Hi all,

+0 (binding)  

would have given a -1, but I'm unsure if this the problem in building is 
sitting in front of the keyboard or if SBT is just not as mature as Maven or if 
there's actually a problem in the project.

Source Bundle:
[OK] Signatures match
[OK] Signatures by apache email address
[OK] Hashes match (Both 256 and 512 versions)
[OK] Source bundle unpacks correctly
[OK] Unpacked sources contain NOTICE, LICENSE, README.md and DISCLAIMER
[OK] Year on NOTICE file correct, LICENSE file correctly lists up the included 
passera and UniquenessCache.scala code
[NOT OK] When running "sbt compile" actually nothing was compiled, when running 
"sbt -mem 6144 test" it compiled however at least one unit test failed: 
                [error] (daffodil-test-ibm1 / Test / test) 
sbt.TestsFailedException: Tests unsuccessful 
                  When re-runnning the "sbt -mem 6144 test" command, the build 
is sort of hanging for more than 40 minutes now ... will abort and re-try a 
third time ...
                  Had to do a "sudo sbt clean" as I was getting permissions 
denied from being a normal user
            When re-running the original build I now got script errors ... so I 
deleted the entire directory, unpacked it again and re-ran the build.
            Unfortunately same error:
                [error] (daffodil-test-ibm1 / Test / test) 
sbt.TestsFailedException: Tests unsuccessful

Binaries:
[NOT OK] The jars should all contain the DISCLAIMER in /META-INF/DISCLAIMER ... 
it seems they are missing this.
[NOT OKisch] The LICENSE files in the jar files all refer to at least some 
files not contained in the jars themselves (Because the same LICENSE text is 
being used for all artifacts) 
[NOT OK] The jars in Nexus don't contain DISCLAIMER files either.


so after validating your release I do have a few comments:

- SHA512 hashes are the default and as far as I recall shorter Hashes are 
considered deprecated (Perhaps drop the SHA265 hashes)
- Does the README.md suggestion to edit "/etc/sbt/sbtopts" also apply for Mac 
and Windows?
- You seem to bundle quite a number of IDE settings files in the release, is 
this intentional? 
- in the Daffodil-Lib module you bundle some code which is BSD-2, BSD-3 
licensed. You include the license text inside the source directory, in other 
projects we usually have a "licenses" directory containing the license texts of 
bundled software, this makes it easier to find for people.
- A number of the test XML files are missing Apache headers even if they could 
contain them
- The .gitignore could use a header.



Am 08.04.20, 00:24 schrieb "Olabusayo Kilo" <ok...@tresys.com>:

    (Resending due to issue with weird formatting/pipes in previous message)
    
    Hi all,
    
    I'd like to call a vote to release Apache Daffodil (incubating) 2.6.0-rc1.
    
    All distribution packages, including signatures, digests, etc. can be
    found at:
    
    https://dist.apache.org/repos/dist/dev/incubator/daffodil/2.6.0-rc1/
    
    Staging artifacts can be found at:
    
    https://repository.apache.org/content/repositories/orgapachedaffodil-1014/
    
    This release has been signed with PGP key 639637FDA8049411, corresponding
    to olabus...@apache.org, which is included in the KEYS file here:
    
    https://downloads.apache.org/incubator/daffodil/KEYS
    
    The release candidate has been tagged in git with v2.6.0-rc1.
    
    For reference, here is a list of all closed JIRAs tagged with 2.0.0:
    
    https://s.apache.org/daffodil-issues-2.6.0
    
    For a summary of the changes in this release, see:
    
    https://daffodil.apache.org/releases/2.6.0/
    
    Please review and vote. The vote will be open for at least 72 hours
    (ends on Friday, 10 April 2020, 3 PM EST).
    
    [ ] +1 approve
    [ ] +0 no opinion
    [ ] -1 disapprove (and reason why)
    
    -- 
    Best Regards
    Lola K.
    
    
    

Reply via email to