Reprising this email from Vinay: > Begin forwarded message: > > From: Vinayakumar B <[email protected]> > Subject: Re: [VOTE] Tuweni 1.2.0-incubating release > Date: September 27, 2020 at 11:48:51 PM PDT > To: [email protected] > Reply-To: [email protected] > > Apologies for late response. > > I checked, > - Incubating in name - Fine. > - LICENCE and DISCLAIMER exists - Fine > - Checked PGP signatures - Fine (Suggestion: Maybe better give option > to download public keys of release owners in website, instead of users > finding them in public servers) The KEYS file is listed in the download section of the website here: http://tuweni.apache.org/download/ <http://tuweni.apache.org/download/> >Choose a source distribution in either tar or zip format, and verify ><https://www.apache.org/dyn/closer.cgi#verify> using the corresponding asc >signature (using the committer file in KEYS ><https://downloads.apache.org/incubator/tuweni/KEYS>). If you cannot do that, >the sha512 hash file may be used to check that the download has completed OK.
Please let me know if that can be made clearer. > - Checked checksums - Fine (Suggestion: .sha512 file contains just > sha512 hash alone. Validation would be easy using tools (sha512sum > --verify) if the format of .sha512 contents is similar to output "sha512 > --tag <file-name>” ) I looked and gradle doesn’t really allow for this. > - Checked tag in github - I think, directly tagging with release > version name (v1.2.0-incubating) and linking all dists, while still VOTE is > in progress may not be good. Its better to tag with RC labels, not give an > impression that its already released. OK, that’s more work for the person cutting releases, but it’s not the first time I hear this. Adding rc to the tag name with commit 4be12dce. > - Tried building from src tar - Generated libs have SNAPSHOT version in > their name. This might not be correct. I saw that only during > BUILD_RELEASE=true or BUILD_TIMESTAMP=true, SNAPSHOT will be avoided. I > think, on the development version SNAPSHOT should be there by default and > on the released version SNAPSHOT should be removed irrespective of whether > publishing to maven or not. Attended to that with e9e09eeb. > > All above points mentioned are not blockers for this release. But they are > something better to be taken care of for upcoming versions. Thanks for the report, appreciated! > > +1 , > > Thanks, > -Vinay
