+1. I've checked a source file with SHA512 sum:
92d2283b10f6952cc070a1dcdd3a35c80f34f77cee1e8d72c5a957554fbfc62af28aca49c8fd173a3ebb6f8c8751dee768593b2202dbe23a1a1a5cd74b04c8c4
Looked at:
-checksum, key (see below)
-rat - looks OK to me.
-builds with
mvn package -DskipTests
-with:
maven install tests fail with:
Tests run: 160, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 27.101
sec <<< FAILURE! - in TestSuite
JSONTest.putPeopleUsesRightMethod(org.netbeans.html.presenters.browser.KOScript)
Time elapsed: 10.182 sec <<< FAILURE!
java.lang.InterruptedException: null
at
net.java.html.json.tests.JSONTest.putPeopleUsesRightMethod(JSONTest.java:366)
at jdk.internal.reflect.GeneratedMethodAccessor1.invoke(Unknown
Source)
at
java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:566)
at
org.netbeans.html.presenters.browser.KOScript.run(KOScript.java:83)
but I am OK with that.
I didn't check the convenience binaries.
On the process side, I think there are two things to fix:
-your (Jarda's) key should be here:
https://www.apache.org/dist/netbeans/KEYS
otherwise ordinary users will not be able to verify the signature (I used
Jarda's key from Whimsy)
This can be fixed together with moving the release to the final location.
-next time, the vote e-mail should explicitly list the SHA512 sum of the
release, otherwise it is not clear on which file the vote is taking place.
Jan
On Mon, Nov 4, 2019 at 7:11 AM Jaroslav Tulach <jaroslav.tul...@gmail.com>
wrote:
> Hi.
> As my first attempt to accommodate the recent donation of various
> presenters by DukeHoff GmbH, and release 1.7 of Apache NetBeans HTML/Java
> API failed with a NOTICE file error and a bug in WebKitPresenterTest, I had
> to prepare new bits. Please find the source at
>
>
> https://dist.apache.org/repos/dist/dev/netbeans/netbeans-html4j/netbeans-html4j-1.7/
>
> and the staged Maven artifacts at
> https://repository.apache.org/content/repositories/orgapachenetbeans-1052/
>
> Please evaluate them again and cast your vote during next 72h. The process
> of evaluation is the same as last week.
>
> What can you verify?
>
> ## checksums, signatures, etc.
>
> Obviously those should be correct.
>
> ## if it builds
>
> I used "mvn package -DskipTests" and I believe the compilation shall
> succeed on JDK8, JDK11, JDK13 (Eric reported problem building on JDK13, but
> it works for me).
>
> ## if the tests pass
>
> Use "mvn package". They may not - especially the newly donated modules are
> very sensitive to surrounding environment. Getting them run on all
> important platforms may take some time.
>
> If you want to report your findings, then please include the version of
> operating system, version of JDK, version of the browser, etc. You may also
> want to run them individually. E.g. "mvn -f browser test" or "mvn -f webkit
> test".
>
> I know that browser doesn't seem to run on Windows 8 with Internet Explorer
> (but it runs with Firefox instead). I know that webkit does run on Ubuntu
> 18.04 with proper libraries ("libwebkit") installed. But those libraries
> aren't present in Ubuntu 19.10 - there is "libwebkit2" and somebody would
> have to adjust the code to use the "libwebkit2" code. In short, there are
> many broken things in the browser and webkit modules - volunteers welcomed
> to make them run!
>
> ## How to cast your vote?
>
> I propose the goal of 1.7 release to be to compile and execute its tests
> (except browser and webkit modules) on all supported platforms. We should
> collect the test failures in browser and webkit modules and find some
> volunteers to fix them for some upcoming update release.
>
