Skip to main content

Notice: this Wiki will be going read only early in 2024 and edits will no longer be possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.

Jump to: navigation, search

Difference between revisions of "Eclipse/PMC"

(→‎Meeting Minutes)
(93 intermediate revisions by 6 users not shown)
Line 1: Line 1:
 
= Documents =
 
= Documents =
−
 
 
Some documents written and/or used by the PMC:
 
Some documents written and/or used by the PMC:
  
Line 11: Line 10:
  
 
= Meeting Minutes =
 
= Meeting Minutes =
 +
'''December 12, 2017''' - McQ, Alex, Lars, Dani, Tom
 +
* P2 capabilities work breaking Oomph and CBI ({{bug|528408}})
 +
** Alex: Hidden bug due to adopters using internals.
 +
** Alex: Code refactoring in p2 pushed to reduce the chance for such breakage.
 +
** Alex: Oomph fixed its codebase to properly check instances so it works with p2 both prior and after adding capabilities.
 +
** McQ: What do we do with p2?
 +
** Alex: Open up the project as much as we can and recruit new committers by being really welcoming to people.
  
−
'''Jan 17, 2017''' - Dani, Sergey, Alex, Lars, Martin, McQ
+
-> Conclusion to not respin M4 due to this bug as it's internal API which projects should not have used.
−
* Sergey: '''Merging platform.resources with platform.ui'''
+
−
** Seems that more contributions to platform.resources are made by platform.ui committers, than others. Only 2 platform.resources committers seem active (and those two are actually platform.ui committers as well)
+
−
*** Opening up to Platform UI would encourage more people to actually contribute to resources. Strive for shared access on components that are interrelated
+
−
** Dani: Merging into platform.ui would mean making Dani and Lars the leaders for signoff - that causes too much burden
+
−
** Sergey: Platform/Team has actually more synergy with Platform/UI than resources
+
−
** Alex: For gcc for example, there is a notion of "maintainers" per git repo for signoffs - that's independent of leadership or commit rights
+
−
*** McQ: Not sure how adding another role would simplify things ?
+
−
** Sergey: The problem for signoffs is not is much the volume of commits, but the volume of failing tests
+
−
** Alex: The manpower of Platform Resources is so low by now that even {{bug|509412}} was not being filed ... existing committers do not really seem to "own" the project any more, this is an indication for merging
+
−
** McQ: Propose applying the recent simplification of gaining commit rights (for "known" committers)
+
−
* Lars: '''Removing javax.xml'''
+
−
** Was introduced only for very old JRE's , it is part of the JRE now
+
−
** Martin: Might cause a uses constraint issue since not exporting the package with version any more, see http://njbartlett.name/2011/09/02/uses-constraints.html
+
−
** Lars: With javax.annotation, that was solved with smart reexporting
+
−
** McQ: OK with removal when we can confirm we're not breaking things
+
−
** Dani: Whoever requested removal should double-check whether the API is the same.
+
−
**'''AI Lars''' send a note to Tom Watson as the submitter.
+
−
* Dani: '''Project Updates'''
+
−
** Deadlines upcoming for CQs and APIs - see Endgame Plan, please provide feedback
+
−
* McQ: '''UI Testing and Jubula'''
+
  
−
<hr/>
+
* Project plan update
−
'''Jan 10, 2017''' - Dani, Lars, Martin, Alex, Sergey, McQ - Special Guest: Alex Schladebeck
+
** Dani: No additional change came from anyone else in the PMC so update of executions environments and adding Java 9 are the only thing due.
−
* Dani: '''Platform Bits on Maven Central''' - Now available thanks to Stephan Herrmann
+
* Next meeting on January 9
−
* Dani: '''Board Committer Rep Elections''' opening
+
−
* Dani: After David Williams' move, Wayne is looking for a Planning Council Chair; Fred Gurr has taken over Simrel Releng
+
−
* Dani: objenesis (Mockito / Easymock dependency) and Java9 - will wait for how upstream is planning to address issues
+
−
* Lars: '''Equinox move to Platform''' - no updates - '''AI Dani''' talk to Tom again
+
−
* McQ / AlexS: '''Jubula'''
+
−
** In the past, with lots of full-time Platform committers, sniff tests just happened automatically giving great confidence. Now, with more part-time work happening, there is more need for end-to-end workflow test automation to give quality confidence
+
−
** Martin: On top of low-level JUnit tests, want at least very few highlevel end-to-end test to avoid "Eclipse looking silly". Need UI tests that are really stable (avoid random failures)
+
−
** Dani: On EGit SWTBot tests, too many libraries to install, how to run on Gerrit HIPP slaves
+
−
** AlexS: Original Jubula goal was to allow non-programmers write tests. Beginning 2015, the client API (for controlling the SUT) was separated from the front-end, was separated from the ITE. This allows writing UI tests programmatically, and avoids the need for a database.
+
−
*** For identifying components/actions to write tests, officially an object mapping needs to be done in the front-end (ITE, integrated test environment). But there may be a shortcut available for getting object IDs
+
−
*** Compared to SWTBot, Jubula is 90% blackbox (code highlevel actions rather than things like calling setText())
+
−
*** UI Tests typically take longer than unittests by definition (due to setup, teardown, ...), so might not be applicable for Gerrit triggers
+
−
*** Dani: Would like to run a first test on CentOS first, as part of the integration tests
+
−
*** Lars: Robustness of the tests? - SWT on Mac is problematic, but Linux and Windows should be OK.
+
−
*** Martin: Dependencies? - Around 5MB Libs, plus the agent needs to be installed (around 100MB). Once that is there, tests are stand-alone on a laptop. There might also be an "embedded agent" but not sure ('''AI AlexS to check''')
+
−
*** Getting a first test up and running ? - Maybe around 1 week, depends on what to do ?
+
−
**** Lars: Would like something for quick access. Martin would something for "fresh download" (fresh config area + workspace) maybe combined with quick access.
+
−
*** '''AI AlexS''' send slides
+
−
*** '''AI Dani + Lars''' propose scenarios. Check possible owners of the initiative on Platform/UI side
+
−
*** '''AI AlexS''' check possible PlatformUI Contributors from Jubula team to get started
+
  
−
== 2016 Minutes ==
+
'''December 05, 2017''' - McQ, Alex, Lars, Dani, Tom
 +
* Vacation: Dani will be in vacation end of week, so other PMC member need to cover for him if something comes up for M4.
  
−
'''Dec 20, 2016''' - Dani, Lars, Alex, Martin, Sergey
+
* Discussion about adding EGit to the SDK? ({{bug|528004}})
−
* Dani: '''{{bug|499768}} Oxygen Plan Update''' - Ubuntu 14.04 removed as a reference platform. CQ deadline added.
+
** Dani using a shared installed to manage the EGit plug-in.
−
* Dani: '''SWTBot Status'''
+
** McQ explains that he thinks EGit is required for a full package but does not feel strong about.
−
** Worked well until Eclipse.org HIPPs were updated - now, with icewm there are focus issues, causing random recognition errors. There is {{bug|506995}} open to provide a better alternative.
+
** Alex: It depends what "SDK" means: SDK to develop plug-ins or SDK to develop Eclipse Platform. We agreed on the former.
−
** CDT has been affected as per {{bug|509248}} - Dani also in contact with egit
+
** Dani, Alex are against adding it because it adds maintaince burden and it is unclear which of the optional packages should be installed.
−
** Until then, we have to wait to look into SWTBot
+
** Tom, Lars have no strong opinions
−
* Sergey: '''Neon.2''' - GA tomorrow, branch is open already
+
** All agree that managing to install this is an easy task
−
* Alex: '''Webkit 2 Update''' - Javascript integration is working; plugins running out-of-process is a huge benefit
+
−
** BrowserFunctions not working yet (callbacks like "onClick" in Browser Widgets) - OK to still make Webkit2 the default ?
+
−
** Should test Javadoc hovers; Help system navigation to Javadoc; Intro pages; expect to be done by M6, probably even M5
+
−
** Martin doesn't want to block downstream adopters; enable fallback by environment variable and announce on cross-project
+
−
** Alex will look at enabling at least onClick(), that's perhaps the 80% case, and do some more testing before switching
+
−
* Dani: '''Eclipse.org Infrastructure Updates'''
+
−
** {{bug|499399}} being discussed on the board level now; especially VNC on the HIPPs, and {{bug|508658}} repo.eclipse.org
+
−
** VNC may be non-trivial to get fixed since SLES is almost entirely targeting back-end services
+
−
** Would need a reasonably recent UI stack - anything like CentOS, Fedora, Mandriva
+
−
** '''Goal would be having the test "close to what end users use"'''
+
−
** Enterprise users tend to use "older" distributions (CentOS, Ubuntu LTS)
+
−
*** Statistics like https://www.linuxcounter.net/statistics/distributions might be misleading since enterprise users typically don't count there
+
−
*** On the other hand, enterprise customers typically also don't use the very latest Eclipse; and have their own support channels to report issues. So for active development, newer distros may be desirable.
+
−
* Will cancel the Dec.27 and Jan.3 PMC calls
+
  
−
<hr/>
+
-> Conclusing to not include EGit in the SDK
−
'''Dec 13, 2016''' - McQ, Lars, Alex, Martin, Dani, Sergey
+
−
* McQ: '''Rebuild for M4''' due to 3 bugs, e.g. {{bug|508717}} - Could testing be improved ?
+
−
** Probably no 1:1 relation, but need a modern story for getting confidence in what we provide to the Community. A massive 7-hours API testing as done today doesn't hurt but isn't the solution
+
−
*** Lars: Test Suite is currently inactive for Quick Access - activation is on the roadmap
+
−
*** Would like some end-to-end scenario testing from the UI
+
−
** Platform UI would like to start using SWTbot - most lightweight - at least some initial tests as examples
+
−
*** '''AI Dani''' talk to Mickael to understand what was blocking from adding it
+
−
** UI tests are good for stable things, problematic for work-in-progress UIs
+
−
** GUIdancer / Jubula looked very impressive - but seems to require a database, so hard to run locally
+
−
** Martin: Reach out to projects on top of the Platform and ask them run on an I-build, at least occasionally?
+
−
*** Sergey thinks that we need to run tests ourselves.
+
−
*** Alex: Linuxtools could easily run tests against I-Builds; Heavy usage of SWTbot in place but it's unreliable - tests are very fragile, even smallest changes break the tests easily, tests were rewritten very often, making this quite expensive
+
−
*** Would only verify SuSE 12 if running on HIPP, but that doesn't tell much about other Platforms
+
−
** McQ would like a "sanity-level check" on commits, to get some level of confidence on the builds ... a quick test which actually checks how people are using Eclipse today
+
−
** Alex thinks that the current test suites are not in a good shape: many tests fail when run locally, even though good on Foundation hardware - cleaning this up is a lot of effort
+
−
** McQ: Interested in the "top 10 ways of working with Eclipse" - could we test the top 10 scenarios ? Can we even find them ... and make those small "top 10" set stable so it can run anywhere ?
+
−
* Lars: '''Push Maven Artifacts to Central'''
+
−
** What should be the group ID ?
+
−
*** Dani thinks we cannot use org.eclipse (too generic) - maybe o.e.platform / o.e.jdt / o.e.pde (Q: Also split out o.e.swt / o.e.ui ?)
+
−
*** Martin wouldn't split out o.e.ui -- would prefer more grouped together in Platform
+
−
*** '''AGREEMENT''' to group by project name: o.e.platform / o.e.jdt / o.e.pde
+
−
** Maven has only 3-part-versions : what to do with the qualifier ?
+
−
*** Alex: Sometime we don't bump the version for Beta releases - cannot override
+
−
**** Dani: releasing with a new qualifier but without a new version is a bug and should be found by reports and fixed
+
−
*** Dani: Using n.n.n-BUILDID is perhaps an issue since it's not treated as a release any more (but a snapshot)
+
−
*** '''AGREEMENT''' to use 3-part versions.
+
−
*** '''AGREEMENT''' to Stefan's proposal only releasing Neon.2 for now.
+
−
*** '''AI Dani''' respond to Stefan.
+
  
−
<hr/>
+
* Impact of cross discussion from David about API changes in which David implied that the platform cannot do API breakages
−
'''Dec 6, 2016''' - Dani, Alex, Martin
+
** Dani clarified with David on the cross-mailing list, no changes for us for the process, as we only remove announced. See https://dev.eclipse.org/mhonarc/lists/cross-project-issues-dev/msg15032.html
−
* McQ talked to Tom Watson about Equinox moving to Platform - Tom wants to discuss with the RT PMC
+
−
* Alex: '''Update on Wayland'''
+
−
** Fedora 25 shipping with Wayland by default
+
−
** Wayland back-end for SWT is a patch in progress, some events not delivered, others incorrectly
+
−
** Hoping to be in good shape by M5.
+
−
** Until then, Gnome can be started on X11 with additional packages installed.
+
−
** If Wayland is needed, the XWayland compatibility layer does work although slow (some drawing can be seen)
+
−
** Some rewrite in the Launcher will be needed due to the security model - each window is isolated
+
−
* Dani: '''How we test milestones''' - people putting in features sometimes don't seem to participate in the milestone week
+
−
** McQ: Agrees that a good deal of manual testing is needed to cover flows that developers actually use
+
−
** Alex: If milestone week overlaps with other duties (outside Platform) those other duties often take precedence
+
−
** Test automation doesn't help for new features that add UI
+
−
** Sergey distributes I-Builds so regressions are often found early - but there's no firm data on quality. Alex also distributes / uses I-Builds.
+
−
** McQ thinks that some confidence in quality is needed, before I-builds are given out
+
−
** Dani: Dogfooding helps a lot more than test plans.
+
−
** ''To be discussed next week when there are more people on the call''
+
−
* 4.6.2 looking good; M4 test pass today for releasing on Friday
+
−
* Oxygen plan hasn't seen comments, so will only put Java 9 updates in there
+
  
−
<hr/>
+
* URL handler registration of Eclipse with the OS
−
'''Nov 29, 2016''' - Dani, McQ, Sergey, Alex, Lars, Martin
+
** SAP plans to contribute a way to register Eclipse to certain URLs. If the user clicks on such a URL, Eclipse would open and be able to show the file. This could be useful for installation of new items in the marketplacen client https://bugs.eclipse.org/bugs/show_bug.cgi?id=527579. VSCode allows this already vscode:extension/eamodio.gitlens.
−
* Sergey: '''Understaffed Projects Announcement''' - should it be sent to a wider audience?
+
** Redhat had similar requirements in the past
−
** Was sent to eclipse-dev with cc: eclipse-pmc
+
** Platform team expects that SAP will also maintain this contribution and take responsibility of it for future releases
−
** Dani suggest checking the mailing list archives or via RSS eg [https://dev.eclipse.org/mailman/listinfo/eclipse-dev eclipse-dev mailing list home]
+
** Dani (after meeting): I think extending the current file based mechanism is OK, but we won't register any (including URL) handler
−
** '''AI Dani''': send Sergey some links
+
−
* Martin: Is it clear enough what contributors could do if they feel not being heard?
+
−
** There is a forum and developer mailing lists, so for the Eclipse project we are probably good
+
−
** Dani: The README.txt on top of the repo looks like a very good place for such kind of information
+
−
** '''AI Martin''' review a couple READMEs and then make suggestions on the eclipse-pmc mailing list
+
−
* Dani: RC4 - no build currently planned will push RC3. There's also Oxygen M4 next week.
+
−
** Need to move out some planned bugzilla's targeted M4 (platform.ui: Lars to help, core: Sergey)
+
−
** JDT UI is tight since Noopur is on vacation and Markus is very busy with Java 9 and HiDPI work
+
−
* New process in place for the IP Log, IP log for Neon Update 2 (4.6.2) got approved today; Review is scheduled and ends on Dec.21
+
−
* JDT Language Server: Still some blockers on the IP Work (hackathon contributions)
+
−
* '''AI All''' Plan Update for Oxygen M4 {{bug|499768}}; Dani will update the Java 9 info
+
−
* Lars: '''Approvals for Equinox'''
+
−
** RT PMC don't have meetings or a process in place for approvals
+
−
** They think that the Eclipse PMC is in a better position to approve Equinox
+
−
** But why is Equinox on the RT PMC after all, does that still make sense? After all builds are done by Eclipse.
+
−
** '''AI Dani''' discuss with Tom
+
  
−
<hr/>
+
-> Conclusion to work on this enhancement together with SAP, Alex plans to dedicate some time for discussing the technical solution. SAP should open a new bug and link all related bugs to this top-level bug so that the technical details can be discussed
−
'''Nov 22, 2016''' - Dani, Martin, Sergey, Alex, McQ (Regrets: Lars - travelling)
+
−
* McQ: Simplify adding committers to understaffed projects
+
−
** '''AI Sergey''' final Polish on Lars' proposal + suggestions as discussed
+
−
* Martin: {{bug|498116}} download servers are hosed every Tuesday
+
−
** PMC agrees to trying a fix for Neon.2 - this kind of issues is important (Android also ran into this kind of issue 2 years ago)
+
−
** Dani: we can encourage them to fix it, but EPP and Equinox are not governed by our PMC
+
−
* Dani: '''Neon.3 Planning'''
+
−
* Dani: '''FEEP Discussions''' on the [https://dev.eclipse.org/mailman/listinfo/eclipse.org-feep-stakeholders eclipse.org-feep-stakeholders] mailing list
+
−
** Voting will be more open this time
+
−
* Dani: '''IP Log for Neon.2''' - Endgame looking good, no planned fixes known at this time, but if there's a fix for {{bug|498116}}, we will consume it
+
  
−
<hr/>
+
* Discussion of building natives
−
'''Nov 15, 2016''' - Dani, Sergey, Alex, Martin, Lars, McQ
+
** moved building of SWT binaries for the 3 primary platforms to the Eclipse Foundation
−
* Dani: Neon.2 Release Review went out, IP Log will come early Dec.
+
** unfortunately after that the Mac went down - fixed now
−
* Dani: RC2 build tomorrow + test pass on Wed - last build before PMC approval is required
+
** M4 SWT natives will be built on the Eclipse Foundation hardware
−
* Dani: Bugzilla 4.6.3 milestone created - currently too many bugs targeted 4.6.2, suggest moving out except if there's a patch attached
+
** Launcher still to be moved
−
* Alex: '''Lucene''' - Using Lucene 6 for Platform.UI Indexing, prebuilt indexes of doc bundles need updating
+
−
** Prebuilt indexes of older versions will not be used, the impact is just a slowdown on first startup of help
+
−
** Old-version prebuilt indexes will trigger an entry in the errorlog, so projects will know
+
−
** Q: How much should we push for having a single Lucene in the Release Train? (In Mars we had two)
+
−
*** Problem: Moving to Lucene 6 is real porting effort, and it is not possible to be backward compatible
+
−
*** EGit or XText, who want to support multiple Eclipse versions, may be allowed to bind against other Lucene by exception; but Planning Council message should be to favor the latest by default
+
−
** Martin: Help Search Result Quality -- Was often critisized due to bad analyzers -- has it improved ?
+
−
*** Alex: Lucene 6 has apparently improved things, but team has not explicitly tested these cases
+
−
* All: '''Understaffed Projects Discussion'''
+
−
** Dani: Would like to send the message in a positive way: As a tool for interested parties to gain access to the code, and a tool for understaffed projects to bring in committers
+
−
** Alex: Would like to send the message, that Eclipse is open to contributions
+
−
*** McQ: Yes, but not at the cost of making existing committers feel bad
+
−
*** '''Lars''' to draft a new note (next week)
+
−
* Lars: '''Mylyn Notification API'''
+
−
** Still on Dani's list of items
+
  
