Re: [Vtd-xml-users] Random Access Proposal (take 2)

[Jimmy Z. <cra...@co...>]

If the goal of unbinding is not for reuse, then it is reasonable that
there is no need for unbinding.. in other words, if you don't want
a bookmark object to bind to another VTDNav object, then
it is very likely that it is the end of the application, and those bookmarks
should be gc'ed ... any comment?

----- Original Message ----- 
From: "Rodrigo Cunha" <rn...@gm...>
To: "Jimmy Zhang" <cra...@co...>
Cc: <vtd...@li...>
Sent: Monday, April 16, 2007 2:14 AM
Subject: Re: [Vtd-xml-users] Random Access Proposal (take 2)


> Sorry about the delay but I've been sick with a nasty upper respiratory 
> tract infection ... back to work now!
>
> Yes, you're right, I can bind to another VTDNav object, but that sounds 
> like a kludge when the objective is really unbinding, and may imply 
> internal bufer allocation inside the bookmark object if the depth 
> increases.
>
>
> Jimmy Zhang wrote:
>> setCursorPosition() and getCursroPosition() have been added in CVS
>>
>> but the benefit of unbind() is still not clear
>>
>> if GC of VTDnav is the goal then one can simply call 
>> "setCursorPosition()" on
>> a different VTDNav object and the mission is accomplished without having 
>> a
>> unbind method for bookMark...
>>
>>
>> ----- Original Message ----- From: "Rodrigo Cunha" <rn...@gm...>
>> To: "Jimmy Zhang" <cra...@co...>
>> Sent: Thursday, April 05, 2007 2:41 AM
>> Subject: Re: [Vtd-xml-users] Random Access Proposal (take 2)
>>
>>
>>> Jimmy Zhang wrote:
>>>
>>> > What if a bookmark object is created, but not yet bind to a  non-null
>>> VTDNav object...
>>>
>>> That's for unbind, right? The unbind rationale is to allow GC of VTDNav 
>>> objects while still allowing easy reusing of bookmark. Obviously you 
>>> could bind them to a dummy VTDNav instead, but that's a kludge, and 
>>> requires more processing. Or you could allow binding to null, but that 
>>> could lead to unnoticed errors.
>>>
>>> > recordCursorPosition() and setCursorPosition() would throw
>>> NullPointerException.. would that bother you?
>>>
>>> No, that's what would happen really, and seems perfectly correct.
>>>
>>> -- 
>>> Rodrigo
>>>
>>>
>>
>>
>>
>
>