2012/11/11 Jason van Zyl <[email protected]>: > > On Nov 11, 2012, at 2:49 AM, Olivier Lamy <[email protected]> wrote: > >> >> Perso I propose a change by pointing you (you means other maven dev >> folks too) to a branch I made somewhere but you commit code without >> listening POV from others. >> If you could wait to hear what other thinks that could be lovely.... > > I believe you do exactly what you accuse me of Olivier. You did not propose a > change, you pointed to your branch with a terse "fixed" as if it were a > foregone conclusion. Oh maybe I should have say "possible fix" using log4j2 sorry for using bad word but I'm a coder not a writer and furthermore I'm not a native english speaker so it can happen (I have updated the jira comment). But I have started a discussion here (AFAIK @apache mailing list is the place to discuss rather than jira) > > I started the SLF4J work, I worked with Ceki to try and minimize the change, > keep the ITs passing while preserving the existing behaviour and keeping the > dependency size and complexity to a minimum. > > I've been working on restoring the behaviour and my goal, at least, was to > reduce the possible complication of using a larger framework. The second I > created the JIRA issue, you point at your branch and say "fixed" without any > explanation. You used the console transfer listener not working -- and I > admit that was annoying and I apologize for leaving it like that so long -- > as a vehicle for adding your preferred logging framework. My goal was to > introduce SLF4J in a minimal way, at least to start. So if that conflicts > with your goal then that's fine but jumping in the middle of the work I'm > doing with a change that proposes to throw away the work I did with SLF4J > Simple is not fine. Couching it as me not taking into account a wider > discussion as a response to me finishing what I started with a veto even less > so.
I don't have any issues using slf4j as logging api but we can go (IMHO) a bit forward with proposing a more advanced logging implementation instead of the choice you made for slf4j-simple (users ask a more advanced logging options for a while) so it's probably the time to do it and take the opportunity of the good changes you made introducing slf4j api > > I didn't change any of the dependencies, completed the work I started and > fixed what I broke which I believe is reasonable. > > If the discussion is now transitioning to users want flexible logging and the > choice of a logging framework that's fine. But I still maintain the CLI use > of logging can be limited and constrained while allowing integrators to make > the small changes necessary to add flexible logging. But if we want to choose > a framework let's look at the options, if people want to go that route, and > select the best option. Integrators ? Again what do you mean with that ? I don't understand why the default Apache Maven couldn't be able to propose a default advanced logging implementation. The size of the jars is around 500K so frankly I don't see that as a blocking issue as we already download internet :-). (and perso I'd like to test some ideas using jansi for possible colorized logging) And I don't understand why we must wait folks doing alternate distributions providing this feature. > > Reverting my commit will break the console transfer listener. The discussion > about the use of a logging framework, and its choice if so decided, is not a > foregone conclusion. So I will revert my commit in the morning when I wake up > if you want the broken behaviour restored. But note I believe you are being > unreasonable in that you haven't said a word until I raised the JIRA issue > today and then took offense to me finishing my work while I was in the > process of correcting what I broke. Obviously you were working on your branch > while I was working on my fixes but nothing was brought up aside from JIRA. > Just a matter of timing as I work on this a bit this week during my holidays (first I asked on log4j mailing list a new feature needed for maven). They fix that very quickly (thanks to log4j dev listening here) so I just finished my work yesterday. Once finished I just put that in a public space for reviews by others and that's it (AFAIK that's the git power) and honestly I'm not sure I'm the good target to be accuse doing stuff in a private area regarding maven... Again as explained previously the goal is to provide an advanced logging implementation in the standard Apache Maven distro. So I started this thread and added a comment in the jira but despite that you committed so even if I don't like that the only solution for me was a veto to be heard. > You have made sweeping changes in the transport and while you have made > improvements, you have introduced several things that don't work as they did > previously -- and I have brought these up with you directly, especially as it > pertains to security -- I have not jumped down your throat with a veto as I > expect you will eventually fix them because you care about users. Please do > me the same courtesy. If you talk about the preemptive for get on wagon I have fixed that. And honestly the issue existed before that even without preemptive for get (see explanation/proposal on this topic here: http://markmail.org/message/7pswshucxc7qwtef) See https://jira.codehaus.org/browse/WAGON-371 (yes maybe I miss to release that I can do it next week) So to be able to move forward I revert my veto. > >> >> >>> >>>> >>>> Thanks >>>> -- >>>> Olivier Lamy >>>> Talend: http://coders.talend.com >>>> http://twitter.com/olamy | http://linkedin.com/in/olamy >>>> >>>> --------------------------------------------------------------------- >>>> To unsubscribe, e-mail: [email protected] >>>> For additional commands, e-mail: [email protected] >>>> >>> >>> Thanks, >>> >>> Jason >>> >>> ---------------------------------------------------------- >>> Jason van Zyl >>> Founder & CTO, Sonatype >>> Founder, Apache Maven >>> http://twitter.com/jvanzyl >>> --------------------------------------------------------- >>> >>> We all have problems. How we deal with them is a measure of our worth. >>> >>> -- Unknown >>> >>> >>> >>> >>> >> >> >> >> -- >> Olivier Lamy >> Talend: http://coders.talend.com >> http://twitter.com/olamy | http://linkedin.com/in/olamy >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> > > Thanks, > > Jason > > ---------------------------------------------------------- > Jason van Zyl > Founder & CTO, Sonatype > Founder, Apache Maven > http://twitter.com/jvanzyl > --------------------------------------------------------- > > > > > > -- Olivier Lamy Talend: http://coders.talend.com http://twitter.com/olamy | http://linkedin.com/in/olamy --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
