You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@directmemory.apache.org by "Maurizio Cucchiara (Created) (JIRA)" <ji...@apache.org> on 2011/12/28 15:22:32 UTC
[jira] [Created] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Cache should allow key objects instead of plain string
------------------------------------------------------
Key: DIRECTMEMORY-43
URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
Project: Apache DirectMemory
Issue Type: Improvement
Reporter: Maurizio Cucchiara
See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Simone Tripodi <si...@apache.org>.
IIUC, shouldn't the allocate(key, size) method be an "internal-only"
method? I mean, it should be used only when a non-null object has to
be stored... or not?
sorry for the silly question!
TIA,
-Simo
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
On Sat, Feb 18, 2012 at 9:14 PM, Simone Tripodi
<si...@apache.org> wrote:
> +1 to Mau, there is no fault here, Raf!!!
>
> just to remark that we have been doing a community work here, joining
> efforts to make a better product than everybody could make
> individually! :)
>
> What I didn't understand, while applying generics, is why at at a
> certain point, Pointer.clazz becomes ByteBuffer.class, when it could
> safety be the type of stored object...
>
> Any hint?
> TIA,
> -Simo
>
> http://people.apache.org/~simonetripodi/
> http://simonetripodi.livejournal.com/
> http://twitter.com/simonetripodi
> http://www.99soft.org/
>
>
>
> On Sat, Feb 18, 2012 at 6:35 PM, Raffaele P. Guidi
> <ra...@gmail.com> wrote:
>> What about NIOTest in org.apache.directmemory.memory?
>>
>> On Sat, Feb 18, 2012 at 6:20 PM, Maurizio Cucchiara
>> <mc...@apache.org>wrote:
>>
>>> Ciao Raffaele,
>>> there is no fault, we are still in time to write a small test :) (and maybe
>>> it would be better if you write it, since you are the one with the deeper
>>> overall view).
>>>
>>> Twitter :http://www.twitter.com/m_cucchiara
>>> G+ :https://plus.google.com/107903711540963855921
>>> Linkedin :http://www.linkedin.com/in/mauriziocucchiara
>>>
>>> Maurizio Cucchiara
>>>
>>>
>>> On 18 February 2012 18:09, Raffaele P. Guidi <raffaele.p.guidi@gmail.com
>>> >wrote:
>>>
>>> > Uhm I guess this is my fault...
>>>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Simone Tripodi <si...@apache.org>.
Hi again guys,
here is my proposal for switching implementations to <K, V>:
https://github.com/apache/directmemory/pull/2
please review!
-Simo
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
On Sat, Feb 18, 2012 at 9:36 PM, Raffaele P. Guidi
<ra...@gmail.com> wrote:
> as I said before it's just a quick hack to allow direct access to cached
> data with a ByteBuffer interface- useful to use it in NIO contexts (it was
> a request made by hiram chirino from the activemq team). Probably it's easy
> to come up with a better implementation. I was also pointing out that
> there's already a test for that (NIOTest.java).
>
> Ciao,
> R
> Il giorno 18/feb/2012 21:15, "Simone Tripodi" <si...@apache.org> ha
> scritto:
>
>> +1 to Mau, there is no fault here, Raf!!!
>>
>> just to remark that we have been doing a community work here, joining
>> efforts to make a better product than everybody could make
>> individually! :)
>>
>> What I didn't understand, while applying generics, is why at at a
>> certain point, Pointer.clazz becomes ByteBuffer.class, when it could
>> safety be the type of stored object...
>>
>> Any hint?
>> TIA,
>> -Simo
>>
>> http://people.apache.org/~simonetripodi/
>> http://simonetripodi.livejournal.com/
>> http://twitter.com/simonetripodi
>> http://www.99soft.org/
>>
>>
>>
>> On Sat, Feb 18, 2012 at 6:35 PM, Raffaele P. Guidi
>> <ra...@gmail.com> wrote:
>> > What about NIOTest in org.apache.directmemory.memory?
>> >
>> > On Sat, Feb 18, 2012 at 6:20 PM, Maurizio Cucchiara
>> > <mc...@apache.org>wrote:
>> >
>> >> Ciao Raffaele,
>> >> there is no fault, we are still in time to write a small test :) (and
>> maybe
>> >> it would be better if you write it, since you are the one with the
>> deeper
>> >> overall view).
