[pool] drop Base(Keyed)ObjectPool?

2011-09-11 Thread Phil Steitz
These classes really do nothing other than maintain the boolean "closed", throwing UnsupportedOperationException or returning nonsense for most methods. The interfaces define contracts, so why do we really need these base classes? Phil

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-11 Thread Gary Gregory
Perhaps we want to keep the ivars in one place since they all have very carefully been decorated with final and volatile just in the right places? What about dropping "Object" from the name? That makes even less sense now that we have generics enabled. Gary On Sun, Sep 11, 2011 at 2:29 PM, Phil

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-11 Thread Phil Steitz
On 9/11/11 12:44 PM, Gary Gregory wrote: > Perhaps we want to keep the ivars in one place since they all have very > carefully been decorated with final and volatile just in the right places? There is only one field being maintained and I think it would actually be clearer to push it into the (sma

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-11 Thread Gary Gregory
On Sun, Sep 11, 2011 at 4:36 PM, Phil Steitz wrote: > On 9/11/11 12:44 PM, Gary Gregory wrote: > > Perhaps we want to keep the ivars in one place since they all have very > > carefully been decorated with final and volatile just in the right > places? > > There is only one field being maintained

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-12 Thread Simone Tripodi
full support from my side on it, +1 I never understood the usefulness of these classes and they don't help on writing concrete pool implementations. Have a nice day, all the best! Simo http://people.apache.org/~simonetripodi/ http://www.99soft.org/ On Sun, Sep 11, 2011 at 8:29 PM, Phil Steitz

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-12 Thread markt
Phil Steitz wrote: >On 9/11/11 12:44 PM, Gary Gregory wrote: >> Perhaps we want to keep the ivars in one place since they all have >very >> carefully been decorated with final and volatile just in the right >places? > >There is only one field being maintained and I think it would >actually be cle

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-18 Thread Phil Steitz
On 9/11/11 12:44 PM, Gary Gregory wrote: > Perhaps we want to keep the ivars in one place since they all have very > carefully been decorated with final and volatile just in the right places? Given that there is only one field being maintained, are you OK with dropping these, Gary? Phil > > What

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-18 Thread Gary Gregory
Sure. I was under the Impression that 2 subclasses reused the fields. Gary On Sep 18, 2011, at 17:39, Phil Steitz wrote: > On 9/11/11 12:44 PM, Gary Gregory wrote: >> Perhaps we want to keep the ivars in one place since they all have very >> carefully been decorated with final and volatile just

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-18 Thread Phil Steitz
On 9/18/11 2:43 PM, Gary Gregory wrote: > Sure. I was under the Impression that 2 subclasses reused the fields. Well, they (GOP, GKOP and others) do use the closed field; but IMO that is not enough justification for the complexity of having them there and the ugliness of the UnsupportedOperationEx

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-19 Thread Jörg Schaible
Phil Steitz wrote: > On 9/18/11 2:43 PM, Gary Gregory wrote: >> Sure. I was under the Impression that 2 subclasses reused the fields. > > Well, they (GOP, GKOP and others) do use the closed field; but IMO > that is not enough justification for the complexity of having them > there and the uglines

Re: [pool] drop Base(Keyed)ObjectPool?

2011-09-20 Thread Phil Steitz
On 9/19/11 1:08 AM, Jörg Schaible wrote: > Phil Steitz wrote: > >> On 9/18/11 2:43 PM, Gary Gregory wrote: >>> Sure. I was under the Impression that 2 subclasses reused the fields. >> Well, they (GOP, GKOP and others) do use the closed field; but IMO >> that is not enough justification for the comp