines is not limited.

The method ''Scope::spawn'' creates a **new coroutine** bound to the specified scope.
Scope is propagated between coroutines.
If a coroutine is launched within a specific Scope, that Scope is considered the current one.
The function like ''spawn(<callable>)'' will create a coroutine within the current Scope.

Coroutines created during the execution of this **new coroutine** will become **sibling tasks**:

<code php>
use function Async\spawn;
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() { // <- new scope

    echo "Sibling task 1\n";

    spawn(function() { // <- $scope is current scope
        echo "Sibling task 2\n";

        spawn(function() { // <- $scope is current scope
            echo "Sibling task 3\n";
        });
    });
});

$scope->awaitCompletion(Async\timeout(60000));
</code>

**Expected output**

<code>
Sibling task 1
Sibling task 2
Sibling task 3
</code>

**Structure:**

<code>
main()                          ← defines a $scope
└── $scope = new Scope()
    ├── task1()                 ← runs in the $scope
    ├── task2()                 ← runs in the $scope
    ├── task3()                 ← runs in the $scope
</code>

Thus, the method ''Scope::spawn'' creates a new branch of sibling coroutines,
where the new coroutine exists at the same level as all subsequent ones.

The code that owns a ''Scope'' object becomes the **Point of responsibility**
for all coroutines executed within that Scope.

> A good practice is to ensure that a Scope object has **only ONE owner**.
> Passing ''$scope'' as a parameter to other functions or assigning it to multiple objects
> is a **potentially dangerous** operation that can lead to complex bugs.
> When a ''$scope'' is owned by multiple modules/classes,
> there is a risk that they may either accidentally extend the Scope’s lifetime
> or accidentally call ''scope::dispose()'' and disrupt its lifecycle.
> Because the code is asynchronous, finding the real cause in such cases is much harder than
> if the Scope had only a single owner.

=== Scope waiting ===

The ''Scope'' class does not implement the ''Awaitable'' interface,
and therefore cannot be used in an ''await'' expression.

Awaiting a ''Scope'' is a potentially **dangerous operation** that should be performed consciously,
not accidentally.

A ''Scope'' can own a coroutine that was created “by someone else” “somewhere else.”
For example, a programmer explicitly defines a ''$scope'',
and then inside a coroutine that belongs to this ''$scope'' calls a function
from a third-party library, which in turn uses ''spawn()''.

<code php>
use function Async\spawn;
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    // Call to third-party library function
    thirdPartyFunction();
});

function thirdPartyFunction() {
    // Library code spawns a coroutine
    spawn(function() {
        sleep(10000); // <- long operation
    });
}
</code>

When you await a ''Scope'', you do **not know** which coroutines you are actually awaiting.
Even if ''$scope'' is used by only one class or module, there is still a chance
that someone accidentally created an “incorrect coroutine.”

That is why awaiting a ''Scope'' must be bounded by a cancellation token.

> To await a group of coroutines,
> a specialized primitive should be used, which will be described in a separate ''RFC''.

There are several Use-Cases where waiting for a ''Scope'' might be necessary:

  * Structured concurrency: when a parent awaits the completion of all child coroutines.
  * Waiting for Scope tasks to complete the cancellation process (see section: [[#cancellation|Cancellation]]).

The structured concurrency pattern with waiting for all child coroutines can be useful
for applications whose lifetime is explicitly limited by external conditions.
For example, the user might stop a console application.

To support a task awaiting in a controlled manner, ''Scope'' provides two specific methods:

  * ''public function awaitCompletion(Awaitable $cancellation): void {}''
  * ''public function awaitAfterCancellation(?callable $errorHandler = null, ?Awaitable $cancellation = null): void {}''

The ''awaitCompletion'' method blocks the execution flow until all tasks within the scope are completed.
The ''awaitAfterCancellation'' method does the same but is intended to be called only after the scope has been cancelled.

<code php>
use function Async\spawn;
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {

    echo "Sibling task 1\n";

    spawn(function() {
        echo "Sibling task 2\n";

        spawn(function() {
            echo "Sibling task 3\n";
        });
    });
});

$scope->awaitCompletion(Async\timeout(60000));
</code>

**Expected output**

<code>
Sibling task 1
Sibling task 2
Sibling task 3
</code>

The ''Scope'' awaiting methods do not capture any task results,
so they cannot be used to await return values.

The ''awaitCompletion'' method can only be used with an explicitly defined cancellation token.
This requirement helps prevent indefinite waiting.

