Surprising behaviour when releasing for 2.10.0-RC1 on Sonatype

[Paul Butcher]

I've just released a build of ScalaMock 3 for 2.10.0-RC1 to Sonatype, and instead of being published to:

https://oss.sonatype.org/content/groups/public/org/scalamock/scalamock_2.10.0-RC1/

as I expected, it's been published to:

https://oss.sonatype.org/content/groups/public/org/scalamock/scalamock_2.10/

Is this what I should expect? And if so, why is a release candidate treated differently from a milestone (previous milestone releases were released to directories named for the milestone).

I'm using SBT 0.12.1 with xsbt-gpg-plugin 0.6. The project definition is here:

https://github.com/paulbutcher/ScalaMock/blob/develop/project/Build.scala

--
paul.butcher->msgCount++

Snetterton, Castle Combe, Cadwell Park...
Who says I have a one track mind?

http://www.paulbutcher.com/
LinkedIn: http://www.linkedin.com/in/paulbutcher
MSN: pa...@paulbutcher.com
AIM: paulrabutcher
Skype: paulrabutcher

[Josh Suereth]

Yes.  if you read the full email about staging, I mentioned how you should:

set scalaBinaryVersion := "2.10.0-RC1".

SBT's "binary version" logic is still dumb about -RC, -SNAPSHOT and -M, so you have to be smart for it.

- Josh

--
You received this message because you are subscribed to the Google Groups "simple-build-tool" group.
To post to this group, send email to simple-b...@googlegroups.com.
To unsubscribe from this group, send email to simple-build-t...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/simple-build-tool?hl=en.

[Josh Suereth]

Not sure.  If you're propagated to maven central, I'd guess you're hosed.   If you can't delete from the oss.sonatype.org web interface, then you might just have to let it be as it is and ward people off of that version....

On Sat, Oct 13, 2012 at 9:11 AM, Paul Butcher <pa...@paulbutcher.com> wrote:

Damn - I missed that.

You don't happen to know if there's any way to delete something from Sonatype after it's been released?

To view this discussion on the web visit https://groups.google.com/d/msg/simple-build-tool/-/-FDjZ_yYl2cJ.

[Paul Butcher]

Fair enough. My fault for not reading your message more carefully (and, for that matter, not checking the staged artefacts before publishing).

--
paul.butcher->msgCount++

Snetterton, Castle Combe, Cadwell Park...
Who says I have a one track mind?

http://www.paulbutcher.com/
LinkedIn: http://www.linkedin.com/in/paulbutcher
MSN: pa...@paulbutcher.com
AIM: paulrabutcher
Skype: paulrabutcher

[Josh Suereth]

Don't worry too much.   My guess is everyone using sbt 0.12 for the first time with 2.10 will experience a wtf.   I'm not sure how to call enough attention to the change though....

[Paul Phillips]

On Sat, Oct 13, 2012 at 6:40 AM, Josh Suereth <joshua....@gmail.com> wrote:

Don't worry too much.   My guess is everyone using sbt 0.12 for the first time with 2.10 will experience a wtf.   I'm not sure how to call enough attention to the change though....

"Not making the change" would have been the way.  I can't fathom it.  We can look forward to many similar mistakes.

Defaults should not burn you like that.

[Johannes Rudolph]

This is what we are considering to use:

scalaBinaryVersion <<= scalaVersion(sV => if (CrossVersion.isStable(sV)) CrossVersion.binaryScalaVersion(sV) else sV)

Wasn't something like this in an earlier sbt 0.12(-RC?) the default?

Johannes

[Paul Phillips]

On Mon, Oct 15, 2012 at 12:47 AM, Johannes Rudolph <johannes...@gmail.com> wrote:

Wasn't something like this in an earlier sbt 0.12(-RC?) the default?

Something like that - not in an RC I don't think but earlier than that.  There is clearly some backstory to the changes but that story did not play out in one of my usual venues.  All I know is that when the music stopped, the wrong monkey was sitting in the chair.