Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Jun 26, 2014

  1. Mark Waite

    Update to JGit 3.4.1 - latest release, match latest git-client

    MarkEWaite authored
  2. Mark Waite

    Update to git-client 1.9.1

    MarkEWaite authored
  3. Mark Waite

    Update to credentials version 1.14

    MarkEWaite authored
  4. Mark Waite

    Update to ssh-credentials 1.7.1

    MarkEWaite authored
  5. Mark Waite

    Update to apache httpcomponents 4.3.4

    MarkEWaite authored

Jun 24, 2014

  1. Jesse Glick

    Avoiding deprecated calls to Run.getEnvironment(); need to pass in a …

    …TaskListener.
    jglick authored

Jun 19, 2014

  1. Jesse Glick

    [JENKINS-9843] Forgot to record what problem this was fixing.

    jglick authored
  2. Jesse Glick

    [FIXED JENKINS-9843] PreBuildMergeOptions.getMergeRemote() cannot be …

    …@Exported, since the return type is not @ExportedBean.
    
    Corrects attempted fix b3ddce6.
    jglick authored
  3. Jesse Glick

    Pointless for GitSCM.getBranches() to be @Exported unless BranchSpec …

    …is an @ExportedBean!
    jglick authored
  4. Jesse Glick

    Removing @Exported from getRepositories() since it does not work anyway.

    RemoteConfig, being in JGit, is not an @ExportedBean.
    Stapler throws NotExportableException, but Property.writeValue ignores it since it is part of a list.
    Thus regardless of how the SCM is configured, you always get the useless
        "repositories" : [
          {
            
          }
        ],
    jglick authored

Jun 16, 2014

  1. Jesse Glick

    [maven-release-plugin] prepare for next development iteration

    jglick authored
  2. Jesse Glick

    [maven-release-plugin] prepare release git-2.3-beta-1

    jglick authored
  3. Jesse Glick

    Next release should be a beta because of new core dep.

    jglick authored
  4. Jesse Glick

    Merge pull request #236 from jglick/SCM-Job

    [JENKINS-23365] Work in a general Job
    jglick authored
  5. Jesse Glick

    Updating Javadoc.

    jglick authored
  6. Jesse Glick

    Deleting apparently half-written comment.

    jglick authored
  7. Jesse Glick

    Nothing “to do” about depending on plugins which were since split fro…

    …m the core.
    jglick authored
  8. Jesse Glick

    Updated to released dependencies.

    jglick authored

Jun 10, 2014

  1. Jesse Glick

    Updated to core API changes.

    jglick authored
  2. Jesse Glick

    Updated to branched multiple-scms-plugin, but [JENKINS-14537] jenkins…

    …ci/multiple-scms-plugin#5 broke MultipleSCMTest, so skipping it for now.
    jglick authored
  3. Jesse Glick

    JUnit 4.

    jglick authored
  4. Jesse Glick

    Comment.

    jglick authored
  5. Jesse Glick

    Typo.

    jglick authored

Jun 09, 2014

  1. Jesse Glick

    Adapted to core changes.

    jglick authored

Jun 07, 2014

  1. Jesse Glick

    Deleting MultipleSCMTest.

    It can no longer pass, at least until multiple-scms-plugin is modified to use Jenkins core as of SCM-Job,
    because MultiSCMChangeLogSet.add throws NoSuchFieldError: build,
    since @AdaptField is not processed on test classpath dependencies but only on code loaded from a plugin class loader.
    Anyway the handling of this plugin has been moved into core so testing its usage from git-plugin is less important.
    jglick authored
  2. Jesse Glick

    Revert "Do not create a zero-length changelog."

    This reverts commit 9a3ba41.
    SCM.checkout does vaguely state that an empty changelog of some kind should be created.
    MultipleSCMTest also requires it, since MultipleSCM reads the delegate changelog without checking whether it exists.
    It is odd in this case because GitSCM does not use XML, yet core gives the changelog a *.xml extension.
    jglick authored
  3. Jesse Glick

    Merge branch 'master' into SCM-Job

    jglick authored
  4. Jesse Glick

    Deleting BrowserChooserTest.

    It does not pass against newer versions of core, and it does not look straightforward to make it pass.
    It is anyway unclear what it is testing, other than that Stapler databinding works in general.
    If you want an integration test for configuration of browsers in the SCM, use JenkinsRule.configRoundTrip.
    Note that the various Descriptor<RepositoryBrowser<?>> implementations in this plugin
    all seem to override newInstance(StaplerRequest, JSONObject), even though there is no apparent reason to do so;
    none of the browsers in mercurial-plugin do so, and they are bound just fine.
    jglick authored
  5. Jesse Glick

    Making test run faster by reducing quiet period from 20s to 3s.

    jglick authored
  6. Jesse Glick

    Commenting out assertions that fail in 1.521+ due to improved core be…

    …havior from JENKINS-18407.
    jglick authored
  7. Jesse Glick

    Deleting apparently senseless file introduced in d98a86c by @ndeloof.

    jglick authored

Jun 06, 2014

  1. Jesse Glick

    Should trigger polling at most once per project.

    jglick authored
  2. Jesse Glick

    Implementing guessBrowser, so far limited to GithubWeb.

    jglick authored
  3. Jesse Glick

    Simpler and more flexible to use changeset ID as the anchor.

    This allows multiple changelogs to be shown in a single page.
    jglick authored
  4. Jesse Glick

    Ability for project-changes.jelly to handle multiple ChangeLogSet’s p…

    …er build.
    jglick authored
Something went wrong with that request. Please try again.