Prepared branch for slf4j 2.0
Hi Ceki I prepared the branch for slf4j 2.0. I cannot make a PR that targets a non existing branch: can you make one? Code is available at https://github.com/ffissore/slf4j/tree/2.0.0-SNAPSHOT I've replaced occurrences of java 1.5 and 1.6 to 1.8 and I've included slf4j fluent Let me know what you think federico
Hi Federico, I just created the 2.0 branch. On 19.02.2019 13:00, Federico Fissore wrote:
Hi Ceki
I prepared the branch for slf4j 2.0. I cannot make a PR that targets a non existing branch: can you make one?
Code is available at https://github.com/ffissore/slf4j/tree/2.0.0-SNAPSHOT
I've replaced occurrences of java 1.5 and 1.6 to 1.8 and I've included slf4j fluent
Let me know what you think
federico
-- Ceki Gülcü
Ceki ha scritto il 19/02/19 alle 14:35:
Hi Federico,
I just created the 2.0 branch.
PR ready https://github.com/qos-ch/slf4j/pull/203 although it fails on code unrelated to the change federico
Hi Federico, Here is a very basic SKELETON implementation of what I would like to see: https://github.com/qos-ch/slf4j/tree/fluent-experiment It's far from complete but gets the picture across. As far as I can tell, the remaining big question is how to handle structured messages. -- Ceki On 20.02.2019 15:37, Federico Fissore wrote:
Ceki ha scritto il 19/02/19 alle 14:35:
Hi Federico,
I just created the 2.0 branch.
PR ready https://github.com/qos-ch/slf4j/pull/203 although it fails on code unrelated to the change
federico
I've updated the PR, merging fluent apis into Logger and hiding details under the spi package I've kept the slf4j-fluent module for testing purposes only: it relies on logback which relies on slf4j-api so having those tests into slf4j-api would have created a circular dependency I'm a basic user so: what are structured messages? federico Ceki ha scritto il 20/02/19 alle 22:20:
Hi Federico,
Here is a very basic SKELETON implementation of what I would like to see:
https://github.com/qos-ch/slf4j/tree/fluent-experiment
It's far from complete but gets the picture across.
As far as I can tell, the remaining big question is how to handle structured messages.
Ceki ha scritto il 20/02/19 alle 22:20:
As far as I can tell, the remaining big question is how to handle structured messages.
Hi Indeed structured messages were not working, but they are working now: I've pushed a fix and travis was able to build the branch this time let me know what you think ps: I've also back-ported the fix to slf4j-fluent 0.10.0, which will be available as soon as maven servers will be synchronized federico
Hi Ceki I see you've added a number of commits on branch 2.0, including some work on the fluent api which are conflicting with the work done in https://github.com/qos-ch/slf4j/pull/203 I assume you're no longer interested, are you? Best federico Federico Fissore ha scritto il 19/02/19 alle 13:00:
Hi Ceki
I prepared the branch for slf4j 2.0. I cannot make a PR that targets a non existing branch: can you make one?
Code is available at https://github.com/ffissore/slf4j/tree/2.0.0-SNAPSHOT
I've replaced occurrences of java 1.5 and 1.6 to 1.8 and I've included slf4j fluent
Let me know what you think
federico _______________________________________________ slf4j-dev mailing list slf4j-dev@qos.ch http://mailman.qos.ch/mailman/listinfo/slf4j-dev
Hi Federico, Your pull request has a number of interesting ideas. I think some of them should be merged into the backward compatible implementation that I just committed a few minutes ago. Backward compatibility is the main hurdle to clear. Once that is done, as now seems to be the case, we can further improve the API piecemeal. Your suggestions for improvement are welcome. In any case, I will be reading your proposal once again even more attentively. -- Ceki On 29.04.2019 08:57, Federico Fissore wrote:
Hi Ceki
I see you've added a number of commits on branch 2.0, including some work on the fluent api which are conflicting with the work done in https://github.com/qos-ch/slf4j/pull/203
I assume you're no longer interested, are you?
Best
federico
Federico Fissore ha scritto il 19/02/19 alle 13:00:
Hi Ceki
I prepared the branch for slf4j 2.0. I cannot make a PR that targets a non existing branch: can you make one?
Code is available at https://github.com/ffissore/slf4j/tree/2.0.0-SNAPSHOT
I've replaced occurrences of java 1.5 and 1.6 to 1.8 and I've included slf4j fluent
Let me know what you think
federico _______________________________________________ slf4j-dev mailing list slf4j-dev@qos.ch http://mailman.qos.ch/mailman/listinfo/slf4j-dev
_______________________________________________ slf4j-dev mailing list slf4j-dev@qos.ch http://mailman.qos.ch/mailman/listinfo/slf4j-dev
-- -- Ceki Gülcü
Ceki ha scritto il 29/04/19 alle 22:55:
Hi Federico,
Your pull request has a number of interesting ideas. I think some of them should be merged into the backward compatible implementation that I just committed a few minutes ago.
Backward compatibility is the main hurdle to clear. Once that is done, as now seems to be the case, we can further improve the API piecemeal.
I'm sure the code in that PR is backwards compatible. Can you point me to where those changes have broken backwards compatibility? best federico
Hi Federico, I mean that it is preferable to make bw compatible changes directly to slf4j-api module instead of introducing a new module. Cheers, -- Ceki On 30/04/2019 10:25, Federico Fissore wrote:
Ceki ha scritto il 29/04/19 alle 22:55:
Hi Federico,
Your pull request has a number of interesting ideas. I think some of them should be merged into the backward compatible implementation that I just committed a few minutes ago.
Backward compatibility is the main hurdle to clear. Once that is done, as now seems to be the case, we can further improve the API piecemeal.
I'm sure the code in that PR is backwards compatible. Can you point me to where those changes have broken backwards compatibility?
best
federico
and now I see that that is what you have done. The new module is for testing only. On 30/04/2019 13:18, Ceki Gülcü wrote:
Hi Federico,
I mean that it is preferable to make bw compatible changes directly to slf4j-api module instead of introducing a new module.
Cheers,
--
Ceki
On 30/04/2019 10:25, Federico Fissore wrote:
Ceki ha scritto il 29/04/19 alle 22:55:
Hi Federico,
Your pull request has a number of interesting ideas. I think some of them should be merged into the backward compatible implementation that I just committed a few minutes ago.
Backward compatibility is the main hurdle to clear. Once that is done, as now seems to be the case, we can further improve the API piecemeal.
I'm sure the code in that PR is backwards compatible. Can you point me to where those changes have broken backwards compatibility?
best
federico
slf4j-dev mailing list slf4j-dev@qos.ch http://mailman.qos.ch/mailman/listinfo/slf4j-dev
participants (3)
-
Ceki -
Ceki Gülcü
-
Federico Fissore