Awaiting the ''$scope'' object also allows handling exceptions from coroutines within the ''$scope'':

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    spawn(function() {
        spawn(function() {
            throw new Exception("Error occurred");
        });
    });
});

try {
    $scope->awaitCompletion(Async\timeout(60000));
} catch (Exception $exception) {
    echo $exception->getMessage()."\n";
}
</code>

**Expected output**

<code>
Error occurred
</code>

Calling the ''awaitCompletion()'' method after the ''Scope'' has been cancelled
will immediately throw a cancellation exception.

<code php>
use Async\Scope;

$scope = new Scope();

try {
    $scope->spawn(task1(...));
    $scope->spawn(task2(...));
    $scope->cancel();
    // Wait all tasks
    $scope->awaitCompletion(Async\timeout(60000));
} catch (Exception $exception) {
    echo "Caught exception: ",$exception->getMessage()."\n";
}
</code>

**Expected output**

<code>
Caught exception: cancelled at ...
</code>

In this example, ''$scope->awaitCompletion(Async\timeout(60000));'' will immediately throw an exception.

If you need to wait for the ''Scope'' to complete after it has been cancelled,
use the special method ''awaitAfterCancellation'', which is designed for this case.

<code php>
use function Async\spawn;
use Async\Scope;

$scope = new Scope();

spawn(function() {
    try {
        $scope->awaitCompletion(Async\timeout(60000));
    } catch (\Async\CancellationError $exception) {
        $scope->awaitAfterCancellation();
        echo "Caught exception: ",$exception->getMessage()."\n";
    }
});

$scope->spawn(function() use ($scope) {
    $scope->cancel();

    try {
        Async\delay(1000);
    } finally {
        sleep(1);
        echo "Finally\n";
    }
});
</code>

**Expected output**

<code>
Finally
Caught exception: cancelled at ...
</code>

In this example, the line ''Finally'' will be printed first because ''$scope->awaitAfterCancellation()''
waits for all coroutines inside the Scope to complete.

The ''awaitAfterCancellation'' method is used in scenarios where final resource cleanup is required
after all child tasks are guaranteed to have finished execution.

There is also a risk of indefinite waiting, so it is **recommended** to explicitly specify a timeout.

=== Scope Hierarchy ===

A hierarchy can be a convenient way to describe an application as a set of dependent tasks:

  * Parent tasks are connected to child tasks and are responsible for their execution time.
  * Tasks on the same hierarchy level are independent of each other.
  * Parent tasks should control their child's tasks.
  * Child tasks MUST NOT control or wait for their parent tasks.
  * It is correct if tasks at the same hierarchy level are only connected to tasks of the immediate child level.

<code>
WebServer
├── Request Worker
│   ├── request1 task
│   │   ├── request1 subtask A
│   │   └── request1 subtask B
│   └── request2 task
│       ├── request2 subtask A
│       └── request2 subtask B
</code>

The work of a web server can be represented as a hierarchy of task groups that are interconnected.
The ''Request Worker'' is a task responsible for handling incoming requests. There can be multiple requests.
Each request may spawn subtasks. On the same level, all requests form a group of request-tasks.

''Scope'' is fit for implementing this concept:

<code>
WebServer
├── Request Worker
│   ├── request1 Scope
│   │   ├── request1 subtask A
│   │   │   └── subtask A Scope
│   │   │       ├── sub-subtask A1
│   │   │       └── sub-subtask A2
│   │   └── request1 subtask B
│   └── request2 Scope
│       ├── request2 subtask A
│       └── request2 subtask B
│           └── subtask B Scope
│               └── sub-subtask B1
</code>

The ''new Scope()'' constructor creates a **root independent Scope**.

To create a **child Scope**, you need to use a special constructor:

''Scope::inherit(?Scope $parentScope = null)''

It returns a new ''Scope'' object that acts as a child.
A coroutine created within the child ''Scope'' can also be considered
a child relative to the coroutines in the parent ''Scope''.

If the ''$parentScope'' parameter is not specified, the current ''Scope'' will be used as the parent.