−
<hr/>
+
'''November 28, 2017''' - McQ, Alex, Lars, Dani
−
'''Nov 8, 2016''' - McQ, Dani, Sergey, Lars, Dani, Alex
+
* Alex: Release Change announcement
−
* Dani: '''Update on N-Builds'''
+
** Agreed that we won't mention the ramp-down process.
−
** Releng team will drop N-builds - will also reduce work on the Releng team
+
** Agreed to add API removal details.
−
** Retention policies will need to be updated on I-Builds (keep weekly I-Builds)
+
** Still work in progress.
−
** Provide a Hudson job which can mark an I-Build as unstable in case it isn't good
+
* Dani: API Removal process for the new release model
−
* Dani: '''Neon.2'''
+
** Agreed to allow removal announcement in ever of the 4 releases. As a consequence every release can have API removals.
−
** Feature Freeze today - Release Review just listing the N&N
+
** https://wiki.eclipse.org/Eclipse/API_Central/Deprecation_Policy already reflects this.
−
** There was some new API, but it was backported mostly for bug fixes ... nothing relevant for the N&N
+
* Dani: Meeting notes
−
** Agreement that release reviews should be minimal effort - there was never community feedback on the reviews.
+
** McQ suggested that who joins last has to do it. Dani: not so good since it won't contain the notes for the discussions that happened before.
−
* Lars: '''Code Reviews on Understaffed Projects'''
+
** Will do it based on rotation and ability to actually write down the notes.
−
** Make it easier for known, good people to get access to the code base for reviewing patches
+
** Dani to do it for this call.
−
** We'd still need to know the people - don't offer commit rights to random people
+
* Dani: Photon plan update
−
** Refactorings: Valuable if done by committers, but reviewing patches that are produced by refactorings is too hard
+
** Will add Java 9 as supported VM.
−
** Making someone a committer is preferred if there is some commitment to doing work moving forward
+
** Anything else? If so, send Dani a note or update {{bug|527964}}
−
** Sergey: There's too many unreviewed patches against Platform in some areas - we need more committers.
+
* McQ: Rename Eclipse Project to Eclipse SDK
−
** Proposal: '''Offer a shortcut for committers on one component who express interest in another component.'''
+
** We should only do it when forced by the Foundation and/or Board
−
*** Criteria: Provided contributions in the past, and shows interest moving forward.
+
** We would need a plan before starting such an effort.
−
*** Don't offer commit rights just for one contribution.
+
−
*** Consider merging components if a component has become so small that it's unnecessarily hard to contribute
+
−
**** On a case by case basis, only if we think that merging components would solve more problems than creating new ones.
+
−
** Dani: Who makes the decision on shortcuts - existing committers, or PMC ?
+
−
*** In most cases, committers should have the say - but PMC may interfere if a project looks dysfunctional.
+
−
*** '''AI Alex''' will write up proposed statement
+
  
−
<hr/>
+
'''November 14, 2017''' - Dani, McQ, Alex, Tom
−
'''Nov 1, 2016''' - McQ, Sergey, Lars, Dani, Alex
+
* Dani:
 +
** Presented our new release model to Planning Council. Was welcomed. No objections. But needs discussion what the Release Train will do.
 +
** RC1 went well, RC2 this week.
  
−
We agreed on the following changes:
+
'''November 07, 2017''' - Dani, McQ, Alex, Tom, Lars
 +
* For the 3 months releases, we have have a development freeze after RC1 until RC3. After RC3 master will be opened again for development. Development is possible until RC1, so this will result in a freeze period of approx. 2 weeks.
 +
* For releases versions of the platform we will use semantic versioning as before (4.8, 4.9, 4.10, ..)
 +
* Eclipse SDK will use versions not branding names unless the release train moves also the same cycle.
 +
* Splash will continue show the release year and month
  
−
* Create maintenance branch immediately after RC1 instead of after the release. Use the maintenance branch to produce the upcoming release.
+
'''October 31, 2017''' - Dani, McQ, Martin, Alex,
−
* Eliminate all code freezes on the main branch for RC releases, except for 1-2 days after RC1 before the maintenance branch is created. Freezes to allow test for milestone builds will still be applied (milestone freezes).
+
* Tom: '''Bugzilla Landing Page''' - EMO said both PMCs need to be contacted
−
* We decided that we will only have ‘I’-like builds, using always signing.
+
** Bug IDs, Equinox product and components remain in place, only the landing page will be updated
−
* We decided to revisit the decision to always ship from master, but that's a longer discussion and won't happen soon.
+
* Dani: '''M3 Respin'''
 +
** Issue had been found by a team member, but got lost between different teams
 +
** Resolved and respun very fast, the fix is out now
 +
* Dani: '''Release Cycles and Java Alignment'''
 +
** Planning Council F2F meeting was at EclipseCon Europe; planning call will be on Thursday
 +
** Initially, felt like aligning with Java 18.3 was a good idea, but for now the plan is keeping the Oxygen SR rhythm and releasing Oxygen.3a aligned with Java
 +
** In case the Oracle release should slip, it will be most likely be named 18.4 (whatever month the release ends up in)
 +
* All: '''Platform Release Rhythm'''
 +
** At ECE, many people were surprised about changing release rhythm but the plan was generally appreciated
 +
** Should stick to the communicated Photon plan though and change only afterwards
 +
** Alex suggests promoting a Milestone build from master (properly marked up as such) aligned with the Java 18 release
 +
*** This would promote master as something that can be used
 +
*** After Photon, we will release 4 times from master - there won't be any maintenance builds any more
 +
*** For the release train, there will also be release candidates.
  
−
<hr/>
+
<strike>October 24, 2017 - cancelled due to ECE</strike>
−
'''Oct 18, 2016''' - McQ, Sergey, Martin, Dani, Alex
+
−
* Dani: '''PMC Approval for Sergey co-lead on Resources''' (Done)
+
−
* Alex: '''Lucene Update'''
+
−
** Looking at updating to 6.x in Oxygen - CQs ongoing for Orbit - hoping to make M4
+
−
** Others like Mylyn, Pydev, ... use Lucene 5.x at this time; others consume the Platform
+
−
** There is some potential breaking things but should be fine since multiple Lucene's in parallel can work.
+
−
** If an old pre-built help index can not be read, a new one is generated (Performance hit on the first run)
+
−
** Dani: No issue, will also update Jetty in M4
+
−
* Dani: Would like to open '''eclipse.jdt''' repository and corresponding sub-project to all JDT committers (currently only used for the JDT feature)
+
−
** The project is almost dead with only 2 active committers, since new JDT leads were not voted in
+
−
** Changes to feature will be rare, and Dani agrees to review
+
−
** Allows to put the Oomph setup file into a shared place where all JDT committers have access
+
−
** PROPOSAL AGREED by all.
+
−
* EclipseCon Europe next week - Dani, Lars and Alex will be there - will '''cancel next week's meeting'''
+
  
 
<hr/>
 
<hr/>
−
'''Oct 11, 2016''' - McQ, Dani, Lars, Alex, Sergey
+
'''October 17, 2017''' - Dani, McQ, Martin, Lars, Alex
−
* Dani: '''Neon Update 2 Endgame Plan sent''' - thanks for the feedback
+
* '''Conf.system:''' Fallback to McQ's number is not a good solution (can't give out the moderator code)
−
* Sergey: We should talk about our release process.
+
** Foundation (Denis) is aware of the problem, looking for a solution sooner than later
−
** McQ asked Sergey to start the discussion via e-mail
+
** Skype group call, or zoom.us would work for Dani and McQ
−
* McQ: What do you think about Devoxx and Eclipse Converge
+
* Dani: '''{{bug|526065}} Java 9 to Oxygen.1a beta update issue'''
 +
** Users who had the Java 9 beta installed and "update" to Oxygen.1a are broken
 +
** McQ: Updating from "beta" to "final" is not necessarily supposed to be working (Lars disagrees)
 +
** Not yet sure what's causing the issue .. p2 bug, or different groupID for Java9beta
 +
** Fix the issue if it looks like a real bug or easy to fix ... otherwise just post a message, preferredly on the Marketplace client (''Dani: that part is done already'')
 +
* Lars: '''Release Cycle Discussion'''
 +
** McQ still in favor of making every milestone a release
 +
*** Alex thinks we're not there yet in terms of releng, there's still too many manual steps. Would prefer slower cadence (eg 3 months) until more automation is there (updating pom.xml, version mgmt, ...)
 +
*** With 4 releases per year (like today) but no more maintenance branches/backports, there's potentially less work than today
 +
*** Try starting with a release every 2nd milestone; fix automation; then strive for releasing every milestone.
 +
*** Dani thinks the model shouldn't change in the middle of Photon; look for feedback then move to the new model next year
 +
** McQ: If we switch to the new model now (Oxygen.2), what do we lose?
 +
*** Martin: Any bigger feature that would need to move into a branch? - Dani: Advanced Source Lookup
 +
*** Dani: Also our rules about "no new features" and "PMC Approval" would need to change
 +
*** '''Continue discussion in 2 weeks''' (skip next week due to EclipseCon)
  
 
<hr/>
 
<hr/>
−
'''Oct 4, 2016''' - McQ, Dani, Martin, Lars, Alex, Sergey
+
'''October 10, 2017''' - Dani, McQ, Lars, Alex, Martin
−
* Dani: '''Neon Update 2 Endgame Plan''' - working on it
+
* Dani: '''Oxygen.1a respin'''
−
* Dani: '''Release Review'''
+
** Primary reason for respin was, that people who have existing XText installed would be broken when updating the Platform
−
** In the past, features were not announced - now for 4.6.2, features + release review must be in by RC1
+
** Opportunity to piggy-back JDT critical fixes was welcome
−
** Will do the minimum necessary / required by EMO, for Update Releases.
+
* Dani: '''Equinox Move''' - Review passed without objections, working on implementation now
−
* Lars: '''p2 merge into Equinox status''' -  
+
** Expecting less trouble than with the Platform and JDT merges recently
−
* Sergey: '''Logistics of dealing with Service Releases'''
+
** Foundation want to keep Equinox as a brand, therefore will be a subproject (beside Platform, JDT, PDE)
−
** Managing version increments is very fragile and error-prone
+
** So to increase the committer base, the fast track option will now apply; also, more alignment in terms of PMC.
−
*** Discussion already happened, see {{bug|499164}} -- we'll not increment minor versions on backports even if API is added
+
** Confirmed again that we will ask Tom to join our PMC. Approval can be found in [https://dev.eclipse.org/mhonarc/lists/eclipse-pmc/msg03166.html this thread].
−
*** '''AI Dani''' send a note out to eclipse-dev again
+
* Dani: '''Planning Council - Rolling Releases Update'''
−
** '''New & Noteworthy / Features for Update Releases:'''
+
** Should remove some of the work maintaining 2 streams - simply release from master regularly
−
*** If there's new features, there must be a New&Noteworthy for users (everyone agrees)
+
** In the past, API breakage (migration guide) was allowed only once per year - how to deal with this moving forward
−
*** What about doing two feature releases per year with one service release each ?
+
** Handle with semantic versioning; how to give adopters sufficient time to adjust
−
**** Dani: Not having features in "some" releases would defeat what the Community wanted
+
** There's usually 2 camps for every topic being discussed ... though practically, Platform sets the pace
−
**** Alex thinks that allowing features every 6 months would calm down most demand for new features
+
** Release Naming - New Name every quarter? Versioning scheme? (like 2017.4 or 17.04)
−
**** PMC Review of feature backports worked fine so far
+
** How to simplify processes (also in terms of IP), can the Foundation scale up? Perhaps not more work, but more deadlines
−
**** '''RESOLUTION''' Will not change the process before having it settle a bit
+
*** In terms of the Release Review, only the N&N and the Migration Guide are relevant; need to satisfy small and big consumers
−
***** McQ would love to see a single development stream all the time ...
+
* Martin: '''Reddeer Contribution'''
−
*** Would the Generic Editor be acceptable in a release before June 2017 ?
+
** When we last discussed this, it was an important criterion that any contributor can run the tests easily
−
**** Tentatively yes, it's a separate bundle not affecting other code, it matters to the Community, so why not...
+
** Alex: Looks good, essentially boils down to running "mvn clean verify" in the PDE repo (or right-click > Run as Reddeer from the IDE) - '''AI Alex''' will have a video sent to the PMC list
−
**** Martin: OK, but what about newly introduced API ... would it be declared "provisional" initially ?
+
* Alex: '''API Removal in Context of Rolling Releases'''
−
**** Dani: There's no such thing as "provisional API". Teams who decide on backporting must also decide on API.
+
** Which Rules to apply for allowing removal of API; Allow removal in any release?
−
**** Sergey: CDT has done full feature release every quarter for some time now and hasn't screwed up API...
+
*** McQ: The real question is, how long it takes consumers to adopt to removal
−
**** Martin,Lars,Sergey: What about marking "fresh" API as "@noreference under construction" until it's hardened...
+
*** Dani: Is more conservative regarding deletion - the value removing something is often smaller than the ripple it causes
−
***** McQ: Then declare APIs as official only once a year ?
+
*** McQ: If there is code in our codebase that significantly slows us down, it should be remove even if API
−
***** Wouldn't want to tie to a yearly cycle again ... but then we're back to the Version Numbering discussion...
+
** Martin: With more releases, there's less push to upgrade ; but will consumers know they are broken? - Semantic versioning should address that, but who really uses upper version ranges ... they have caused more pain than benefit in the past
−
***** '''AI All''' continue discussion on mailing list or next week.
+
** Dani: In the past, with our 3-year deprecation plan, Major versions have not been updated even on announced removals...
−
 
+
** McQ: If we are earnest about faster velocity, we '''have to''' fully adopt semantic versioning practices. If we as producers have done what we could, our job is done - consumers who don't use upper version ranges are on their own
  
 
<hr/>
 
<hr/>
−
'''Sep 27, 2016''' - McQ, Dani, Lars, Alex + Sergey Prigogin (guest)
+
'''October 3, 2017''' - Dani, McQ, Alex, Martin
−
* Discussion about PMC upstaffing
+
* '''Conf system woes''' - Dani couldn't dial in, Denis restarted the bridge leaving Martin and Alex in a Zombie bridge
 +
** '''ACTION McQ''' will send out conf.details using IBM's bridge
 +
* {{bug|520720}} Mickael's question on Performance improvement
 +
** Dani: There is already an asynchronous RequestLayout API for clients, and APIs already suggest clients use the async one
 +
*** Therefore, changing the synchronous API to work asynchronously doesn't look right.
 +
** Alex: We seem to be mixing two aspects here -- agree regarding RequestLayout concerns, but why should the ProgressMonitor one be a problem?
 +
*** Dani: Agree that the ProgressMonitor approach makes sense, but a very careful review is needed - existing code already subclasses ProgressMonitor, need to make sure that expected methods are still called
 +
*** '''ACTION Dani''' will do a careful review of the ProgressMonitor part and deny the generic RequestLayout part
 +
* Dani: '''Equinox Restructuring''' request for the move is in, will work with Tom and Foundation forward
 +
* Dani: '''Oxygen.1a''' - RC2 is in for JUnit5 and Java9
 +
** Respin of Oxygen.1 for {{bug|520176}} Mac 10.13 High Sierra problem? - Not currently planned, there will be 1a instead
 +
** A feature patch is another option, providing patched SWT; and, there's the workarounds
 +
* Dani: '''Rolling Release''' - 2 topics: rolling release, and alignment with Java Update Releases
 +
** Some projects on the train have alignment goals with other projects outside Java; will discuss again in the Planning Council
 +
* Alex: '''{{bug|522733}} webkit crash''' - A workaround is available, in contact with webkit devs for a complete fix to keep Photon usable
 +
** Workaround would introduce a small memory leak, but only on the first instance of webkit created
  
 
<hr/>
 
<hr/>
−
'''Sep 20, 2016''' - Martin, McQ, Dani, Lars, Alex
 
−
* Resources Leadership - '''AI Dani''' reach out to Szymon
 
−
* PMC upstaffing: Focus on large installs; UI Freeze monitor, code cleanup, solid solutions, focus on code reviews
 
−
* Dani: '''Java 9 Moved to July 2017''' - AI Dani talk to the Planning Council, proposing an "Update 1" in July
 
−
** Better for consumers, available via "just update", though little bit extra work (CQ + Release Review)
 
−
** Schedule looks still aggressive when looking at specifications; need to keep the team motivated
 
−
* Dani: '''SelectionListener SWT Change''' - reverted for M2 since more clients were affected than expected
 
−
** Proposed change turned out to be source-incompatible - will look at a different approach
 
  
−
<hr/>
+
'''September 26, 2017''' - Dani, McQ, Alex, Martin
−
'''Sep 13, 2016''' - Dani, McQ, Lars
+
* {{bug|520176}} '''MacOSX 10.13 (High Sierra) Menus Disabled'''
−
* Upstaffing the PMC
+
** Only non-English Locales are affected
−
* Build failures due to an SWT Change
+
** For 4.7.1a and Photon, a fix is in
 +
** Foundation [https://www.eclipse.org/org/press-release/20170925criticalbug.php Press Release (Newsletter)] is out with information how to work around
 +
** Tweets are around as well -- keep spreading the message
 +
* '''Equinox Move Review''' - progressing
 +
* Alex: {{bug|522733}} '''GTK Crash on Close preventing IDE Restart''' - can it go into 4.7.1a ?
 +
** Seen with Fedora 27 (Webkitgtk 2.18)
 +
** Foundation wanted to keep Java9 stuff separate, so pushing it there might not even help
 +
** The issue is serious (hard crash preventing restart of the IDE), though fortunately not many people have that particular webkitgtk version yet
 +
** Investigate details and possible workarounds, then consider our options -- maybe a Foundation Newsletter similar to the OSX 10.13 issue
  
 
<hr/>
 
<hr/>
−
'''Sep 6, 2016''' - Dani, Lars, Alex, Martin
+
'''September 19, 2017''' - McQ, Alex, Martin, Lars
−
* Dani: Thanks for approving Noopur
+
* '''Async APIs request from the eclipse-pmc list''' (not discussed today)
−
* Dani: Java Language Server Proposal - at the Foundation, will be called "JDT Language Server"
+
* Lars: '''Java 6-month release cycles'''
−
* Dani: RC4 - Fix for an Equinox p2 issue, only known issue being fixed, looking good
+
** Alex likes the 6 month idea - less backporting
−
* Dani: Looking at the generic code editor, would prefer moving some things to M3 (talking to Sopot and Mickael)
+
** McQ would ideally like full releases every 3 months (except not releasing anything unfinished)
−
* LTS: Market for paid fixes on old versions not quite as good as expected
+
** Dani: If Java really delivers every 6 months (will they?), Eclipse could consider rolling releases every milestone. There is a doc from the Planning Council (Mélanie Bats) in that regard.
−
* Release Reviews: When is a feature considered an "important feature" ?
+
** Dani: Support for the public JDKs will be limited to the current release only (6mo)
−
** Important Features require a full release review thus lots of effort, avoided for Neon.1
+
** Risk: Oracle Java may be working on big features "in the background", if there is very little time adopting new features (current license doesn't allow releasing stuff built on Beta's) it's hard for Eclipse to keep up
−
** '''AI Dani''' discuss in Planning Council call tomorrow
+
** Lars: Some teams already moving to Kotlin...
 +
** Dani: More discusssions planned at ECE, AC etc
 +
** '''AGREEMENT''' to wait for now for (a) the Planning Council announcement and (b) how Java will actually implement and live up to their plans. Work towards faster release cycles (that's a good message to the consumers anyways).
 +
** Will need to work on details. If there's more releases, and each one needs a review, it's more work on release records and reviews. Already on a good path here, paperwork reduced to a matter of an hour.
 +
* Dani: Java 9 (Oxygen.1a) round the door. Plan is a full release.
 +
* McQ: {{bug|520176}} '''MacOS 10.13 menubar issue''' people at Apple are also looking at it. Biggest problem is that it's not 100% reproducible. Might end up with some System Properties to be set to resolve the problem.
  
