[ 
http://jira.codehaus.org/browse/SCM-444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=189809#action_189809
 ] 

Mark Struberg commented on SCM-444:
-----------------------------------

Petter, pushing to the current branch should already be the default behaviour. 

We also had a discussion about whether we should add more support for 
distributed SCMs in maven-scm-api in the future, because bazaar, hg, etc works 
the same way basically. 

There are 2 issues:
* The current way of always pushing is good for projects like we have here at 
Apache and a few other pretty centralised projects where basically only 1 main 
repository exists. But it the project is more a forked 
pull-it-from-wherever-you-need approach, this probably wont work for the long 
term. Another idea is that it would make testing releases a lot easier. You do 
a 'local' release and push it to a staging repo. If all works out fine then you 
can push the tag to the central repo...

* For release:perform we currently do a git-clone from the upstream repo. This 
may cause some unnecessary traffic (and time). This is not such a big problem 
for git but for hg and bazaar which performs a lot slower. Since with all DCMS 
we have the full repo available locally, we may easily checkout into 
target/checkout from our _local_ repo instead of going to the remote.


> Git provider does 'git push' during 'mvn release:prepare' which causes 
> unwanted problems
> ----------------------------------------------------------------------------------------
>
>                 Key: SCM-444
>                 URL: http://jira.codehaus.org/browse/SCM-444
>             Project: Maven SCM
>          Issue Type: Bug
>          Components: maven-scm-provider-git
>    Affects Versions: 1.1
>            Reporter: Petter Måhlén
>            Priority: Minor
>
> When doing 'mvn release:prepare' with a Git provider, a 'git push' command is 
> executed. This is not ideal because the push command can fail or push things 
> from the local repository that are not needed/wanted in the remote 
> repository. Some examples are:
> 1. The local repository has two branches: master (tracking origin/master) and 
> dummy (tracking origin/dummy). The release is being made on the master 
> branch, and the dummy and origin/dummy branches have diverged. Running 
> 'release:prepare' causes a 'git push', which will succeed for the master 
> branch (assuming that the release preparation has been made correctly) and 
> fail for the dummy branch (the two branches have diverged and need to be 
> merged or rebased). The release preparation aborts and the directory is left 
> in a somewhat inconsistent state where manual cleaning up is needed (removing 
> pom.xml.next files, changing versions to <new>-SNAPSHOT, etc.)
> 2. The local repository has two branches: master (tracking origin/master) and 
> localtest (not in the origin repository). The localtest branch shouldn't be 
> published because it is just used for some temporary testing and doesn't even 
> work. It will be pushed during 'release:prepare'.
> Suggested behaviour: use 'git push origin <currentbranch>:<currentbranch>', 
> or even better, query for which remote repository to push to (found in 
> .git/config) and which branch to push from and to. For me, it would be great 
> to have a 'confirm push' before doing it so as to keep things clean, but 
> maybe that is quite complex.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira


Reply via email to