**An example:**
<code php>
use function Async\spawn;
use Async\Scope;
use Async\CancellationError;

// New independent root scope
$scope = new Scope();

// child scope from $scope number 1
$childScope1 = Scope::inherit($scope);

$scope->spawn(function() {
    $childScope2 = Scope::inherit(); // child scope from $scope number 2
});
</code>

**WebServer example:**

<code php>
use function Async\await;
use Async\Scope;
use Async\CancellationError;

function webServer(): void
{
    // Creating a new Root Scope
    $scope = new Scope();

    // socket server code that accepts connections...
    while ($socket = acceptConnection()) {
        $scope->spawn(connectionHandler(...), $socket);
    }
}

function connectionHandler($socket): void
{
    // Current scope is the Root Scope
    // $scope is inherited from the current Scope.
    $scope = Scope::inherit();

    // We do not provide direct access to the Scope object in other functions
    $cancelToken = fn(string $message) => $scope->cancel(new CancellationError($message));

    $scope->spawn(function() use ($socket, $cancelToken) {

        $limiterScope = Scope::inherit(); // child scope for connectionLimiter

        $limiterScope->spawn(function() {
            Async\delay(10000);
            $cancelToken("The request processing limit has been reached.");
        }, $cancelToken);

        handleRequest($socket, $cancelToken);
    });
}
</code>
Let's examine how this example works.

  - The ''webServer'' function creates a new ''Root Scope'' for the entire server.
  - Each time a new connection is accepted, a new coroutine is created within the ''Root Scope''.
  - ''connectionHandler'' creates an inherited ''Scope'' from the current context.
  - A ''$cancelToken'' closure is created to provide a simple way to cancel the entire scope with a custom message.
  - A child ''$limiterScope'' is created for timeout management.
  - A timeout coroutine automatically cancels the request after 10 seconds with the message "The request processing limit has been reached."

<code>
Root <- Root Scope
│
├── connectionHandler (Scope) <- inherited scope
│   │
│   ├── main request handling (Coroutine) <- $scope
│   │   └── timeout limiter (Coroutine) <- $limiterScope
│   │
</code>

The ''connectionHandler'' uses structured concurrency principles where the timeout limiter coroutine
is automatically managed within its dedicated ''$limiterScope''. When the request processing completes
or times out, all associated coroutines are properly cleaned up through the scope hierarchy.

''$limiterScope'' is used to isolate the timeout management logic from the main request handling
that should be cancelled when the request is completed.

=== Scope cancellation ===

The ''cancel'' method cancels all child coroutines and all child ''Scopes'' of the current ''Scope''.:

<code php>
use function Async\spawn;
use Async\Scope;

echo "Start\n";

$scope = new Scope();

$scope->spawn(function() {
    spawn(function() {
        Async\delay(1000);
        echo "Task 1\n";
    });

    spawn(function() {
        Async\delay(2000);
        echo "Task 2\n";
    });
});

$scope->cancel();
echo "End\n";
</code>

**Expected output**

<code>
Start
End
</code>

Another simple cancellation example:

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    sleep(1);
    echo "Task 1\n";
});

$scope->cancel(new Async\CancellationError('Task was cancelled'));
</code>

Canceling a ''Scope'' triggers the cancellation of all coroutines
within that ''Scope'' and all child ''Scopes'' in hierarchical order.

=== Self-cancellation of Scope ===

A coroutine can cancel its own ''Scope'', which **does** affect execution by cancelling all coroutines in the scope, including itself:

<code php>
use Async\Scope;

$scope = new Scope();
$scope->spawn(function() use ($scope) {
    $scope->cancel(new \Async\CancellationError("Scope cancelled"));
    echo "This executes\n";
});
</code>

=== Scope disposal ===

**Coroutine Scope** has several resource cleanup strategies
that can be triggered either explicitly, on demand,
or implicitly when the ''Scope'' object loses its last reference.

There are three available strategies for ''Scope'' termination:

| **Method**              |                                              |
| ''disposeSafely''       | Marks as zombie coroutines, does not cancel  |
| ''dispose''             | Cancels with a warning                       |
| ''disposeAfterTimeout'' | Issues a warning, then cancels after a delay |

The main goal of all three methods is to terminate the execution of coroutines
that belong to the ''Scope'' or its child Scopes.
However, each method approaches this task slightly differently.

The ''disposeSafely'' method is used by default in the destructor of the ''Async\Scope'' class.
Its key feature is transitioning coroutines into a **zombie coroutine** state.
A **zombie coroutine** continues execution but is tracked by the system differently than regular coroutines.
(See section: [[#zombie_coroutine_policy|Zombie coroutine policy]]).

A warning is issued when a **zombie coroutine** is detected.