>> >>
>> >> Twitter :http://www.twitter.com/m_cucchiara
>> >> G+ :https://plus.google.com/107903711540963855921
>> >> Linkedin :http://www.linkedin.com/in/mauriziocucchiara
>> >>
>> >> Maurizio Cucchiara
>> >>
>> >>
>> >> On 18 February 2012 18:09, Raffaele P. Guidi <
>> raffaele.p.guidi@gmail.com
>> >> >wrote:
>> >>
>> >> > Uhm I guess this is my fault...
>> >>
>>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by "Raffaele P. Guidi" <ra...@gmail.com>.
as I said before it's just a quick hack to allow direct access to cached
data with a ByteBuffer interface- useful to use it in NIO contexts (it was
a request made by hiram chirino from the activemq team). Probably it's easy
to come up with a better implementation. I was also pointing out that
there's already a test for that (NIOTest.java).
Ciao,
R
Il giorno 18/feb/2012 21:15, "Simone Tripodi" <si...@apache.org> ha
scritto:
> +1 to Mau, there is no fault here, Raf!!!
>
> just to remark that we have been doing a community work here, joining
> efforts to make a better product than everybody could make
> individually! :)
>
> What I didn't understand, while applying generics, is why at at a
> certain point, Pointer.clazz becomes ByteBuffer.class, when it could
> safety be the type of stored object...
>
> Any hint?
> TIA,
> -Simo
>
> http://people.apache.org/~simonetripodi/
> http://simonetripodi.livejournal.com/
> http://twitter.com/simonetripodi
> http://www.99soft.org/
>
>
>
> On Sat, Feb 18, 2012 at 6:35 PM, Raffaele P. Guidi
> <ra...@gmail.com> wrote:
> > What about NIOTest in org.apache.directmemory.memory?
> >
> > On Sat, Feb 18, 2012 at 6:20 PM, Maurizio Cucchiara
> > <mc...@apache.org>wrote:
> >
> >> Ciao Raffaele,
> >> there is no fault, we are still in time to write a small test :) (and
> maybe
> >> it would be better if you write it, since you are the one with the
> deeper
> >> overall view).
> >>
> >> Twitter :http://www.twitter.com/m_cucchiara
> >> G+ :https://plus.google.com/107903711540963855921
> >> Linkedin :http://www.linkedin.com/in/mauriziocucchiara
> >>
> >> Maurizio Cucchiara
> >>
> >>
> >> On 18 February 2012 18:09, Raffaele P. Guidi <
> raffaele.p.guidi@gmail.com
> >> >wrote:
> >>
> >> > Uhm I guess this is my fault...
> >>
>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Simone Tripodi <si...@apache.org>.
+1 to Mau, there is no fault here, Raf!!!
just to remark that we have been doing a community work here, joining
efforts to make a better product than everybody could make
individually! :)
What I didn't understand, while applying generics, is why at at a
certain point, Pointer.clazz becomes ByteBuffer.class, when it could
safety be the type of stored object...
Any hint?
TIA,
-Simo
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
On Sat, Feb 18, 2012 at 6:35 PM, Raffaele P. Guidi
<ra...@gmail.com> wrote:
> What about NIOTest in org.apache.directmemory.memory?
>
> On Sat, Feb 18, 2012 at 6:20 PM, Maurizio Cucchiara
> <mc...@apache.org>wrote:
>
>> Ciao Raffaele,
>> there is no fault, we are still in time to write a small test :) (and maybe
>> it would be better if you write it, since you are the one with the deeper
>> overall view).
>>
>> Twitter :http://www.twitter.com/m_cucchiara
>> G+ :https://plus.google.com/107903711540963855921
>> Linkedin :http://www.linkedin.com/in/mauriziocucchiara
>>
>> Maurizio Cucchiara
>>
>>
>> On 18 February 2012 18:09, Raffaele P. Guidi <raffaele.p.guidi@gmail.com
>> >wrote:
>>
>> > Uhm I guess this is my fault...
>>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by "Raffaele P. Guidi" <ra...@gmail.com>.
What about NIOTest in org.apache.directmemory.memory?
On Sat, Feb 18, 2012 at 6:20 PM, Maurizio Cucchiara
<mc...@apache.org>wrote:
> Ciao Raffaele,
> there is no fault, we are still in time to write a small test :) (and maybe
> it would be better if you write it, since you are the one with the deeper
> overall view).
