From Sergey.Bylokhov at oracle.com  Tue Feb  9 15:45:44 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Tue, 9 Feb 2016 18:45:44 +0300
Subject: <Beans Dev> [9] Review Request: 8136382 SimpleBeanInfo.loadImage
 succeeds when running with a security manager
Message-ID: <56BA09A8.7030304@oracle.com>

Hello.

Please review the fix for jdk9.
This fix revert back the fix for JDK-8131926:
http://hg.openjdk.java.net/jdk9/jdk9/jdk/rev/58fde919b6c3
http://mail.openjdk.java.net/pipermail/beans-dev/2015-August/000178.html

In the fix of JDK-8131926 we replaced getResource to 
getResourceAsStream, because "resources are private to a named module 
and as things stand, there isn't a way to a URL to a resource in a named 
module." But in current implementation "Class::getResource" will returns 
a URL to a resource in a module, when invoked from code within that 
module (0fbe4c72638a). And it also apply security checks.

Bug: https://bugs.openjdk.java.net/browse/JDK-8136382
Webrev can be found at: http://cr.openjdk.java.net/~serb/8136382/webrev.00

-- 
Best regards, Sergey.

From Sergey.Bylokhov at oracle.com  Tue Feb  9 17:17:27 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Tue, 9 Feb 2016 20:17:27 +0300
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
Message-ID: <56BA1F27.4090002@oracle.com>

Hello.
Please review the fix for jdk9.

The specification is updated, the test will use IllegalArgumentException 
instead of IllegalAccessError.
ccc will be filed after the technical review.

Bug: https://bugs.openjdk.java.net/browse/JDK-8130061
Webrev can be found at: http://cr.openjdk.java.net/~serb/8130061/webrev.00

-- 
Best regards, Sergey.

From Alan.Bateman at oracle.com  Tue Feb  9 20:40:29 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Tue, 9 Feb 2016 21:40:29 +0100
Subject: <Beans Dev> [9] Review Request: 8136382
 SimpleBeanInfo.loadImage succeeds when running with a security manager
In-Reply-To: <56BA09A8.7030304@oracle.com>
References: <56BA09A8.7030304@oracle.com>
Message-ID: <56BA4EBD.3010809@oracle.com>



On 09/02/2016 16:45, Sergey Bylokhov wrote:
> Hello.
>
> Please review the fix for jdk9.
> This fix revert back the fix for JDK-8131926:
> http://hg.openjdk.java.net/jdk9/jdk9/jdk/rev/58fde919b6c3
> http://mail.openjdk.java.net/pipermail/beans-dev/2015-August/000178.html
>
> In the fix of JDK-8131926 we replaced getResource to 
> getResourceAsStream, because "resources are private to a named module 
> and as things stand, there isn't a way to a URL to a resource in a 
> named module." But in current implementation "Class::getResource" will 
> returns a URL to a resource in a module, when invoked from code within 
> that module (0fbe4c72638a). And it also apply security checks.
>
> Bug: https://bugs.openjdk.java.net/browse/JDK-8136382
> Webrev can be found at: 
> http://cr.openjdk.java.net/~serb/8136382/webrev.00
>
Sergey - have you consider clarifying the javadoc as part of this change?

-Alan

From Sergey.Bylokhov at oracle.com  Tue Feb  9 20:49:36 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Tue, 9 Feb 2016 23:49:36 +0300
Subject: <Beans Dev> [9] Review Request: 8136382
 SimpleBeanInfo.loadImage succeeds when running with a security manager
In-Reply-To: <56BA4EBD.3010809@oracle.com>
References: <56BA09A8.7030304@oracle.com> <56BA4EBD.3010809@oracle.com>
Message-ID: <56BA50E0.30101@oracle.com>

On 09.02.16 23:40, Alan Bateman wrote:
> Sergey - have you consider clarifying the javadoc as part of this change?

I can add clarification that this is a shortcut to the 
getResourceAsStream/getResource which simplify loading of images, but 
the current implementations of getResourceAsStream/getResource change 
from time to time, so I am not sure that I can describe some stable 
behavior.

-- 
Best regards, Sergey.