<code php>
use function Async\spawn;
use function Async\await;
use Async\Scope;

$scope = new Scope();

await($scope->spawn(function() {
    spawn(function() {
        Async\delay(1000);
        echo "Task 1\n";
    });

    spawn(function() {
        Async\delay(2000);
        echo "Task 2\n";
    });

    echo "Root task\n";
}));

$scope->disposeSafely();
</code>

**Expected output**

<code>
Root task
Warning: Coroutine is zombie at ... in Scope disposed at ...
Warning: Coroutine is zombie at ... in Scope disposed at ...
Task 1
Task 2
</code>

The ''$scope'' variable is released immediately after the coroutine ''Root task'' completes execution,
so the child coroutine ''Task 1'' does not have time to execute
before the ''disposeSafely'' method is called.

''disposeSafely'' detects this and signals it with a warning but allows the coroutine to complete.

The ''Scope::dispose'' method differs from ''Scope::disposeSafely'' in that it does not leave **zombie coroutines**.
It cancels **all coroutines**.
When coroutines are detected as unfinished, a warning is issued.

**Example**

<code php>
use function Async\spawn;
use function Async\await;
use Async\Scope;

$scope = new Scope();

await($scope->spawn(function() {
    spawn(function() {
        Async\delay(1000);
        echo "Task 1\n";
    });

    spawn(function() {
        Async\delay(2000);
        echo "Task 2\n";
    });

    echo "Root task\n";
}));

$scope->dispose();
</code>

**Expected output**

<code>
Warning: Coroutine is zombie at ... in Scope disposed at ...
Warning: Coroutine is zombie at ... in Scope disposed at ...
Warning: Coroutine is zombie at ... in Scope disposed at ...
</code>

The ''disposeAfterTimeout'' method is a delayed version of the ''disposeSafely'' method.
The ''$timeout'' parameter must be greater than zero but less than 10 minutes.

<code php>
use Async\Scope;

class Service
{
    private Scope $scope;

    public function __construct() {
        $this->scope = new Scope();
    }

    public function __destruct() {
        $this->scope->disposeAfterTimeout(5000);
    }

    public function run(): void {
        $this->scope->spawn(function() {

            spawn(function() {
                Async\delay(1000);
                echo "Task 2\n";
                Async\delay(5000);
                echo "Task 2 next line never executed\n";
            });

            echo "Task 1\n";
        });
    }
}

$service = new Service();
$service->run();

sleep(1);
unset($service);
</code>

**Expected output**

<code>
Task 1
Warning: Coroutine is zombie at ... in Scope disposed at ...
Task 2
</code>

The ''dispose*()'' methods can be called multiple times, which is not considered an error.

If the ''Scope::cancel()'' method is called with a parameter after the ''Scope'' has already been cancelled,
**PHP** will emit a warning indicating that the call will be ignored.

=== Scope cancellation/disposal order ===

If a ''Scope'' has child ''Scopes'', the coroutines in the child ''Scopes'' will be canceled or disposed first,
followed by those in the parent — from the bottom up in the hierarchy.
This approach increases the likelihood that resources will be released correctly.
However, it does not guarantee this,
since the exact order of coroutines in the execution queue cannot be determined with 100% certainty.

During the release of child ''Scopes'',
the same cleanup strategy is used that was applied to the parent ''Scope''.

If the ''disposeSafely'' method is called, the child Scopes will also be released using the ''disposeSafely'' strategy.
If the ''dispose'' method is used, the child Scopes will use the same method for cleanup.

The ''disposeAfterTimeout'' method will delay the execution of ''disposeSafely'' for the specified time.

=== Spawn with disposed scope ===

When the ''cancel()'' or ''dispose()'' method is called, the ''Scope'' is marked as closed.
Attempting to launch a coroutine with this Scope will result in a fatal exception.

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    echo "Task 1\n";
});

$scope->cancel();

$scope->spawn(function() { // <- AsyncException: Coroutine scope is closed
    echo "Task 2\n";
});
</code>

=== Exception Handling ===

The ''Scope'' class provides a method for handling exceptions:

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    throw new Exception("Task 1");
});

$scope->setExceptionHandler(function (Exception $e) {
    echo "Caught exception: {$e->getMessage()}\n";
});

$scope->awaitCompletion(Async\timeout(60000));
</code>

An exception handler has the right to suppress the exception.
However, if the exception handler throws another exception,
the exception propagation algorithm will continue.

==== Scope Inspection Tools ====

The ''Scope'' class provides methods for inspecting the state and hierarchy of scopes:

| Method                                | Description                                                                        |
| **''getChildScopes():array''**        | Returns an array of all child scopes of the current scope                          |
| **''getCoroutines():array''**         | Returns a list of coroutines that are registered within the specified ''Scope''   |

These methods are useful for debugging and monitoring the state of scopes and their associated coroutines.

