Skip to main content
Breadcrumbs
Date: 2024-06-11
Present:
- Ed Burns (MSFT)
- Arjan Tijms (OmniFish)
- Jared Anderson (IBM)
- Kyle Aure (IBM)
- Emily Jiang (IBM)
- Jim Krueger (IBM)
- Riva Philip (IBM)
- Anand NK (IBM)
- Nathan Rauh (IBM)
- Thomas Watson (IBM)
- Adam Yoho (IBM)
- James Perkins (Red Hat)
- John Clingan (Red Hat)
- Scott Marlow (Red Hat)
- Brian Stansberry (Red Hat)
- Ed Bratt (Oracle)
- Lukas Jungmann (Oracle)
- Dmitry Kornilov (Oracle)
- Jan Westerkamp (iJUG)
- Petr Aubrecht (Payara)
- Werner Keil (Ambassador/Committer)
Agenda and Minutes
Top-of-mind from Ed and Arjan
- Security Spec going to ballot?
- Arjan asserts the mentor need to take actions to start the ballot
- ACTION: Discuss at Specification committee tomorrow.
- Platform TCK
- What happened at the call last week?
- Top priority for the Platform TCK is to understand how much time is required to complete the refactoring.
- Still not in a position to give a date for a date.
- Plan is to build a way to scale up and parallelize the work.
- There was no plan all along for a plan B to go back to the previous way of doing things.
- Status of “have a meeting every week from now until EE 11 is done”?
- Updated the calendar for links to these meetings.
- ACTION: Scott Marlow to update the calendar of record.
- In the meantime, the meetings will be at 11am EDT Wednesdays at the same Zoom number as usual:
Join Zoom Meeting
- https://eclipse.zoom.us/j/96059419718
- Emily offered some suggestions on the way forward, in terms of project management and prioritization.
- Let’s keep the focus on what we need to know to get clarity on the schedule.
- The marketing concerns will adjust accordingly, and push back if necessary.
- Arjan suggested more transparency regarding the remaining TCKs that need to be completed. Perhaps a Google sheet. This would give us a better ability to understand the size of the remaining work.
- Tanja
- Can the TCK team enumerate the set of skills the individuals need to have in order to be productive?
- EE component TCK team members.
- Include this in the notification of the schedule slippage to the Steering Committee.
- Ed and Arjan to go to Steering committee next week with Scott Marlow’s estimation worksheet and the outer bound slippage estimate, currently pegged at one quarter.
- Steering committee
- What happened at the call last week?
- They asked for what date is feasible.
- Arjan reported that the work is done for specs, APIs, and implementations.
- Summer holidays were discussed, regarding the schedule slippage.
- Actions for 2024-03-29 and M3
- Ed, Ivar, Arjan
- M3 release
- Lingering, in staging
- Waiting: servlet and pages
- Will be released shortly after
- Pages and Servlet to hit maven central.
- Data 1.0 is done on schedule.
- Data impl for GlassFish 8.0-JDK17.
Previous wave
- Wave 5 release review initiated by 2024-03-29
- Spec projects
- **Jakarta Authorization **
- Ballot passed and API available in Maven Central
- CCR approved and closed \
- Jakarta Persistence*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- Jakarta Validation*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- Jakarta WebSocket*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- Jakarta Servlet*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- (some final small steps remaining)
- Jakarta Pages*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- (some final small steps remaining)
- Service releases
- XML Web Services
- Tag Libraries
- Need to update dependencies.
- Actions for 2024-03-29 and M3
- Ed, Ivar, Arjan
- M3 release
- Lingering, in staging
- Waiting: servlet and pages
- Will be released shortly after
Current wave
- Wave 6, 7 specs release review initiated by 2024-04-27
- Spec projects
- Jakarta Faces*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- Speedbump
- Ticket filed. See JEA-378 for link to GitLab ticket.
- Jakarta Data* (release review initiated by 2024-05-31)
- Ballot passed and API available in Maven Central
- CCR approved and closed
- ACTION: Ed needs to update the platform release plan to refer to version: 1.0.1.
- Jakarta Concurrency*
- Ballot passed and API available in Maven Central
- CCR approved and closed
- **We need the TCK 3.1.1 service release with high urgency **
- ACTION: Ed needs to update the platform release plan to refer to the service release to depend on 3.1.1.
- Kyle asked: do we really need to have a different version for the service release of the TCK as for the spec?
- Nathan observed other specs do this differently. There is not a uniform way to do this.
- Jakarta Authentication*
- Ballot started. Running till June 19th
- 7 +1s already
- Jakarta Security*
Risks
- Finish the platform / profile TCKs
- When will the refactoring be done?
- Iteration 1: when an impl can run and pass the tests
- Not an option to stall the refactoring mid-way
- When can we ask for a date? Or a date for a date?
- Soon at at point where we can onboard more people from the spec project teams to help with the work
- Open ReWrite recipes to refactor
- Platform TCK call last week.
- Jan suggested completing the Core Profile TCK first for shorter feedback loop. Volunteers wanted.
- Implementation that passes everything
- GlassFish close
- Jakarta Concurrency
- Passes the not yet released TCK 3.1.1
- Jakarta Data
- Scott Stark’s Hibernate Runner for GlassFish
- Based on annotation processor and Hibernate libs copied to /lib of a GlassFish installation
- **Update of spec text in platform **
- Spec integration text added to the platform / profile specs?
- May be platform / profile specific tests needed to be added
- Action: Now that the components are done, it is time to focus on the Platform and profile spec updates and tests.
- Action for all: Read through the changes made since EE 10
- E.g.
- platform change for Persistence?
- CDI Integration
- **Questions for the TCK Project: **
- Have the CDI tests been moved from the CDI TCK to the Platform TCK/Web Profile TCK/Core Profile TCK?
- Will the Persistence tests run on an actual Server/container?
- Persistence TCK was passed running EclipseLink and Hibernate in Java SE mode. No server of any kind (no GlassFish, no WildFly, etc) was ever even started to pass these tests.
- Draft PR that runs SE Mode persistence tests from within GlassFish (passed mostly, except for a specific SE test that tests that Jakarta Validation is NOT there)
- TCK needs to be updated to support “platform mode”, but is this even possible without a re-release of the TCK???
Back to the top