From Alan.Bateman at oracle.com  Thu Feb 11 15:22:23 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Thu, 11 Feb 2016 15:22:23 +0000
Subject: <Beans Dev> [9] Review Request: 8136382
 SimpleBeanInfo.loadImage succeeds when running with a security manager
In-Reply-To: <56BA50E0.30101@oracle.com>
References: <56BA09A8.7030304@oracle.com> <56BA4EBD.3010809@oracle.com>
	<56BA50E0.30101@oracle.com>
Message-ID: <56BCA72F.80401@oracle.com>



On 09/02/2016 20:49, Sergey Bylokhov wrote:
> On 09.02.16 23:40, Alan Bateman wrote:
>> Sergey - have you consider clarifying the javadoc as part of this 
>> change?
>
> I can add clarification that this is a shortcut to the 
> getResourceAsStream/getResource which simplify loading of images, but 
> the current implementations of getResourceAsStream/getResource change 
> from time to time, so I am not sure that I can describe some stable 
> behavior.
>
With the updated implementation then you are isolated from possible 
changes. You are right that there may be changes on resource 
encapsulation and okay with me if you put all changing the spec until then.

One thing that could be done, separate issue if you want, is update the 
@return to say that it returns null if the resource is not found or the 
resource could not be loaded as an Image. That might be clearer than the 
existing wording.

-Alan.

From Alan.Bateman at oracle.com  Thu Feb 11 15:26:20 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Thu, 11 Feb 2016 15:26:20 +0000
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BA1F27.4090002@oracle.com>
References: <56BA1F27.4090002@oracle.com>
Message-ID: <56BCA81C.4010502@oracle.com>



On 09/02/2016 17:17, Sergey Bylokhov wrote:
> Hello.
> Please review the fix for jdk9.
>
> The specification is updated, the test will use 
> IllegalArgumentException instead of IllegalAccessError.
> ccc will be filed after the technical review.
>
> Bug: https://bugs.openjdk.java.net/browse/JDK-8130061
> Webrev can be found at: 
> http://cr.openjdk.java.net/~serb/8130061/webrev.00
>
Should this say newProxyInstance rather than getProxyClass?

-Alan.

From Sergey.Bylokhov at oracle.com  Thu Feb 11 15:49:55 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Thu, 11 Feb 2016 18:49:55 +0300
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BCA81C.4010502@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
Message-ID: <56BCADA3.20600@oracle.com>

On 11.02.16 18:26, Alan Bateman wrote:
> On 09/02/2016 17:17, Sergey Bylokhov wrote:
>> Hello.
>> Please review the fix for jdk9.
>>
>> The specification is updated, the test will use
>> IllegalArgumentException instead of IllegalAccessError.
>> ccc will be filed after the technical review.
>>
>> Bug: https://bugs.openjdk.java.net/browse/JDK-8130061
>> Webrev can be found at:
>> http://cr.openjdk.java.net/~serb/8130061/webrev.00
>>
> Should this say newProxyInstance rather than getProxyClass?

The Proxy.newProxyInstance() has not the description of why the 
IllegalArgumentException can be thrown, it points to Proxy.getProxyClass():

      * <p>{@code Proxy.newProxyInstance} throws
      * {@code IllegalArgumentException} for the same reasons that
      * {@code Proxy.getProxyClass} does.
......
      * @throws  IllegalArgumentException if any of the restrictions on the
      *          parameters that may be passed to {@code getProxyClass}
      *          are violated
     public static Object newProxyInstance(ClassLoader loader,
                                           Class<?>[] interfaces,
                                           InvocationHandler h)



-- 
Best regards, Sergey.

From Alan.Bateman at oracle.com  Thu Feb 11 15:52:41 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Thu, 11 Feb 2016 15:52:41 +0000
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BCADA3.20600@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com>
Message-ID: <56BCAE49.8070808@oracle.com>



On 11/02/2016 15:49, Sergey Bylokhov wrote:
>
> The Proxy.newProxyInstance() has not the description of why the 
> IllegalArgumentException can be thrown, it points to 
> Proxy.getProxyClass():
>
>      * <p>{@code Proxy.newProxyInstance} throws
>      * {@code IllegalArgumentException} for the same reasons that
>      * {@code Proxy.getProxyClass} does.
> ......
>      * @throws  IllegalArgumentException if any of the restrictions on 
> the
>      *          parameters that may be passed to {@code getProxyClass}
>      *          are violated
>     public static Object newProxyInstance(ClassLoader loader,
>                                           Class<?>[] interfaces,
>                                           InvocationHandler h)
>
>
Proxy.getProxyClass is proposed to be deprecated in Java SE 9 - you'll 
see the changes in the jake forest, they just haven't got to JDK 9 main 
line yet. There is also a very clear list of restrictions in the updated 
javadoc. Since you are going into jdk9/client then maybe the simplest is 
to just reference the restrictions in the Proxy class.