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    echo "Task 1\n";
});

$scope->spawn(function() {
    echo "Task 2\n";
});

// Get all coroutines in this scope
$coroutines = $scope->getCoroutines();
echo "Number of coroutines in scope: " . count($coroutines) . "\n";

// Create child scope
$childScope = Scope::inherit($scope);

// Get all child scopes
$childScopes = $scope->getChildScopes();
echo "Number of child scopes: " . count($childScopes) . "\n";
</code>

==== Error detection ====

Detecting erroneous situations when using coroutines is an important part of analyzing an application's reliability.

**The following scenarios are considered potentially erroneous**

  - A coroutine belongs to a global scope and is not awaited by anyone (a **zombie coroutine**).
  - The root scope has been destroyed (its destructor was called), but no one awaited it or ensured that its resources were explicitly cleaned up (e.g., by calling ''$scope->cancel()'' or ''$scope->dispose()'').
  - An attempt to await a coroutine from within itself.
  - Awaiting ''$scope'' from within itself or from one of its child scopes.
  - Stuck tasks in the cancellation state.
  - Deadlocks caused by circular dependencies between coroutines.

**PHP** will respond to such situations by issuing **warnings**, including debug information about the involved coroutines.
Developers are expected to write code in a way that avoids triggering these warnings.

An attempt to use the expression ''await($coroutine)'' from within the same coroutine throws an exception.

<code php>
use function Async\spawn;
use function Async\await;

$coroutine = null;
$coroutine = spawn(function() use (&$coroutine) {
    await($coroutine); // <- AsyncException: A coroutine cannot await itself. Coroutine spawned at ...
});
</code>

Using the ''Scope::awaitCompletion()'' from a coroutine that belongs to the same ''$scope'' or
to one of its child scopes will throw a fatal exception.
This condition makes it impossible to perform the ''$globalScope->awaitCompletion'' call.

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() use ($scope) {
    $scope->awaitCompletion(Async\timeout(1000)); // <- AsyncException: Awaiting a scope from within itself or
                                                  // its child scope would cause a deadlock. Scope created at ...
});
</code>

=== Error mitigation strategies ===

To avoid accidentally hanging coroutines whose lifetimes were not correctly limited, follow these rules:

  * Use **separate Scopes** for different coroutines. This is the best practice, as it allows explicitly defining lifetime dependencies between Scopes.
  * Use ''Scope::dispose()''. The ''dispose()'' method cancels coroutine execution and logs an error.
  * Don't mix semantically different coroutines within the same ''Scope''.
  * Avoid building hierarchies between ''Scopes'' with complex interdependencies.
  * Do not use cyclic dependencies between ''Scopes''.
  * The principle of single point of responsibility and ''Scope'' ownership. Do not pass the ''Scope'' object to different coroutine functions (unless the action happens in a closure). Do not store ''Scope'' objects in different places. Violating this rule can lead to manipulations with ''Scope'', which may cause a deadlock or disrupt the application's logic.
  * Child coroutines should not wait for their parents. Child Scopes should not wait for their parents.

<code php>
namespace ProcessPool;

use Async\Scope;

final class ProcessPool
{
    private Scope $watcherScope;
    private Scope $jobsScope;
    private Scope $pool;

    /**
     * List of pipes for each process.
     * @var array
     */
    private array $pipes = [];
    /**
     * Map of process descriptors: pid => bool
     * If the value is true, the process is free.
     * @var array
     */
    private array $descriptors = [];

    public function __construct(readonly public string $entryPoint, readonly public int $max, readonly public int $min)
    {
        // Define the coroutine scopes for the pool, watcher, and jobs
        $this->watcherScope = new Scope();
        $this->jobsScope = new Scope();
        $this->pool = new Scope();
    }

    public function __destruct()
    {
        $this->watcherScope->dispose();
        $this->pool->dispose();
        $this->jobsScope->dispose();
    }

    public function start(): void
    {
        $this->watcherScope->spawn($this->processWatcher(...));

        for ($i = 0; $i < $this->min; $i++) {
            $this->pool->spawn($this->startProcess(...));
        }
    }

    public function stop(): void
    {
        $this->watcherScope->cancel();
        $this->pool->cancel();
        $this->jobsScope->cancel();
    }

    private function processWatcher(): void
    {
        while (true) {
            try {
                $this->pool->awaitCompletion(Async\timeout(60000));
            } catch (StopProcessException $exception)  {
                echo "Process was stopped with message: {$exception->getMessage()}\n";

                if($exception->getCode() !== 0 || count($this->descriptors) < $this->min) {
                    $this->pool->spawn($this->startProcess(...));
                }
            }
        }
    }
}
</code>

The example above demonstrates how splitting coroutines into
Scopes helps manage their interaction and reduces the likelihood of errors.

