* [JENKINS-58734] Use SHA-256 for crumbs
Signed-off-by: Matt Sicker <boards@gmail.com>
* Use SHA-256 more consistently
Signed-off-by: Matt Sicker <boards@gmail.com>
* [58571] Check if name is editable on item before doing rename.
* [JENKINS-58571]Form failure when rename item with not editable name.
* [JENKINS-58571] Change to IOException when isNameEditable is false.
* [JENKINS-58571] Validate test harness furure version.
* [JENKINS-58571] Validate test harness 2.55.
* [JENKINS-58548] Remove extra whitespace from scriptText
This removes the unnecessary whitespace being output in scriptText
and anywhere else that has non-XML whitespace semantics in its output
when using <l:view>.
Signed-off-by: Matt Sicker <boards@gmail.com>
* Fix test name
Co-Authored-By: Daniel Beck <1831569+daniel-beck@users.noreply.github.com>
(cherry picked from commit dae53088b7)
* [JENKINS-58595] Add missing ${h} in rss and atom views
Now that l:view was fixed not to output whitespace, these views can be
updated to directly use l:view. Otherwise, they needed to initialize
the ${h} Jelly variable which is done via l:view normally.
This fixes a regression introduced in SECURITY-534.
Signed-off-by: Matt Sicker <boards@gmail.com>
* Fix the fix
Signed-off-by: Matt Sicker <boards@gmail.com>
* [JENKINS-58548] Remove extra whitespace from scriptText
This removes the unnecessary whitespace being output in scriptText
and anywhere else that has non-XML whitespace semantics in its output
when using <l:view>.
Signed-off-by: Matt Sicker <boards@gmail.com>
* Fix test name
Co-Authored-By: Daniel Beck <1831569+daniel-beck@users.noreply.github.com>
* Batch up dynamic plugin installations from setup wizard.
* Reworked to thread the list of batched-up plugins properly and introduced CompleteBatchJob to force the setup wizard to wait.
* UpdateCenterPluginInstallTest.test_installKnownPlugins failure.
* @amuniz notes that an existing message could now be considered misleading.
* Introduced UpdateCenter.DownloadJob.getDisplayName for better consistency in logging, as suggested by @Vlatombe.
* Reducing log levels of messages from dynamicLoad which are not of great interest, as suggested by @Vlatombe.
* Reducing log level of another message of little interest.
* Adding classic UI for CompleteBatchJob.
* CompleteBatchJob should be used only when dynamicLoad.
* Null defense on UpdateCenterJob.site.
Otherwise can get:
java.lang.NullPointerException
at hudson.model.UpdateCenter.getConnectionCheckJob(UpdateCenter.java:850)
at hudson.model.UpdateCenter.addConnectionCheckJob(UpdateCenter.java:817)
at hudson.model.UpdateCenter.addJob(UpdateCenter.java:804)
at hudson.model.UpdateSite$Plugin.deploy(UpdateSite.java:1367)
at hudson.PluginManager.install(PluginManager.java:1537)
at hudson.PluginManager.install(PluginManager.java:1496)
at hudson.PluginManager.doInstall(PluginManager.java:1445)
at …
This change catches the exception and returns false, invalidating the user session. This behavior is sensible, because if we are unable to validate the user we must assume the session is invalid.
Catching the error at this point reportedly helps with some scenarios and shouldn't cause any further harm.
(cherry picked from commit 8349cebb61)
Kohsuke Kawaguchi