Le jeu. 6 nov. 2025 à 22:12, Vladimir Sitnikov <[email protected]> a écrit :
> Romain, > > I said "Both exec plugin and *maven-dependency-plugin resolve to the same > classpath" > Then you said "showed your it was not the case". > > Then I *proved *classpaths are the same. That proves you have not shown the > classpaths differ. > Vladimir, with all the respect can have, I explained you why it was not and proved you it was not and that your example code was misleading in that regard very early in this regard. I understand your obstination to request proof and not want to check as a "I don't care" and therefore I'll stop on this thread but please take my advice to think about it over the week end for your use case and check/debug for your example. > > Do you finally agree that dependency and exec resolve the same classpaths? > Obviously not since I know it is wrong and technically it can't be. > > Very same thing happens with all the other examples you throw. > > > gave several and even the ones you quoted about jackson (I added spark - > >with aws client for ex - which is a trivial one) or commons. > > You did mention jackson, spark and commons. > However, you have never mentioned exact issues. > I did, you just never read answers people give you, dependency convergence would be conflicting with your proposal or not the right version until semver enters into play as explains gradle doc. > > Hey Romain, there's no weight in just saying "there were issues with > Jackson, so Vladimir's proposal is bad". > Justify your claims please. > Same for you, there is no point having broken example and say they are right so please take some time for you and to think about the comments you got - not only from me. > > Vladimir >