>
> Twitter :http://www.twitter.com/m_cucchiara
> G+ :https://plus.google.com/107903711540963855921
> Linkedin :http://www.linkedin.com/in/mauriziocucchiara
>
> Maurizio Cucchiara
>
>
> On 18 February 2012 18:09, Raffaele P. Guidi <raffaele.p.guidi@gmail.com
> >wrote:
>
> > Uhm I guess this is my fault...
>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Maurizio Cucchiara <mc...@apache.org>.
Ciao Raffaele,
there is no fault, we are still in time to write a small test :) (and maybe
it would be better if you write it, since you are the one with the deeper
overall view).
Twitter :http://www.twitter.com/m_cucchiara
G+ :https://plus.google.com/107903711540963855921
Linkedin :http://www.linkedin.com/in/mauriziocucchiara
Maurizio Cucchiara
On 18 February 2012 18:09, Raffaele P. Guidi <ra...@gmail.com>wrote:
> Uhm I guess this is my fault...
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by "Raffaele P. Guidi" <ra...@gmail.com>.
Uhm I guess this is my fault...
Il giorno 18/feb/2012 18:01, "Maurizio Cucchiara (Commented) (JIRA)" <
jira@apache.org> ha scritto:
>
> [
> https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13211018#comment-13211018]
>
> Maurizio Cucchiara commented on DIRECTMEMORY-43:
> ------------------------------------------------
>
> Those are exactly the problems I have encountered last time.
> To make the thing worse, I didn't find any test (or docs) which documents
> the use of directBuffer.
> So, in order to avoid breaking silently the code, I preferred to leave
> everything everything as it was.
> What about to first write (in case they don't already exists) some tests
> which demonstrate the use of directBuffer?
>
>
> > Cache should allow key objects instead of plain string
> > ------------------------------------------------------
> >
> > Key: DIRECTMEMORY-43
> > URL:
> https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> > Project: Apache DirectMemory
> > Issue Type: Improvement
> > Reporter: Maurizio Cucchiara
> > Assignee: Maurizio Cucchiara
> > Attachments: DIRECTMEMORY-43.patch,
> DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch,
> patch.org.apache.directmemory.cache.zip
> >
> >
> > See http://goo.gl/D4SRN
>
> --
> This message is automatically generated by JIRA.
> If you think it was sent incorrectly, please contact your JIRA
> administrators:
> https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
> For more information on JIRA, see: http://www.atlassian.com/software/jira
>
>
>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by "Raffaele P. Guidi" <ra...@gmail.com>.
This is an exception to the general logic that gives direct access to a
clone of the underlying ByteBuffer - basically used to enable NIO access
On Wed, Jan 25, 2012 at 5:38 AM, Maurizio Cucchiara (Commented) (JIRA) <
jira@apache.org> wrote:
>
> [
> https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13192857#comment-13192857]
>
> Maurizio Cucchiara commented on DIRECTMEMORY-43:
> ------------------------------------------------
>
> Sorry guys,
> I would have liked to change the current api layout to <K,V> generics.
> Unfortunately there is a logic that I can't understand behind the code.
> If you take a look at CacheServiceImpl (http://goo.gl/Ge8X9) you will see
> that the method retrieve can return a different kind of object than the
> expected one (see line 206).
> Generally speaking, I don't like the api methods which force the user to
> check what he get, and I would like to get rid of them.
> But unfortunately I didn't find a test which documents the need of this
> case.
> Do we really need it?
>
> > Cache should allow key objects instead of plain string
> > ------------------------------------------------------
> >
> > Key: DIRECTMEMORY-43
> > URL:
> https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> > Project: Apache DirectMemory
> > Issue Type: Improvement
> > Reporter: Maurizio Cucchiara
> > Assignee: Maurizio Cucchiara
> > Attachments: DIRECTMEMORY-43.patch
> >
> >
> > See http://goo.gl/D4SRN
>
> --
> This message is automatically generated by JIRA.