Here, ''watcherScope'' monitors tasks in ''pool''.
When a process finishes, the watcher detects this event and, if necessary, starts a new process or not.
The monitoring logic is completely separated from the process startup logic.

The lifetime of ''watcherScope'' matches that of ''pool'', but not longer than the lifetime of the watcher itself.

The overall lifetime of all coroutines in the ''ProcessPool'' is determined by the lifetime of the ''ProcessPool''
object or by the moment the ''stop()'' method is explicitly called.

==== Zombie coroutine policy ====

Coroutines whose lifetime extends beyond the boundaries of their parent ''Scope''
are handled according to a separate **policy**.

This policy aims to strike a balance between uncontrolled resource leaks and the need to abruptly
terminate coroutines, which could lead to data integrity violations.

If there are no active coroutines left in the execution queue and no events to wait for,
the application is considered complete.

**Zombie coroutines** differ from regular ones in that they are not counted as active.
Once the application is considered finished,
zombie coroutines are given a time limit within which they must complete execution.
If this limit is exceeded, all zombie coroutines are canceled.

The delay time for handling zombie coroutines can be configured using
a constant in the ''php.ini'' file: ''async.zombie_coroutine_timeout'', which is set to two seconds by default.

If a coroutine is created within a user-defined ''Scope'', the programmer
can set a custom timeout for that specific ''Scope'' using the ''Scope::disposeAfterTimeout(int $ms)'' method.

=== Structured concurrency ===

**Structured concurrency** allows organizing coroutines into a group or hierarchy
to manage their lifetime or exception handling.

The parent task is required to take responsibility for its child tasks and
must not complete before the children have finished their execution.

To implement structured concurrency, it is recommended to use the ''Scope'' class with proper hierarchy management.

The following code implements this idea:

<code php>
use Async\Scope;

function copyFile(string $sourceFile, string $targetFile): void
{
    $source = fopen($sourceFile, 'r');
    $target = fopen($targetFile, 'w');

    $buffer = null;

    try {
        // Child scope
        $tasks = new \Async\CoroutineGroup(Scope::inherit());

        // Read data from the source file
        $tasks->spawn(function() use (&$buffer, $source) {
            while (!feof($source)) {
                if ($buffer === null) {
                    $chunk = fread($source, 1024);
                    $buffer = $chunk !== false && $chunk !== '' ? $chunk : null;
                }

                suspend();
            }

            $buffer = '';
        });

        // Write data to the target file
        $tasks->spawn(function() use (&$buffer, $target) {
            while (true) {
                if (is_string($buffer)) {
                    if ($buffer === '') {
                        break; // End of a file
                    }

                    fwrite($target, $buffer);
                    $buffer = null;
                }

                suspend();
            }

            echo "Copy complete.\n";
        });

        await($tasks);
    } finally {
        fclose($source);
        fclose($target);
    }
}

$copyTasks = new \Async\Scope();

$copyTasks->spawn(copyFile(...), 'source.txt', 'target.txt');
$copyTasks->spawn(copyFile(...), 'source2.txt', 'target2.txt');

$copyTasks->awaitCompletion(Async\timeout(60000));
</code>

The example creates two task groups: a parent and a child.
The parent task group handles the copy operations directly, while the child tasks perform file reading and writing.
File descriptors will not be closed until all child copy tasks have completed.
The main code will not finish until all copy operations are completed.

==== Timer functions ====

The standard async library includes two functions similar to ''usleep()'':
* ''Async\delay(int $ms): void''
* ''Async\timeout(int $ms): Awaitable''

The ''delay'' function suspends the execution of a coroutine for the specified number of milliseconds.
Unlike ''usleep'', the ''delay'' function will throw a cancellation exception if the coroutine is cancelled.

