Why use jQuery.isObject in jQuery.extend

[lrbabe]

In the current implementation of jQuery.extend we can find the
following code:

// Recurse if we're merging object values
if ( deep && copy && typeof copy === "object" && !copy.nodeType ) {
        var clone;

        if ( src ) {
                clone = src;
        } else if ( jQuery.isArray(copy) ) {
                clone = [];
        } else if ( jQuery.isObject(copy) ) {
                clone = {};
        } else {
                clone = copy;
        }

        // Never move original objects, clone them
        target[ name ] = jQuery.extend( deep, clone, copy );
}

If the first test passes, we know that copy is of type object. So what
is this jQuery.isObject exactly supposed to test?
It's probably meant to be more precise than the first test, excluding
objects such as "new Date()". What is strange is that the current
implementation of isObject always returns false (and seems therefore
useless, as suggested in http://dev.jquery.com/ticket/4946) without
causing any test to fail, while replacing its code with
function( obj ) {
return Object.prototype.toString.call( obj );
}
will cause one test to fail (test 10 with custom objects)...

Could someone light me up?

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

Yehuda was working through that code and didn't get a chance to finish
it up. Yehuda, any thoughts?

--John



On Sat, Nov 7, 2009 at 12:05 AM, lrbabe <lrbabe@...> wrote:

> In the current implementation of jQuery.extend we can find the
> following code:
>
> // Recurse if we're merging object values
> if ( deep && copy && typeof copy === "object" && !copy.nodeType ) {
>        var clone;
>
>        if ( src ) {
>                clone = src;
>        } else if ( jQuery.isArray(copy) ) {
>                clone = [];
>        } else if ( jQuery.isObject(copy) ) {
>                clone = {};
>        } else {
>                clone = copy;
>        }
>
>        // Never move original objects, clone them
>        target[ name ] = jQuery.extend( deep, clone, copy );
> }
>
> If the first test passes, we know that copy is of type object. So what
> is this jQuery.isObject exactly supposed to test?
> It's probably meant to be more precise than the first test, excluding
> objects such as "new Date()". What is strange is that the current
> implementation of isObject always returns false (and seems therefore
> useless, as suggested in http://dev.jquery.com/ticket/4946) without
> causing any test to fail, while replacing its code with
> function( obj ) {
> return Object.prototype.toString.call( obj );
> }
> will cause one test to fail (test 10 with custom objects)...
>
> Could someone light me up?
>
> --
>
> You received this message because you are subscribed to the Google Groups "jQuery Development" group.
> To post to this group, send email to jquery-dev@....
> To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
> For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.
>
>
>

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[DBJDBJ]

For IE, isObject(), is perhaps not that trivial ... DBJ

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

Maybe something like http://gist.github.com/229188 would be enough.

On Nov 7, 12:57 am, John Resig <jere...@...> wrote:

> Yehuda was working through that code and didn't get a chance to finish
> it up. Yehuda, any thoughts?
>
> --John
>
> On Sat, Nov 7, 2009 at 12:05 AM, lrbabe <lrb...@...> wrote:
> > In the current implementation of jQuery.extend we can find the
> > following code:
>
> > // Recurse if we're merging object values
> > if ( deep && copy && typeof copy === "object" && !copy.nodeType ) {
> >        var clone;
>
> >        if ( src ) {
> >                clone = src;
> >        } else if ( jQuery.isArray(copy) ) {
> >                clone = [];
> >        } else if ( jQuery.isObject(copy) ) {
> >                clone = {};
> >        } else {
> >                clone = copy;
> >        }
>
> >        // Never move original objects, clone them
> >        target[ name ] = jQuery.extend( deep, clone, copy );
> > }
>
> > If the first test passes, we know that copy is of type object. So what
> > is this jQuery.isObject exactly supposed to test?
> > It's probably meant to be more precise than the first test, excluding
> > objects such as "new Date()". What is strange is that the current
> > implementation of isObject always returns false (and seems therefore
> > useless, as suggested inhttp://dev.jquery.com/ticket/4946) without
> > causing any test to fail, while replacing its code with
> > function( obj ) {
> > return Object.prototype.toString.call( obj );
> > }
> > will cause one test to fail (test 10 with custom objects)...
>
> > Could someone light me up?
>
> > --
>
> > You received this message because you are subscribed to the Google Groups "jQuery Development" group.
> > To post to this group, send email to jquery-dev@....
> > To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
> > For more options, visit this group athttp://groups.google.com/group/jquery-dev?hl=en.

