Sure! Please go ahead On Mon, Jan 29, 2024 at 8:30 AM Francisco Javier Tirado Sarti < ftira...@redhat.com> wrote:
> Ok, > Can I try removing the driver without removing the code myself? > I understood that as far as there is not references to the driver in the > code base, we are fine, isnt it? > > On Mon, Jan 29, 2024 at 1:58 PM Alex Porcelli <porce...@apache.org> wrote: > > > Francisco, > > > > Modifying the scope to test won’t work, the compilation will still fail. > > > > But even if it could work, having test scope is still a LEGAL violation, > so > > we can’t have it. > > > > Regarding us moving this code somewhere else, it creates a huge grey area > > in my opinion. We, as Apache Committers have responsibilities with > Apache. > > > > Last option you mentioned was about having it in the codebase without > being > > complied, this will create dead code and I’m heavily -1 with this option > > (to not mention that this still a LEGAL issue - as the source code would > > still not be complying with Apache Foundation legal requirements). > > > > > > On Mon, Jan 29, 2024 at 6:33 AM Francisco Javier Tirado Sarti < > > ftira...@redhat.com> wrote: > > > > > And finally, as another alternative that was suggested in the past (I > > think > > > was proposed by Enrique and Ricardo, but I do not recall the details > > right > > > now), we might set up a "non supported extension repo", where we can > move > > > all addons we have "issues" with, like this one. > > > The reason I am reluctant to completely remove this one is that it was > > > originally added by community folks in their spare time and I worried > > about > > > how it would look like for them ( I don't really know is they are still > > > using it or not, to be honest) > > > > > > > > > On Mon, Jan 29, 2024 at 12:27 PM Francisco Javier Tirado Sarti < > > > ftira...@redhat.com> wrote: > > > > > > > If that is not not working, then I suggest removing Oracle related > > > project > > > > from the module list of the parent poms. That will prevent its > > > deployment, > > > > but community will still have the code available for community usage. > > > > > > > > On Mon, Jan 29, 2024 at 12:14 PM Francisco Javier Tirado Sarti < > > > > ftira...@redhat.com> wrote: > > > > > > > >> Hi Alex, > > > >> I think Quarkus-jdbc-oracle can be scoped to test. Can you try it? > > > >> > > > >> > > > >> On Mon, Jan 29, 2024 at 9:32 AM Enrique Gonzalez Martinez < > > > >> egonza...@apache.org> wrote: > > > >> > > > >>> It is fine then to remove it. That means it is required to remove > > > >>> those modules we support using oracle as well (unfortunately). > > > >>> I am ok removing it all at once. > > > >>> +1 > > > >>> > > > >>> El vie, 26 ene 2024 a las 19:08, Tibor Zimányi (< > tzima...@apache.org > > >) > > > >>> escribió: > > > >>> > > > > >>> > Thank you for the investigation Alex! With what you wrote, I > > > personally > > > >>> > agree we should remove all the relevant code. I think even if it > > > would > > > >>> > somehow work, it would still be a potential legal problem, so > > better > > > to > > > >>> > remove all of it. It is in git history, so if someone would like > to > > > >>> revive > > > >>> > it, it is there. > > > >>> > > > > >>> > Best regards, > > > >>> > Tibor > > > >>> > > > > >>> > Dňa pi 26. 1. 2024, 19:03 Alex Porcelli <a...@porcelli.me> > > > napísal(a): > > > >>> > > > > >>> > > Coming back here with an updated PR [1] [2] set following the > > > >>> proposed > > > >>> > > concept to exclude the minimal possible amount of code.... > > > >>> > > Unfortunately it won't work. > > > >>> > > > > > >>> > > Although pointed out that the existing effective code (not > test) > > > >>> > > didn't require the JDBC, is not exactly accurate as you can see > > > here > > > >>> > > [3] with the quarkus-jdbc-oracle (which brings the oracle > driver > > as > > > >>> > > dependency). Once this dependency is removed, the module fails > in > > > >>> > > compile time (quarkus/hibernate process the > > application.properties > > > in > > > >>> > > compile time). I've tried all sorts of combinations in the > > > >>> > > application.properties, but all of them result in the same > > > >>> compilation > > > >>> > > failure by Quarkus. > > > >>> > > > > > >>> > > If the above description is true and there is no alternative, I > > > think > > > >>> > > we can agree that preserving the Oracle code won't have only no > > > >>> tests, > > > >>> > > but will also mislead users with a jar that says > "oracle-quarkus" > > > >>> that > > > >>> > > won't work. If this situation holds true, I'd prefer we > > completely > > > >>> > > remove the Oracle dependency (code and tests). > > > >>> > > > > > >>> > > I'd love to hear your thoughts. > > > >>> > > > > > >>> > > [1] - > > > >>> https://github.com/apache/incubator-kie-kogito-runtimes/pull/3376 > > > >>> > > [2] - > > > https://github.com/apache/incubator-kie-kogito-apps/pull/1969 > > > >>> > > [3] - > > > >>> > > > > > >>> > > > > > > https://github.com/apache/incubator-kie-kogito-apps/pull/1969/files#diff-65c393eaf9cd1897cec903ed11bee1706e53f938eb78d80520dcfbe364738ad6L44-L47 > > > >>> > > > > > >>> > > On Wed, Jan 24, 2024 at 5:39 AM Francisco Javier Tirado Sarti > > > >>> > > <ftira...@redhat.com> wrote: > > > >>> > > > > > > >>> > > > Hi Alex, > > > >>> > > > I agree with Enrique. Removing the driver from poms would > have > > > been > > > >>> > > enough. > > > >>> > > > Im afraid we will have to redo the oracle stuff (it was not > > easy > > > >>> to avoid > > > >>> > > > code duplication with postgres) at some moment in the future. > > > >>> > > > I commented on the PR on that regard. > > > >>> > > > So, Im really wondering, and please do not take it wrong, > given > > > >>> than > > > >>> > > > Enrique and myself are the most active committer touching > that > > > code > > > >>> > > > section, who agreed on that? > > > >>> > > > > > > >>> > > > On Wed, Jan 24, 2024 at 10:50 AM Alex Porcelli < > > > >>> porce...@apache.org> > > > >>> > > wrote: > > > >>> > > > > > > >>> > > > > Enrique, > > > >>> > > > > > > > >>> > > > > I might be interpreting the Category X dependency way too > > > >>> strictly; but > > > >>> > > > > it’s my current understanding that we cannot have such > > > >>> dependency in > > > >>> > > our > > > >>> > > > > codebase at all. > > > >>> > > > > > > > >>> > > > > The direction that we agreed is to avoid endless discussion > > and > > > >>> just > > > >>> > > remove > > > >>> > > > > for now; once we past the critical phase of our first > > release- > > > >>> we can > > > >>> > > try > > > >>> > > > > to explore and discuss different angles to tackle the > > Category > > > X > > > >>> > > > > dependency; it’s just not worth spend time right now on it. > > > >>> > > > > > > > >>> > > > > Hope it clarifies > > > >>> > > > > > > > >>> > > > > > > > >>> > > > > On Wed, Jan 24, 2024 at 3:21 AM Enrique Gonzalez Martinez < > > > >>> > > > > elguard...@gmail.com> wrote: > > > >>> > > > > > > > >>> > > > > > Hi Alex, > > > >>> > > > > > I don't understand why the code and scripts from Oracle > are > > > >>> being > > > >>> > > > > > removed. If the only problem is the binary dependency I > > don't > > > >>> see > > > >>> > > > > > remove code for oracle is required unless Oracle word is > > > >>> something > > > >>> > > > > > subject to not being used in Apache. > > > >>> > > > > > > > > >>> > > > > > I would suggest only to remove the dependencies and for > the > > > >>> test just > > > >>> > > > > > execute it conditional only if the driver is available. > > > >>> > > > > > > > > >>> > > > > > Cheers :) > > > >>> > > > > > > > > >>> > > > > > El mié, 24 ene 2024 a las 0:00, Alex Porcelli (< > > > >>> a...