> If you think it was sent incorrectly, please contact your JIRA
> administrators:
> https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
> For more information on JIRA, see: http://www.atlassian.com/software/jira
>
>
>
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Benoit Perroud (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13178336#comment-13178336 ]
Benoit Perroud commented on DIRECTMEMORY-43:
--------------------------------------------
Thanks for the link. Yes for counting the hits and the accesses, but Pointer doesn't need to know anything about serialization. It deals only with byte[] or ByteBuffer.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (DIRECTMEMORY-43) Cache should allow
key objects instead of plain string
Posted by "Michael André Pearce (Issue Comment Edited JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210861#comment-13210861 ]
Michael André Pearce edited comment on DIRECTMEMORY-43 at 2/18/12 7:50 AM:
---------------------------------------------------------------------------
It is pluggable, this is just improvements suggested to make to the core to support generics, and as such achieves the above jira issue at the same time, added patch files.
was (Author: teaandcoffee):
It is pluggable, this is just improvements suggested to make to the core to support generics, and as such achieves the above jira issue at the same time, i will make a patch file later this week.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, org.apache.directmemory.cache.zip, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (DIRECTMEMORY-43) Cache should allow
key objects instead of plain string
Posted by "Michael André Pearce (Issue Comment Edited JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210861#comment-13210861 ]
Michael André Pearce edited comment on DIRECTMEMORY-43 at 2/18/12 7:53 AM:
---------------------------------------------------------------------------
It is pluggable, this is just improvements suggested to make to the core to support generics, and as such achieves the above jira issue at the same time, added patch files. see https://issues.apache.org/jira/browse/DIRECTMEMORY-60 for ehcache stuff.
was (Author: teaandcoffee):
It is pluggable, this is just improvements suggested to make to the core to support generics, and as such achieves the above jira issue at the same time, added patch files.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Tommaso Teofili (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13209276#comment-13209276 ]
Tommaso Teofili commented on DIRECTMEMORY-43:
---------------------------------------------
bq . didn't find a test which documents the need of this case.
If you take a look at the examples module you can find:
{noformat}
@Test
public void testPut()
{
try
{
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
}
catch ( Exception e )
{
e.printStackTrace();
fail( e.getLocalizedMessage() );
}
}
{noformat}
inside SolrOffHeapCacheTest under examples test sources.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Maurizio Cucchiara (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13192857#comment-13192857 ]
Maurizio Cucchiara commented on DIRECTMEMORY-43:
------------------------------------------------
Sorry guys,
I would have liked to change the current api layout to <K,V> generics.
Unfortunately there is a logic that I can't understand behind the code.
If you take a look at CacheServiceImpl (http://goo.gl/Ge8X9) you will see that the method retrieve can return a different kind of object than the expected one (see line 206).
Generally speaking, I don't like the api methods which force the user to check what he get, and I would like to get rid of them.
But unfortunately I didn't find a test which documents the need of this case.
Do we really need it?
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Assigned] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Maurizio Cucchiara (Assigned) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Maurizio Cucchiara reassigned DIRECTMEMORY-43:
----------------------------------------------
Assignee: Maurizio Cucchiara
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Simone Tripodi (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210927#comment-13210927 ]
Simone Tripodi commented on DIRECTMEMORY-43:
--------------------------------------------
Thank you Michael, I cannot apply the patch ATM because there is a false statement:
{code}
@@ -214,11 +214,11 @@
if ( ptr.clazz == ByteBuffer.class )
{
// skip serialization if it is a bytebuffer
- return ptr.directBuffer;
+ return (V) ptr.directBuffer;
}
{code}
as you can see, we cannot blindly cast {{ByteBuffer}} to {{V}}.
I am studying a deeper type inference in the whole class set. Thanks you for contributing!
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Simone Tripodi (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210859#comment-13210859 ]
Simone Tripodi commented on DIRECTMEMORY-43:
--------------------------------------------
Michael, good, but can you please provide a plain text patch, made against the latest committed trunk? That would make much more easier understand how to integrate your modifications.
I haven't had a look yet at your contribution, my question is: does EHCache needs to be integrated directly in the DM core, or it is something pluggable?
TIA,
-Simo
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Michael André Pearce (Updated JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Michael André Pearce updated DIRECTMEMORY-43:
---------------------------------------------
Attachment: org.apache.directmemory.cache.zip
Have added some work which i needed to make generic these classes as your comments suggest to integrate nicely with EHCache.
Generics Version of CacheService, CacheServiceImpl so that key can be any object, also so that value stored is generic.
