You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@jena.apache.org by Andy Seaborne <an...@apache.org> on 2017/10/27 13:53:11 UTC
Jena 3.5.0 RC2 plan
There are test problems with TestProcessFileLock (on windows,
consistently) and TestDatabaseOps (intermittent but seems to like
picking on Bruno).
There a couple of other small PRs for fixes which look safe as well.
PR #294 AdapterFileManager fix
(Rob - JENA-1405 can be resolved?)
PR #297 Elephas testing speed up
(slight discussion ongoing about details)
TestDatabaseOps:
PR #295 Use @Rule in testing to isolate tests.
PR #296 Control the test in disk-touching DBOE modules.
TestProcessFileLock:
PR #298 Isolate tests witha @Rule and improve/fix lock release.
I don't intend to rebuild the javadoc. As you may have noticed, I had
some "fun" trying to get it staged. This cutting corners but no public
API is changing. If necessary, it can be done after pushing the release out.
Master is at version 3.6.0-SNAPSHOT so it goes back.
(and switching a unified versions number makes that easier)
To smooth the reset, mainly for any of us who have switched to the
latest snapshot, I'll apply the PR's, we can test the snapshot, then
reset back to 3.5.0-SNAPSHOT just before the RC2 release build.
Please don't push to master without also letting dev@ know to make sure
I don't miss anything.
Good plan? Anything missing?
Andy
Re: Jena 3.5.0 RC2 plan
Posted by aj...@apache.org.
Back from Vienna!
Master just built beautifully for me on Mac OSX, from commit 92c793b67dbb4138858106774d57b23418dd4ae5.
ajs6f
Andy Seaborne wrote on 10/27/17 4:42 PM:
> Currently on master:
>
> All the PRs are integrated (minimal version of #297 - tests are much faster, with a build taking about 12-13 minutes).
>
> It's run a couple of times on Apache jenkins as well.
>
> If anyone gets the chance to run from master on Windows or OSX, that would be great. TestDatabaseOps should be fixed.
> TestProcessFileLock on Windows may well be but IMo is not a blocker to a release - its a test setup issue.
>
> Andy
>
> On 27/10/17 14:53, Andy Seaborne wrote:
>> There are test problems with TestProcessFileLock (on windows, consistently) and TestDatabaseOps (intermittent but
>> seems to like picking on Bruno).
>>
>> There a couple of other small PRs for fixes which look safe as well.
>>
>> PR #294 AdapterFileManager fix
>> (Rob - JENA-1405 can be resolved?)
>> PR #297 Elephas testing speed up
>> (slight discussion ongoing about details)
>>
>> TestDatabaseOps:
>> PR #295 Use @Rule in testing to isolate tests.
>> PR #296 Control the test in disk-touching DBOE modules.
>>
>> TestProcessFileLock:
>> PR #298 Isolate tests witha @Rule and improve/fix lock release.
>>
>>
>> I don't intend to rebuild the javadoc. As you may have noticed, I had some "fun" trying to get it staged. This cutting
>> corners but no public API is changing. If necessary, it can be done after pushing the release out.
>>
>>
>> Master is at version 3.6.0-SNAPSHOT so it goes back.
>> (and switching a unified versions number makes that easier)
>>
>> To smooth the reset, mainly for any of us who have switched to the latest snapshot, I'll apply the PR's, we can test
>> the snapshot, then reset back to 3.5.0-SNAPSHOT just before the RC2 release build.
>>
>> Please don't push to master without also letting dev@ know to make sure I don't miss anything.
>>
>> Good plan? Anything missing?
>>
>> Andy
>>
Re: Jena 3.5.0 RC2 plan
Posted by Andy Seaborne <an...@apache.org>.
Currently on master:
All the PRs are integrated (minimal version of #297 - tests are much
faster, with a build taking about 12-13 minutes).
It's run a couple of times on Apache jenkins as well.
If anyone gets the chance to run from master on Windows or OSX, that
would be great. TestDatabaseOps should be fixed. TestProcessFileLock on
Windows may well be but IMo is not a blocker to a release - its a test
setup issue.
Andy
On 27/10/17 14:53, Andy Seaborne wrote:
> There are test problems with TestProcessFileLock (on windows,
> consistently) and TestDatabaseOps (intermittent but seems to like
> picking on Bruno).
>
> There a couple of other small PRs for fixes which look safe as well.
>
> PR #294 AdapterFileManager fix
> (Rob - JENA-1405 can be resolved?)
> PR #297 Elephas testing speed up
> (slight discussion ongoing about details)
>
> TestDatabaseOps:
> PR #295 Use @Rule in testing to isolate tests.
> PR #296 Control the test in disk-touching DBOE modules.
>
> TestProcessFileLock:
> PR #298 Isolate tests witha @Rule and improve/fix lock release.
>
>
> I don't intend to rebuild the javadoc. As you may have noticed, I had
> some "fun" trying to get it staged. This cutting corners but no public
> API is changing. If necessary, it can be done after pushing the release
> out.
>
>
> Master is at version 3.6.0-SNAPSHOT so it goes back.
> (and switching a unified versions number makes that easier)
>
> To smooth the reset, mainly for any of us who have switched to the
> latest snapshot, I'll apply the PR's, we can test the snapshot, then
> reset back to 3.5.0-SNAPSHOT just before the RC2 release build.
>
> Please don't push to master without also letting dev@ know to make sure
> I don't miss anything.
>
> Good plan? Anything missing?
>
> Andy
>