<AWT Dev> [9] Review request for 7081580: Specification for MouseInfo.getNumberOfButtons() doesn't contain info about "awt.mouse.numButtons"

Wed Mar 18 00:18:30 UTC 2015

The answer is simple, usually awt/swing used unspecified properties as 
way to add some functionality in update release. And later, in the major 
release this functionality implemented by the new methods with proper 
specification(and a desktop property usually mapped to this method, for 
compatibility reason).
Note that this property was approved already by the ccc request.

17.03.15 16:54, Phil Race wrote:
> But going back to my first question, why does the app need to know 
> about the
> mouseButtons property if there's a method ?  I am not sure what we - 
> or the
> user - gain by documenting it.
>
> If it was a "standard property" then it might make sense to list it on 
> that page
> but if its only purpose is as how the implementation communicates 
> internally,
> why expose it ?
>
> -phil.
>
>
> On 3/17/2015 4:48 PM, Phil Race wrote:
>> OK .. yes, it does say that. So there is no central place to find all 
>> supported
>> desktop properties ?
>>
>> -phil.
>>
>> On 3/17/2015 4:37 PM, Sergey Bylokhov wrote:
>>> Hi, Phil.
>>> This document is for " desktop properties supported by the AWT but 
>>> not documented elsewhere - typically because there is no suitable 
>>> method or class - are documented here."
>>>
>>> 17.03.15 16:28, Phil Race wrote:
>>>> Unless you intend to document that property here :-
>>>> http://docs.oracle.com/javase/8/docs/api/java/awt/doc-files/DesktopProperties.html 
>>>>
>>>>
>>>> maybe it should not be mentioned ..
>>>>
>>>> -phil.
>>>>
>>>> On 3/17/2015 4:16 PM, Sergey Bylokhov wrote:
>>>>> +1
>>>>>
>>>>> 16.03.15 7:32, Anton V. Tarasov wrote:
>>>>>> Hi Semyon,
>>>>>>
>>>>>> As a minimalistic description of the property, this looks ok to 
>>>>>> me. So, if there's nothing else to say about it, I'm fine with 
>>>>>> the fix.
>>>>>>
>>>>>> Regards,
>>>>>> Anton.
>>>>>>
>>>>>> On 16.03.2015 16:22, Semyon Sadetsky wrote:
>>>>>>> Hi!
>>>>>>>
>>>>>>> Thank you Anton!
>>>>>>> The updated webrev is: 
>>>>>>> http://cr.openjdk.java.net/~alexsch/semyon-sadetsky/7081580/webrev.01/ 
>>>>>>>
>>>>>>>
>>>>>>> --Semyon
>>>>>>>
>>>>>>>
>>>>>>> On 3/12/2015 1:42 PM, Anton V. Tarasov wrote:
>>>>>>>> Hi Semyon, Sergey,
>>>>>>>>
>>>>>>>> I agree with that the modified javadoc is not good.
>>>>>>>>
>>>>>>>> 1. When you say something is done "by calling A.b()", it means 
>>>>>>>> I can write exactly "A.b()" in my code and this will do the 
>>>>>>>> job. However, that's not the case with 
>>>>>>>> Toolkit.getDesktopProperty() (it won't be compiled).
>>>>>>>>
>>>>>>>> In order to refer to a method, you can use either of the 
>>>>>>>> following constructions:
>>>>>>>>
>>>>>>>> a) the A.b() method
>>>>>>>> b) {@link A#b}
>>>>>>>> c) the {@link A#b} method
>>>>>>>>
>>>>>>>> b/c is preferrable.
>>>>>>>>
>>>>>>>> 2. When you say "The value is obtained by calling something", 
>>>>>>>> it's not quite clear what or who obtains the value. The method 
>>>>>>>> itself? Or this is an alternative way to get it for a user?
>>>>>>>>
>>>>>>>> 3. If this is the only place in the spec where the property is 
>>>>>>>> introduced, then you should somehow reflect this fact. For 
>>>>>>>> instance, like this:
>>>>>>>>
>>>>>>>> The value is set by the "awt.mouse.numButtons" property, which 
>>>>>>>> can be obtained directly with the {@link 
>>>>>>>> Toolkit#getDesktopProperty} method.
>>>>>>>>
