What is with everyone's aversion to using the Maven Release Plugin?  I
realize that it may not do exactly what we need out of the box, but it's a
very useful tool.  At home, I push a button in my Jenkins setup and it cuts
a new release to the Nexus OSS staging repository awaiting me to finalize
it.  Can we not do a process like that?  Perhaps we just create our own
variant of the release plugin in commons to do our release process?

On Fri, Apr 15, 2016 at 11:33 AM Benson Margulies <bimargul...@gmail.com>
wrote:

> On Fri, Apr 15, 2016 at 11:29 AM, sebb <seb...@gmail.com> wrote:
> > Thanks!
> >
> > It also occurs to me that having the RC tag in the POM is not
> > necessarily a problem.
>
>
> If you prefer to modify the doc to tell people how to use the plugin
> so that the net result is the RC tag in the POM, OK. That satisfies my
> ask for a set of instructions that an RM can follow without getting
> into a controversy.
>
> All this will be moot if some other voters don't show up.
>
>
>
>
> >
> > So long as the tag is retained after a successful vote, then does it
> > matter whether the tag in the POM is IO-1.2.3-RC4 or IO-1.2.3 ?
> >
> > On 15 April 2016 at 16:02, Brent Worden <brent.wor...@gmail.com> wrote:
> >> I probably won't be much help either as I have never used the plugin
> but,
> >> there is a tag option for the plugin that might help control the SCM tag
> >> that is used.  Of all the options for the plugin listed on
> >>
> https://maven.apache.org/maven-release/maven-release-plugin/prepare-mojo.html
> ,
> >> the tag* options might meet your needs.
> >>
> >>
> >> Thanks,
> >>
> >> Brent
> >>
> >> On Fri, Apr 15, 2016 at 5:16 AM, Benson Margulies <
> bimargul...@gmail.com>
> >> wrote:
> >>
> >>> On Thu, Apr 14, 2016 at 10:33 PM, sebb <seb...@gmail.com> wrote:
> >>> > On 15 April 2016 at 02:19, Benson Margulies <bimargul...@gmail.com>
> >>> wrote:
> >>> >> Sebb,
> >>> >>
> >>> >> I don't know why you think I could distinguish the following
> possible
> >>> >> behaviors of prior RMs with a reasonable level of effort:
> >>> >
> >>> > As I have already said, I don't use the plugin.
> >>> > However I have followed release votes and AFAICR the final tag never
> >>> > been created before the vote completed.
> >>> > In all cases I can remember, the final tag is created once the vote
> >>> > has completed.
> >>> > And I am pretty certain other RMs have used the release plugin.
> >>> >
> >>> > So it seems clear to me that something has gone wrong here.
> >>> >
> >>> > I don't know what, so I think we need input from an RM who has used
> >>> > the release plugin.
> >>> > They should be able to point out what is missing from the
> instructions.
> >>>
> >>> This started when I sent email two days ago saying that I was
> >>> perplexed by the instructions. I only proceeded when no prior RM, or
> >>> anyone else, answered.
> >>>
> >>> Maybe they used -DdryRun=true and then edited. Maybe I'm just being
> >>> obtuse. We'll see.
> >>>
> >>>
> >>>
> >>> >
> >>> >> 1: didn't use the maven-release-plugin
> >>> >> 2: did use the maven-release-plugin and then used svn mvn to rename
> >>> >> the tag it created
> >>> >> 3: did use the maven-release-plugin, typed 'commons-io-2.x-RCy' at
> the
> >>> >> prompt, and released with a pom with an incorrect scm element.
> >>> >> 4: something else I didn't think of.
> >>> >>
> >>> >> I didn't volunteer to be an archaeologist, I volunteered to be a
> >>> >> release manager..
> >>> >>
> >>> >> At best, the doc is, as I wrote to start this conversation,
> >>> >> incomplete, in that it does not give specific instructions for
> >>> >> achieving the PMC's goals using the plugin.
> >>> >>
> >>> >> To me, the following sequence is inoffensive:
> >>> >>
> >>> >> 1: run release:prepare, creating a premature 'real' tag.
> >>> >> 2: svn mv to convert that tag to an -RC tag.
> >>> >> 3: svn mv again to convert it to a (conventionally) immutable 'real
> >>> >> tag' if the vote passes.
> >>> >>
> >>> >> However, what I find inoffensive isn't important here. I'm not a PMC
> >>> >> member here. If this PMC has a strong policy of tag immutability,
> then
> >>> >> this PMC needs to document a procedure that both treats tags as
> >>> >> immutable and creates completely correct poms, including the scm
> >>> >> element that has to anticipate the eventual tag. It could create
> that
> >>> >> policy by erasing any mention of release:prepare, or by filling in
> the
> >>> >> missing details (though I continue to believe that there is no way
> to
> >>> >> make it come out the way you want.)
> >>> >>
> >>> >> Believe me, if I ever RM another commons release, I won't use
> >>> >> release:prepare until and unless someone writes up a step-by-step
> >>> >> guide that leads me to do so in an acceptable way.
> >>> >>
> >>> >> If you examine the 'tags' directory in svn right now, you will see
> >>> >> that it contains what I believe that you want it to contain: no
> >>> >> commons-io-2.5 tag, but rather a commons-io-2.5-RC4 tag. So, I
> >>> >> respectfully submit that we could proceed to decide if this release
> is
> >>> >> good enough, and sort out the documentation later.
> >>> >>
> >>> >>
> ---------------------------------------------------------------------
> >>> >> To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> >>> >> For additional commands, e-mail: dev-h...@commons.apache.org
> >>> >>
> >>> >
> >>> > ---------------------------------------------------------------------
> >>> > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> >>> > For additional commands, e-mail: dev-h...@commons.apache.org
> >>> >
> >>>
> >>> ---------------------------------------------------------------------
> >>> To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> >>> For additional commands, e-mail: dev-h...@commons.apache.org
> >>>
> >>>
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> > For additional commands, e-mail: dev-h...@commons.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> For additional commands, e-mail: dev-h...@commons.apache.org
>
>

Reply via email to