+1 as always (but more than usual) documentation needs improvement

Anyhow I passed the patch yesterday eveninv

On Wed, Oct 3, 2012 at 9:38 AM, Jeff MAURY <[email protected]> wrote:

> Le 2 oct. 2012 16:19, "Min Zhou" <[email protected]> a écrit :
> >
> > Hi, Jeff.
> >
> > Thanks for your quick reply. I written the test code according to the
> >  guide at http://directmemory.apache.org/simple-usage.html .
> > There is no refer that tell me I should close the CacheService at the
> >  end.  IMHO, creating a instance of CacheService following a builder
> >  pattern, which is not different from the construction of other normal
> >  Object instances. There no open() method after construction, users
> >  can't realize that they should close something if there is no
> > document on that. The timer is internally started by directmemory,
> > make it  a daemon thread should be better.
> > How do you think about that? Do you think that my patch will make
> > things getting better?
> Min,
>
> Yes, I think your patch adds a protection against incorrect use but does
> not prevent against leaked resources (at least until full gc occurs)
> We should open a jira regarding the documentation
>
> Jeff
> >
> >
> > Thanks,
> > Min
> >
> >
> > On Tue, Oct 2, 2012 at 7:54 PM, Jeff MAURY <[email protected]>
> wrote:
> >
> > > Hello,
> > >
> > > there is also another workaround: close the CacheService at the end of
> your
> > > test, this is the recommended way of correct use.
> > > I may perhaps find time to work on it this night.
> > >
> > > Jeff
> > >
> > > On Tue, Oct 2, 2012 at 12:56 PM, Min Zhou <[email protected]> wrote:
> > >
> > > > Hi, all,
> > > >
> > > > Is there anybody can take a look at DIRECTMEMORY-101? This is really
> a
> > > bug.
> > > >
> > > > Thanks,
> > > > Min
> > > > --
> > > > My research interests are distributed systems, parallel computing and
> > > > bytecode based virtual machine.
> > > >
> > > > My profile:
> > > > http://www.linkedin.com/in/coderplay
> > > > My blog:
> > > > http://coderplay.javaeye.com
> > > >
> > >
> > >
> > >
> > > --
> > > Jeff MAURY
> > >
> > >
> > > "Legacy code" often differs from its suggested alternative by actually
> > > working and scaling.
> > >  - Bjarne Stroustrup
> > >
> > > http://www.jeffmaury.com
> > > http://riadiscuss.jeffmaury.com
> > > http://www.twitter.com/jeffmaury
> > >
> >
> >
> >
> > --
> > My research interests are distributed systems, parallel computing and
> > bytecode based virtual machine.
> >
> > My profile:
> > http://www.linkedin.com/in/coderplay
> > My blog:
> > http://coderplay.javaeye.com
>

Reply via email to