substr passing null...

16 years ago by Nathanael D. Noblet — view source — reply

unread

Hello,
I just have a question, often there are 'optional' parameters to
functions, I've always thought that most of the time I could pass null
to these if I wanted to leave one empty. This has as far as I can tell
always worked except recently I was using substr(). If I pass null to
the third param, I get nothing back. Isn't passing null the same as not
passing anything? Is this a bug? Or is my logic flawed?

--
Nathanael d. Noblet
Gnat Solutions, Inc
T: 403.875.4613

16 years ago by Christopher Jones — view source — reply

unread

Nathanael D. Noblet wrote:

Hello,
I just have a question, often there are 'optional' parameters to
functions, I've always thought that most of the time I could pass null
to these if I wanted to leave one empty. This has as far as I can tell
always worked except recently I was using substr(). If I pass null to
the third param, I get nothing back. Isn't passing null the same as not
passing anything? Is this a bug? Or is my logic flawed?

I know Ilia recently fixed a few functions that didn't ignore null
parameters, but it doesn't appear that substr() was one of them.

Triple check the problem reproduces with the latest snapshot
(snaps.php.net). If it's still broken then log a bug (bugs.php.net).

Chris

--
Email: christopher.jones@oracle.com Tel: +1 650 506 8630
Twitter: http://twitter.com/ghrd Free PHP Book: http://tinyurl.com/UGPOM

16 years ago by johannes@php.net — view source — reply

unread

Hi,

Nathanael D. Noblet wrote:

Hello,
I just have a question, often there are 'optional' parameters to
functions, I've always thought that most of the time I could pass null
to these if I wanted to leave one empty. This has as far as I can tell
always worked except recently I was using substr(). If I pass null to
the third param, I get nothing back. Isn't passing null the same as not
passing anything? Is this a bug? Or is my logic flawed?

I know Ilia recently fixed a few functions that didn't ignore null
parameters, but it doesn't appear that substr() was one of them.

I guess that's part of the zend_parse_parameters changes. As we now use
the "s" modifier for instance which might better be a "s!". We should
certainly spent some time on checking these things.

All remember: We're feature freezing, so spend time on finding and
fixing bugs instead of adding new ones! :-)
(this includes users who can at least report such issues ... )

johannes

16 years ago by Nathanael D. Noblet — view source — reply

unread

Peter Walther wrote:

it's only a bug if you always expect optional NULL parameters to be
ignored.

Well that's the question I'm asking. I assumed optional parameters
passed null should be ignored. If this isn't the behaviour php uses,
then I'm good, but I've used passing null to optional params all over
the place, and just recently had this one not work. If my logic /
assumptions are flawed, that's fine otherwise I can file a bug.

--
Nathanael d. Noblet
Gnat Solutions, Inc
T: 403.875.4613

16 years ago by Scott MacVicar — view source — reply

unread

Nathanael D. Noblet wrote:

Peter Walther wrote:

it's only a bug if you always expect optional NULL parameters to be
ignored.

Well that's the question I'm asking. I assumed optional parameters
passed null should be ignored. If this isn't the behaviour php uses,
then I'm good, but I've used passing null to optional params all over
the place, and just recently had this one not work. If my logic /
assumptions are flawed, that's fine otherwise I can file a bug.

You're assumption is wrong then, NULL isn't treated as not passing a
value. The reason it worked with substr was by pure chance.

Scott

16 years ago by dan@dancryer.com — view source — reply

unread

You're assumption is wrong then, NULL isn't treated as not passing a
value. The reason it worked with substr was by pure chance.

Out of interest, is there a reason that that is the case? Surely passing
null would be best treated as the same as passing nothing?

16 years ago by Paul Biggar — view source — reply

unread

You're assumption is wrong then, NULL isn't treated as not passing a
value. The reason it worked with substr was by pure chance.

Out of interest, is there a reason that that is the case? Surely passing
null would be best treated as the same as passing nothing?

NULL is a value. There is no way to tell that when you pass NULL, you
actually intended to pass nothing and use the default value.