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[lrbabe]

It's actually not clear where the problem lies, I'll test your code.

On Nov 8, 10:31 am, Robert Katić <robert.ka...@...> wrote:

> Maybe something likehttp://gist.github.com/229188would be enough.
>
> On Nov 7, 12:57 am, John Resig <jere...@...> wrote:
>
> > Yehuda was working through that code and didn't get a chance to finish
> > it up. Yehuda, any thoughts?
>
> > --John
>
> > On Sat, Nov 7, 2009 at 12:05 AM, lrbabe <lrb...@...> wrote:
> > > In the current implementation of jQuery.extend we can find the
> > > following code:
>
> > > // Recurse if we're merging object values
> > > if ( deep && copy && typeof copy === "object" && !copy.nodeType ) {
> > >        var clone;
>
> > >        if ( src ) {
> > >                clone = src;
> > >        } else if ( jQuery.isArray(copy) ) {
> > >                clone = [];
> > >        } else if ( jQuery.isObject(copy) ) {
> > >                clone = {};
> > >        } else {
> > >                clone = copy;
> > >        }
>
> > >        // Never move original objects, clone them
> > >        target[ name ] = jQuery.extend( deep, clone, copy );
> > > }
>
> > > If the first test passes, we know that copy is of type object. So what
> > > is this jQuery.isObject exactly supposed to test?
> > > It's probably meant to be more precise than the first test, excluding
> > > objects such as "new Date()". What is strange is that the current
> > > implementation of isObject always returns false (and seems therefore
> > > useless, as suggested inhttp://dev.jquery.com/ticket/4946) without
> > > causing any test to fail, while replacing its code with
> > > function( obj ) {
> > > return Object.prototype.toString.call( obj );
> > > }
> > > will cause one test to fail (test 10 with custom objects)...
>
> > > Could someone light me up?
>
> > > --
>
> > > You received this message because you are subscribed to the Google Groups "jQuery Development" group.
> > > To post to this group, send email to jquery-dev@....
> > > To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
> > > For more options, visit this group athttp://groups.google.com/group/jquery-dev?hl=en.

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[DBJDBJ]

This version of isObject, in IE returns true, when tested with
window.alert ...

--DBJ

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

Native functions in IE is not handled correctly by isFunction either,
but it was decided that that cases are not significant...

On Nov 8, 12:23 pm, DBJDBJ <dbj...@...> wrote:
> This version of isObject, in IE returns true, when tested with
> window.alert ...
>
> --DBJ

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

In case everyone missed it - this discussion happened a few months ago
where we debated what exactly a good solution would look like:
http://gist.github.com/153271

Right now the:
return /^function Object/.test( obj.constructor );

solution works the best but it doesn't work in Opera Mini (which
doesn't have function decompilation) - nor is it particularly fast.

As I remember it, when we left off working on that function we were
trying to find a better solution.

--John



On Sun, Nov 8, 2009 at 12:29 PM, Robert Katić <robert.katic@...> wrote:

> Native functions in IE is not handled correctly by isFunction either,
> but it was decided that that cases are not significant...
>
> On Nov 8, 12:23 pm, DBJDBJ <dbj...@...> wrote:
>> This version of isObject, in IE returns true, when tested with
>> window.alert ...
>>
>> --DBJ
>
> --
>
> You received this message because you are subscribed to the Google Groups "jQuery Development" group.
> To post to this group, send email to jquery-dev@....
> To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
> For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.
>
>
>

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

> Right now the:
> return /^function Object/.test( obj.constructor );
>
> solution works the best but it doesn't work in Opera Mini (which
> doesn't have function decompilation) - nor is it particularly fast.

constructor property is often changed.
I hope you want isObject(jQuery()) == false.

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

> constructor property is often changed.
> I hope you want isObject(jQuery()) == false.

Yep, and that'll work as we expect it to. The one case where it won't
is if you do obj.constructor = Object;

If anyone thinks of an alternative solution please feel free to modify
the gist and put your code up there - something that covers all the
test cases would be great.

--John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

/^function Object/.test( jQuery().constructor ) == true

You don't need to explicitly set constructor to change it.
Redefining prototype, constructor is automatically altered (==Object).