@porcelli.me>) > > > >>> > > > > > escribió: > > > >>> > > > > > > > > > >>> > > > > > > Hi Francisco and team, > > > >>> > > > > > > > > > >>> > > > > > > Here are the links for the PRs that remove references > for > > > >>> Oracle > > > >>> > > > > > > database dependencies: > > > >>> > > > > > > > > > >>> > > > > > > - > > > >>> > > > > https://github.com/apache/incubator-kie-kogito-runtimes/pull/3369 > > > >>> > > > > > > - > > > >>> https://github.com/apache/incubator-kie-kogito-apps/pull/1968 > > > >>> > > > > > > > > > >>> > > > > > > On Fri, Jan 12, 2024 at 6:36 AM Francisco Javier Tirado > > > Sarti > > > >>> > > > > > > <ftira...@redhat.com> wrote: > > > >>> > > > > > > > > > > >>> > > > > > > > Hi Alex, > > > >>> > > > > > > > Thanks for the info.k > > > >>> > > > > > > > Quarkus-oracle-jdbc dependency includes ojdbc11 and > > > ori18n, > > > >>> > > which I'm > > > >>> > > > > > not > > > >>> > > > > > > > sure are affected by this policy. > > > >>> > > > > > > > However, kogito-test-utils includes oracle-xe, which > > > seems > > > >>> to be > > > >>> > > > > > affected, > > > >>> > > > > > > > so I'm wondering which should be the approach? > Probably > > > >>> removing > > > >>> > > the > > > >>> > > > > > driver > > > >>> > > > > > > > and the test associated with it that will start > failing > > > >>> should be > > > >>> > > > > > enough. > > > >>> > > > > > > > Problem is that we will leave some functionality not > > > >>> tested but I > > > >>> > > > > guess > > > >>> > > > > > > > that's ok, since, after > > > >>> > > > > > > > > > > >>> https://github.com/apache/incubator-kie-kogito-apps/pull/1941, > > > >>> > > > > oracle > > > >>> > > > > > addon > > > >>> > > > > > > > delegates to generic jpa addon. > > > >>> > > > > > > > > > > >>> > > > > > > > On Thu, Jan 11, 2024 at 11:17 PM Alex Porcelli < > > > >>> a...@porcelli.me > > > >>> > > > > > > >>> > > > > > wrote: > > > >>> > > > > > > > > > > >>> > > > > > > > > Based on this thread [1] and this legal issue > > feedback > > > >>> [2] - > > > >>> > > remove > > > >>> > > > > > > > > Oracle driver is required to comply with Apache > > > >>> Foundation > > > >>> > > > > > > > > requirements. > > > >>> > > > > > > > > > > > >>> > > > > > > > > I created the issue [3] and this is a blocker for > > > >>> release. I'm > > > >>> > > > > > > > > planning to work on this myself. > > > >>> > > > > > > > > > > > >>> > > > > > > > > [1] > > > >>> > > > > > > > https://lists.apache.org/thread/1t29vxrnhfxk63j4n6do6g8jhl6l858j > > > >>> > > > > > > > > [2] > https://issues.apache.org/jira/browse/LEGAL-663 > > > >>> > > > > > > > > [3] > > > >>> https://github.com/apache/incubator-kie-issues/issues/817 > > > >>> > > > > > > > > > > > >>> > > > > > > > > > > > >>> > > > > > > > >>> > --------------------------------------------------------------------- > > > >>> > > > > > > > > To unsubscribe, e-mail: > > dev-unsubscr...@kie.apache.org > > > >>> > > > > > > > > For additional commands, e-mail: > > > dev-h...@kie.apache.org > > > >>> > > > > > > > > > > > >>> > > > > > > > > > > > >>> > > > > > > > > > >>> > > > > > > > > > >>> > > > > > --------------------------------------------------------------------- > > > >>> > > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > >>> > > > > > > For additional commands, e-mail: > dev-h...@kie.apache.org > > > >>> > > > > > > > > > >>> > > > > > > > > >>> > > > > > > > > >>> > > > > > -- > > > >>> > > > > > Saludos, Enrique González Martínez :) > > > >>> > > > > > > > > >>> > > > > > > > > >>> > --------------------------------------------------------------------- > > > >>> > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > >>> > > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > >>> > > > > > > > > >>> > > > > > > > > >>> > > > > > > > >>> > > > > > >>> > > > > > --------------------------------------------------------------------- > > > >>> > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > >>> > > For additional commands, e-mail: dev-h...@kie.apache.org > > > >>> > > > > > >>> > > > > > >>> > > > >>> > --------------------------------------------------------------------- > > > >>> To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > >>> For additional commands, e-mail: dev-h...@kie.apache.org > > > >>> > > > >>> > > > > > >