>>>>>>>> You don't have to _specify_ the way getNumberOfButtons() 
>>>>>>>> obtains the property, unless this implementation detail should 
>>>>>>>> really be specified. (For instance, if it was obtained by a 
>>>>>>>> method which could be overriden in an application.)
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Anton.
>>>>>>>>
>>>>>>>>
>>>>>>>> On 12.03.2015 11:42, Semyon Sadetsky wrote:
>>>>>>>>> Sorry, Sergey. Still don't understand what you mean.
>>>>>>>>> The issue is about*to do**mention* "awt.mouse.numButtons".
>>>>>>>>> Now you are saying that there is no value to mention it for 
>>>>>>>>> the first time in this spec. Doesn't it contradict to the 
>>>>>>>>> request itself?
>>>>>>>>> You couldn't be more specific on what do you want, could you?
>>>>>>>>> The fix just adds one short statement to the spec. Maybe 
>>>>>>>>> you'll find it to be more productive to just rephrase as you 
>>>>>>>>> want and write here.
>>>>>>>>>
>>>>>>>>> Thank you!
>>>>>>>>> --Semyon
>>>>>>>>>
>>>>>>>>> On 3/12/2015 11:11 AM, Sergey Bylokhov wrote:
>>>>>>>>>> Hi, Semyon.
>>>>>>>>>> That's a specification which should be read as written. But 
>>>>>>>>>> if you mean this is not the same things, then it is unclear 
>>>>>>>>>> what value will be added to the description of 
>>>>>>>>>> "awt.mouse.numButtons" property, which mentions in the 
>>>>>>>>>> specification for the first time. Since getNumberOfButtons 
>>>>>>>>>> obtain something not specified from the getToolkit, modify it 
>>>>>>>>>> somehow(w/o specification) and returns. See for example 
>>>>>>>>>> Toolkit.getToolkit and Toolkit. 
>>>>>>>>>> areExtraMouseButtonsEnabled(). It is not necessary write so 
>>>>>>>>>> specific specification but at least it should be clear.
>>>>>>>>>>
>>>>>>>>>> It would be good to rephrase it somehow.
>>>>>>>>>>
>>>>>>>>>> 12.03.15 0:09, Semyon Sadetsky wrote:
>>>>>>>>>>> Hi Sergey,
>>>>>>>>>>>
>>>>>>>>>>> I didn't find any mention in the new text that the method 
>>>>>>>>>>> returns the same value as Tolkit.get... returns.
>>>>>>>>>>> I'm not an expert in English but in my opinion "obtained by" 
>>>>>>>>>>> verb doesn't state that the same value is returned without 
>>>>>>>>>>> any handling.
>>>>>>>>>>> Maybe you've mixed it up with "proxy"?
>>>>>>>>>>>
>>>>>>>>>>> Thanks,
>>>>>>>>>>> --Semyon
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On 3/12/2015 9:47 AM, Sergey Bylokhov wrote:
>>>>>>>>>>>> Hi, Semyon.
>>>>>>>>>>>> The fix in general is correct, but it adds an assertion 
>>>>>>>>>>>> that this method should return the same values as 
>>>>>>>>>>>> Toolkit.get...
>>>>>>>>>>>> And this is incorrect, and we can get a new CR that 
>>>>>>>>>>>> implementation don't follow the specification. Probably we 
>>>>>>>>>>>> can simplify it and state that we use numeric value from 
>>>>>>>>>>>> desktop property or something like that?
>>>>>>>>>>>>
>>>>>>>>>>>> 11.03.15 22:52, Semyon Sadetsky wrote:
>>>>>>>>>>>>> Hello,
>>>>>>>>>>>>>
>>>>>>>>>>>>> please review fix for jdk9.
>>>>>>>>>>>>>
>>>>>>>>>>>>> Webrev: 
>>>>>>>>>>>>> http://cr.openjdk.java.net/~alexsch/semyon-sadetsky/7081580/webrev.00/ 
>>>>>>>>>>>>>
>>>>>>>>>>>>> Bug: https://bugs.openjdk.java.net/browse/JDK-7081580
>>>>>>>>>>>>>
>>>>>>>>>>>>> Thanks,
>>>>>>>>>>>>> --Semyon
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> -- 
>>>>>>>>>> Best regards, Sergey.
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> -- 
>>>>> Best regards, Sergey.
>>>>
>>>
>>>
>>> -- 
>>> Best regards, Sergey.
>>
>

-- 
Best regards, Sergey.