|Pete Muir||Aug 17, 2009 10:05 am|
|Ralph Goers||Aug 17, 2009 2:40 pm|
|Pete Muir||Aug 18, 2009 6:37 am|
|Ralph Goers||Aug 18, 2009 7:10 am|
|近藤 健||Aug 18, 2009 9:59 am|
|Pete Muir||Aug 19, 2009 8:31 am|
|Ralph Goers||Aug 19, 2009 9:15 am|
|Ceki Gulcu||Aug 19, 2009 11:17 am|
|Pete Muir||Aug 19, 2009 11:20 am|
|Pete Muir||Aug 19, 2009 11:29 am|
|Ceki Gulcu||Aug 19, 2009 11:42 am|
|Pete Muir||Aug 19, 2009 11:51 am|
|Ceki Gulcu||Aug 19, 2009 12:38 pm|
|Ralph Goers||Aug 19, 2009 1:42 pm|
|Ceki Gulcu||Aug 19, 2009 1:58 pm|
|Ceki Gulcu||Aug 19, 2009 2:15 pm|
|Ralph Goers||Aug 19, 2009 2:21 pm|
|Ralph Goers||Aug 19, 2009 2:31 pm|
|Ceki Gulcu||Aug 19, 2009 2:40 pm|
|近藤 健||Aug 20, 2009 8:22 am|
|Ralph Goers||Aug 20, 2009 8:35 am|
|Takeshi Kondo||Aug 20, 2009 10:07 am|
|ralp...@dslextreme.com||Aug 20, 2009 10:20 am|
|Ceki Gulcu||Aug 20, 2009 1:58 pm|
|Ceki Gulcu||Aug 20, 2009 2:05 pm|
|Takeshi Kondo||Aug 21, 2009 10:28 pm|
|Takeshi Kondo||Aug 22, 2009 10:32 pm||.jar, .jar|
|Ralph Goers||Aug 23, 2009 8:21 am|
|Takeshi Kondo||Aug 23, 2009 8:40 am|
|Ceki Gulcu||Aug 23, 2009 10:38 am|
|Takeshi Kondo||Aug 23, 2009 4:59 pm|
|Ralph Goers||Aug 23, 2009 9:56 pm|
|Ceki Gulcu||Aug 24, 2009 6:14 am|
|Takeshi Kondo||Aug 24, 2009 10:02 am|
|Ceki Gulcu||Aug 24, 2009 10:22 am|
|Takeshi Kondo||Aug 24, 2009 11:05 am|
|Ceki Gulcu||Aug 24, 2009 11:27 am|
|Takeshi Kondo||Aug 24, 2009 12:36 pm|
|Ceki Gulcu||Aug 24, 2009 12:56 pm|
|Takeshi Kondo||Aug 24, 2009 1:15 pm|
|Ceki Gulcu||Aug 24, 2009 1:24 pm|
|Ralph Goers||Aug 24, 2009 1:33 pm|
|Takeshi Kondo||Aug 24, 2009 2:02 pm|
|Ceki Gulcu||Aug 25, 2009 1:32 am|
|Subject:||Re: [slf4j-dev] slf4j i8ln|
|From:||Pete Muir (pmu...@redhat.com)|
|Date:||Aug 19, 2009 8:31:11 am|
Whether or not resource bundles suck in our opinion, they are the standard approach to this so I believe we can't just dismiss them.
I'm also unsure how, in your approach, a framework would provide i8ln'ized log messages which would be used?
On 17 Aug 2009, at 22:41, Ralph Goers wrote:
My 2 cents.
ResourceBundles suck. Even in Java 1.6 it is difficult to change the implementation and it only works if the application cooperates. The default implementation finds the bundles in the classpath which makes it difficult if you like to store the files outside of the application. Also, since they are loaded on the classpath they aren't automatically reloaded when modified. My organization also has "special" needs when it comes to internationalization - a single application can support thousands of clients each of which may want to override some of the keys in the bundles.
In short, it seems to me to make far more sense to use a separate I18n framework to deal with the actual message text and then just make sure that SLF4J can accept the Locale as a parameter to be passed to the Appender.
Another option along the same lines would be to use the message field as the message key, along with the parameters and pass those to the Appender along with the locale. There again, an I18N framework would deal with the messages.
In short, SLF4J should support I18N but not implement it.
FWIW - I have a need to implement an I18N framework based on Apache Commons Configuration to support the needs I discussed in the first paragraph. I'm considering doing it in the existing I18N project in the Apache Commons Sandbox.
On Aug 17, 2009, at 10:05 AM, Pete Muir wrote:
As discussed here https://jira.jboss.org/jira/browse/WBRI-290, we would like to switch to slf4j as our logger (it offers a logging facade, supports MDC/NDC and parameter replacement).
However, as Takeshi highlights here https://jira.jboss.org/jira/browse/WBRI-214 , a needed feature is explicit i8n support, and it sounds like Ceki would be happy to accept a contribution for this directly to the slf4j.
Perhaps, to get started, we should discuss the overall design and aims. In the linked issue, Takeshi adds three features in the patch:
- ability to localize a message by providing a resource bundle, which has the same name as the class using the logger (the declaring class) - the ability to log an enum value (rather than using a static to hold the message/key) - the ability to associate the level at which to log with the message with the enum (via an annotation) rather than in the call from the declaring class
(Takeshi, correct me if this is incorrect). I think we can probably separate these features out when discussing.
I think we would also need:
- ability to specify the resource bundle to use when getting the logger - ability to use statics fields or just a string id embedded in call to logger
But I'm sure others have given this more thought than me!