Bug #81177 [Com]: Typehint parameter is ignored when passing NULL directly. Is it correct?

From: Date: Sat, 19 Jun 2021 21:14:05 +0000
Subject: Bug #81177 [Com]: Typehint parameter is ignored when passing NULL directly. Is it correct?
References: 1  Groups: php.bugs 
Request: Send a blank email to php-bugs+get-234516@lists.php.net to get a copy of this message
Edit report at https://bugs.php.net/bug.php?id=81177&edit=1

 ID:                 81177
 Comment by:         rtrtrtrtrt at dfdfdfdf dot dfd
 Reported by:        6562680 at gmail dot com
 Summary:            Typehint parameter is ignored when passing NULL
                     directly. Is it correct?
 Status:             Not a bug
 Type:               Bug
 Package:            *General Issues
 Operating System:   Win10
 PHP Version:        Irrelevant
 Block user comment: N
 Private report:     N

 New Comment:

> My problem? No, seems it's like PHP problem

no

> about we don't want the "undefined" type

besides that NULL is not undefined just use a typehint and strict-types
PEBCAK


Previous Comments:
------------------------------------------------------------------------
[2021-06-19 20:55:34] a at b dot c

Pull requests are welcome.

------------------------------------------------------------------------
[2021-06-19 20:38:57] 6562680 at gmail dot com

Last 5 years you solving problems with a words.

Even then once i say "would be great to use starts()" function and after 2 years FINALLY
you decide to create it... and even then it returns false creates one more if.

Speaking speaking speaking explaining explaining...

In that time we've got PHP8 with union types without array types, still havent generics, still
havent nested classes and readonly properties, havent internal dependency injector, havent that damn
decorators...

But well, we have something "CurlHandle" that interfaced resource handler, some
ArrayPrototypiers and you have discussion to multiline call syntax
"
attr
> call
> call
> call
"

that expected only in SCRIPTS deployed on microservices without OOP and patterns...

Continue speaking. Time is on other side

------------------------------------------------------------------------
[2021-06-19 20:18:53] requinix@php.net

https://www.php.net/manual/en/functions.arguments.php#functions.arguments.default

Every variable must have a value. Attempting to use a variable that does not have a value (because
it hasn't been created yet) will "return" null, but that in no way means a null value
represents undefined.

Specifying =null as the optional value for a parameter necessarily means that the parameter supports
it, therefore calling the function with null as an explicit value is not prohibited.

------------------------------------------------------------------------
[2021-06-19 19:54:56] 6562680 at gmail dot com

My problem? No, seems it's like PHP problem, about we don't want the "undefined"
type (i agree with)

Need some way in typehint that means "default value" if null passed or constant like
infinity "UNDEFINED" that pass from parent function to child... not sure it will solve.

Currently we can do like "section 2" but when phpstorm autocompletes arguments - we wont
see default values.

In my project its not a problem. When you prefer vendor libraries - its mandatory.

In short - typehint suggestions:
1. way to mark php null as undefined
2. way to detect empty string in type hint

Would like anybody finally suggests AGAIN class nesting and readonly properties and that request
wont be ignored AGAIN.
But its another story...

------------------------------------------------------------------------
[2021-06-19 18:54:36] rtrtrtrtrt at dfdfdfdf dot dfd

> 1) If use just "$a" - it will fail is i didnt pass the argument

correct - it isn't an optional param

> 2) If use "int $a" - it will fail if argument is not an integer

correct

> 3) If use "$a = 123" - if wont fail and if i DIDNT PASS 
> argument - it will be 123, but i still can pass NULL

what else? it's an optional, NON-TYPED param

so WHAT is your problem?

------------------------------------------------------------------------


The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at

    https://bugs.php.net/bug.php?id=81177


--
Edit this bug report at https://bugs.php.net/bug.php?id=81177&edit=1


Thread (9 messages)

« previous php.bugs (#234516) next »