 
<hr/>
 
<hr/>
−
<strike>Aug 30, 2016 - No topics</strike>
+
'''September 12, 2017''' - McQ, Alex, Dani
 +
* Dani asked Alex to look into {{bug|517063}} with high prio. We need this for M2.
 +
* Dani: sub-project merge
 +
** McQ and Lars approved my message regarding sub-project. Will send out the note today.
 +
** Still an issue with IP log not being merged
 +
** Release engineers can no longer restart the HIPPs
 +
* Started to discuss [https://dev.eclipse.org/mhonarc/lists/eclipse-pmc/msg03116.html Advise for performance enhancement making API methods behave async]
 +
** We need more time to verify the fix and look at the patch. Will discuss and decide next week.
  
−
<strike>Aug 23, 2016 - No topics</strike>
+
'''September 5, 2017''' - McQ, Alex, Martin
 +
* Alex: {{bug|521639}} '''Mac Signing Issues for Oxygen.1'''
 +
** 3 requests out of 5 failing, don't know yet why - webmasters assume hardware issues, trying to move to a different machine
 +
** RC4:
 +
*** Security fix in p2 ({{bug|518031}})
 +
*** Changes in Releng scripts to run tests on different Mac machine
  
 
<hr/>
 
<hr/>
−
'''Aug 16, 2016''' - Dani, Martin, Lars
+
'''August 29, 2017''' - McQ, Dani, Martin
−
* Dani will not always be available next two weeks (Eclipse Summit India and IBM Bangalore team visit)
+
* All: '''Conference System Woes''' - 3 tries for McQ (race conditions?), 2 tries for Martin (silence first), 10 tries for Dani
−
** Please split the load on +1'ing backport/API requests - please do review the code and set the pmc+1 flag
+
* Dani: '''Subproject Merge''' done after some hiccups, no complaints since Fri
−
* Dani: '''{{bug|498106}} Oxygen Plan'''
+
** Subprojects are now "archived" with a comment that they got merged into the parent project
−
** Agreed by all so far, including removal of REL6 as a reference platform
+
** '''AI Dani''' planning to send another notification to mailing lists:
−
** Individual subproject Wikis look pretty consistent now
+
*** former PLs to become Repo Maintainers responsible for signoff and planning
−
** UNIX Platforms (Solaris,AIX,HP-UX) have been removed for M1
+
*** Bugzilla remaining the same
−
*** PPC-Linux Releng was also removed but accidentally (due to a hardware failure), will be restored for M2.
+
*** Some former subproject leads already stepped up as combined project co-leads (Lars, Lakshmi, Sarika)
−
** No consensus on Ubuntu yet - will keep 14.04 in for now, consider taking out on the next update
+
** McQ: Move forward towards the new structure, but there's also need to push for responsibility among the subcomponents (finding go-to persons...)
−
* Dani: '''{{bug|499164}} Version Numbering for Neon.1'''
+
* Dani: '''StyledText improvement request on the ML''' (Pluggable selection model)
−
** The original "+5" or "+10" proposal showed some issues:
+
** point to the e4 incubator for experiments, encourage innovation though might be quite hard and pervasive
−
** Update with skipping would propagate to features; no tooling; no real benefit of semantic versioning
+
* RC3 looking good (except for build issues on the Foundation side, no I-Build since Aug.22 - Mac DMG is at risk for RC3)
−
** '''AGREEMENT''' to keeping using the status quo for now ("allow small updates even with micro uprev only")
+
* Dani away next week, please watch the list for approvals towards RC4 next week
−
* Lars: '''p2 pending patches'''
+
 
−
** '''AI Dani''' send E-Mail to Tom - suggest reviews or merger p2 back to Platform
+
−
* Dani: '''{{bug|499399}} Eclipse Infrastructure Stability'''
+
−
** Denis is looking at it - maybe too many requests coming in ?
+
−
* Lars: '''e4 spy and wizard migration to PDE'''
+
−
** Dani: Are wizards e4 specific, or could they be used with JFace ?
+
−
** New SWT Layout Spy should go to SWT, rather then keeping in PDE's Build
+
−
** Some spies usable also in RCP Apps - for example the e4 model spy - 6 independent projects, dependency on emf.edit
+
−
*** Looks like a big extension to PDE -- consider some consolidation and cleanup (tests, String externalization, ...)
+
−
*** '''AI Lars''' to check if contributors would be willing to clean up as discussed, then moving into pde.ui would be fine.
+
 
<hr/>
 
<hr/>
 +
'''August 22, 2017''' - Alex, Dani, Martin
 +
* All: '''Conference System Woes''' - Dani had 7 minutes trouble dialing in; Alex was kicked out
 +
* Dani: '''Plan''' - '''AI Martin''' review
 +
** Target Platform reduced - removed too much? - and Subproject plans. Rest is copy-and-paste.
 +
* Dani: '''Subproject Merge''' - Everything approved, waiting for implementation by EMO.
 +
* Dani: '''RC2 this week''' - Still many Platform/UI bugs, Dani will make a pass moving out since Lars is on vacation
 +
* Martin: '''feature/bundle version issues'''
 +
** Dani: Checking versions is quite time consuming, there's no tooling solution for the workspace right now (but there are reports).
 +
** For branding bundles and features, rules are different than for others - Dani currently doing those, since few people have access
 +
** Currently we're looking good, but automation would be better than manual nagging
  
−
'''Aug 09, 2016''' - Dani
 
−
* no one joined - had private conversation with McQ later
 
−
** '''guys, please announce when you can't join. I can make better use of my time than waiting for you in the call ;-) - thanks.'''
 
 
<hr/>
 
<hr/>
−
 
+
'''August 15, 2017''' - Alex, Dani
−
'''Aug 02, 2016''' - Dani, Martin, Alex
+
* Dani: '''subproject merge''' initiated - waiting for webmasters
−
* Dani: New [https://www.eclipse.org/projects/project-plan.php?planurl=http://www.eclipse.org/eclipse/development/plans/eclipse_project_plan_4_7_draft.xml Oxygen Plan Format]
+
* Dani: Provided Photon (4.8) release record and plan - please provide comments
−
** '''AI Dani''' ask Denis about aggregating the Sub Wiki's into the master page
+
* Alex and Dani discussed feature/bundle version issues - next builds should be better
−
* Alex: SWT Fragments for ARM64 etc
+
−
** Tycho doesn't support auto-selecting fragments - in contact with Jan and Tobias
+
−
** Dani: Would like to revert changes - don't want to punish people who are not interested in fragments
+
−
** Not really an SWT problem, so editing .target files would be a clean solution
+
−
** Will probably go with empty (dummy) fragments without natives, as that's the simplest solution
+
−
* Dani: SWT team - Lakshmi going to Sweden, will not be available until end 2016
+
−
** Arun is mostly back, Sravan is now on Releng
+
−
** Need to push further on building at the Foundation!!
+
−
** Alex: Getting CentOS for building ... but test machines are on SLES11, which is too old, will need at least SLES12
+
−
*** Dani: Will need an owner at the Eclipse Foundation to ensure this proceeds ! Alex agreed to further drive this.
+
−
* Target Env: Dropping REL6 -- reference platform for testing will only be REL7 (though will continue to run on REL6)
+
−
** Reducing support level this year, in order to prepare for only supporting GTK3 next year
+
  
 
<hr/>
 
<hr/>
  
−
'''July 26, 2016''' - no meeting
+
'''August 8, 2017''' - Alex, Martin, Dani
−
 
+
* Eclipse [[Asterisk]] system is not working reliably (again). It took Dani 6 minutes until he could dial in (using the Swiss number - US number failed repeatedly). McQ can't dial in until 10 min later.
−
'''July 19, 2016''' - no meeting
+
* Dani: starting the '''planning record for Photon''' to announce participation in M1
 +
** Toplevel items: JUnit5, Java9. Subplans on the Wiki.
 +
* Dani: '''subproject merge''' - wanted to wait for M1, but should be done this week
 +
* Alex: '''ppc64be''' - Sravan is waiting on final decision whether it will be removed in Photon
 +
** Dani: PMC decision was settled, so removed from Eclipse plan - IBM ''might'' still continue building it for internal use
  
 
<hr/>
 
<hr/>
−
'''July 12, 2016''' - McQ, Lars, Alex
+
'''August 1, 2017''' - no call
−
* Dani: Hackathon summary
+
  
 +
'''July 25, 2017''' - Alex, Lars, Dani
 +
* project merge got go, Dani to implement it with webmaster and then send out the announcement this or next week. E-mail will also mention that existing project leads can become co-leads on Eclipse Platform
 +
* Discussion if CVS should be removed from the SDK build, Alex +1, Lars +1, Dani +1
 +
* Discussion if the remaining 2.0 compatibility layer can be removed, Dani: OK to remove but other plug-ins might still have dependencies to it, must the also updated, as well as the documentation
 +
 +
'''July 18, 2017''' - McQ, Alex, Lars, Dani
 +
* project merge review awaiting go from Wayne - Dani pinged again in the bug report
 +
* Dani will kick off planning for Photon a bit earlier since he will be in Bangalore the next 12 days. Platform UI already started.
 +
* M1 on August 11
 
<hr/>
 
<hr/>
−
'''July 5, 2016''' - Lars, McQ, Alex, Dani, Martin
+
'''July 11, 2017''' - no meeting
−
* Lars: SWT move to Java 8
+
 
−
* Alex: Initial builds of GTK 2.24 for AIX - planning to bump minimum GTK version to 2.24
+
'''July 4, 2017''' - no meeting
−
** Will remove non-cairo drawing, this is where most of the breakage happens
+
−
** Ubuntu 12.04: 2.24.10 / 3.4.5
+
−
** RHEL 6: 2.24.23-6
+
−
** AIX, HP-UX, Solaris ? -- Oxygen will only focus on Linux, Windows, Mac
+
−
** Secondary Platforms: Not built on EF Hardware, but interested parties can contribute binaries
+
−
** Trying to support Platforms which can't move up to modern GTK at a similar pace as Linux is hard
+
−
** '''AI Alex''' write up a message to the Community
+
−
* Lars: {{bug|486961}} extending the Eclipse Text Editor
+
−
** Dani: Text Editor must remain unchanged, but feel free to provide a new different code editor
+
−
** https://bugs.eclipse.org/bugs/show_bug.cgi?id=496114#c7
+
−
** Not against adding extension points ... but having the standard text editor adopt those is a no-go
+
−
* Dani: Build, and Hackathon
+
  
 
<hr/>
 
<hr/>
−
'''June 28, 2016''' - McQ, Martin, Alex
+
'''June 27, 2017''' - McQ, Dani, Martin, Alex
−
* Discussion about Language Server Protocol
+
* '''Equinox Merge to the Eclipse TLP'''
−
** MS sponsoring hackathon in Zurich, Dani expected to go there
+
** Keep Equinox as a separate Project, to allow independent Equinox releases without Platform (and keep Bugzilla simple)
−
** The protocol has an Open Source license, but there is no Open Source Governance ... who owns the specification?
+
** The only argument for stronger integration would be getting more committers, but there is no strong need and we have the FastTrack process in place for that
 +
* '''Eclipse Merge of Component (Sub)Projects into Platform'''
 +
** Just 4 Projects: Platform, Equinox, JDT, PDE (plus the incubators for e4 and JDT)
 +
** Former (Sub)Project Leads to become "Component Leads", "Technology Owners" or "Repository Owners"
 +
*** Repository Ownership is appealing as the structure is very clear (+1 Dani,Alex,McQ)
 +
** Platform Lead initially Dani, more Co-Leads could step up if they want
 +
*** Main work is helping component leads if they can't come to a consensus - usually less work than PMC duties
 +
* Dani: '''JSR 376 Ballot'''
 +
** Public review reconsideration ballot got approved by all except Red Hat who abstained
 +
** Much progress in the past weeks, consensus except RH abstained
 +
** Looks like the OOBE of Java9 would tolerate reflection, currently looks like on track for a September Release
 +
* '''Vacations''' many people will be off next week, especially in the U.S.
 +
** PMC call will be skipped next week
  
 
<hr/>
 
<hr/>
−
'''June 21, 2016''' -  
+
'''June 20, 2017''' - MQ, Dani, Lars, Alex
−
*  
+
* Attracting new contributors discussions - code cleanup and deletion of old paths needed
 +
** Alex: Big issue to get people working on codebase and considering many now irrelevant paths in the codebase
 +
** Dani: Cleanups should be done by the one starting them through the whole SDK
 +
** Lars: Not feasible as some projects are slower pace moving and the change might require more effort than one can or is willing to spend in his free time
 +
** Alex: Great opportunity to have new people joining by doing simple tasks and grow them into full committers
 +
** McQ: We have to come up with list of items we would like to get cleaned and properly follow our deprecation policy even if the migration guide for Photon get really big with things we schedule for removal in the future.
 +
* Dani: Sub-projects merge
 +
** Name of the new project - McQ: Everyone refers to it as Platform so natural choice it is
 +
** JDT, PDE, Platform (subprojects) to become single project
 +
** Person responsible for some code area in the merged project - to be further dicussed
 +
* Dani: Dropping Xulrunner support - PMC to approve it in the bug
  
 
<hr/>
 
<hr/>
−
'''June 14, 2016''' - Dani, McQ, Martin, Lars
+
'''Juni 13, 2017''' - MQ, Dani, Lars
−
* McQ: '''Mattermost / Slack''' for communication
+
* Dani: Release review for 4.7 was successful
−
** not much different than newsgroup or IRC, but very dynamic when everybody is on it
+
* Dani: Rebuild for 4.7RCa required due to new EMF version and a fix for a critical bug which resulted in duplicated menu entries
−
* Lars: '''Master is Open'''
+
−
** Dani: Simple story - if we do need an emergency build we would create the maintenance branch on demand based on the tag
+
−
** Only limitation: We can't update the parent pom's at this time
+
−
** Don't have tools to check if the MANIFEST.MF needs to update - only API Tooling for API changes, but nothing for micro changes ({{bug|360703}}) is open to improve that)
+
−
** We do have an alert if there's no version/source change but the binaries changed (e.g. due to a compiler update) but that's not in the workspace
+
−
* Lars: '''Upgrade SWT to Java 8 for Oxygen ?''' - to be decided by the SWT team, the PMC is OK with the update if the SWT team decide so
+
−
* McQ: '''Language API -- Overlap between Orion and Che''' (from [https://dev.eclipse.org/mhonarc/lists/ide-dev/ ide-dev] mailing list and {{bug|496114}})
+
−
** The protocol is pretty big, but well documented - looks like an improved version of Flux
+
−
** The protocol is generic, implementations need to be specific -- some implementations already ongoing (Javascript)
+
−
*** People start working on editors hooking up to the protocol -- does it make sense building a JDT Language Server ?
+
−
** The tricky part is, that the Che workspace model is very different than the Eclipse workspace
+
−
** How to create a small lightweight server that would scale -- one workspace per user may not scale
+
−
*** At the beginning, it would be less powerful than JDT today since context would be lacking
+
−
*** But the direction is very interesting ... though maybe too many people are implementing the same thing right now
+
−
* Dani: '''What goes into the Neon Update Releases'''
+
−
** as discussed on May 17 (see below): Must be in master first, consider for backport after that
+
−
** '''AI: Dani''' will prepare a note
+
−
* Dani: '''Using Sound in Eclipse'''
+
−
** Eric Rizzo uploaded a feature to provide different sounds for successful or failed JUnit tests (see {{bug|496003}})
+
−
** Martin: CDT has allowed sound for breakpoint actions since long
+
−
** Lars: Moving the Mylyn notification framework to Platform UI was on the plan already, now considered for Oxygen
+
−
** McQ: Every Platform has an OS-Level Notification system so '''that''' should really be leveraged
+
−
** Martin: Mylyn notifications can be "active", can OS notifications also be active ? - McQ thinks yes
+
−
** Could work on SWT implementation of host platform notifications in parallel to pushing down the Mylyn API
+
−
** Request on specific JUnit sounds not yet decided.
+
  
−
'''June 7, 2016''' - Dani, McQ, Alex
+
* Merge of the sub-projects
−
* no topics
+
** Planned for M1, desired to be coordinated with the move of Equinox to Platform
 +
** We prefer to keep the project pages and wikis so that all the information is kept
 +
** Preferable we want to have one committer list for Eclipse platform
 +
** Dani to check with foundation how the merge can be done
  
−
<hr/>
+
'''May 30, 2017''' - Martin, Dani, Lars
−
'''May 31, 2016''' - Dani, Martin, Lars, McQ, Alex
+
* Dani: Please approve or disapprove the release review
−
* Dani: '''IP Log'''
+
* Dani: '''Starting the Debugger doesn't work always'''
−
** The tool did not work, but it was manually submitted and approved
+
** Occurs for Lars, but isn't reproducible - one of the top 3 things to fix if steps to reproduce are found
−
** In previous years, some people wanted the IP Log to not be so publicly visible, so now only project leads can generate and submit it
+
** Lars: Equinox replaced the DS implementation in 4.7, the issue is probably related
−
* Dani: '''Release Review'''
+
** Failure is totally random: On a very slow Virtual Machine, it failed almost always (9 failures out of 10); on other machine it works most of the time; even after working fine 3 times, it may fail the 4th time; seems to occur more frequently on older machines than newer ones.
−
** Reply to the public E-Mail with comments or +1
+
** Martin: try with a reverse debugger like [http://chrononsystems.com/products/chronon-time-travelling-debugger/download Chronon] ?
−
* Dani: '''Update Jetty to 3.9.3'''
+
*** Should help by just collecting a log; and when the failure occurs, "play back" the log to understand under what condition the race condition occurs; the a reproducible case is available, and the log can also be sent around between different developers for investigation, potentiall also to Apache Felix upstream
−
** Consensus to update - Jetty team working on fixing their update site
+
* Dani: '''Equinox Launcher Issues'''
−
* Dani: '''RC4 Status Update'''
+
** During M7, a change was made to ignore certain command-line arguments
−
** Cooling down - maybe 2 fixes from SWT, 1 in Platform/Text Search, and the approved doc fix for deprecating the mac command
+
** This caused 2 regressions - {{bug|517013}} and {{bug|516349}} MacOSX restart regression
 +
** Java 9 will probably change the parameters again, see {{bug|516911}} :(
 +
** It's getting too late to make any more changes to the Launcher. Re-compiling on different hardware has already caused {{bug|517013}}, it's possible that more issues (not yet known) might be caused by the rebuild. With Java 9 probably changing again, we don't win much by rebuilding to try and align with something that's not yet released.
 +
*** Option A: Put options into the eclipse.ini -- today it works for both Java 8 and Java 9
 +
*** Option B: Add the "probably proposed" new option to the launcher, will need to rebuild the launcher again
 +
*** Option C: Revert the launcher to what it was before (and well tested) back to M6 - for running with Java 9, the options have to be added manually
 +
** Users will have to read on a webpage what needs to be done for Java 9 launching; for Eclipse SDK, a single option is needed - some add-on components
 +
*** Additional complexity: Different JVMs need different options (IBM, Oracle...)
 +
*** Martin: {{bug|517452}} Introduce a fallback mechanism to the launcher, which reads the Java ID and loads 'eclipse-javaID.ini' with a fallback to 'eclipse.ini', but does not seem reasonable for Oxygen since too late
 +
*** Better to have a clean story: At IBM, the entire build environment changed ... reverting gets us back to a bullet-proof launcher that was tested for 7 milestones ... will work safely with Java 8, for Java 9 people will have to read the N&N or StackOverflow to find out about the right arguments - same as in Neon
 +
* '''AGREEMENT to revert the Launcher to the M6 version'''
  