On Nov 8, 2:13 pm, John Resig <jere...@...> wrote:

> > constructor property is often changed.
> > I hope you want isObject(jQuery()) == false.
>
> Yep, and that'll work as we expect it to. The one case where it won't
> is if you do obj.constructor = Object;
>
> If anyone thinks of an alternative solution please feel free to modify
> the gist and put your code up there - something that covers all the
> test cases would be great.
>
> --John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

http://gist.github.com/229254

On Nov 8, 2:13 pm, John Resig <jere...@...> wrote:

> > constructor property is often changed.
> > I hope you want isObject(jQuery()) == false.
>
> Yep, and that'll work as we expect it to. The one case where it won't
> is if you do obj.constructor = Object;
>
> If anyone thinks of an alternative solution please feel free to modify
> the gist and put your code up there - something that covers all the
> test cases would be great.
>
> --John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

> http://gist.github.com/229254

Does this handle all the tests on that page? and in IE 6/7/8 as well?

If so, let's land it.

--John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

I thing so, but not sure 100%. This seam a good task for TestSwarms.

On Nov 8, 4:38 pm, John Resig <jere...@...> wrote:
> >http://gist.github.com/229254
>
> Does this handle all the tests on that page? and in IE 6/7/8 as well?
>
> If so, let's land it.
>
> --John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

http://browsershots.org/http://bender.fesb.hr/~robert/scripts/isObjectLiteral.html

Let's see...

On Nov 8, 4:42 pm, Robert Katić <robert.ka...@...> wrote:

> I thing so, but not sure 100%. This seam a good task for TestSwarms.
>
> On Nov 8, 4:38 pm, John Resig <jere...@...> wrote:
>
> > >http://gist.github.com/229254
>
> > Does this handle all the tests on that page? and in IE 6/7/8 as well?
>
> > If so, let's land it.
>
> > --John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

> I thing so, but not sure 100%. This seam a good task for TestSwarms.

Well, we can do that eventually - but in the short term: Does it work on IE?

--John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

Yes, it passes all these tests.

On Nov 8, 4:57 pm, John Resig <jere...@...> wrote:
> > I thing so, but not sure 100%. This seam a good task for TestSwarms.
>
> Well, we can do that eventually - but in the short term: Does it work on IE?
>
> --John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

Nice! Just make a bug and link to your commit on your Github fork and
I'll land it (also, be sure to get the spacing right - and be sure to
use {} braces and ===).

--John



On Sun, Nov 8, 2009 at 5:19 PM, Robert Katić <robert.katic@...> wrote:

> Yes, it passes all these tests.
>
> On Nov 8, 4:57 pm, John Resig <jere...@...> wrote:
>> > I thing so, but not sure 100%. This seam a good task for TestSwarms.
>>
>> Well, we can do that eventually - but in the short term: Does it work on IE?
>>
>> --John
>
> --
>
> You received this message because you are subscribed to the Google Groups "jQuery Development" group.
> To post to this group, send email to jquery-dev@....
> To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
> For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.
>
>
>

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[Robert Katić]

http://dev.jquery.com/ticket/5474

On Nov 8, 2:13 pm, John Resig <jere...@...> wrote:

> > constructor property is often changed.
> > I hope you want isObject(jQuery()) == false.
>
> Yep, and that'll work as we expect it to. The one case where it won't
> is if you do obj.constructor = Object;
>
> If anyone thinks of an alternative solution please feel free to modify
> the gist and put your code up there - something that covers all the
> test cases would be great.
>
> --John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.

[John Resig]

Thanks!

On Sunday, November 8, 2009, Robert Katić <robert.katic@...> wrote:

> http://dev.jquery.com/ticket/5474
>
> On Nov 8, 2:13 pm, John Resig <jere...@...> wrote:
>> > constructor property is often changed.
>> > I hope you want isObject(jQuery()) == false.
>>
>> Yep, and that'll work as we expect it to. The one case where it won't
>> is if you do obj.constructor = Object;
>>
>> If anyone thinks of an alternative solution please feel free to modify
>> the gist and put your code up there - something that covers all the
>> test cases would be great.
>>
>> --John
>
> --
>
> You received this message because you are subscribed to the Google Groups "jQuery Development" group.
> To post to this group, send email to jquery-dev@....
> To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
> For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.
>
>
>

--
--John

--

You received this message because you are subscribed to the Google Groups "jQuery Development" group.
To post to this group, send email to jquery-dev@....
To unsubscribe from this group, send email to jquery-dev+unsubscribe@....
For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en.