Hi Bruce,
For those who use callbacks now, how is this in any way better? They will
> eventually end up using an OOP approach anyway (as that's the strategic
> goal).
> Migrating from `session_set_handler_callbacks()` would still be
> (potentially)
> non-trivial. And what's the point migrating *to
On Wed, 21 Jun 2023 at 09:43, Máté Kocsis wrote:
> The reason why I think it's a good approach to have an intermediate state
> is to give
> these people the possibility to defer the actual migration until the
> very end.
>
Isn't that exactly what a deprecation period is for?
If we want to give
On Wed, Jun 21, 2023, at 11:57 AM, Rowan Tommins wrote:
> On Wed, 21 Jun 2023 at 09:43, Máté Kocsis wrote:
>
>> The reason why I think it's a good approach to have an intermediate state
>> is to give
>> these people the possibility to defer the actual migration until the
>> very end.
>>
>
>
> Isn'
Hi internals,
Today PHP is more typed than ever, however, one feature is still missing:
ability to check complex variable types using standard type declarations.
For example, to check that $var is a nullable string one has to do
`assert($var === null || is_string($var))`. To check that $var is fa
On Wed, Jun 21, 2023, at 5:44 PM, Удальцов Валентин wrote:
> Hi internals,
>
> Today PHP is more typed than ever, however, one feature is still missing:
> ability to check complex variable types using standard type declarations.
>
> For example, to check that $var is a nullable string one has to do