 
<hr/>
 
<hr/>
−
'''May 24, 2016''' - Dani, Lars, Martin
+
'''May 23, 2017''' - Alex, Martin, Dani, McQ
−
* Dani: '''IP Log Generator''' - New version is unable to deal with Eclipse Project, Wayne is working on it
+
* Dani: '''Apache Batik XXE Vulnerability'''
−
* Dani: '''Release Review''' - Deadline next week (June 2)
+
** Alex: Moving to Batik 1.9 would be good, but there is so much UI breakage that it's not viable for Oxygen
−
* Dani: '''Deprecation Request from Brian'''
+
* Martin: '''Bugs discovered during RC test cycle'''
−
** Command org.eclipse.ui.cocoa.fullscreenWindow -- all PMC agree on the deprecation although late, since it only affects docs
+
** Surprised that some low-priority issues do get attention ({{bug|517108}}) while others get no response at all ({{bug|517013}}
−
* Dani: '''Copyright End Date'''
+
** Dani: Only interested in severe issues and regressions - send message to platform-releng-dev with severe regressions found
−
** '''AGREEMENT''' to no longer require updating the end date (though committers are still ALLOWED to update it)
+
** Dani: Please +1 the Gerrit for {{bug|517108}}
−
** Will no longer run the automated copyright tool (actually haven't run it the past 3 releases)
+
−
** Will live with the source code notices being inaccurate / incomplete - the accurate state is in Git
+
−
** This is acknowledged to not be ideal, but the best compromise we can come up with since "bulk changes" are not OK
+
−
* Dani: '''Neon Status'''
+
−
** Java 9 launching - affects all Java consumers, not clear yet what Oracle will do, no plan changing Equinox/Launcher at this time
+
  
 
<hr/>
 
<hr/>
−
'''May 17, 2016''' - Dani, Lars, Martin, Alex
+
'''May 16, 2017''' - Alex, Martin, Dani, Lars
−
* Dani: '''{{Bug|475184}} Neon Plan Update 2'''
+
* Plan Update - updated on Wiki, main plan updated, please provide feedback until tomorrow latest:
−
** Windows 10 and Ubuntu 16.04 will be added as "active" Platforms that people do work on
+
** Has Updates to delivered / not delivered, and Target Operating Environment
−
** SWT would like to remove Windows 8 and Ubuntu 12.04 from the list of reference Platforms
+
* {{bug|509922}} Performance Tests: Header is wrong, but it really compares against 4.6.1: results are created but report is wrong
−
*** Alex: Ubuntu 12.04 has GTK 3.10 which is hard to support
+
** Code should have been improved through many action enablement changes and startup improvments
−
**** Correction (Markus): Ubuntu 12.04 has GTK 3.4.2. Ubuntu 14.04 has GTK 3.10.8
+
** Sravan will work on making a proper comparison against 4.6
−
*** Message is clear: "Supposed to run on older Platforms, but these are the primary ones" - including last 2 LTS ones
+
* '''Test failures:''' Some in Platform/UI fail frequently, the most frequent ones should be considered for fixing
−
** Agreement to keep Windows 7 and remove Windows 8
+
** Browser tests: pick "an available" server, the M7 run was on a newly provisioned server which didn't have all the stuff for Webkit. This should be fixed now.
−
** '''AI Dani''' to check if SuSE 11 is right (12 been out since 2 years)
+
** CVS tests fail from time to time since the CVS server isn't stable at the foundation
−
* Dani: '''Copyright End Date''' - EMO: To be decided per project
+
** UI tests may fail if Mozilla or similar show a pop-up dialog requesting an update
−
** IBMers will have to continue updating the copyright end date, regardless of what other companies do
+
* {{bug|516114}} Tabbed Properties View Styling Change
−
** Copyright tool: Only the person who made a change is allowed to update the Copyright notice
+
** Due to the code change, the Tabbed Properties View is now loaded even if it's not used; that may degrade performance
−
*** gray area: actually fixing the date can be considered a change
+
** The issue is due to how styling is architected; an OSGi flag activates the plugin if any class is "used", the CCS engine just checks for a class without even instantiating, that already triggers the bundle activation
−
** Alex is in favor of consistency across the project
+
** That should be discussed for change in 4.7.1 or 4.8 ... maybe use reflection to avoid this, or avoid the need for eager activation of tabbed properties ... Lars did that in the past, but it's a lot of effort
−
* Dani: '''Policy for Neon.Update Releases''' - What to allow in, which rules to apply on the endgame
+
** Activation by itself would be relatively cheap, but then the code goes through all Singleton instantiations - hard to estimate
−
** Martin: Fact is, that the Update stream is built less often, and thus also tested less often
+
*** The impact is likely small compared to PDE for example, which takes 1.5 seconds out of 6 seconds startup time
−
** Rule: Any fix or feature must be in master first (and may be backported thereafter)
+
** '''AGREEMENT:''' In order to be fair to other people who worked on improving startup performance, the change should be reverted. We have a change that benefits few people, but impacts everyone here.
−
** Will want "some" features in Update, but probably not all of them - '''decision''': discuss feature by feature at PMC
+
*** Lars: OK to revert, but in the next cycle we should look at improving activations in general. Opening the Resource perspective should not activate PDE + JDT (today, this is likely triggered due to some decorators).
−
** Stakeholder who's interested in a backport must send a quick note to the PMC (just like for API change requests)
+
*** McQ would love to see an "optimistic UI" which can render at least the shell even if not all menus/etc are yet filled in
−
** In case too many requests should come in, could still discuss updating the procedure
+
* Dani is out Thu-Sat, please jump in and help the team if something is needed
−
* Dani: '''Neon RC2 Builds Update'''
+
* Alex: '''Project Structure'''
−
** Currently looking good, but latest Java 9 Launchers broke the Annotation support - investigating
+
** Platform/Resources doesn't have a lead at the moment. Would like to merge more and more into Platform/UI. It doesn't make sense keeping separate structures just for the sake of creating barriers to contributions.
−
** Quality generally looking good - "Stock GTK3" very stable now
+
** Eventually we'll just have the Platform/UI morph into "The Platform Project" with more co-leads
−
** GTK3 on SuSE, Fedora looking much better than Mars (at least with a GNOME shell)
+
** In the end it's not about positions and authorities .. it's all about people who are good
−
** Lars: {{bug|493477}} detached editors broken on GTK2 nominating for RC
+
** Dani: Agrees; in experience, every 2nd contribution is causing some issues, but we do have good people even if they are not official leads
 +
** McQ: What we really need is high-level overview and authority to recognize changes that might negatively impact others..
 +
** Could eventually end up with JDT/PDE/SWT/Platform. Lars and McQ are not so sure any more if SWT really needs to be separate, had several overlapping contributions - 9 active SWT committers also work on Platform code. Consider moving forward with this after the Oxygen release...
 +
** '''AGREEMENT''' to move forward on this, with a ''Platform Leadership'' group. We need to advance to leadership of Eclipse along with advancing the code. Leadership change should also lead to growing community.
  
 
<hr/>
 
<hr/>
−
'''May 10, 2016''' - Dani, Martin, Alex, McQ
+
'''May 9, 2017''' - Alex, Martin, Dani, McQ
−
* Dani: '''Policy for what should go into Neon.Update Releases'''
+
* Dani: '''JSR376 Jigsaw''': 10 said yes and 13 said no; a new spec to be provided within 30 days
−
** Would like to be more flexible than in the past - discussion point for next week
+
** From a technical perspective, should be possible to resolve within 30 days
−
* Martin: '''RHEL7 -- GTK3 versus --launcher.GTK_Version 2'''
+
** JDT team implementing a compiler; IBM implementing a JVM, that's completely separate concerns!
−
** GTK2 rendering still much more reliable than GTK 3 - forced to release commercial software with the GTK2 fallback for now
+
** For implementing a compiler, the last spec is quite old, that is not good enough
−
*** Drawback: Even with MATE desktop using GTK2 (for accessibility), the Browser widget webkitgtk2 is not easy to get since not in the official repos
+
** JDT team votes via Mike Milinkovich (Eclipse Foundation), Mike considered JDT team and RH team and probably others
−
** Alex: Desktops are moving faster - adopting change very quickly - cannot keep up with the investment that the Community currently makes in SWT
+
* Dani: '''M7/RC0''' - test passes upcoming; compressed scheduled due to the moved date for Devoxx
−
*** Heavily limiting what we support looks like the only viable approach at the moment...
+
** When using the RC, watch out for 2 things:
−
*** There's 2 big groups of bugs: (a) "Next" version of GTK, and (b) "Old" version like Ubuntu 13.04 which has 3.10 - can we limit support to "latest" ?
+
**# Action contribution enablement fix -- look out for menu and toolbar items not properly enabled
−
*** Martin: Need "LTS" Linux (RHEL 7, RHEL 6, Ubuntu 12, 14 - soon 16).
+
**# Section forms layout
−
** Alex: Gnome people have come up with a "Container environment" for GTK - could ship Eclipse that way ?
+
* Dani: '''httpcore 4.4.1 -&gt; 4.4.6 ECF Update and API Compatibility'''
−
*** Limitations: Updates must come through this application - already using Gimp from such an environment
+
** Annotations got removed - no binary breakage, but breaks source compatibility
−
*** Might be able to deliver Eclipse that way next year (Oxygen)
+
** People will prefer newer httpcore; API breakage due to annotations is unlikely
 +
** Also, source breakage gives developers a chance to just recompile, so accepting the new version is better
 +
** Alex: Are we forcing our dependencies to not break API? - We don't have that power anyways?
 +
** Alex: httpclient has security issues every few months ... better consume the latest now, and we'll likely have to update again. So if breaking compatibility, better do it now
 +
** '''AGREEMENT''' to do the update and thus accept the source breakage.
 +
* Flooding in Ottawa: Restricted to a fairly small area
 +
* Dani: '''Release Review''', does anybody look at the full document or can it be simplified?
 +
** All of the information in the RV is public (eg bug statistics, # contributors, ...)
 +
** McQ: It's useful for consumers to see that new things are happening, that is the N&N; beyond that, the use is limited. Effort around half a day.
 +
** Many bloggers talk about new things anyways
 +
** Migration Guide and N&N will get into the RV Material; beyond that, Dani offers adding stuff that people request
 +
** '''AGREEMENT''' to keep it simple, and send add-on requests to Dani if they come up.
  
 
<hr/>
 
<hr/>
−
'''May 3, 2016''' - Dani, Martin, Alex, McQ
+
'''May 2, 2017''' - Dani, Lars, Martin, Alex, McQ
−
* [https://dev.eclipse.org/mhonarc/lists/eclipse-pmc/msg02598.html Brian's request on the PMC list]
+
* Lars: '''Regex Evaluator Plugin'''
−
** McQ: Why would we care about adding small enhancements in RC1 when we should focus on testing and critical fixes
+
** Basically a view where one can input a regex and text, and see if it matches
−
** Dani: Two parts in the change - the original one (approvable), but another change also sneaked in (enable the prefix search) - not approvable
+
** Would like to add to e4 incubator, goal to eventually add into Platform/Text or similar
−
** Alex: People don't want to wait for getting in an improvement...
+
** McQ: Where it the boundary between e4 and Platform? How do people kick off little projects .. e4 incubator?
−
** McQ: OK for adding as feature to Neon.1 , but Neon.0 should be focused on bug fixes
+
** '''AGREEMENT''' to move code to Eclipse.org (into the incubator) and proceed from there. Might not end up in Platform, but will be good having the code at Eclipse.org.
−
** Keep in Gerrit for now - Neon.1 stream is typically opened when no new changes are expected. Rebasing is relatively easy from Gerrit.
+
* Dani: Endgame plan sent - Oxygen looking good, not too many bugs open
−
** '''ACTION''' McQ to -1 and argue why.
+
* Dani: '''JSR Jigsaw Ballot'''
−
 
+
** Asking for finalizing the Specs on JSR376 before voting Yes, since otherwise life is hard
−
* Alex: '''Update on Accessibility'''
+
−
** broken since GTK 3.8 where accessibility is no longer a module but shipped with GTK itself
+
−
** due to that, the way accessibility is implemented is totally different - no more factories - experimenting with possible approaches
+
−
** Discussing the addition of Eclipse to their test infrastructure, at a meeting a month from now
+
−
*** and even so, not sure how well accessibility is tested in GTK itself
+
−
** Launching with -gtk_version 2 is not a workaround, because the v2 GTK doesn't register as an accessible application in newer desktops !
+
  
 
<hr/>
 
<hr/>
−
'''April 26, 2016''' - Dani, Lars, Alex, Martin, McQ
+
'''Apr 25, 2017''' - Dani, Alex, Martin
−
* Mickael's API change request ... tried to address feedback from Markus but then got no additional feedback
+
* Dani: '''Neon.3a'''
−
* Alex: '''Accessibility''' - broken on recent GNOME and probably others
+
** Respin done -- problem: Updating from Neon.3 to Neon.3a , the problematic bundles are not removed
−
** Accessibility is important. Have reports from users who say "Eclipse works for them"
+
*** '''AI Alex''' tell Jeff to ask Pascal or Tom about how to disable the bundles on update
−
** McQ: Need to keep some focus and look at what we can do for Neon.1
+
*** Martin: If this problem can't be resolved could live with it ... better not take too much risk trying to resolve this very special problem
−
* Alex: '''Platform.ui generification of project'''
+
−
** Dani: Won't get a consistent story when done piecemeal - needs to be done "per framework" and not necessarily "per plugin"
+
−
** McQ: It's a huge effort - enough by itself for one whole release !
+
−
** Dani: If a meaningful separate unit is found that can be done, but things like e.g. LabelProvider, ContentProvider and TreeViewer can't be done isolated
+
−
** Lars: If a smaller unit of generification work is possible, that's permitted when somebody signs up for the work and it has been discussed in the public
+
−
** '''ACTION Alex''' to prepare a message for review and then send that.
+
−
* Lars: '''Copyright discussion'''
+
−
** Not having the end date in the copyright notice makes the copyright invalid.
+
−
** The IP Advisory Board will discuss it again.
+
−
* Dani: '''Update on FEEP'''
+
−
** Paul not reached out yet
+
−
** A list now shows what's done or in progress - but no time
+
−
* Dani: '''Update on M7'''
+
−
** Last 4 builds failed due to infrastructure issues - Maven stuff could not be accessed
+
−
** Will try to enable cache on the build side and re-try until the cache is full
+
−
 
+
−
<br/>
+
−
'''April 19, 2016''' - McQ, Dani, Martin, Alex (regrets: Lars)
+
−
* Dani: '''IP Stuff''' on Copyright Updates pinged the Lawyers but no update yet, need to wait
+
−
* Dani: '''FEEP''' discussed on AC, People were sympathetic to our issues.
+
−
** General direction that FEEP sponsored work is no different than other contributions
+
−
** Wayne took the Action Item setting up a meeting with Paul and interested parties.
+
−
** Website with ongoing FEEP still pending - Dani is tracking
+
−
* Dani: '''All PMC please review and +1 API requests''' when you find time
+
−
** McQ does review and typically ask question but not close enough to the code to +1
+
−
** '''AGREEMENT''' on "+0" votes to express "no objections but others should approve too
+
−
* '''Reviews vs. Features'''
+
−
** Alex: Badge System was quite successful at Fedora
+
−
 
+
−
<br/>
+
−
'''April 12, 2016''' - McQ, Dani, Martin, Lars, Alex
+
−
* Lars: '''Get rid of the Copyright Year Update Requirement'''
+
−
** [https://bugs.eclipse.org/bugs/show_bug.cgi?id=485257#c13 bug 485257_c13] EMO: All information is in the git repos
+
−
** McQ: OK lifting the requirement but doesn't want wrong information in files .. get rid of all copyright years ?
+
−
** Dani: Replacing the end-date by "[see git]" might be an option
+
−
** '''ACTION''' McQ or Dani talk to IBM Legal, then decide next week
+
−
 
+
−
* Lars: '''{{bug|490012}} Remove Project Settings to force strictly compatible JRE'''
+
−
** Example: When don't have admin rights on a customer machine, can't install anything
+
−
** Dani: With the Gerrit Build Validation, the risk of getting unbuildable code is no longer there
+
−
** Proposal: Project requests a particular JRE, but the workspace configures treating this as an error or warning
+
−
** '''DECISION''' to get rid of the strict error per project.
+
−
 
+
−
* Dani: '''{{bug|475934}} API Change on model processing order'''
+
−
** Dani: Even if this fixes an issue, M7 is late in the game
+
−
** McQ: Does API specify the order of events? - No; only on internal method.
+
−
** McQ: Is this transparent to most consumers, or will 5% have to react in their code ?
+
−
*** Lars has a few customers hit by this, but had to do a few tricks to mitigate the issue;
+
−
*** Making the change won't break them, but they could now remove their workaround
+
−
*** Martin: Not a formal API change and feels like cleanup/polish - M7 as the "performance milestone" feels just right for this
+
−
** Dani: Can be convinced making the change now if it is widely announced; convinced that this will hit more people than helping
+
−
** '''DECISION''' to allow making the change and announce widely.
+
−
 
+
−
* Dani: '''FEEP and EMO Hiring Update'''
+
−
** EMO working on website update to announce the work being done
+
−
** Didn't talk about the specific SWT issue around "FEEP should provide clean stuff rather than quick fix"
+
−
** Already agreed on hiring a person, communication will come in June/July
+
−
 
+
−
* Dani: '''IDE-dev proposal'''
+
−
** People don't find our tips and tricks; suggestion to add to the welcome page
+
−
** Solstice L&F issue on old Browser
+
−
*** McQ: Eclipse is a developer tool, and developers run on current browsers ... OK if only old browsers are affected
+
−
*** OK if there is a way out (such as choosing a different Theme) - not blocking anyone.
+
  