Cache class is included which uses <String, Object> , I have also got a non singleton cache, which is required for EHCache, but please see the EHCache Item i will create in a bit.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Michael André Pearce (Updated JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Michael André Pearce updated DIRECTMEMORY-43:
---------------------------------------------
Attachment: (was: org.apache.directmemory.cache.zip)
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Hudson (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13215144#comment-13215144 ]
Hudson commented on DIRECTMEMORY-43:
------------------------------------
Integrated in directmemory-trunk #115 (See [https://builds.apache.org/job/directmemory-trunk/115/])
[DIRECTMEMORY-43] Cache should allow key objects instead of plain string (Revision 1292990)
Result = SUCCESS
olamy :
Files :
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/cache/Cache.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/cache/CacheService.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/cache/CacheServiceImpl.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/AbstractOffHeapMemoryBuffer.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/AllocationPolicy.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/MemoryManager.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/MemoryManagerService.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/MemoryManagerServiceImpl.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/MemoryManagerServiceWithAllocationPolicyImpl.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/OffHeapMemoryBuffer.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/OffHeapMemoryBufferImpl.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/OffHeapMergingMemoryBufferImpl.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/Pointer.java
* /incubator/directmemory/trunk/directmemory-cache/src/main/java/org/apache/directmemory/memory/RoundRobinAllocationPolicy.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/cache/CacheConcurrentTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/cache/CacheLightConcurrentTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/AbstractOffHeapMemoryBufferTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/BaseTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/Concurrent2Test.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/Concurrent3Test.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/ConcurrentTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/MallocTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/MemoryManagerServiceImplTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/MemoryManagerServiceImplWithMerginOHMBAndAllocationPolicyTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/MemoryManagerServiceImplWithMerginOHMBTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/MemoryManagerTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/NIOTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/OffHeapMemoryBufferTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/OffHeapMergingMemoryBufferTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/RoundRobinAllocationPolicyTest.java
* /incubator/directmemory/trunk/directmemory-cache/src/test/java/org/apache/directmemory/memory/Starter.java
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Simone Tripodi
> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (DIRECTMEMORY-43) Cache should allow
key objects instead of plain string
Posted by "Tommaso Teofili (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13209276#comment-13209276 ]
Tommaso Teofili edited comment on DIRECTMEMORY-43 at 2/16/12 11:14 AM:
-----------------------------------------------------------------------
bq . didn't find a test which documents the need of this case.
If you take a look at the integrations/solr module you can find the testPut() method which does:
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
inside SolrOffHeapCacheTest.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
was (Author: teofili):
bq . didn't find a test which documents the need of this case.
If you take a look at the examples module you can find:
{noformat}
@Test
public void testPut()
{
try
{
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
}
catch ( Exception e )
{
e.printStackTrace();
fail( e.getLocalizedMessage() );
}
}
{noformat}
inside SolrOffHeapCacheTest under examples test sources.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (DIRECTMEMORY-43) Cache should allow
key objects instead of plain string
Posted by "Michael André Pearce (Issue Comment Edited JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210861#comment-13210861 ]
Michael André Pearce edited comment on DIRECTMEMORY-43 at 2/18/12 7:18 AM:
---------------------------------------------------------------------------
It is pluggable, this is just improvements suggested to make to the core to support generics, and as such achieves the above jira issue at the same time, i will make a patch file later this week.
was (Author: teaandcoffee):
It is pluggable, this is just improvements suggested to make to the core to support generics, i will make a patch file later this week.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Michael André Pearce (Updated JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Michael André Pearce updated DIRECTMEMORY-43:
---------------------------------------------
Attachment: patch.org.apache.directmemory.cache.zip
adding patch.org.apache.directmemory.cache.zip
contains svn patch's to apply to svn head.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, org.apache.directmemory.cache.zip, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Maurizio Cucchiara (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13211018#comment-13211018 ]
Maurizio Cucchiara commented on DIRECTMEMORY-43:
------------------------------------------------
Those are exactly the problems I have encountered last time.
To make the thing worse, I didn't find any test (or docs) which documents the use of directBuffer.
So, in order to avoid breaking silently the code, I preferred to leave everything everything as it was.
What about to first write (in case they don't already exists) some tests which demonstrate the use of directBuffer?
