Thanks for investigating further on this Basil! /James
On Fri, 11 Jun 2021 at 16:00, Basil Crow <[email protected]> wrote: > On Fri, Jun 11, 2021 at 2:19 AM Robert Sandell <[email protected]> > wrote: > > > > Some historical context to know where we "old timers" are coming from :) > > https://kohsuke.org/2012/03/03/potd-package-renamed-asm/ > > Thanks for providing this context, Robert! I have a tremendous amount > of respect for all the old-timers in this project. Thanks for having > the patience to explain this to a newcomer like me. > > I dug up the original post [1] where Kohsuke laid out his complaints > about ASM. All these complaints seem completely legitimate in the > context of ASM 3 in 2010, which definitely seems like it broke > compatibility with ASM 2. > > I know these scars run deep, but I think we should re-examine things > in 2021. As of commit 6a0e7842de436225f3866d3567834c6285107114 in ASM > 5.2, ASM added signature tests to avoid breaking backward binary > compatibility with any version >= 4.0. These signature tests are still > present in "src/test/resources" on the latest version of ASM. So ASM > appears to take compatibility seriously these days. > > We shouldn't be as nervous about compatibility between ASM 5 and ASM 9 > in 2021 as we were about compatibility between ASM 2 and ASM 3 in > 2010. Things have changed in the intervening decade. > > > That's why me and James and others are very very wary of bumping the asm > dependencies, I think even more wary than for Guava, because there isn't > enough code coverage due to how the classpath is during unit testing. > > Again, I know these scars run deep, but allow me to reiterate that > core has _already_ bumped ASM to latest in December 2020, in my > opinion unintentionally, as a result of bumping JNR to latest. I > didn't review that change, but I am not casting aspersions on those > who reviewed and merged it either. It is a particularly easy mistake > to make. But the fact is, I think we are unlikely to roll it back now. > So whether we like it or not, ASM 9 as a core dependency seems like it > is here to stay. > > And since it is here to stay, and since compatibility is taken > seriously in recent versions of ASM, I think it is OK for plugins to > use the version from core by excluding the ASM dependency. > > Basil > > [1] > https://web.archive.org/web/20120701173200/http://weblogs.java.net/blog/kohsuke/archive/2010/02/12/asm-incompatible-changes > [2] > https://gitlab.ow2.org/asm/asm/-/commit/6a0e7842de436225f3866d3567834c6285107114 > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CAFwNDjpBuUqLfb7McWCZ6Ue2eMCGDj2vff9atO0_94VLykXCeQ%40mail.gmail.com > . > -- You received this message because you are subscribed to the Google Groups "Jenkins Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-dev/CAPzq3pd_MpUKje_8dcQjhB4tptTYKb6KteuuqKw-RNJHJNxxcw%40mail.gmail.com.