 
<hr/>
 
<hr/>
−
'''April 5, 2016''' - McQ, Dani, Alex, Martin, Lars
+
'''Apr 18, 2017''' - Dani, Lars, Martin, Alex, McQ
−
* Dani: '''FEEP Update'''
+
* Dani: '''RH and Java9 Jigsaw'''
−
** Foundation pays Patrik Suzzi as a contractor to look at Platform bugs (triage, and fix)
+
** Very good write-up linked from the [https://dev.eclipse.org/mhonarc/lists/eclipse.org-architecture-council/msg03385.html AC Mailinglist] - see also [https://dev.eclipse.org/mhonarc/lists/eclipse.org-architecture-council/msg03389.html follow-up]
−
** Will focus on critical / blocker bugs initially, or bugs already tagged by committers for 4.6 M7 and 4.6
+
** Eclipse JDT team concerns that the Spec is not complete brought forward to Wayne
−
** Mickael Barbero still quite involved in Releng - Foundation looking for Releng replacement so Mickael can develop (already reviewed autosave - but might be on vacation at the moment)
+
* Dani: '''Neon.3 Respin'''
−
* Dani: '''Solaris 32-bit'''
+
** Last action was with Ed Merks to validate that the Fix from Tom Watson actually works - no update so far
−
** Will provide Solaris 64-bit binaries, but stop building Solaris 32-bit with Neon
+
* Lars: '''[https://marketplace.eclipse.org/content/glance#group-details Glance] Migration''' to e4 and/or Platform
−
** Open question: what does this mean in terms of GTK versions ? (GTK3 would be nice)
+
** e4 incubator was always intended to have very low entry barrier - can give the Glance team access quickly
−
*** Alex may start asking questions as soon as the binary fragments are in the repo
+
* Lars
−
** McQ: Windows 32-bit is still active - future change probably tied to
+
** JUnit 5 licensing - EPL + Apache
−
* Dani: '''New Solstice Welcome'''
+
−
** Some issues with Logo, sometimes it's there sometimes not; Brian working on it, Solstice will become the new default in M7
+
−
** Currently, ONE product logo must work on all Themes - bad design - probably to be changed
+
−
** McQ: Do we mirror the welcome for BIDI ? (Dani thinks yes, saw bugs come along)
+
−
* Alex: '''FEEP Process Discussion (around Themed Scrollbars)'''
+
−
** "Scrollbar Theming": Contractor doing work without SWT committers knowing, and disregarding suggestions from the committers
+
−
*** Looks like the proper fix from committers was not even tried ?
+
−
*** It's OK putting adding a layer if it solves a problem ... but if the proper solution is suggested, it should at least be tried.
+
−
*** McQ: Accessibility - with 10000 items it's getting slow - should try the native way
+
−
** Lars: Transparency? - No insight on what FEEP's are actually ongoing
+
−
*** '''AI Dani''' send note to Wayne and Mike asking for location of FEEP Project List
+
−
*** '''AI Martin''' compare against AC's requested list, and put on agenda for the next AC call
+
−
** Lars: Styling is actually an important business driver!
+
−
** Alex: FEEP feels like "quick and dirty" is favored at the moment over sustainable solutions !
+
−
*** McQ: Perhaps a problem with the funding model ?
+
−
*** Dani: Lack of people who want to fix issues on the list of the AC ?
+
−
*** McQ: FEEP is still early - time to provide feedback now. Goal '''Transparency''' on what's done, and why design decisions are taken.
+
−
*** McQ: Who cares for StyledText - have Markus have a conversation with the submitter ?
+
−
**** Lars: We have no API for styling scrollbars (background, foreground) - Alex: '''THIS''' is what the FEEP should have adressed !
+
−
** Martin: Next step should be conversation with EMO about toplevel transparency, and basic guidelines ("sustainable solution over quick fix").
+
−
** Alex: Need something like JEP (Java Enhancement Process) for FEEP.
+
−
*** McQ: Agree this is what we want, but would it discourage submitters ? The lighter-weight the process the better...
+
−
** Alex: If anyone stepped in and did that work it's fine ... but such work sponsored via FEEP seems putting resources in a wrong direction.
+
−
* Lars '''Question: Google work on improved Type Hierarchy'''
+
−
** Main work is a new Index, Type hierarchy will be the first client, planned past Neon (contributor: Stefan Xenos).
+
  
 
<hr/>
 
<hr/>
−
'''March 29, 2016''' - McQ, Lars, Martin, Dani, Alex
+
'''Apr 11, 2017''' - McQ, Lars, Alex, Dani
−
* Lars: '''PDE Build in Maintenance''' - Officially state this ?
+
* Dani: Update from last Planning Council call:
−
** PDE Build still works; and we're not yet in a position to replace it (it's needed for manual export)
+
** There will be a special drop for Java 9 in July, but it has not yet been decided whether it will be a release where users automatically get the update, or a special p2 repository or Marketplace entry where the update can be done on demand.
−
** Alex: What about not quite declaring maintenance mode, but state something like "for a new project, you might consider Tycho or something else"
+
** Issues with Neon.3. A respin might be necessary. For details see https://dev.eclipse.org/mhonarc/lists/eclipse.org-planning-council/msg02741.html.
−
** McQ: '''DECISION''' Ensure that on the PDE Build Homepage, we clarify the relationship between PDE Build and other technologies (like Tycho)
+
* Dani: Asked Lars to review the Dark theme bug. Lars, said, he's not a user of it, but has one in his team. Dani: Ask him to review the change.
−
* Lars: '''Integrate more projects into Platform UI?'''
+
−
** Having few commits on a component, or overlapping committer list, is not an argument for merging projects
+
−
** Arguments for merging would only be if existing committers struggle with handling incoming contributions (or if component leads want to get merged) ... but that's not quite the case right now. Being able to look at a component by itself has also some value.
+
−
** Alex: Being able to handle Team/UI would make sense for Platform/UI ... but not for the native resources
+
−
** '''DECISION''' to not merge projects at this time.
+
−
* Lars: '''Request for API Changes on eclipse-pmc mailing list'''
+
−
** McQ: OK for a single PMC member to +1 the obvious ones, but should do due diligence on API changes (eg ping consumers etc)
+
−
** Alex: OK for API Additions, but should discuss API changes
+
−
** Compromise: API change requests must be requested on eclipse-pmc mailing list; PMC members must "speak up" on the mailing list if they want an item discussed, otherwise any PMC member can just approve. '''AI Dani''' notify mailing list.
+
−
* Dani: 900 Bugs targeted Neon
+
−
** Need to move away from "just moving bugs from one milestone to the next"
+
  
 
<hr/>
 
<hr/>
−
'''March 22, 2016''' - McQ, Dani, Alex, Martin
+
'''Apr 04, 2017''' - Dani, Martin, Alex
−
* Dani: we shipped M6 on time. Watch out for API change requests on the PMC mailing list.
+
* Martin: {{bug|514257}} and [https://dev.eclipse.org/mhonarc/lists/cross-project-issues-dev/msg14316.html cross-project Neon.3 wiring issue]
−
* Dani: public holiday on Friday (Good Friday) and/or Monday (Easter Monday)
+
** Problem: External libs don't deal with versions the same way we do
−
* Alex: gave an update on tests running on CentOS with GTK+ 3
+
** Dani thinks that the problem is that Release Train projects don't agree on one version to use
−
* Discussion about Che
+
** Martin saw a similar problem related to some bundles doing import-package and others doing require-bundle:
−
** McQ: Che still feels monolithic - unfortunate that the Eclipse Resource Model was not considered
+
*** Commons Logging is not a Singleton
−
** Dani: Preferences in Che vs. the IDE
+
** Dani has a call with Tom later today; hoping a respin can be done after fixing in Equinox
−
** McQ: What would it take to unify Che, Orion and the Eclipse IDE ?
+
* Dani: Heads up on upcoming Planning Council discussion re: Java 9 and Junit 5
 +
* Dani: '''PPC''' - No response on McQ's message regarding PPC, Dani will update the Plan after another week
  
 
<hr/>
 
<hr/>
−
'''March 15, 2016''' - McQ, Dani, Alex, Martin
+
'''Mar 28, 2017''' - McQ, Alex, Lars, Dani
−
* Dani: '''Platform/UI got awarded as the Most Open Project :)'''
+
* Replacement for Sergey
−
* Dani: Mention PDE/Build as in maintenance? - discuss next week when Lars is back
+
** let's wait and see how things go
−
* McQ: Merge Subproject Committers Groups? - Membership based on expertise is important
+
** next candidate would be Tom Watson once Equinox is moved back to the Eclipse top-level project
−
* Dani: Test Failures due to hidpi work - not entirely done yet but looking good for M7
+
* Dani: PPC strategy: drop 32-bit, provide 64-bit with Oxygen (4.7) and drop 64-bit BE with Photon (4.8)
−
* Dani: Lots of failures due to "Widget Disposed" failures
+
** all agree
−
** Gerrit tests didn't fail, probably sine the event timing on VNC is a little bit different than running manually
+
** McQ will send a note to cross-project-issues-dev
−
** Alex: VNC/Gerrit tests run on GTK2 whereas all the rest defaults to GTK3 now
+
* Lars will nominate Mikaël Barbero for Platform UI
−
** Looking for a machine with GTK3 to enable them -- working with Mikael Barbero to get this for David
+
* Lars plans to bring a IDE search plug-in to e4 (http://ystrot.github.io/glance/)
−
* VS Code Open Sourced - plugin model is very similar to Orion
+
−
** Excellent with TypeScript, not quite as good as Orion for JavaScript
+
−
** Ideally, would love to see some unification of web-based IDEs (like Che already using Orion editor)
+
−
 
+
−
'''March 8, 2016''' - No call
+
  
 
<hr/>
 
<hr/>
−
'''March 1, 2016''' - Dani, Martin
+
'''Mar 21, 2017''' - Dani, Martin, Alex
−
* Dani OOO next week
+
* Dani: '''Dirk Fauth''' [https://dev.eclipse.org/mhonarc/lists/eclipse-pmc/msg02891.html Requests on eclipse-pmc]
−
* Alex: '''{{bug|179924}} SSH-agent authentication''' 3rd party code for SSH using JVM internals
+
** Alex agrees with Dani
−
** Dani: Suggest asking on the bug for confirmation what's the JNA team's plan ... they have a full year time
+
** Where does the "0.x version before graduation" requirement come from? - Development Process, but maybe didn't apply to plugin versions
−
** Martin: JNA seems quite widely used these days...
+
** Dani thinks that forcing a 0.x version causes forced breakage once something goes API thus doesn't make much sense
−
* EclipseCon next week - Dani and Alex out - Martin will open the Bridge but might be canceled.
+
* Dani: '''Szymon Ptasiewicz'''
  
 
<hr/>
 
<hr/>
−
'''February 23, 2016''' - McQ, Martin, Dani, Alex
+
'''Mar 14, 2017''' - Dani, Martin, Alex
−
* Dani: '''Mars.2 is on track'''
+
* Dani: '''Java 9 readiness''' - see [https://dev.eclipse.org/mhonarc/lists/cross-project-issues-dev cross-project-issues-dev]
−
* Dani: '''FEEP Update'''
+
** Step 1: For testing, just download a Java 9 beta JVM, add the module -- all testing is possible from commandline
−
** EMO still looking for people bidding on items prioritized by the AC - Pascal to look at some p2ish issues in April
+
*** Platform team has an install, but Mike was concerned about legal issues - a bug is open to allow all simrel access
−
* Dani: '''Update on Welcome'''
+
** Step 2: Running the JDeps tools as per the [[Java 9 Readiness]] wiki can show violations
−
** Brian de Alwis contributed Solstice based Welcome - looking nice and done right, just a 4th theme to select (some artwork needs IP review)
+
*** Can even run with Java 8, Dani ran it on simrel, it's looking mostly good, only Orbit shows violations
−
* McQ: '''Scrollbars on Windows''' and their appearance
+
*** Platform team has some scripts; for Hudson there's even a Mojo, but it aborts when it finds a violation.
−
** Dark Theme Scrollbars don't currently look good on Windows - proposal to paint in SWT - currently proposed on StyledText only
+
** Step 3: Reflection - Will only figure out when knowing the code or testing
−
** Dani agrees with Alex that this would open the door to hell
+
** All these 3 tests can be done without the Java 9 patch in PDE. Only for debugging, one needs to self-host.
−
** McQ: Having seen AWT take ages in terms of emulating OS appearance makes the idea of painting ourselves not too appealing
+
*** This can be done either with Ed's Oomph support, or via the Marketplace
−
** Also concerned about special cases (like performance with ultra-long lines etc)
+
** Platform team is still negotiating options for better support with Oracle
−
** Alex: Any approach in this direction will have bad effects in terms of SWT maintainability and its cross-platform support (think Wayland, RAP, ...). API additions designed with not at least 2 out of the 3 WSs supported are probably not good cross platform APIs considering how different the WSs are.
+
* Dani: '''JUnit 5''' considers cutting down features for aligning with Oxygen
−
** Could a GTK version on Windows perhaps be an alternative? Alex(after meeting additions): GTK is supposed to run fine there and Broadway on Windows can be driving for it as figuring the build on Windows should give working SWT (thanks to Wayland work making GTK port quite crossplatform).
+
** Decision to align, contributing support to Oxygen Update 1 in September
 +
* Dani: '''2 Platform Issues in M6'''
 +
** platform.ui reexports everything including jface - lower bundles not updated - JDT may fail if installed (fixed for M7)
 +
** major version uprev: interface not marked as @noimplement; reverting the version, people who loaded Friday's M6 would never get any updates any more; decision to stick to the version uprev, since the respective bundle is not used a lot
 +
*** In general, accidental major version uprev should be reverted because the error would stay around for long and might break existing binary bundles who can't be fixed any more
 +
*** In this case, it was OK to keep since the bundle isn't used much
 +
* Alex: '''Generic Editor and Testing'''
 +
** Looking very good already - only 1/3 of the code needed for getting the same feature set as with existing methods for creating editors
 +
** Looking for features to add missing documentation (workflows, use-cases like converting existing editors): End goal is to provide everything that the JDT Editor is providing
 +
** Dani: Likes replacement of pages on multipage editors, and some new editors; would not consider replacing the JDT editor
 +
** In the Platform.ISV docs bundle, there should be a section for "how to write your own editor"
  
 
<hr/>
 
<hr/>
−
 
+
'''Mar 7, 2017''' - Dani, Martin, Alex, Sergey, Lars
−
'''February 16, 2016''' - Alex, Dani, Martin
+
* Dani: '''4.6.3 Updates''' - Looking good, Equinox bugs moved out by Tom
−
* no topics
+
* Dani: '''4.7M7 Update''' - Infrastructure issues but testing looks OK (except Mac which is not built yet)
 +
* Dani: '''Java 9 / Planning Council Update''' - Postponing Oxygen was discussed but denied; late July update being considered
 +
** See the [https://dev.eclipse.org/mhonarc/lists/eclipse.org-planning-council/ eclipse.org-planning-council archives]
 +
* Lars: '''Platform.Resources Leadership''' after Sergey Leaving
 +
** Dani pinged Szymon, will continue discussion when we hear back; merging with platform.ui is one option
 +
** Sergey: The "Team" and "Resources/Filesystem" parts could be considered separately. Consider merging Team into UI.
 +
** Alex: One top Platform Project would make sense now...
 +
** Dani: reconsider when McQ is back. In the past, SWT was seen as separate.
 +
** Dani: Releng needs to be separate and protected because Releng committers can also change Hudson jobs, shut down servers etc where no history is kept
  
 
<hr/>
 
<hr/>
−
 
+
'''Feb 28, 2017''' - Sergey, Alex, Martin, McQ
−
'''February 9, 2016''' - Alex, McQ, Dani
+
* Alex: '''Short update on Neon.3'''
−
* Dani: looking good for Mars.2 (4.5.2) - no RC4 planned
+
** 2 jdt.core bugs asking for PMC approval - regressions compared to Neon.2, approved
−
* Dani: please take a look at {{bug|466370}} Welcome Screen redesign
+
** PPC rebuild is in, so looking good
−
* Dani: please take a look at {{bug|472614}} Smart Project Importer
+
* Sergey will leave Eclipse work after end of March
 +
* Martin: '''RCPTT / Reddeer / Testing'''
 +
** Reddeer: No tests that check Platform only - everything is on higher level today (WTP, server connectors etc)
 +
** Runs on Neon.2 right now inside RH - the infrastructure isn't built for consuming daily builds
 +
** Would take some effort reducing to Platform only - no time for this in the RH team at the moment
 +
** '''AI Martin''' will look at integrating RCPTT as time permits; would like {{bug|505826}} resolved ideally
 +
*** See https://github.com/moberhuber/eclipse.platform.rcptt-tests
 +
*** And https://github.com/eclipse-testing/eclipse-platform
  
 
<hr/>
 
<hr/>
−
 
+
'''Feb 21, 2017''' - Dani, Alex, Martin, Lars, Sergey
−
'''February 2, 2016''' - Dani, Alex, Martin, McQ
+
* Lars: {{Bug|512273}} Allow any committer to retrigger Gerrit validation, and {{Bug|512319}} allow rebase
−
* Dani: '''Foundation Hires''' - Dani hopes they'll find someone in Europe to deal with Releng issues
+
** '''Agreement''' to allow retrigger for anyone, but rebase should be done by committers working on a contribution
−
* Dani: '''Neon M5''' went smoothly - M6 will be the API freeze; tomorrow is the last planned RC for Mars.2
+
** Dani: Not clear where to retrigger (Sergey: it's discoverable once permission is there)
−
* Dani: '''CQ Deadline''' on Feb.12 for Neon
+
** '''AI Lars''' follow up on the bug
−
* Dani: {{bug|486647}} for changing the strategy when opening files not associated with Eclipse
+
* Martin: '''UI Testing'''
−
** With M5, can now use the System Editor (default) or Text Editor - there's a request to always use the text editor
+
** [http://eclipse.org/rcptt RCPTT] is completely Open Source; commercial server could be used for load balancing tests but this is not necessary
−
** Dani thinks that for JPG one wants to use the OS
+
** Alex: Interested in Reddeer for upstreaming tests that already exist at RH; good reports about stability and scalability at RH
−
*** Martin agrees, OpenWith and choosing an editor remembers the choice, and one can also remember;
+
** When we decide which way we go, we need a plan who's doing the work (create the tests). Maintenance would depend on their structure.
−
*** We also show the System Icon in Project Explorer, so using the System Editor by default is natural
+
** Some Reddeer tests already upstreamed with WTP and Linuxtools
−
*** Alex set it to "ask via popup" - can be quite noisy, but avoids trouble when KDevelop or something associated itself with Makefiles or so
+
** Lars: Where would the tests live? - In a common project accessible to all, or with the component they test ?
−
** Agreement that flexibility is fine - see no case for changing the default
+
*** Lars, Dani: Since these are functional end-to-end, should live in their own repository, accessible by all committers.
 +
** '''AI Martin''' follow-up with Jubula. Try RCPTT on Windows and Linux. Give feedback till next week.
 +
** '''AI Alex''' share list of Reddeer tests existing at RH
 +
** '''AI Lars''' play with the Reddeer API
 +
* Dani: '''Update on PPC''' drop 32-bit, build 64-bit both le and be on RH VMs {{bug|512224}}
 +
** Alex: ppc64 currently built on Fedora 25; could build on REL7 if foundation requests a machine from RH Brno Farm '''AI Alex and Dani''' ask Denis
 +
* Dani: '''Neon.3 RC3''' please keep an eye on builds
 +
* Dani: Away this Friday afternoon and whole next week
  
 
<hr/>
 
<hr/>
−
'''January 26, 2016''' - Dani, McQ, Alex, Martin, Lars
+
'''Feb 15, 2017'''
−
* Official PMC representation on AC calls - keep McQ listed since he is interested but nominate Dani instead of John
+
−
* Next round of FEEP coming up - discussed pro and cons
+
−
* Dani: busy week for the team: Mars.2 (4.5.2) RC2 and Neon (4.6) M5 due this Friday - on track
+
  