The ''timeout'' function is similar to ''delay'', but it returns an ''Awaitable'' object:

<code php>
use function Async\spawn;
use function Async\await;
use function Async\timeout;
use function Async\delay;

try {
    delay(1000); // suspends the coroutine for 1 second
    // Try to fetch data from the URL within 1 second
    echo await(spawn('file_get_content', 'https://php.net/'), timeout(1000));
} catch (\Async\AwaitCancelledException) {
    echo "Operation was cancelled by timeout\n";
} catch (\Async\CancellationError) {
    echo "Operation was cancelled by user\n";
}
</code>

=== Responsibility points ===

A **responsibility point** is code that explicitly waits for the completion of a coroutine or a 'Scope':

<code php>
$scope = new Scope();

$scope->spawn(function() {
  throw new Exception("Task 1");
});

try {
    $scope->awaitCompletion(Async\timeout(60000));
} catch (\Throwable $e) {
     echo "Caught exception: {$e->getMessage()}\n";
}
</code>


==== Error Handling ====

An uncaught exception in a coroutine follows this flow:

  - If the coroutine is awaited using the ''await'' keyword, the exception is propagated to the awaiting points. If multiple points are awaiting, each will receive the same exception (**Each await point will receive the exact same exception object, not cloned**).
  - The exception is passed to the ''Scope''.
  - If the ''Scope'' has an exception handler defined, it will be invoked.
  - If the ''Scope'' does not have an exception handler, the ''cancel()'' method is called, canceling all coroutines in this scope, including all child scopes.
  - If the ''Scope'' has responsibility points, i.e., the construction ''Scope::awaitCompletion'', all responsibility points receive the exception.
  - Otherwise, the exception is passed to the parent scope if it is defined.
  - If there is no parent scope, the exception falls into ''globalScope'', where the same rules apply as for a regular scope.

> **Note:** In addition to regular exception handling, when a deadlock condition is detected
> (circular dependencies between coroutines), a ''DeadlockError'' is thrown as a fatal error
> at the end of the application lifecycle. This error is not intended for normal exception handling
> but serves as a diagnostic tool to identify architectural problems.

{{ :rfc:true_async:exception_flow.svg |}}

**Example**

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    throw new Exception("Task 1");
});

$exception1 = null;
$exception2 = null;

$scope2 = new Scope();

$scope2->spawn(function() use ($scope, &$exception1) {
    try {
        $scope->awaitCompletion(Async\timeout(60000));
    } catch (Exception $e) {
        $exception1 = $e;
        echo "Caught exception1: {$e->getMessage()}\n";
    }
});

$scope2->spawn(function() use ($scope, &$exception2) {
    try {
        $scope->awaitCompletion(Async\timeout(60000));
    } catch (Exception $e) {
        $exception2 = $e;
        echo "Caught exception2: {$e->getMessage()}\n";
    }
});

$scope2->awaitCompletion(Async\timeout(60000));

echo $exception1 === $exception2 ? "The same exception\n" : "Different exceptions\n";
</code>

If an exception reaches ''globalScope'' and is not handled in any way,
it triggers **Graceful Shutdown Mode**, which will terminate the entire application.

The ''Scope'' class allows defining an exception handler that can prevent exception propagation.

For this purpose, two methods are used:
  - **''setExceptionHandler''** – triggers for any exceptions thrown within this **Scope**.
  - **''setChildScopeExceptionHandler''** – triggers for exceptions from **child Scopes**.

> The methods ''setExceptionHandler'' and ''setChildScopeExceptionHandler'' cannot be used with the ''globalScope''.
> If an attempt is made to do so, an exception will be thrown.

**Example**

<code php>
$scope = new Scope();
$scope->setExceptionHandler(function (Async\Scope $scope, Async\Coroutine $coroutine, Throwable $e) {
    echo "Caught exception: {$e->getMessage()}\n in coroutine: {$coroutine->getSpawnLocation()}\n";
});

$scope->spawn(function() {
    throw new Exception("Task 1");
});

$scope->awaitCompletion(Async\timeout(60000));
</code>

Using these handlers,
you can implement the **Supervisor** pattern, i.e.,
a **Scope** that will not be canceled when an exception occurs in coroutines.

> If the ''setExceptionHandler'' or ''setChildScopeExceptionHandler'' handlers throw an exception,
> it will be propagated to the **parent Scope** or the **global Scope**.

The **''setChildScopeExceptionHandler''** method allows handling exceptions only from **child Scopes**,
which can be useful for implementing an algorithm where the **main Scope** runs core tasks,
while **child Scopes** handle additional ones.

For example:

<code php>

use Async\Scope;
use Async\Coroutine;

final class Service
{
    private Scope $scope;

    public function __construct()
    {
        $this->scope = new Scope();

        $this->scope->setChildScopeExceptionHandler(
        static function (Scope $scope, Coroutine $coroutine, \Throwable $exception): void {
            echo "Occurred an exception: {$exception->getMessage()} in Coroutine {$coroutine->getSpawnLocation()}\n";
        });
    }

    public function start(): void
    {
        $this->scope->spawn($this->run(...));
    }

    public function stop(): void
    {
        $this->scope->cancel();
    }

    private function run(): void
    {
        while (($socket = $this->service->receive()) !== null) {

            $scope = Scope::inherit($this->scope);

            // supervisor pattern
            ($scope->spawn($this->handleRequest(...), $socket)->onFinally(
                static function () use ($scope) {
                    $scope->disposeSafely();
                }
            );
        }
    }
}
</code>

''$this->scope'' listens for new connections on the server socket.
Canceling ''$this->scope'' means shutting down the entire service.