Paul

--
Paul Biggar
paul.biggar@gmail.com

16 years ago by Kenan R Sulayman — view source — reply

unread

Why don't make the function return FALSE if parameter equals NULL ?

--
(c) Kenan Sulayman
Freelance Designer and Programmer

Life's Live Poetry

2009/1/28 Paul Biggar paul.biggar@gmail.com

You're assumption is wrong then, NULL isn't treated as not passing a
value. The reason it worked with substr was by pure chance.

Out of interest, is there a reason that that is the case? Surely passing
null would be best treated as the same as passing nothing?

NULL is a value. There is no way to tell that when you pass NULL, you
actually intended to pass nothing and use the default value.

Paul

--
Paul Biggar
paul.biggar@gmail.com

16 years ago by Kalle Sommer Nielsen — view source — reply

unread

Hello

2009/1/28 Kenan R Sulayman kureal@kkooporation.de:

Why don't make the function return FALSE if parameter equals NULL ?

Well the $start parameter is required for substr(), so I don't see the
deal here? But with now initalizers being added in the manual by Jakub
its not much of a deal to use those values if you wanna use the
default anyway. For example mysql_connect:
http://www.php.net/mysql_connect

But I do agree that in the end it would be easier to just be able to
pass NULL, but it can just have a different meaning ;)

--
(c) Kenan Sulayman
Freelance Designer and Programmer

Life's Live Poetry

2009/1/28 Paul Biggar paul.biggar@gmail.com

You're assumption is wrong then, NULL isn't treated as not passing a
value. The reason it worked with substr was by pure chance.

Out of interest, is there a reason that that is the case? Surely passing
null would be best treated as the same as passing nothing?

NULL is a value. There is no way to tell that when you pass NULL, you
actually intended to pass nothing and use the default value.

Paul

--
Paul Biggar
paul.biggar@gmail.com

--

--
Kalle Sommer Nielsen

16 years ago by david.zuelke@bitextender.com — view source — reply

unread

Am 28.01.2009 um 18:41 schrieb Scott MacVicar:

Nathanael D. Noblet wrote:

Peter Walther wrote:

it's only a bug if you always expect optional NULL parameters to be
ignored.

Well that's the question I'm asking. I assumed optional parameters
passed null should be ignored. If this isn't the behaviour php uses,
then I'm good, but I've used passing null to optional params all over
the place, and just recently had this one not work. If my logic /
assumptions are flawed, that's fine otherwise I can file a bug.

You're assumption is wrong then, NULL isn't treated as not passing a
value. The reason it worked with substr was by pure chance.

Out of curiosity - how hard would it be to implement a "default"
keyword that can be passed in function/method calls with the new
parameter parsing API?

David

16 years ago by johannes@schlueters.de — view source — reply

unread

Hi,

Out of curiosity - how hard would it be to implement a "default"
keyword that can be passed in function/method calls with the new
parameter parsing API?

It would mean to introduce a new value type or tons of rewrites. But as
I said there's all we need with "s!" as modifier then we can handle
NULL-Params independently from a String-Parameter - where this makes
sense.

With other functions not NULL is the default but the value of some
Flag-Parameter (const DEFAULT_VALUE = 1024; function foo($param =
DEFAULT_VALUE);) so this is a case by case thing ...

But as said before: We should check functions where we changed the
parameter parsing to be as less BC as it makes sense.

johannes

16 years ago by Lukas Kahwe Smith — view source — reply

unread

showing that passing null as the third param for substr is somehow
interpreted to mean no length or something. Is this indeed a bug
then? and if so where can I submit it?

Please submit a documentation bug if this change is not yet covered in
the documentation. If you are very motivated to help us in finding and
documenting such issues, I also welcome you to register on the wiki
for an account so that you can work on:
http://wiki.php.net/doc/scratchpad/upgrade/53

You might also want to contact phpdoc@lists.php.net in that case.

regards,
Lukas Kahwe Smith
mls@pooteeweet.org