 
<hr/>
 
<hr/>
−
'''January 19, 2016''' - Dani, Alex, Martin
+
'''Feb 8, 2017''' - Dani, Alex, Martin, Lars
−
* Upstaffing PMC - meet next week
+
* Dani: '''{{bug|509412}}''' - Adopt newer JSch for Platform 4.6.3
−
* Update Eclipse PMC representation on the AC
+
** Alex is fully consumed with Webkit work for Oxygen
−
** Since only Alex will represent the Eclipse PMC at EclipseCon, should push for not making formal decisions unless there is remote attendance support
+
** Need to move target platform to the latest Orbit in order to pick up the new JSch; that might cause other changes, since Orbit retention policy keeps only one version
−
* Dani: {{bug|485429}} '''PDE Build''' - looks like a lot of people still use it; OK if moving to a different component, but there's still a dependency:
+
** Dani: Cherry-pick the new JSch recipe to the Orbit Maintenance Branch ?
−
** Plugin or feature export still needs PDE Build -- Whoever wants to remove it, will need to implement a replacement '''in the Platform'''
+
** Martin: Looks like in the past, new Orbit R-Builds were made for Maintenance ... but with the move of Orbit to git, it is unclear how branch builds for Neon.x would be made
−
** Alex: Platform build currently does double bootstrapping Tycho already ... making this more complex would be a problem
+
* Dani: '''Update on RT PMC (Equinox) coming to Platform'''
−
** Introducing other build system into the Platform would be a wrong approach
+
** Tom Watson is in favor, reached out to Wayne for process
−
** Will go ahead with what we have for 4.6
+
** Potentially move entire IP log, but IP team is currently busy with CQs for Oxygen
−
* Alex: As of today's GTK3 build, Eclipse is working fine on Wayland :) thanks to a number of fixes in GTK
+
* Dani: '''Finally Unblocked on Java Language Server'''
−
** Some GTK developers are now considering Eclipse as part of their regression suite :)
+
** Dirk from Microsoft signed the committer agreement
−
* SWT move to Java 7 was much appreciated; moving forward, how to deal with functional interfaces ?
+
* Dani: '''Scenarios for UI Testing'''
−
* Alex would be interested in some styling support in SWT - would simplify some things, at least on GTK
+
** Initial Scenario from the PMC Notes can already be used; once that works, could extend to more
−
** Could be a very simple API like setStyle(String) - win32 also seems to have some styling support
+
* Dani: '''Sign Up for Security Rep'''
−
* Martin: '''Stability of GTK3 on Mars.2 ?'''
+
** See [https://dev.eclipse.org/mhonarc/lists/tools-pmc/msg03927.html Wayne's message to PMCs] requesting PMC attendance
−
** Alex: Many distros do their own hacks, so hard to tell... Mars.2 should be stable on every distro that has a stable GTK3 distro without too many patches
+
** John Arthorne used to be the Security Rep;
−
** Can't recommend GTK3 on Ubuntu yet; few things improved lately, but still causes troubles whenever Ubuntu updates their GTK, requires more fixes in SWT; Kubunutu and derivatives might be even worse
+
** Tasks: Follow the security mailing list; currently low volume but might get higher with IoT getting up speed
−
*** Good news is that GTK3.16 dropped support for custom theming engines - causes some ugliness but at least it's stable
+
** '''RESOLUTION: Alex agrees to sign up''' since already following security issues
−
** Debian or SuSE should be OK;
+
* Dani+Alex: '''libswt-gtk3 natives for PPC'''
−
** Expect Neon GTK3 to be more stable everywhere than GTK2
+
** Alex: Trying to limit the list of supported GTK versions to maximum 2; now at the latest GTK2 version
 +
** Currently, only Intel builds do both GTK2 and GTK3; PowerPC only does GTK2; the current hardware is unable to install REL7
 +
** Would prefer building at the Foundation over using custom company hardware; cost of hardware is around 50K
 +
** Primary Platforms should be built at the Foundation, secondary builds can be contributed ... should Linux-PPC be considered primary ?
 +
** Alex could easily provide a Fedora Platform, but it would be much newer than anything else that's around; ARM32 and ARM64 builds are already provided by Fedora
 +
** If somebody (IBM) needs an older version like CentOS, it would need to be provided by such adopters
 +
** '''AI Dani+McQ''' find out on priority of Linux-PPC (32bit, 64bit)
 +
* Alex: '''Update on Reddeer Testing Framework'''
 +
** A new version is being written, that no longer uses SWTBot underneath; there's a plan making this an Eclipse project
 +
** Engineers claim that after the rewrite it's a lot more stable recognizing widgets than on top of SWTBot
 +
** Martin has been looking at [http://eclipsesource.com/blogs/tutorials/rcp-testing-tool-rcptt-basic-tutorial/ RCPTT] which looks promising initially (easy to record tests) but unclear how stable such tests would be.
 +
** (call had to be closed at this point)
  
 
<hr/>
 
<hr/>
−
'''January 12, 2016''' - McQ, Dani, Alex, Martin
+
<strike>Feb 1, 2017 - cancelled</strike>
−
* Dani: '''Upstaffing the PMC'''
+
−
** Considering "one-time invitation" to get to know candidates better; not so comfortable with a "trial period"
+
−
* Dani: '''Remove Kim Moir from Platform Releng''' - ideally talk to Kim before moving forward
+
−
* Dani: {{bug|485429}} '''Remove PDE Build from our drops?'''
+
−
** Removing PDE Build is one thing - adding a different technology would be wrong, as wrong as adding EGit
+
−
** McQ and Dani will reach out inside IBM;
+
−
** Patches are still being submitted; middle ground would be remove from the delivery but keep in repo (and deprecate since not adding features)
+
−
* Dani: {{bug|485257}} '''Copyright Policy Change''' - waiting on EMO/Legal input
+
−
* Alex: '''Updating SWT to Java7'''
+
−
* McQ: '''Travel for EclipseCon'''
+
−
 
+
 
<hr/>
 
<hr/>
−
'''January 5, 2016''' - McQ, Dani
+
 
−
* Dani: '''PMC Membership'''
+
'''Jan 24, 2017''' - Dani, Lars, Martin, Alex
 +
* Dani: '''New Jetty Version''' - Alex will look at it after FOSDEM (Feb 7)
 +
* Dani: '''javax.xml''' - Plan to announce removal from the feature on cross-project for M6
 +
** Cannot remove from the compile prerequisites, because plugins still on Java 5 need it
 +
* Alex: '''[http://jboss-reddeer.github.io/reddeer/ Reddeer Testing Framework]'''
 +
** JBoss Tools uses Reddeer for overall user story validation
 +
** Lars: Why wasn't the framework added to SWTBot? -
 +
*** Reddeer is not a framework, it's meant as a test harness library - smaller scope than SWTBot or Jubula
 +
*** API Stability across multiple Eclipse versions is out of scope, thus created outside Eclipse for now
 +
** Martin: Any improvements in object recognition compared to plain SWTBot?
 +
*** Reddeer tries to make use of the best mechanisms from SWTBot for object recognition (Finder, or callbots). That saves users from making newbie mistakes in SWTBot, but apart from that it's no better
 +
*** No solution for external (non-Eclipse) dialogs ... that's not doable in SWT
 +
** Lars: EclipseSource people have been talking about [http://eclipsesource.com/blogs/tutorials/rcp-testing-tool-rcptt-basic-tutorial/ RCPTT], it might be able to deal with native tooltips etc
 +
** Dani and Lars think Jubula is too heaviweight; might work for Integration Builds, but not for individual developer's builds. Local setup must be really easy. Also, setup looks non-trivial, needs resources to work on ... and, doesn't seem to add much benefit over SWTBot (assuming that Jubula agent doesn't hook into native libs like win32, Cocoa or GTK).
 +
*** Alex: Especially integration in Maven may be hard.
 +
** Alex: Main limitation is that we don't have resources to actually create tests. Suggest reaching out for help asking people who would actually create tests ... then use whatever framework that people would like to use.
 +
** Dani and Lars won't be able to come up with a workflow scenario before end of next week --&gt; Alex will come up with an idea since he has meetings on Monday.
 +
*** Dani: Launch Eclipse, choose a workspace, open the package explorer, create a Java project (helloworld). Expand later, maybe into quick assist.
 +
*** '''AI Alex''' to demo Reddeer in 2 weeks
 +
*** '''AI Martin''' to try installing Jubula until next week
 +
*** '''AI Martin''' ping Sergey re: asking for help on Platform/Resources.
 +
 
 +
<hr>
 +
'''Jan 17, 2017''' - Dani, Sergey, Alex, Lars, Martin, McQ
 +
* Sergey: '''Merging platform.resources with platform.ui'''
 +
** Seems that more contributions to platform.resources are made by platform.ui committers, than others. Only 2 platform.resources committers seem active (and those two are actually platform.ui committers as well)
 +
*** Opening up to Platform UI would encourage more people to actually contribute to resources. Strive for shared access on components that are interrelated
 +
** Dani: Merging into platform.ui would mean making Dani and Lars the leaders for signoff - that causes too much burden
 +
** Sergey: Platform/Team has actually more synergy with Platform/UI than resources
 +
** Alex: For gcc for example, there is a notion of "maintainers" per git repo for signoffs - that's independent of leadership or commit rights
 +
*** McQ: Not sure how adding another role would simplify things ?
 +
** Sergey: The problem for signoffs is not is much the volume of commits, but the volume of failing tests
 +
** Alex: The manpower of Platform Resources is so low by now that even {{bug|509412}} was not being filed ... existing committers do not really seem to "own" the project any more, this is an indication for merging
 +
** Martin: Can't see how merging into a bigger entity would resolve the "ownership" problem. Perhaps make interested contributors Platform/Resources committers instead, and see how to resolve the ownership issue on the smaller scope ?
 +
** McQ: Propose applying the recent simplification of gaining commit rights (for "known" committers)
 +
* Lars: '''Removing javax.xml'''
 +
** Was introduced only for very old JRE's , it is part of the JRE now
 +
** Martin: Might cause a uses constraint issue since not exporting the package with version any more, see http://njbartlett.name/2011/09/02/uses-constraints.html
 +
** Lars: With javax.annotation, that was solved with smart reexporting
 +
** McQ: OK with removal when we can confirm we're not breaking things
 +
** Dani: Whoever requested removal should double-check whether the API is the same.
 +
**'''AI Lars''' send a note to Tom Watson as the submitter.
 +
* Dani: '''Project Updates'''
 +
** Deadlines upcoming for CQs and APIs - see Endgame Plan, please provide feedback
 +
* McQ: '''UI Testing and Jubula'''
  
 
<hr/>
 
<hr/>
 +
'''Jan 10, 2017''' - Dani, Lars, Martin, Alex, Sergey, McQ - Special Guest: Alex Schladebeck
 +
* Dani: '''Platform Bits on Maven Central''' - Now available thanks to Stephan Herrmann
 +
* Dani: '''Board Committer Rep Elections''' opening
 +
* Dani: After David Williams' move, Wayne is looking for a Planning Council Chair; Fred Gurr has taken over Simrel Releng
 +
* Dani: objenesis (Mockito / Easymock dependency) and Java9 - will wait for how upstream is planning to address issues
 +
* Lars: '''Equinox move to Platform''' - no updates - '''AI Dani''' talk to Tom again
 +
* McQ / AlexS: '''Jubula'''
 +
** In the past, with lots of full-time Platform committers, sniff tests just happened automatically giving great confidence. Now, with more part-time work happening, there is more need for end-to-end workflow test automation to give quality confidence
 +
** Martin: On top of low-level JUnit tests, want at least very few highlevel end-to-end test to avoid "Eclipse looking silly". Need UI tests that are really stable (avoid random failures)
 +
** Dani: On EGit SWTBot tests, too many libraries to install, how to run on Gerrit HIPP slaves
 +
** AlexS: Original Jubula goal was to allow non-programmers write tests. Beginning 2015, the client API (for controlling the SUT) was separated from the front-end, was separated from the ITE. This allows writing UI tests programmatically, and avoids the need for a database.
 +
*** For identifying components/actions to write tests, officially an object mapping needs to be done in the front-end (ITE, integrated test environment). But there may be a shortcut available for getting object IDs
 +
*** Compared to SWTBot, Jubula is 90% blackbox (code highlevel actions rather than things like calling setText())
 +
*** UI Tests typically take longer than unittests by definition (due to setup, teardown, ...), so might not be applicable for Gerrit triggers
 +
*** Dani: Would like to run a first test on CentOS first, as part of the integration tests
 +
*** Lars: Robustness of the tests? - SWT on Mac is problematic, but Linux and Windows should be OK.
 +
*** Martin: Dependencies? - Around 5MB Libs, plus the agent needs to be installed (around 100MB). Once that is there, tests are stand-alone on a laptop. There might also be an "embedded agent" but not sure ('''AI AlexS to check''')
 +
*** Getting a first test up and running ? - Maybe around 1 week, depends on what to do ?
 +
**** Lars: Would like something for quick access. Martin would something for "fresh download" (fresh config area + workspace) maybe combined with quick access.
 +
*** '''AI AlexS''' send slides
 +
*** '''AI Dani + Lars''' propose scenarios. Check possible owners of the initiative on Platform/UI side
 +
*** '''AI AlexS''' check possible PlatformUI Contributors from Jubula team to get started
  
 
= Archive =
 
= Archive =
 +
* [[Eclipse/PMC/Minutes 2016 | Archive of Meeting Minutes from 2016]]
 
* [[Eclipse/PMC/Minutes 2015 | Archive of Meeting Minutes from 2015]]
 
* [[Eclipse/PMC/Minutes 2015 | Archive of Meeting Minutes from 2015]]
 
* [[Eclipse/PMC/Minutes 2014 | Archive of Meeting Minutes from 2014]]
 
* [[Eclipse/PMC/Minutes 2014 | Archive of Meeting Minutes from 2014]]

Revision as of 10:56, 13 December 2017

Documents

Some documents written and/or used by the PMC:

Meeting Schedule

The Eclipse Project PMC has a weekly phone meeting every Tuesday at 11.00am EST.

Meeting Minutes

December 12, 2017 - McQ, Alex, Lars, Dani, Tom

  • P2 capabilities work breaking Oomph and CBI (bug 528408)
    • Alex: Hidden bug due to adopters using internals.
    • Alex: Code refactoring in p2 pushed to reduce the chance for such breakage.
    • Alex: Oomph fixed its codebase to properly check instances so it works with p2 both prior and after adding capabilities.
    • McQ: What do we do with p2?
    • Alex: Open up the project as much as we can and recruit new committers by being really welcoming to people.

-> Conclusion to not respin M4 due to this bug as it's internal API which projects should not have used.

  • Project plan update
    • Dani: No additional change came from anyone else in the PMC so update of executions environments and adding Java 9 are the only thing due.
  • Next meeting on January 9

December 05, 2017 - McQ, Alex, Lars, Dani, Tom

  • Vacation: Dani will be in vacation end of week, so other PMC member need to cover for him if something comes up for M4.
  • Discussion about adding EGit to the SDK? (bug 528004)
    • Dani using a shared installed to manage the EGit plug-in.
    • McQ explains that he thinks EGit is required for a full package but does not feel strong about.
    • Alex: It depends what "SDK" means: SDK to develop plug-ins or SDK to develop Eclipse Platform. We agreed on the former.
    • Dani, Alex are against adding it because it adds maintaince burden and it is unclear which of the optional packages should be installed.
    • Tom, Lars have no strong opinions
    • All agree that managing to install this is an easy task

-> Conclusing to not include EGit in the SDK

  • URL handler registration of Eclipse with the OS
    • SAP plans to contribute a way to register Eclipse to certain URLs. If the user clicks on such a URL, Eclipse would open and be able to show the file. This could be useful for installation of new items in the marketplacen client https://bugs.eclipse.org/bugs/show_bug.cgi?id=527579. VSCode allows this already vscode:extension/eamodio.gitlens.
    • Redhat had similar requirements in the past
    • Platform team expects that SAP will also maintain this contribution and take responsibility of it for future releases
    • Dani (after meeting): I think extending the current file based mechanism is OK, but we won't register any (including URL) handler

-> Conclusion to work on this enhancement together with SAP, Alex plans to dedicate some time for discussing the technical solution. SAP should open a new bug and link all related bugs to this top-level bug so that the technical details can be discussed

  • Discussion of building natives
    • moved building of SWT binaries for the 3 primary platforms to the Eclipse Foundation
    • unfortunately after that the Mac went down - fixed now
    • M4 SWT natives will be built on the Eclipse Foundation hardware
    • Launcher still to be moved

November 28, 2017 - McQ, Alex, Lars, Dani

  • Alex: Release Change announcement
    • Agreed that we won't mention the ramp-down process.
    • Agreed to add API removal details.
    • Still work in progress.
  • Dani: API Removal process for the new release model
  • Dani: Meeting notes
    • McQ suggested that who joins last has to do it. Dani: not so good since it won't contain the notes for the discussions that happened before.
    • Will do it based on rotation and ability to actually write down the notes.
    • Dani to do it for this call.
  • Dani: Photon plan update
    • Will add Java 9 as supported VM.
    • Anything else? If so, send Dani a note or update bug 527964
  • McQ: Rename Eclipse Project to Eclipse SDK
    • We should only do it when forced by the Foundation and/or Board
    • We would need a plan before starting such an effort.

November 14, 2017 - Dani, McQ, Alex, Tom

  • Dani:
    • Presented our new release model to Planning Council. Was welcomed. No objections. But needs discussion what the Release Train will do.
    • RC1 went well, RC2 this week.

November 07, 2017 - Dani, McQ, Alex, Tom, Lars

  • For the 3 months releases, we have have a development freeze after RC1 until RC3. After RC3 master will be opened again for development. Development is possible until RC1, so this will result in a freeze period of approx. 2 weeks.
  • For releases versions of the platform we will use semantic versioning as before (4.8, 4.9, 4.10, ..)
  • Eclipse SDK will use versions not branding names unless the release train moves also the same cycle.
  • Splash will continue show the release year and month

October 31, 2017 - Dani, McQ, Martin, Alex,

  • Tom: Bugzilla Landing Page - EMO said both PMCs need to be contacted
    • Bug IDs, Equinox product and components remain in place, only the landing page will be updated
  • Dani: M3 Respin
    • Issue had been found by a team member, but got lost between different teams
    • Resolved and respun very fast, the fix is out now
  • Dani: Release Cycles and Java Alignment
    • Planning Council F2F meeting was at EclipseCon Europe; planning call will be on Thursday
    • Initially, felt like aligning with Java 18.3 was a good idea, but for now the plan is keeping the Oxygen SR rhythm and releasing Oxygen.3a aligned with Java
    • In case the Oracle release should slip, it will be most likely be named 18.4 (whatever month the release ends up in)
  • All: Platform Release Rhythm
    • At ECE, many people were surprised about changing release rhythm but the plan was generally appreciated
    • Should stick to the communicated Photon plan though and change only afterwards
    • Alex suggests promoting a Milestone build from master (properly marked up as such) aligned with the Java 18 release
      • This would promote master as something that can be used
      • After Photon, we will release 4 times from master - there won't be any maintenance builds any more
      • For the release train, there will also be release candidates.