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Maurizio Cucchiara (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Maurizio Cucchiara updated DIRECTMEMORY-43:
-------------------------------------------
Attachment: DIRECTMEMORY-43.patch
The attached patch change the key from String to generic Object.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (DIRECTMEMORY-43) Cache should allow
key objects instead of plain string
Posted by "Tommaso Teofili (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13209276#comment-13209276 ]
Tommaso Teofili edited comment on DIRECTMEMORY-43 at 2/17/12 11:19 AM:
-----------------------------------------------------------------------
bq. didn't find a test which documents the need of this case.
If you take a look at the integrations/solr module you can find the testPut() method which does:
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
inside SolrOffHeapCacheTest.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
was (Author: teofili):
bq . didn't find a test which documents the need of this case.
If you take a look at the integrations/solr module you can find the testPut() method which does:
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
inside SolrOffHeapCacheTest.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Raffaele P. Guidi (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13178341#comment-13178341 ]
Raffaele P. Guidi commented on DIRECTMEMORY-43:
-----------------------------------------------
@maurizio it almost does, but it is called pointer because MemoryManager aim is to have a (kind of) malloc() implementation that can be used in a cache service but also in other cases (as the backend for a virtual file system, as someone pointed out, or generally as a fast(er) byte buffer), so separation of concerns should be applied (all things cache-related - eviction and expiry policies included - into the CacheService and the lower level memory management in MemoryService) @benoit it contains the class because it is needed for protostuff serialization, but it could be not needed if the buffer is used as a byte array (or directly as a ByteBuffer i.e. by a NIO server). Sure some order should be done. Proposals?
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Maurizio Cucchiara (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13178334#comment-13178334 ]
Maurizio Cucchiara commented on DIRECTMEMORY-43:
------------------------------------------------
@Benoit,
ATM I think that a Pointer partially acts as CacheEntry (http://goo.gl/gdXGi).
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Simone Tripodi (Resolved) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Simone Tripodi resolved DIRECTMEMORY-43.
----------------------------------------
Resolution: Fixed
Assignee: Simone Tripodi (was: Maurizio Cucchiara)
Fixed on [r1292990|http://svn.apache.org/viewvc?rev=1292990&view=rev]
thanks Olivier for pushing it - had some troubles with git (still in n00b mode :P)
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Simone Tripodi
> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Michael André Pearce (Commented JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210861#comment-13210861 ]
Michael André Pearce commented on DIRECTMEMORY-43:
--------------------------------------------------
It is pluggable, this is just improvements suggested to make to the core to support generics, i will make a patch file later this week.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Maurizio Cucchiara (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13177336#comment-13177336 ]
Maurizio Cucchiara commented on DIRECTMEMORY-43:
------------------------------------------------
Hi Simo,
you're right that's a very good point, let me do some checks.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Tommaso Teofili (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13177338#comment-13177338 ]
Tommaso Teofili commented on DIRECTMEMORY-43:
---------------------------------------------
big +1 for <K,V>
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Olivier Lamy (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Olivier Lamy updated DIRECTMEMORY-43:
-------------------------------------
Fix Version/s: 0.6.0
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Simone Tripodi
> Fix For: 0.6.0
>
> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Issue Comment Edited] (DIRECTMEMORY-43) Cache should allow
key objects instead of plain string
Posted by "Tommaso Teofili (Issue Comment Edited) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13209276#comment-13209276 ]
Tommaso Teofili edited comment on DIRECTMEMORY-43 at 2/17/12 11:20 AM:
-----------------------------------------------------------------------
If you take a look at the integrations/solr module you can find the testPut() method which does:
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
inside SolrOffHeapCacheTest.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
was (Author: teofili):
bq. didn't find a test which documents the need of this case.
If you take a look at the integrations/solr module you can find the testPut() method which does:
QueryResultKey queryResultKey =...
DocValues docValues = ...
solrOffHeapCache.put( queryResultKey, docValues );
inside SolrOffHeapCacheTest.
The SolrCache<K,V> interface has generics so the impl using DM is SolrOffHeapCache<K,V> and consequently it'd be very good if we could have coherent behavior in the underlying DM Cache
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Michael André Pearce (Commented JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210935#comment-13210935 ]
Michael André Pearce commented on DIRECTMEMORY-43:
--------------------------------------------------
This is fine though Simone.
