I did ask Greg Luck to remove the hibernate ehcache provider from the
ehcache distribution when I did commit the new one. but I had no answer.
Gavin is a commiter for ehcache, he could do the job.
Emmanuel Bernard
01.55.21.52.14
Fnac - DSI
Direction France - Nouvelles technologies
67, boulevard
Hi guys,
Have anyone told the ehcache guys to update their cacheprovider ?
As you can see in http://forum.hibernate.org/viewtopic.php?t=938297,
we are now bundling our own ehcache provider to work around the need for
having
start/stop on the providers.
This makes it an requirement that hibernate
EHCache 0.5 was released today. It has the backport to JDK1.3 and
JDK1.2 that has been discussed in this thread included.
The Changelog follows:
EHCache-0.5
===
Bug Fixes
-
- Implemented not creating a Disk Store when overflowToDisk="false" is
set in ehcache.xml
- toString() met
Great :)
Greg Luck wrote:
Gavin
Done.
I have changed logging to commons-logging and have added LRUHashMap
from commons-collections. This, along with some packaging changes will
be in CVS later today.
Greg Luck
On 15/11/2003, at 10:08 AM, Gavin King wrote:
We already have a dependency to
Gavin
Done.
I have changed logging to commons-logging and have added LRUHashMap
from commons-collections. This, along with some packaging changes will
be in CVS later today.
Greg Luck
On 15/11/2003, at 10:08 AM, Gavin King wrote:
We already have a dependency to commons-collections, so perh
Luck" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, November 15, 2003 1:10 AM
Subject: [Hibernate] EHCache and JDK1.4
> Hi
>
> EHCache does require the use of JDK1.4.
>
> Specifically we use:
>
> - LinkedHashMap, which provides us with an LRU Hash
We already have a dependency to commons-collections, so perhaps just use
commons SequencedHashMap instead of LinkedHashMap.
Greg Luck wrote:
Hi
EHCache does require the use of JDK1.4.
Specifically we use:
- LinkedHashMap, which provides us with an LRU HashMap
- JDK1.4 logging
- SoftRefererenc
On 15 Nov (09:10), Greg Luck wrote:
> I was pretty keen to use the java.util.LinkedHashMap, because the
> combination of a doubly-linked
> list combined with a Hashmap in JCS was the source of the memory leak
> in it. The memory store code
> went from around 1500 lines down to a few hundred and
Hi
EHCache does require the use of JDK1.4.
Specifically we use:
- LinkedHashMap, which provides us with an LRU HashMap
- JDK1.4 logging
- SoftRefererences
I guess there is a lot of production code out there on JDK1.3 and
earlier. I could replace
the JDK1.4 logging with commons logging and the j
is already deprecated in 2.1 and replaced by
David Channon wrote:
Now what does this change (and the generalisation) mean to the DTD and all
those JCS cache
tags?
--David.
---
This SF.Net email sponsored by: ApacheCon 2003,
16-19 November in
cache
tags?
--David.
- Original Message -
From: "Gavin King" <[EMAIL PROTECTED]>
To: "hibernate list" <[EMAIL PROTECTED]>
Sent: Friday, November 14, 2003 11:17 AM
Subject: [Hibernate] EHCache
>
> Greg Luck has forked JCS and created a new project
ate be 1.4+ only?
-Original Message-
From: Gavin King [mailto:[EMAIL PROTECTED]
Sent: Thursday, November 13, 2003 5:18 PM
To: hibernate list
Subject: [Hibernate] EHCache
Greg Luck has forked JCS and created a new project on sf.net. IMO we
should replace our JCS integration with EHCache. D
On 13 Nov (19:28), Sandeep Dath wrote:
> Perhaps there could also be a *little* more documentation (all supporting
> caching frameworks) on caching, and caching best practices?
On my desk and will soon be in CVS.
--
Christian Bauer
[EMAIL PROTECTED]
---
PM
To: hibernate list
Subject: [Hibernate] EHCache
Greg Luck has forked JCS and created a new project on sf.net. IMO we
should replace our JCS integration with EHCache. Does anyone object to
simply removing JCS support in 2.1? I think we have all seen how there
are problems with robustness
Greg Luck has forked JCS and created a new project on sf.net. IMO we
should replace our JCS integration with EHCache. Does anyone object to
simply removing JCS support in 2.1? I think we have all seen how there
are problems with robustness / performance in the build we are currently
using.
htt
15 matches
Mail list logo