October 24, 2017 - cancelled due to ECE

October 17, 2017 - Dani, McQ, Martin, Lars, Alex

  • Conf.system: Fallback to McQ's number is not a good solution (can't give out the moderator code)
    • Foundation (Denis) is aware of the problem, looking for a solution sooner than later
    • Skype group call, or zoom.us would work for Dani and McQ
  • Dani: bug 526065 Java 9 to Oxygen.1a beta update issue
    • Users who had the Java 9 beta installed and "update" to Oxygen.1a are broken
    • McQ: Updating from "beta" to "final" is not necessarily supposed to be working (Lars disagrees)
    • Not yet sure what's causing the issue .. p2 bug, or different groupID for Java9beta
    • Fix the issue if it looks like a real bug or easy to fix ... otherwise just post a message, preferredly on the Marketplace client (Dani: that part is done already)
  • Lars: Release Cycle Discussion
    • McQ still in favor of making every milestone a release
      • Alex thinks we're not there yet in terms of releng, there's still too many manual steps. Would prefer slower cadence (eg 3 months) until more automation is there (updating pom.xml, version mgmt, ...)
      • With 4 releases per year (like today) but no more maintenance branches/backports, there's potentially less work than today
      • Try starting with a release every 2nd milestone; fix automation; then strive for releasing every milestone.
      • Dani thinks the model shouldn't change in the middle of Photon; look for feedback then move to the new model next year
    • McQ: If we switch to the new model now (Oxygen.2), what do we lose?
      • Martin: Any bigger feature that would need to move into a branch? - Dani: Advanced Source Lookup
      • Dani: Also our rules about "no new features" and "PMC Approval" would need to change
      • Continue discussion in 2 weeks (skip next week due to EclipseCon)

October 10, 2017 - Dani, McQ, Lars, Alex, Martin

  • Dani: Oxygen.1a respin
    • Primary reason for respin was, that people who have existing XText installed would be broken when updating the Platform
    • Opportunity to piggy-back JDT critical fixes was welcome
  • Dani: Equinox Move - Review passed without objections, working on implementation now
    • Expecting less trouble than with the Platform and JDT merges recently
    • Foundation want to keep Equinox as a brand, therefore will be a subproject (beside Platform, JDT, PDE)
    • So to increase the committer base, the fast track option will now apply; also, more alignment in terms of PMC.
    • Confirmed again that we will ask Tom to join our PMC. Approval can be found in this thread.
  • Dani: Planning Council - Rolling Releases Update
    • Should remove some of the work maintaining 2 streams - simply release from master regularly
    • In the past, API breakage (migration guide) was allowed only once per year - how to deal with this moving forward
    • Handle with semantic versioning; how to give adopters sufficient time to adjust
    • There's usually 2 camps for every topic being discussed ... though practically, Platform sets the pace
    • Release Naming - New Name every quarter? Versioning scheme? (like 2017.4 or 17.04)
    • How to simplify processes (also in terms of IP), can the Foundation scale up? Perhaps not more work, but more deadlines
      • In terms of the Release Review, only the N&N and the Migration Guide are relevant; need to satisfy small and big consumers
  • Martin: Reddeer Contribution
    • When we last discussed this, it was an important criterion that any contributor can run the tests easily
    • Alex: Looks good, essentially boils down to running "mvn clean verify" in the PDE repo (or right-click > Run as Reddeer from the IDE) - AI Alex will have a video sent to the PMC list
  • Alex: API Removal in Context of Rolling Releases
    • Which Rules to apply for allowing removal of API; Allow removal in any release?
      • McQ: The real question is, how long it takes consumers to adopt to removal
      • Dani: Is more conservative regarding deletion - the value removing something is often smaller than the ripple it causes
      • McQ: If there is code in our codebase that significantly slows us down, it should be remove even if API
    • Martin: With more releases, there's less push to upgrade ; but will consumers know they are broken? - Semantic versioning should address that, but who really uses upper version ranges ... they have caused more pain than benefit in the past
    • Dani: In the past, with our 3-year deprecation plan, Major versions have not been updated even on announced removals...
    • McQ: If we are earnest about faster velocity, we have to fully adopt semantic versioning practices. If we as producers have done what we could, our job is done - consumers who don't use upper version ranges are on their own

October 3, 2017 - Dani, McQ, Alex, Martin

  • Conf system woes - Dani couldn't dial in, Denis restarted the bridge leaving Martin and Alex in a Zombie bridge
    • ACTION McQ will send out conf.details using IBM's bridge
  • bug 520720 Mickael's question on Performance improvement
    • Dani: There is already an asynchronous RequestLayout API for clients, and APIs already suggest clients use the async one
      • Therefore, changing the synchronous API to work asynchronously doesn't look right.
    • Alex: We seem to be mixing two aspects here -- agree regarding RequestLayout concerns, but why should the ProgressMonitor one be a problem?
      • Dani: Agree that the ProgressMonitor approach makes sense, but a very careful review is needed - existing code already subclasses ProgressMonitor, need to make sure that expected methods are still called
      • ACTION Dani will do a careful review of the ProgressMonitor part and deny the generic RequestLayout part
  • Dani: Equinox Restructuring request for the move is in, will work with Tom and Foundation forward
  • Dani: Oxygen.1a - RC2 is in for JUnit5 and Java9
    • Respin of Oxygen.1 for bug 520176 Mac 10.13 High Sierra problem? - Not currently planned, there will be 1a instead
    • A feature patch is another option, providing patched SWT; and, there's the workarounds
  • Dani: Rolling Release - 2 topics: rolling release, and alignment with Java Update Releases
    • Some projects on the train have alignment goals with other projects outside Java; will discuss again in the Planning Council
  • Alex: bug 522733 webkit crash - A workaround is available, in contact with webkit devs for a complete fix to keep Photon usable
    • Workaround would introduce a small memory leak, but only on the first instance of webkit created

September 26, 2017 - Dani, McQ, Alex, Martin

  • bug 520176 MacOSX 10.13 (High Sierra) Menus Disabled
    • Only non-English Locales are affected
    • For 4.7.1a and Photon, a fix is in
    • Foundation Press Release (Newsletter) is out with information how to work around
    • Tweets are around as well -- keep spreading the message
  • Equinox Move Review - progressing
  • Alex: bug 522733 GTK Crash on Close preventing IDE Restart - can it go into 4.7.1a ?
    • Seen with Fedora 27 (Webkitgtk 2.18)
    • Foundation wanted to keep Java9 stuff separate, so pushing it there might not even help
    • The issue is serious (hard crash preventing restart of the IDE), though fortunately not many people have that particular webkitgtk version yet
    • Investigate details and possible workarounds, then consider our options -- maybe a Foundation Newsletter similar to the OSX 10.13 issue

September 19, 2017 - McQ, Alex, Martin, Lars

  • Async APIs request from the eclipse-pmc list (not discussed today)
  • Lars: Java 6-month release cycles
    • Alex likes the 6 month idea - less backporting
    • McQ would ideally like full releases every 3 months (except not releasing anything unfinished)
    • Dani: If Java really delivers every 6 months (will they?), Eclipse could consider rolling releases every milestone. There is a doc from the Planning Council (Mélanie Bats) in that regard.
    • Dani: Support for the public JDKs will be limited to the current release only (6mo)
    • Risk: Oracle Java may be working on big features "in the background", if there is very little time adopting new features (current license doesn't allow releasing stuff built on Beta's) it's hard for Eclipse to keep up
    • Lars: Some teams already moving to Kotlin...
    • Dani: More discusssions planned at ECE, AC etc
    • AGREEMENT to wait for now for (a) the Planning Council announcement and (b) how Java will actually implement and live up to their plans. Work towards faster release cycles (that's a good message to the consumers anyways).
    • Will need to work on details. If there's more releases, and each one needs a review, it's more work on release records and reviews. Already on a good path here, paperwork reduced to a matter of an hour.
  • Dani: Java 9 (Oxygen.1a) round the door. Plan is a full release.
  • McQ: bug 520176 MacOS 10.13 menubar issue people at Apple are also looking at it. Biggest problem is that it's not 100% reproducible. Might end up with some System Properties to be set to resolve the problem.

September 12, 2017 - McQ, Alex, Dani

  • Dani asked Alex to look into bug 517063 with high prio. We need this for M2.
  • Dani: sub-project merge
    • McQ and Lars approved my message regarding sub-project. Will send out the note today.
    • Still an issue with IP log not being merged
    • Release engineers can no longer restart the HIPPs
  • Started to discuss Advise for performance enhancement making API methods behave async
    • We need more time to verify the fix and look at the patch. Will discuss and decide next week.

September 5, 2017 - McQ, Alex, Martin

  • Alex: bug 521639 Mac Signing Issues for Oxygen.1
    • 3 requests out of 5 failing, don't know yet why - webmasters assume hardware issues, trying to move to a different machine
    • RC4:
      • Security fix in p2 (bug 518031)
      • Changes in Releng scripts to run tests on different Mac machine

August 29, 2017 - McQ, Dani, Martin

  • All: Conference System Woes - 3 tries for McQ (race conditions?), 2 tries for Martin (silence first), 10 tries for Dani
  • Dani: Subproject Merge done after some hiccups, no complaints since Fri
    • Subprojects are now "archived" with a comment that they got merged into the parent project
    • AI Dani planning to send another notification to mailing lists:
      • former PLs to become Repo Maintainers responsible for signoff and planning
      • Bugzilla remaining the same
      • Some former subproject leads already stepped up as combined project co-leads (Lars, Lakshmi, Sarika)
    • McQ: Move forward towards the new structure, but there's also need to push for responsibility among the subcomponents (finding go-to persons...)
  • Dani: StyledText improvement request on the ML (Pluggable selection model)
    • point to the e4 incubator for experiments, encourage innovation though might be quite hard and pervasive
  • RC3 looking good (except for build issues on the Foundation side, no I-Build since Aug.22 - Mac DMG is at risk for RC3)
  • Dani away next week, please watch the list for approvals towards RC4 next week

August 22, 2017 - Alex, Dani, Martin

  • All: Conference System Woes - Dani had 7 minutes trouble dialing in; Alex was kicked out
  • Dani: Plan - AI Martin review
    • Target Platform reduced - removed too much? - and Subproject plans. Rest is copy-and-paste.
  • Dani: Subproject Merge - Everything approved, waiting for implementation by EMO.
  • Dani: RC2 this week - Still many Platform/UI bugs, Dani will make a pass moving out since Lars is on vacation
  • Martin: feature/bundle version issues
    • Dani: Checking versions is quite time consuming, there's no tooling solution for the workspace right now (but there are reports).
    • For branding bundles and features, rules are different than for others - Dani currently doing those, since few people have access
    • Currently we're looking good, but automation would be better than manual nagging

August 15, 2017 - Alex, Dani

  • Dani: subproject merge initiated - waiting for webmasters
  • Dani: Provided Photon (4.8) release record and plan - please provide comments
  • Alex and Dani discussed feature/bundle version issues - next builds should be better

August 8, 2017 - Alex, Martin, Dani

  • Eclipse Asterisk system is not working reliably (again). It took Dani 6 minutes until he could dial in (using the Swiss number - US number failed repeatedly). McQ can't dial in until 10 min later.
  • Dani: starting the planning record for Photon to announce participation in M1
    • Toplevel items: JUnit5, Java9. Subplans on the Wiki.
  • Dani: subproject merge - wanted to wait for M1, but should be done this week
  • Alex: ppc64be - Sravan is waiting on final decision whether it will be removed in Photon
    • Dani: PMC decision was settled, so removed from Eclipse plan - IBM might still continue building it for internal use

August 1, 2017 - no call

July 25, 2017 - Alex, Lars, Dani

  • project merge got go, Dani to implement it with webmaster and then send out the announcement this or next week. E-mail will also mention that existing project leads can become co-leads on Eclipse Platform
  • Discussion if CVS should be removed from the SDK build, Alex +1, Lars +1, Dani +1
  • Discussion if the remaining 2.0 compatibility layer can be removed, Dani: OK to remove but other plug-ins might still have dependencies to it, must the also updated, as well as the documentation

July 18, 2017 - McQ, Alex, Lars, Dani

  • project merge review awaiting go from Wayne - Dani pinged again in the bug report
  • Dani will kick off planning for Photon a bit earlier since he will be in Bangalore the next 12 days. Platform UI already started.
  • M1 on August 11

July 11, 2017 - no meeting

July 4, 2017 - no meeting

June 27, 2017 - McQ, Dani, Martin, Alex

  • Equinox Merge to the Eclipse TLP
    • Keep Equinox as a separate Project, to allow independent Equinox releases without Platform (and keep Bugzilla simple)
    • The only argument for stronger integration would be getting more committers, but there is no strong need and we have the FastTrack process in place for that
  • Eclipse Merge of Component (Sub)Projects into Platform
    • Just 4 Projects: Platform, Equinox, JDT, PDE (plus the incubators for e4 and JDT)
    • Former (Sub)Project Leads to become "Component Leads", "Technology Owners" or "Repository Owners"
      • Repository Ownership is appealing as the structure is very clear (+1 Dani,Alex,McQ)
    • Platform Lead initially Dani, more Co-Leads could step up if they want
      • Main work is helping component leads if they can't come to a consensus - usually less work than PMC duties
  • Dani: JSR 376 Ballot
    • Public review reconsideration ballot got approved by all except Red Hat who abstained
    • Much progress in the past weeks, consensus except RH abstained
    • Looks like the OOBE of Java9 would tolerate reflection, currently looks like on track for a September Release
  • Vacations many people will be off next week, especially in the U.S.
    • PMC call will be skipped next week

June 20, 2017 - MQ, Dani, Lars, Alex

  • Attracting new contributors discussions - code cleanup and deletion of old paths needed
    • Alex: Big issue to get people working on codebase and considering many now irrelevant paths in the codebase
    • Dani: Cleanups should be done by the one starting them through the whole SDK
    • Lars: Not feasible as some projects are slower pace moving and the change might require more effort than one can or is willing to spend in his free time
    • Alex: Great opportunity to have new people joining by doing simple tasks and grow them into full committers
    • McQ: We have to come up with list of items we would like to get cleaned and properly follow our deprecation policy even if the migration guide for Photon get really big with things we schedule for removal in the future.
  • Dani: Sub-projects merge
    • Name of the new project - McQ: Everyone refers to it as Platform so natural choice it is
    • JDT, PDE, Platform (subprojects) to become single project
    • Person responsible for some code area in the merged project - to be further dicussed
  • Dani: Dropping Xulrunner support - PMC to approve it in the bug

Juni 13, 2017 - MQ, Dani, Lars

  • Dani: Release review for 4.7 was successful
  • Dani: Rebuild for 4.7RCa required due to new EMF version and a fix for a critical bug which resulted in duplicated menu entries
  • Merge of the sub-projects
    • Planned for M1, desired to be coordinated with the move of Equinox to Platform
    • We prefer to keep the project pages and wikis so that all the information is kept
    • Preferable we want to have one committer list for Eclipse platform
    • Dani to check with foundation how the merge can be done

May 30, 2017 - Martin, Dani, Lars

  • Dani: Please approve or disapprove the release review
  • Dani: Starting the Debugger doesn't work always
    • Occurs for Lars, but isn't reproducible - one of the top 3 things to fix if steps to reproduce are found
    • Lars: Equinox replaced the DS implementation in 4.7, the issue is probably related
    • Failure is totally random: On a very slow Virtual Machine, it failed almost always (9 failures out of 10); on other machine it works most of the time; even after working fine 3 times, it may fail the 4th time; seems to occur more frequently on older machines than newer ones.
    • Martin: try with a reverse debugger like Chronon ?
      • Should help by just collecting a log; and when the failure occurs, "play back" the log to understand under what condition the race condition occurs; the a reproducible case is available, and the log can also be sent around between different developers for investigation, potentiall also to Apache Felix upstream
  • Dani: Equinox Launcher Issues
    • During M7, a change was made to ignore certain command-line arguments
    • This caused 2 regressions - bug 517013 and bug 516349 MacOSX restart regression
    • Java 9 will probably change the parameters again, see bug 516911 :(
    • It's getting too late to make any more changes to the Launcher. Re-compiling on different hardware has already caused bug 517013, it's possible that more issues (not yet known) might be caused by the rebuild. With Java 9 probably changing again, we don't win much by rebuilding to try and align with something that's not yet released.
      • Option A: Put options into the eclipse.ini -- today it works for both Java 8 and Java 9
      • Option B: Add the "probably proposed" new option to the launcher, will need to rebuild the launcher again
      • Option C: Revert the launcher to what it was before (and well tested) back to M6 - for running with Java 9, the options have to be added manually
    • Users will have to read on a webpage what needs to be done for Java 9 launching; for Eclipse SDK, a single option is needed - some add-on components
      • Additional complexity: Different JVMs need different options (IBM, Oracle...)
      • Martin: bug 517452 Introduce a fallback mechanism to the launcher, which reads the Java ID and loads 'eclipse-javaID.ini' with a fallback to 'eclipse.ini', but does not seem reasonable for Oxygen since too late
      • Better to have a clean story: At IBM, the entire build environment changed ... reverting gets us back to a bullet-proof launcher that was tested for 7 milestones ... will work safely with Java 8, for Java 9 people will have to read the N&N or StackOverflow to find out about the right arguments - same as in Neon
  • AGREEMENT to revert the Launcher to the M6 version

May 23, 2017 - Alex, Martin, Dani, McQ

  • Dani: Apache Batik XXE Vulnerability
    • Alex: Moving to Batik 1.9 would be good, but there is so much UI breakage that it's not viable for Oxygen
  • Martin: Bugs discovered during RC test cycle
    • Surprised that some low-priority issues do get attention (bug 517108) while others get no response at all (bug 517013
    • Dani: Only interested in severe issues and regressions - send message to platform-releng-dev with severe regressions found
    • Dani: Please +1 the Gerrit for bug 517108

May 16, 2017 - Alex, Martin, Dani, Lars

  • Plan Update - updated on Wiki, main plan updated, please provide feedback until tomorrow latest:
    • Has Updates to delivered / not delivered, and Target Operating Environment
  • bug 509922 Performance Tests: Header is wrong, but it really compares against 4.6.1: results are created but report is wrong
    • Code should have been improved through many action enablement changes and startup improvments
    • Sravan will work on making a proper comparison against 4.6
  • Test failures: Some in Platform/UI fail frequently, the most frequent ones should be considered for fixing
    • Browser tests: pick "an available" server, the M7 run was on a newly provisioned server which didn't have all the stuff for Webkit. This should be fixed now.
    • CVS tests fail from time to time since the CVS server isn't stable at the foundation
    • UI tests may fail if Mozilla or similar show a pop-up dialog requesting an update
  • bug 516114 Tabbed Properties View Styling Change
    • Due to the code change, the Tabbed Properties View is now loaded even if it's not used; that may degrade performance
    • The issue is due to how styling is architected; an OSGi flag activates the plugin if any class is "used", the CCS engine just checks for a class without even instantiating, that already triggers the bundle activation
    • That should be discussed for change in 4.7.1 or 4.8 ... maybe use reflection to avoid this, or avoid the need for eager activation of tabbed properties ... Lars did that in the past, but it's a lot of effort
    • Activation by itself would be relatively cheap, but then the code goes through all Singleton instantiations - hard to estimate
      • The impact is likely small compared to PDE for example, which takes 1.5 seconds out of 6 seconds startup time
    • AGREEMENT: In order to be fair to other people who worked on improving startup performance, the change should be reverted. We have a change that benefits few people, but impacts everyone here.
      • Lars: OK to revert, but in the next cycle we should look at improving activations in general. Opening the Resource perspective should not activate PDE + JDT (today, this is likely triggered due to some decorators).
      • McQ would love to see an "optimistic UI" which can render at least the shell even if not all menus/etc are yet filled in
  • Dani is out Thu-Sat, please jump in and help the team if something is needed
  • Alex: Project Structure
    • Platform/Resources doesn't have a lead at the moment. Would like to merge more and more into Platform/UI. It doesn't make sense keeping separate structures just for the sake of creating barriers to contributions.
    • Eventually we'll just have the Platform/UI morph into "The Platform Project" with more co-leads
    • In the end it's not about positions and authorities .. it's all about people who are good
    • Dani: Agrees; in experience, every 2nd contribution is causing some issues, but we do have good people even if they are not official leads
    • McQ: What we really need is high-level overview and authority to recognize changes that might negatively impact others..
    • Could eventually end up with JDT/PDE/SWT/Platform. Lars and McQ are not so sure any more if SWT really needs to be separate, had several overlapping contributions - 9 active SWT committers also work on Platform code. Consider moving forward with this after the Oxygen release...
    • AGREEMENT to move forward on this, with a Platform Leadership group. We need to advance to leadership of Eclipse along with advancing the code. Leadership change should also lead to growing community.