-Alan


From Sergey.Bylokhov at oracle.com  Thu Feb 11 15:58:44 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Thu, 11 Feb 2016 18:58:44 +0300
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BCAE49.8070808@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com> <56BCAE49.8070808@oracle.com>
Message-ID: <56BCAFB4.4000903@oracle.com>

On 11.02.16 18:52, Alan Bateman wrote:
> Proxy.getProxyClass is proposed to be deprecated in Java SE 9 - you'll
> see the changes in the jake forest, they just haven't got to JDK 9 main
> line yet. There is also a very clear list of restrictions in the updated
> javadoc. Since you are going into jdk9/client then maybe the simplest is
> to just reference the restrictions in the Proxy class.

Right, the new javadoc of newProxyInstance() in the jake is quite 
detailed, I'll update the fix.


-- 
Best regards, Sergey.

From mandy.chung at oracle.com  Thu Feb 11 19:53:02 2016
From: mandy.chung at oracle.com (Mandy Chung)
Date: Thu, 11 Feb 2016 11:53:02 -0800
Subject: <Beans Dev> [9] Review Request: 8130061
	java.beans.EventHandler.create does not specify how it fails
	when an EventHandler cannot be created
In-Reply-To: <56BCAFB4.4000903@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com> <56BCAE49.8070808@oracle.com>
	<56BCAFB4.4000903@oracle.com>
Message-ID: <F17EF010-1CEB-46A5-85CC-743B7104420A@oracle.com>


> On Feb 11, 2016, at 7:58 AM, Sergey Bylokhov <Sergey.Bylokhov at oracle.com> wrote:
> 
> On 11.02.16 18:52, Alan Bateman wrote:
>> Proxy.getProxyClass is proposed to be deprecated in Java SE 9 - you'll
>> see the changes in the jake forest, they just haven't got to JDK 9 main
>> line yet. There is also a very clear list of restrictions in the updated
>> javadoc. Since you are going into jdk9/client then maybe the simplest is
>> to just reference the restrictions in the Proxy class.
> 
> Right, the new javadoc of newProxyInstance() in the jake is quite detailed, I'll update the fix.


In case, anyone is interested in the jake?s updated spec of newProxyInstance:

http://download.java.net/jigsaw/docs/api/java/lang/reflect/Proxy.html#newProxyInstance-java.lang.ClassLoader-java.lang.Class:A-java.lang.reflect.InvocationHandler-

Mandy

From Sergey.Bylokhov at oracle.com  Thu Feb 11 21:00:03 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Fri, 12 Feb 2016 00:00:03 +0300
Subject: <Beans Dev> [9] Review Request: 8136382
 SimpleBeanInfo.loadImage succeeds when running with a security manager
In-Reply-To: <56BCA72F.80401@oracle.com>
References: <56BA09A8.7030304@oracle.com> <56BA4EBD.3010809@oracle.com>
	<56BA50E0.30101@oracle.com> <56BCA72F.80401@oracle.com>
Message-ID: <56BCF653.4040204@oracle.com>

On 11.02.16 18:22, Alan Bateman wrote:
>
> One thing that could be done, separate issue if you want, is update the
> @return to say that it returns null if the resource is not found or the
> resource could not be loaded as an Image. That might be clearer than the
> existing wording.

Something like this?
http://cr.openjdk.java.net/~serb/8136382/webrev.01


-- 
Best regards, Sergey.

From Sergey.Bylokhov at oracle.com  Thu Feb 11 21:19:03 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Fri, 12 Feb 2016 00:19:03 +0300
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BCAFB4.4000903@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com> <56BCAE49.8070808@oracle.com>
	<56BCAFB4.4000903@oracle.com>
Message-ID: <56BCFAC7.7080906@oracle.com>

On 11.02.16 18:58, Sergey Bylokhov wrote:
> On 11.02.16 18:52, Alan Bateman wrote:
>> Proxy.getProxyClass is proposed to be deprecated in Java SE 9 - you'll
>> see the changes in the jake forest, they just haven't got to JDK 9 main
>> line yet. There is also a very clear list of restrictions in the updated
>> javadoc. Since you are going into jdk9/client then maybe the simplest is
>> to just reference the restrictions in the Proxy class.
>
> Right, the new javadoc of newProxyInstance() in the jake is quite
> detailed, I'll update the fix.