As the generics are forcing the code stored in the buffer to only be of V type. As such this code would only execute if V is a ByteBuffer class.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Simone Tripodi (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Simone Tripodi updated DIRECTMEMORY-43:
---------------------------------------
Attachment: DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch
Hi again guys,
starting from the patch that Michael kindly provided, I tried to enrich the type inference in the whole system, I got anyway confused by {{ByteBuffer}} types mixed in the cache {{V}} type... can anyone rescue me?
I am attaching anyway my version of the patch - pay attention that it cannot compile due to not assignable ByteBuffer type...
I hope Mau/Raf have clearer ideas about it!
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Simone Tripodi (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13177334#comment-13177334 ]
Simone Tripodi commented on DIRECTMEMORY-43:
--------------------------------------------
I like it, I just wonder if we can change the cache signature of type <K, V>
We could use it in the future also to be integrated in OGNL - and so in Struts...
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Benoit Perroud (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13192923#comment-13192923 ]
Benoit Perroud commented on DIRECTMEMORY-43:
--------------------------------------------
This is a hack introduced to be able to return directly a ByteBuffer, and play with the bytebuffer to avoid byte[] copy.
Altough I find the feature interesting, I also would prefer a documented function getByteBufferForKey(key) to access directly the bytebuffer.
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Simone Tripodi <si...@apache.org>.
there is a reason why I adopted it: Pointer contains the stored Class
type reference, so having the generic there, helps on avoiding
unneeded and useless casts and SuppressWarnings - indeed, type are
enforced at higher level, so why casting?
best,
-Simo
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
On Sat, Feb 18, 2012 at 2:54 PM, Michael André Pearce
<mi...@me.com> wrote:
> Also to note, there is no need to make your pointer generic (this can be sorted at a later date), as a user of the cache the generics enforce what goes in and out the cache.
>
>
> On 18 Feb 2012, at 13:53, Michael André Pearce wrote:
>
>> Simon,
>>
>> Taking your Cache<String,String>
>>
>> So K = String, and V = String therefor ptr.clazz would = String not ByteBuffer , java.lang.String doesn't extend ByteBuffer, thus this code would execute.
>>
>>
>> Mike
>>
>>
>> On 18 Feb 2012, at 13:40, Simone Tripodi (Commented) (JIRA) wrote:
>>
>>>
>>> [ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210943#comment-13210943 ]
>>>
>>> Simone Tripodi commented on DIRECTMEMORY-43:
>>> --------------------------------------------
>>>
>>> Michael, I had a deep review in the code, and IIUC, we can have a {{Cache<String, String>}}, that would mean:
>>>
>>> {code}
>>> @@ -214,11 +214,11 @@
>>> if ( ptr.clazz == ByteBuffer.class )
>>> {
>>> // skip serialization if it is a bytebuffer
>>> - return ptr.directBuffer;
>>> + return (String) ptr.directBuffer;
>>> }
>>> {code}
>>>
>>> that is not correct
>>>
>>>> Cache should allow key objects instead of plain string
>>>> ------------------------------------------------------
>>>>
>>>> Key: DIRECTMEMORY-43
>>>> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
>>>> Project: Apache DirectMemory
>>>> Issue Type: Improvement
>>>> Reporter: Maurizio Cucchiara
>>>> Assignee: Maurizio Cucchiara
>>>> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>>>>
>>>>
>>>> See http://goo.gl/D4SRN
>>>
>>> --
>>> This message is automatically generated by JIRA.
>>> If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
>>> For more information on JIRA, see: http://www.atlassian.com/software/jira
>>>
>>>
>>
>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Michael André Pearce <mi...@me.com>.
Also to note, there is no need to make your pointer generic (this can be sorted at a later date), as a user of the cache the generics enforce what goes in and out the cache.
On 18 Feb 2012, at 13:53, Michael André Pearce wrote:
> Simon,
>
> Taking your Cache<String,String>
>
> So K = String, and V = String therefor ptr.clazz would = String not ByteBuffer , java.lang.String doesn't extend ByteBuffer, thus this code would execute.
