Have you tried generators? It looks like you are trying to implement
coroutines.
You can learn more about cooperative multitasking in PHP on this awesome
post written by Nikic:
https://nikic.github.io/2012/12/22/Cooperative-multitasking-using-coroutines-in-PHP.html
- Marcos
On Wed, Sep 26, 2018 at 16:47 David Rodrigues
wrote:
> I do not know if I can get the same result with the current PHP features
> without taking many turns. So I thought of a feature that I'm initially
> calling "interruptions" (similar to those that occur on a CPU).
>
> Nowadays we have the Exceptions, which stop the execution of a function and
> initiate a process of "catch" and treatment of the same. So I thought of
> something similar, but did not break the execution flow, allowing reactions
> to depend on what was happening inside a function.
>
> It would basically work according to the following flow (the square
> brackets number is the execution order):
>
> function sum(int $a, int $b): int {
> [2] interrupts with new SumInterruption($a, $b);
> [5] return $a + $b;
> }
>
> interruptable {
> [6] $sum = [1] sum(1, 2);
> [7] printf($sum);
> }
> [3] catch (SumInterruption $interruption) {
> [4] printf('Calculating: %d + %d = ', $interruption->a,
> $interruption->b);
> }
>
> Using current PHP features I can do like that:
> https://pastebin.com/Bci6BBfi
>
> Note that all code will be executed, and the interpection will only
> redirect temporarily the execution flow to the "catch" block, then will
> back to "sum()" block to return the sum. Like Exceptions, an Interruption
> will traverse the code execution tree until find a interruption catch
> block, but if it doesn't exists, just not will happen (or maybe throw
> InterruptionNotHandledException or something like it).
>
> In one of my real example cases, I have a code that could be manipulated by
> another method. Currently I need argument the self instance to this method,
> so it could run another method from the caller method to make some
> adjustments, which is a bit confuses.
>
> I hope you understand my point, and I am open to discuss that.
>
> Thanks!
>
> --
> David Rodrigues
>