The new version:
http://cr.openjdk.java.net/~serb/8130061/webrev.01

-- 
Best regards, Sergey.

From Alan.Bateman at oracle.com  Fri Feb 12 08:47:38 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Fri, 12 Feb 2016 08:47:38 +0000
Subject: <Beans Dev> [9] Review Request: 8136382
 SimpleBeanInfo.loadImage succeeds when running with a security manager
In-Reply-To: <56BCF653.4040204@oracle.com>
References: <56BA09A8.7030304@oracle.com> <56BA4EBD.3010809@oracle.com>
	<56BA50E0.30101@oracle.com> <56BCA72F.80401@oracle.com>
	<56BCF653.4040204@oracle.com>
Message-ID: <56BD9C2A.9050006@oracle.com>



On 11/02/2016 21:00, Sergey Bylokhov wrote:
>
> Something like this?
> http://cr.openjdk.java.net/~serb/8136382/webrev.01
>
@return looks good.

-Alan

From Alan.Bateman at oracle.com  Fri Feb 12 09:08:06 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Fri, 12 Feb 2016 09:08:06 +0000
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BCFAC7.7080906@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com> <56BCAE49.8070808@oracle.com>
	<56BCAFB4.4000903@oracle.com> <56BCFAC7.7080906@oracle.com>
Message-ID: <56BDA0F6.3070703@oracle.com>



On 11/02/2016 21:19, Sergey Bylokhov wrote:
>
> The new version:
> http://cr.openjdk.java.net/~serb/8130061/webrev.01
>
This is okay although the following might be a bit clearer (up to you):

"if creating a Proxy for {@code listenerInterface} fails for any of the 
restrictions specified by Proxy#newProxyInterface".

-Alan


From Sergey.Bylokhov at oracle.com  Fri Feb 12 11:08:14 2016
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Fri, 12 Feb 2016 14:08:14 +0300
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BDA0F6.3070703@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com> <56BCAE49.8070808@oracle.com>
	<56BCAFB4.4000903@oracle.com> <56BCFAC7.7080906@oracle.com>
	<56BDA0F6.3070703@oracle.com>
Message-ID: <56BDBD1E.2000402@oracle.com>

On 12.02.16 12:08, Alan Bateman wrote:
> This is okay although the following might be a bit clearer (up to you):
>
> "if creating a Proxy for {@code listenerInterface} fails for any of the
> restrictions specified by Proxy#newProxyInterface".

I agree, it looks better:
http://cr.openjdk.java.net/~serb/8130061/webrev.02


-- 
Best regards, Sergey.

From Alan.Bateman at oracle.com  Fri Feb 12 11:24:04 2016
From: Alan.Bateman at oracle.com (Alan Bateman)
Date: Fri, 12 Feb 2016 11:24:04 +0000
Subject: <Beans Dev> [9] Review Request: 8130061
 java.beans.EventHandler.create does not specify how it fails when an
 EventHandler cannot be created
In-Reply-To: <56BDBD1E.2000402@oracle.com>
References: <56BA1F27.4090002@oracle.com> <56BCA81C.4010502@oracle.com>
	<56BCADA3.20600@oracle.com> <56BCAE49.8070808@oracle.com>
	<56BCAFB4.4000903@oracle.com> <56BCFAC7.7080906@oracle.com>
	<56BDA0F6.3070703@oracle.com> <56BDBD1E.2000402@oracle.com>
Message-ID: <56BDC0D4.3090303@oracle.com>



On 12/02/2016 11:08, Sergey Bylokhov wrote:
> On 12.02.16 12:08, Alan Bateman wrote:
>> This is okay although the following might be a bit clearer (up to you):
>>
>> "if creating a Proxy for {@code listenerInterface} fails for any of the
>> restrictions specified by Proxy#newProxyInterface".
>
> I agree, it looks better:
> http://cr.openjdk.java.net/~serb/8130061/webrev.02
>
Looks good to me, thank you for taking the suggestions.

-Alan