Each new connection is handled in a separate **Scope**, which is inherited from ''$this->scope''.
If an exception occurs in a coroutine created within a **child Scope**,
it will be passed to the ''setChildScopeExceptionHandler'' handler and will not affect
the operation of the service as a whole.

{{ :rfc:true_async:supervisor.svg |}}

=== Scope::onFinally ===

For ''Scope'', the callback receives the completed scope as a parameter:

<code php>
use Async\Scope;

$scope = new Scope();

$scope->spawn(function() {
    throw new Exception("Task 1");
});

$scope->onFinally(function (Scope $completedScope) {
    echo "Scope " . spl_object_id($completedScope) . " completed\n";
});

$scope->awaitCompletion(Async\timeout(60000));
</code>

=== Critical section ===

Sometimes it's necessary to execute a **critical section** of code
that must not be cancelled via ''CancellationError''.

For example, this could be a sequence of write operations or a transaction.

For this purpose, the ''Async\protect'' function is used,
which allows executing a closure in a non-cancellable (silent) mode.

<code php>
use function Async\spawn;

function task(): void
{
    Async\protect(fn() => fwrite($file, "Critical data\n"));
}

spawn(task(...));
</code>

If a ''CancellationError'' was sent to a coroutine during ''protect()'',
the exception will be thrown immediately after the execution of ''protect()'' completes.

The use of loops or unsafe operations inside a critical section can be checked by static analyzers.


===== Backward Incompatible Changes =====

This RFC introduces new functionality and does not break existing code.

The ''Scope'' class and structured concurrency are opt-in features.

===== Proposed PHP Version(s) =====

PHP 8.6

It is proposed to approve this RFC together with or after the [[rfc:true_async|Base RFC]].

===== RFC Impact =====

==== To SAPIs ====

No direct impact. Scope management is implemented at the engine level and works with all SAPIs.

==== To Existing Extensions ====

Extensions can optionally use Scope for managing concurrent operations.

==== To Opcache ====

No impact.

==== New Constants ====

The following ''php.ini'' constants are introduced:

  * ''async.zombie_coroutine_timeout'' - timeout for zombie coroutine detection (default: 2 seconds)

==== To the Ecosystem ====

Provides standard structured concurrency primitives that async libraries can build upon.

==== php.ini Defaults ====

  * ''async.zombie_coroutine_timeout = 2'' (seconds)

===== Open Issues =====

None at this time.

===== Future Scope =====

Potential future enhancements:

  * Additional scope lifetime strategies
  * Scope-local storage (like thread-local but for scope)
  * Advanced cancellation policies

===== Proposed Voting Choices =====

This RFC requires a 2/3 majority to pass.

Voting options:
  * Yes - Accept Scope and Structured Concurrency functionality
  * No - Reject this RFC

===== Patches and Tests =====

Implementation is part of the true-async project:
  * https://github.com/true-async/php-src

===== References =====

  * [[rfc:true_async|PHP True Async (Base RFC)]] - Must be read first
  * [[https://doi.org/10.1145/3547276.3548519|Structured Concurrency - A Review]]
  * [[https://en.wikipedia.org/wiki/Structured_concurrency|Structured Concurrency]] (Wikipedia)
  * [[https://kotlinlang.org/docs/coroutines-basics.html|Kotlin Coroutines]]: Structured concurrency in production
  * [[https://docs.swift.org/swift-book/LanguageGuide/Concurrency.html|Swift Concurrency]]: Task groups and structured concurrency
  * [[https://trio.readthedocs.io/|Trio]] (Python): Pioneering structured concurrency library

===== Implementation =====

The current implementation of the project is located here:
https://github.com/true-async

The code will be split into several PRs upon agreement.

===== Changelog =====

==== Version 1.0 (November 2025) ====
   * Initial version: Split from base RFC
   * Comprehensive Scope and structured concurrency functionality
   * Zombie coroutine detection and prevention
   * Critical sections with protect() function
   * Scope hierarchy and propagation

===== Rejected Features =====

Keep this updated with features that were discussed on the mail lists.