May 9, 2017 - Alex, Martin, Dani, McQ

  • Dani: JSR376 Jigsaw: 10 said yes and 13 said no; a new spec to be provided within 30 days
    • From a technical perspective, should be possible to resolve within 30 days
    • JDT team implementing a compiler; IBM implementing a JVM, that's completely separate concerns!
    • For implementing a compiler, the last spec is quite old, that is not good enough
    • JDT team votes via Mike Milinkovich (Eclipse Foundation), Mike considered JDT team and RH team and probably others
  • Dani: M7/RC0 - test passes upcoming; compressed scheduled due to the moved date for Devoxx
    • When using the RC, watch out for 2 things:
      1. Action contribution enablement fix -- look out for menu and toolbar items not properly enabled
      2. Section forms layout
  • Dani: httpcore 4.4.1 -> 4.4.6 ECF Update and API Compatibility
    • Annotations got removed - no binary breakage, but breaks source compatibility
    • People will prefer newer httpcore; API breakage due to annotations is unlikely
    • Also, source breakage gives developers a chance to just recompile, so accepting the new version is better
    • Alex: Are we forcing our dependencies to not break API? - We don't have that power anyways?
    • Alex: httpclient has security issues every few months ... better consume the latest now, and we'll likely have to update again. So if breaking compatibility, better do it now
    • AGREEMENT to do the update and thus accept the source breakage.
  • Flooding in Ottawa: Restricted to a fairly small area
  • Dani: Release Review, does anybody look at the full document or can it be simplified?
    • All of the information in the RV is public (eg bug statistics, # contributors, ...)
    • McQ: It's useful for consumers to see that new things are happening, that is the N&N; beyond that, the use is limited. Effort around half a day.
    • Many bloggers talk about new things anyways
    • Migration Guide and N&N will get into the RV Material; beyond that, Dani offers adding stuff that people request
    • AGREEMENT to keep it simple, and send add-on requests to Dani if they come up.

May 2, 2017 - Dani, Lars, Martin, Alex, McQ

  • Lars: Regex Evaluator Plugin
    • Basically a view where one can input a regex and text, and see if it matches
    • Would like to add to e4 incubator, goal to eventually add into Platform/Text or similar
    • McQ: Where it the boundary between e4 and Platform? How do people kick off little projects .. e4 incubator?
    • AGREEMENT to move code to Eclipse.org (into the incubator) and proceed from there. Might not end up in Platform, but will be good having the code at Eclipse.org.
  • Dani: Endgame plan sent - Oxygen looking good, not too many bugs open
  • Dani: JSR Jigsaw Ballot
    • Asking for finalizing the Specs on JSR376 before voting Yes, since otherwise life is hard

Apr 25, 2017 - Dani, Alex, Martin

  • Dani: Neon.3a
    • Respin done -- problem: Updating from Neon.3 to Neon.3a , the problematic bundles are not removed
      • AI Alex tell Jeff to ask Pascal or Tom about how to disable the bundles on update
      • Martin: If this problem can't be resolved could live with it ... better not take too much risk trying to resolve this very special problem

Apr 18, 2017 - Dani, Lars, Martin, Alex, McQ

  • Dani: RH and Java9 Jigsaw
    • Very good write-up linked from the AC Mailinglist - see also follow-up
    • Eclipse JDT team concerns that the Spec is not complete brought forward to Wayne
  • Dani: Neon.3 Respin
    • Last action was with Ed Merks to validate that the Fix from Tom Watson actually works - no update so far
  • Lars: Glance Migration to e4 and/or Platform
    • e4 incubator was always intended to have very low entry barrier - can give the Glance team access quickly
  • Lars
    • JUnit 5 licensing - EPL + Apache

Apr 11, 2017 - McQ, Lars, Alex, Dani

  • Dani: Update from last Planning Council call:
  • Dani: Asked Lars to review the Dark theme bug. Lars, said, he's not a user of it, but has one in his team. Dani: Ask him to review the change.

Apr 04, 2017 - Dani, Martin, Alex

  • Martin: bug 514257 and cross-project Neon.3 wiring issue
    • Problem: External libs don't deal with versions the same way we do
    • Dani thinks that the problem is that Release Train projects don't agree on one version to use
    • Martin saw a similar problem related to some bundles doing import-package and others doing require-bundle:
      • Commons Logging is not a Singleton
    • Dani has a call with Tom later today; hoping a respin can be done after fixing in Equinox
  • Dani: Heads up on upcoming Planning Council discussion re: Java 9 and Junit 5
  • Dani: PPC - No response on McQ's message regarding PPC, Dani will update the Plan after another week

Mar 28, 2017 - McQ, Alex, Lars, Dani

  • Replacement for Sergey
    • let's wait and see how things go
    • next candidate would be Tom Watson once Equinox is moved back to the Eclipse top-level project
  • Dani: PPC strategy: drop 32-bit, provide 64-bit with Oxygen (4.7) and drop 64-bit BE with Photon (4.8)
    • all agree
    • McQ will send a note to cross-project-issues-dev
  • Lars will nominate Mikaël Barbero for Platform UI
  • Lars plans to bring a IDE search plug-in to e4 (http://ystrot.github.io/glance/)

Mar 21, 2017 - Dani, Martin, Alex

  • Dani: Dirk Fauth Requests on eclipse-pmc
    • Alex agrees with Dani
    • Where does the "0.x version before graduation" requirement come from? - Development Process, but maybe didn't apply to plugin versions
    • Dani thinks that forcing a 0.x version causes forced breakage once something goes API thus doesn't make much sense
  • Dani: Szymon Ptasiewicz

Mar 14, 2017 - Dani, Martin, Alex

  • Dani: Java 9 readiness - see cross-project-issues-dev
    • Step 1: For testing, just download a Java 9 beta JVM, add the module -- all testing is possible from commandline
      • Platform team has an install, but Mike was concerned about legal issues - a bug is open to allow all simrel access
    • Step 2: Running the JDeps tools as per the Java 9 Readiness wiki can show violations
      • Can even run with Java 8, Dani ran it on simrel, it's looking mostly good, only Orbit shows violations
      • Platform team has some scripts; for Hudson there's even a Mojo, but it aborts when it finds a violation.
    • Step 3: Reflection - Will only figure out when knowing the code or testing
    • All these 3 tests can be done without the Java 9 patch in PDE. Only for debugging, one needs to self-host.
      • This can be done either with Ed's Oomph support, or via the Marketplace
    • Platform team is still negotiating options for better support with Oracle
  • Dani: JUnit 5 considers cutting down features for aligning with Oxygen
    • Decision to align, contributing support to Oxygen Update 1 in September
  • Dani: 2 Platform Issues in M6
    • platform.ui reexports everything including jface - lower bundles not updated - JDT may fail if installed (fixed for M7)
    • major version uprev: interface not marked as @noimplement; reverting the version, people who loaded Friday's M6 would never get any updates any more; decision to stick to the version uprev, since the respective bundle is not used a lot
      • In general, accidental major version uprev should be reverted because the error would stay around for long and might break existing binary bundles who can't be fixed any more
      • In this case, it was OK to keep since the bundle isn't used much
  • Alex: Generic Editor and Testing
    • Looking very good already - only 1/3 of the code needed for getting the same feature set as with existing methods for creating editors
    • Looking for features to add missing documentation (workflows, use-cases like converting existing editors): End goal is to provide everything that the JDT Editor is providing
    • Dani: Likes replacement of pages on multipage editors, and some new editors; would not consider replacing the JDT editor
    • In the Platform.ISV docs bundle, there should be a section for "how to write your own editor"

Mar 7, 2017 - Dani, Martin, Alex, Sergey, Lars

  • Dani: 4.6.3 Updates - Looking good, Equinox bugs moved out by Tom
  • Dani: 4.7M7 Update - Infrastructure issues but testing looks OK (except Mac which is not built yet)
  • Dani: Java 9 / Planning Council Update - Postponing Oxygen was discussed but denied; late July update being considered
  • Lars: Platform.Resources Leadership after Sergey Leaving
    • Dani pinged Szymon, will continue discussion when we hear back; merging with platform.ui is one option
    • Sergey: The "Team" and "Resources/Filesystem" parts could be considered separately. Consider merging Team into UI.
    • Alex: One top Platform Project would make sense now...
    • Dani: reconsider when McQ is back. In the past, SWT was seen as separate.
    • Dani: Releng needs to be separate and protected because Releng committers can also change Hudson jobs, shut down servers etc where no history is kept

Feb 28, 2017 - Sergey, Alex, Martin, McQ

  • Alex: Short update on Neon.3
    • 2 jdt.core bugs asking for PMC approval - regressions compared to Neon.2, approved
    • PPC rebuild is in, so looking good
  • Sergey will leave Eclipse work after end of March
  • Martin: RCPTT / Reddeer / Testing

Feb 21, 2017 - Dani, Alex, Martin, Lars, Sergey

  • Lars: bug 512273 Allow any committer to retrigger Gerrit validation, and bug 512319 allow rebase
    • Agreement to allow retrigger for anyone, but rebase should be done by committers working on a contribution
    • Dani: Not clear where to retrigger (Sergey: it's discoverable once permission is there)
    • AI Lars follow up on the bug
  • Martin: UI Testing
    • RCPTT is completely Open Source; commercial server could be used for load balancing tests but this is not necessary
    • Alex: Interested in Reddeer for upstreaming tests that already exist at RH; good reports about stability and scalability at RH
    • When we decide which way we go, we need a plan who's doing the work (create the tests). Maintenance would depend on their structure.
    • Some Reddeer tests already upstreamed with WTP and Linuxtools
    • Lars: Where would the tests live? - In a common project accessible to all, or with the component they test ?
      • Lars, Dani: Since these are functional end-to-end, should live in their own repository, accessible by all committers.
    • AI Martin follow-up with Jubula. Try RCPTT on Windows and Linux. Give feedback till next week.
    • AI Alex share list of Reddeer tests existing at RH
    • AI Lars play with the Reddeer API
  • Dani: Update on PPC drop 32-bit, build 64-bit both le and be on RH VMs bug 512224
    • Alex: ppc64 currently built on Fedora 25; could build on REL7 if foundation requests a machine from RH Brno Farm AI Alex and Dani ask Denis
  • Dani: Neon.3 RC3 please keep an eye on builds
  • Dani: Away this Friday afternoon and whole next week

Feb 15, 2017

Feb 8, 2017 - Dani, Alex, Martin, Lars

  • Dani: bug 509412 - Adopt newer JSch for Platform 4.6.3
    • Alex is fully consumed with Webkit work for Oxygen
    • Need to move target platform to the latest Orbit in order to pick up the new JSch; that might cause other changes, since Orbit retention policy keeps only one version
    • Dani: Cherry-pick the new JSch recipe to the Orbit Maintenance Branch ?
    • Martin: Looks like in the past, new Orbit R-Builds were made for Maintenance ... but with the move of Orbit to git, it is unclear how branch builds for Neon.x would be made
  • Dani: Update on RT PMC (Equinox) coming to Platform
    • Tom Watson is in favor, reached out to Wayne for process
    • Potentially move entire IP log, but IP team is currently busy with CQs for Oxygen
  • Dani: Finally Unblocked on Java Language Server
    • Dirk from Microsoft signed the committer agreement
  • Dani: Scenarios for UI Testing
    • Initial Scenario from the PMC Notes can already be used; once that works, could extend to more
  • Dani: Sign Up for Security Rep
    • See Wayne's message to PMCs requesting PMC attendance
    • John Arthorne used to be the Security Rep;
    • Tasks: Follow the security mailing list; currently low volume but might get higher with IoT getting up speed
    • RESOLUTION: Alex agrees to sign up since already following security issues
  • Dani+Alex: libswt-gtk3 natives for PPC
    • Alex: Trying to limit the list of supported GTK versions to maximum 2; now at the latest GTK2 version
    • Currently, only Intel builds do both GTK2 and GTK3; PowerPC only does GTK2; the current hardware is unable to install REL7
    • Would prefer building at the Foundation over using custom company hardware; cost of hardware is around 50K
    • Primary Platforms should be built at the Foundation, secondary builds can be contributed ... should Linux-PPC be considered primary ?
    • Alex could easily provide a Fedora Platform, but it would be much newer than anything else that's around; ARM32 and ARM64 builds are already provided by Fedora
    • If somebody (IBM) needs an older version like CentOS, it would need to be provided by such adopters
    • AI Dani+McQ find out on priority of Linux-PPC (32bit, 64bit)
  • Alex: Update on Reddeer Testing Framework
    • A new version is being written, that no longer uses SWTBot underneath; there's a plan making this an Eclipse project
    • Engineers claim that after the rewrite it's a lot more stable recognizing widgets than on top of SWTBot
    • Martin has been looking at RCPTT which looks promising initially (easy to record tests) but unclear how stable such tests would be.
    • (call had to be closed at this point)

Feb 1, 2017 - cancelled

Jan 24, 2017 - Dani, Lars, Martin, Alex

  • Dani: New Jetty Version - Alex will look at it after FOSDEM (Feb 7)
  • Dani: javax.xml - Plan to announce removal from the feature on cross-project for M6
    • Cannot remove from the compile prerequisites, because plugins still on Java 5 need it
  • Alex: Reddeer Testing Framework
    • JBoss Tools uses Reddeer for overall user story validation
    • Lars: Why wasn't the framework added to SWTBot? -
      • Reddeer is not a framework, it's meant as a test harness library - smaller scope than SWTBot or Jubula
      • API Stability across multiple Eclipse versions is out of scope, thus created outside Eclipse for now
    • Martin: Any improvements in object recognition compared to plain SWTBot?
      • Reddeer tries to make use of the best mechanisms from SWTBot for object recognition (Finder, or callbots). That saves users from making newbie mistakes in SWTBot, but apart from that it's no better
      • No solution for external (non-Eclipse) dialogs ... that's not doable in SWT
    • Lars: EclipseSource people have been talking about RCPTT, it might be able to deal with native tooltips etc
    • Dani and Lars think Jubula is too heaviweight; might work for Integration Builds, but not for individual developer's builds. Local setup must be really easy. Also, setup looks non-trivial, needs resources to work on ... and, doesn't seem to add much benefit over SWTBot (assuming that Jubula agent doesn't hook into native libs like win32, Cocoa or GTK).
      • Alex: Especially integration in Maven may be hard.
    • Alex: Main limitation is that we don't have resources to actually create tests. Suggest reaching out for help asking people who would actually create tests ... then use whatever framework that people would like to use.
    • Dani and Lars won't be able to come up with a workflow scenario before end of next week --> Alex will come up with an idea since he has meetings on Monday.
      • Dani: Launch Eclipse, choose a workspace, open the package explorer, create a Java project (helloworld). Expand later, maybe into quick assist.
      • AI Alex to demo Reddeer in 2 weeks
      • AI Martin to try installing Jubula until next week
      • AI Martin ping Sergey re: asking for help on Platform/Resources.

Jan 17, 2017 - Dani, Sergey, Alex, Lars, Martin, McQ

  • Sergey: Merging platform.resources with platform.ui
    • Seems that more contributions to platform.resources are made by platform.ui committers, than others. Only 2 platform.resources committers seem active (and those two are actually platform.ui committers as well)
      • Opening up to Platform UI would encourage more people to actually contribute to resources. Strive for shared access on components that are interrelated
    • Dani: Merging into platform.ui would mean making Dani and Lars the leaders for signoff - that causes too much burden
    • Sergey: Platform/Team has actually more synergy with Platform/UI than resources
    • Alex: For gcc for example, there is a notion of "maintainers" per git repo for signoffs - that's independent of leadership or commit rights
      • McQ: Not sure how adding another role would simplify things ?
    • Sergey: The problem for signoffs is not is much the volume of commits, but the volume of failing tests
    • Alex: The manpower of Platform Resources is so low by now that even bug 509412 was not being filed ... existing committers do not really seem to "own" the project any more, this is an indication for merging
    • Martin: Can't see how merging into a bigger entity would resolve the "ownership" problem. Perhaps make interested contributors Platform/Resources committers instead, and see how to resolve the ownership issue on the smaller scope ?
    • McQ: Propose applying the recent simplification of gaining commit rights (for "known" committers)
  • Lars: Removing javax.xml
    • Was introduced only for very old JRE's , it is part of the JRE now
    • Martin: Might cause a uses constraint issue since not exporting the package with version any more, see http://njbartlett.name/2011/09/02/uses-constraints.html
    • Lars: With javax.annotation, that was solved with smart reexporting
    • McQ: OK with removal when we can confirm we're not breaking things
    • Dani: Whoever requested removal should double-check whether the API is the same.
    • AI Lars send a note to Tom Watson as the submitter.
  • Dani: Project Updates
    • Deadlines upcoming for CQs and APIs - see Endgame Plan, please provide feedback
  • McQ: UI Testing and Jubula

Jan 10, 2017 - Dani, Lars, Martin, Alex, Sergey, McQ - Special Guest: Alex Schladebeck

  • Dani: Platform Bits on Maven Central - Now available thanks to Stephan Herrmann
  • Dani: Board Committer Rep Elections opening
  • Dani: After David Williams' move, Wayne is looking for a Planning Council Chair; Fred Gurr has taken over Simrel Releng
  • Dani: objenesis (Mockito / Easymock dependency) and Java9 - will wait for how upstream is planning to address issues
  • Lars: Equinox move to Platform - no updates - AI Dani talk to Tom again
  • McQ / AlexS: Jubula
    • In the past, with lots of full-time Platform committers, sniff tests just happened automatically giving great confidence. Now, with more part-time work happening, there is more need for end-to-end workflow test automation to give quality confidence
    • Martin: On top of low-level JUnit tests, want at least very few highlevel end-to-end test to avoid "Eclipse looking silly". Need UI tests that are really stable (avoid random failures)
    • Dani: On EGit SWTBot tests, too many libraries to install, how to run on Gerrit HIPP slaves
    • AlexS: Original Jubula goal was to allow non-programmers write tests. Beginning 2015, the client API (for controlling the SUT) was separated from the front-end, was separated from the ITE. This allows writing UI tests programmatically, and avoids the need for a database.
      • For identifying components/actions to write tests, officially an object mapping needs to be done in the front-end (ITE, integrated test environment). But there may be a shortcut available for getting object IDs
      • Compared to SWTBot, Jubula is 90% blackbox (code highlevel actions rather than things like calling setText())
      • UI Tests typically take longer than unittests by definition (due to setup, teardown, ...), so might not be applicable for Gerrit triggers
      • Dani: Would like to run a first test on CentOS first, as part of the integration tests
      • Lars: Robustness of the tests? - SWT on Mac is problematic, but Linux and Windows should be OK.
      • Martin: Dependencies? - Around 5MB Libs, plus the agent needs to be installed (around 100MB). Once that is there, tests are stand-alone on a laptop. There might also be an "embedded agent" but not sure (AI AlexS to check)
      • Getting a first test up and running ? - Maybe around 1 week, depends on what to do ?
        • Lars: Would like something for quick access. Martin would something for "fresh download" (fresh config area + workspace) maybe combined with quick access.
      • AI AlexS send slides
      • AI Dani + Lars propose scenarios. Check possible owners of the initiative on Platform/UI side
      • AI AlexS check possible PlatformUI Contributors from Jubula team to get started

Archive

Retrieved from "https://wiki.eclipse.org/index.php?title=Eclipse/PMC&oldid=421235"

Back to the top