>
>
> Mike
>
>
> On 18 Feb 2012, at 13:40, Simone Tripodi (Commented) (JIRA) wrote:
>
>>
>> [ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210943#comment-13210943 ]
>>
>> Simone Tripodi commented on DIRECTMEMORY-43:
>> --------------------------------------------
>>
>> Michael, I had a deep review in the code, and IIUC, we can have a {{Cache<String, String>}}, that would mean:
>>
>> {code}
>> @@ -214,11 +214,11 @@
>> if ( ptr.clazz == ByteBuffer.class )
>> {
>> // skip serialization if it is a bytebuffer
>> - return ptr.directBuffer;
>> + return (String) ptr.directBuffer;
>> }
>> {code}
>>
>> that is not correct
>>
>>> Cache should allow key objects instead of plain string
>>> ------------------------------------------------------
>>>
>>> Key: DIRECTMEMORY-43
>>> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
>>> Project: Apache DirectMemory
>>> Issue Type: Improvement
>>> Reporter: Maurizio Cucchiara
>>> Assignee: Maurizio Cucchiara
>>> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>>>
>>>
>>> See http://goo.gl/D4SRN
>>
>> --
>> This message is automatically generated by JIRA.
>> If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
>> For more information on JIRA, see: http://www.atlassian.com/software/jira
>>
>>
>
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Simone Tripodi <si...@apache.org>.
Hi,
> Taking your Cache<String,String>
>
> So K = String, and V = String therefor ptr.clazz would = String not ByteBuffer , java.lang.String doesn't extend ByteBuffer, thus this code would execute.
the would be in any case wrong - wouldn't be enough having the <V>?
there is a reason if ByteBuffer case is considered special, or not?
Simo
PS: I know that sounds strange, but my name is SimonE :) Italy is the
only country in the world AFAIK where that name si male :D
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
Re: [jira] [Commented] (DIRECTMEMORY-43) Cache should allow key
objects instead of plain string
Posted by Michael André Pearce <mi...@me.com>.
Simon,
Taking your Cache<String,String>
So K = String, and V = String therefor ptr.clazz would = String not ByteBuffer , java.lang.String doesn't extend ByteBuffer, thus this code would execute.
Mike
On 18 Feb 2012, at 13:40, Simone Tripodi (Commented) (JIRA) wrote:
>
> [ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210943#comment-13210943 ]
>
> Simone Tripodi commented on DIRECTMEMORY-43:
> --------------------------------------------
>
> Michael, I had a deep review in the code, and IIUC, we can have a {{Cache<String, String>}}, that would mean:
>
> {code}
> @@ -214,11 +214,11 @@
> if ( ptr.clazz == ByteBuffer.class )
> {
> // skip serialization if it is a bytebuffer
> - return ptr.directBuffer;
> + return (String) ptr.directBuffer;
> }
> {code}
>
> that is not correct
>
>> Cache should allow key objects instead of plain string
>> ------------------------------------------------------
>>
>> Key: DIRECTMEMORY-43
>> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
>> Project: Apache DirectMemory
>> Issue Type: Improvement
>> Reporter: Maurizio Cucchiara
>> Assignee: Maurizio Cucchiara
>> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>>
>>
>> See http://goo.gl/D4SRN
>
> --
> This message is automatically generated by JIRA.
> If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
> For more information on JIRA, see: http://www.atlassian.com/software/jira
>
>
[jira] [Commented] (DIRECTMEMORY-43) Cache should allow key objects
instead of plain string
Posted by "Simone Tripodi (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DIRECTMEMORY-43?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13210943#comment-13210943 ]
Simone Tripodi commented on DIRECTMEMORY-43:
--------------------------------------------
Michael, I had a deep review in the code, and IIUC, we can have a {{Cache<String, String>}}, that would mean:
{code}
@@ -214,11 +214,11 @@
if ( ptr.clazz == ByteBuffer.class )
{
// skip serialization if it is a bytebuffer
- return ptr.directBuffer;
+ return (String) ptr.directBuffer;
}
{code}
that is not correct
> Cache should allow key objects instead of plain string
> ------------------------------------------------------
>
> Key: DIRECTMEMORY-43
> URL: https://issues.apache.org/jira/browse/DIRECTMEMORY-43
> Project: Apache DirectMemory
> Issue Type: Improvement
> Reporter: Maurizio Cucchiara
> Assignee: Maurizio Cucchiara
> Attachments: DIRECTMEMORY-43.patch, DIRECTMEMORY-43_wider-type-inference_NOT-WORKING.patch, patch.org.apache.directmemory.cache.zip
>
>
> See http://goo.gl/D4SRN
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira