Le jeu. 9 juil. 2020 à 12:43, Tibor Digana <[email protected]> a écrit :
> @Romain it is very strange that the internal person says that you want to > revise the existence of Surefire and also we have a colleague Christian who > creates a competitor plugin and again an internal new person, very strange. > Not really,we already discussed it here and I agree we already disagree on that topic but point is not surefire which is just one detail there but maven modularity and adaptability to today's IT ecosystem, dont get it wrong please. Btw using the exec plugin would make the users unhappy. And in my > experiences with the users, they consider the Surefire as part of Maven > because it is part of it. > Agree, it is also why default should stay relevant. > Regarding the Scala testing it is not true, as you said, that > scalatest-maven-plugin is popular. No! Another build process is popular, it > is SBT. > I didnt get that part but it is common to move away from sbt to use maven too - think spark or flink do? And there was one more plugin you mentioned but it is 6 years out of > maintenance, i do not remember the name of that plugin. > Frontend maven plugin? It is very up to date. Spock? It runs in jupiter platform these days. BDD (cucumber, jbehave, ...) ones or perf ones (soapui, jmeter) are also regularly used and are still about testing. > On Wed, Jul 8, 2020 at 1:58 PM Romain Manni-Bucau <[email protected]> > wrote: > > > https://github.com/apache/maven-surefire/pull/305 mitigates the flush > > issue > > - understand it solves the user blocker but brings a new config which can > > be outdated in another milestone (not a big deal, milestones are there > for > > that too IMHO) - but a clean solution is indeed to just be aligned on > > server flushing (doable wrapping the buffer itself) or just not needing > it > > as it was in 2.x thanks to process piping. > > > > Long term we should revise surefire responsibilities IMHO because today > it > > is trivial to supersede it with an exec-maven-plugin execution (you only > > lose maven site integration but gain in simplicity and site integration > is > > not that much used today outside commons projects I think). > > > > > > Romain Manni-Bucau > > @rmannibucau <https://twitter.com/rmannibucau> | Blog > > <https://rmannibucau.metawerx.net/> | Old Blog > > <http://rmannibucau.wordpress.com> | Github < > > https://github.com/rmannibucau> | > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | Book > > < > > > https://www.packtpub.com/application-development/java-ee-8-high-performance > > > > > > > > > Le mer. 8 juil. 2020 à 13:42, Tibor Digana <[email protected]> a > > écrit : > > > > > If you guys mean the improvements seriously then we have to work on a > PR > > > together. But this work is not for one hour. I am mainly repairing the > > unit > > > tests in the late evenings when I am watching the TV. So this is my > daily > > > workflow. > > > > > > The logs are flushed when BufferedOutputStream overflows, this means > 64KB > > > buffer or when some control event is emitted. This makes such an effect > > on > > > the console that a big batch of logs suddenly appear. > > > > > > Currently I am merging SUREFIRE-1809 which is in the interest of Camel > > and > > > Quarkus projects. Also a blocker issue which has however the > workaround. > > > > > > > > > > > > On Wed, Jul 8, 2020 at 9:07 AM Romain Manni-Bucau < > [email protected] > > > > > > wrote: > > > > > > > +1 perf and lost of output flushing are big drawbacks we should fix > > asap > > > > without regression (reverting m5 upgrade to m4 brings other issues) > > > > > > > > Le mer. 8 juil. 2020 à 07:32, Enrico Olivelli <[email protected]> > a > > > > écrit : > > > > > > > > > Olivier, > > > > > If it is only a matter of doing the release I can pack it today or > > > > tomorrow > > > > > (not Friday). > > > > > > > > > > What's the status of master branch? I mean, are there any other > > blocker > > > > > issues waiting for fix or for review? > > > > > > > > > > Enrico > > > > > > > > > > Il Mer 8 Lug 2020, 01:40 Falko Modler <[email protected]> ha > scritto: > > > > > > > > > > > Hi Oliver, > > > > > > > > > > > > which performance issue do you mean? > > > > > > > > > > > > I was actually hoping to revive > > > > > > https://github.com/apache/maven-surefire/pull/169 for M6 but I > > > haven't > > > > > > had time thus far. > > > > > > > > > > > > Cheers, > > > > > > > > > > > > Falko > > > > > > > > > > > > Am 08.07.2020 um 00:38 schrieb Olivier Lamy: > > > > > > > Hi, > > > > > > > I wonder if there is a plan to release M6 ASAP with the > > performance > > > > > issue > > > > > > > fixed? > > > > > > > TBH M5 is definitely NOT usable but it contains some > interesting > > > > fixes > > > > > > but > > > > > > > performance is awful. > > > > > > > I'm definitely happy to help. > > > > > > > let me know. > > > > > > > > > > > > > > cheers > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > To unsubscribe, e-mail: [email protected] > > > > > > For additional commands, e-mail: [email protected] > > > > > > > > > > > > > > > > > > > > > > > > > > >
