Implement default methods in TaskListener and BuildListener #3122

jglick · 2017-10-31T23:08:48Z

Would avoid the need for LessAbstractTaskListener in jenkinsci/workflow-support-plugin#15, for example.

@reviewbybees

ghost · 2017-10-31T23:15:14Z

This pull request originates from a CloudBees employee. At CloudBees, we require that all pull requests be reviewed by other CloudBees employees before we seek to have the change accepted. If you want to learn more about our process please see this explanation.

svanoort · 2017-11-01T00:19:47Z

core/src/main/java/hudson/util/LogTaskListener.java

@@ -82,12 +70,13 @@ public void close() {
        private final StackTraceElement caller;
        private final ByteArrayOutputStream baos = new ByteArrayOutputStream();

-        public LogOutputStream(Logger logger, Level level, StackTraceElement caller) {
+        LogOutputStream(Logger logger, Level level, StackTraceElement caller) {


🐛 - Potential breaker for code depending on this constructor.

Not true; the class was private already, so the public modifier on its constructor was just misleading.

svanoort · 2017-11-01T00:21:27Z

core/src/main/java/hudson/util/StreamTaskListener.java

@@ -58,7 +52,7 @@
 * 
 * @author Kohsuke Kawaguchi
 */
-public class StreamTaskListener extends AbstractTaskListener implements Serializable, Closeable {
+public class StreamTaskListener implements TaskListener, Closeable {


🐛 You can no longer cast to AbstractTaskListener or check for instanceof AbstractTaskListener.

That is true, but AbstractTaskListener was only ever a convenience base class, so there was never a legitimate reason to do so to begin with.

I was unable to find the implementation, which may be affected. But I agree that keeping the inheritance from a deprecated class would be the safest option for now. This old class is not doing anything wrong anyway

As in #2879 I would prefer to eliminate all usages of the deprecated type to reduce confusion, since there is no plausible risk from the technical API change here.

svanoort

A couple small compatibility issues but otherwise looks reasonable AFAICT.

jglick

Not issues I think.

jglick · 2017-11-02T01:22:09Z

core/src/main/java/hudson/util/LogTaskListener.java

@@ -82,12 +70,13 @@ public void close() {
        private final StackTraceElement caller;
        private final ByteArrayOutputStream baos = new ByteArrayOutputStream();

-        public LogOutputStream(Logger logger, Level level, StackTraceElement caller) {
+        LogOutputStream(Logger logger, Level level, StackTraceElement caller) {


Not true; the class was private already, so the public modifier on its constructor was just misleading.

jglick · 2017-11-02T01:23:05Z

core/src/main/java/hudson/util/StreamTaskListener.java

@@ -58,7 +52,7 @@
 * 
 * @author Kohsuke Kawaguchi
 */
-public class StreamTaskListener extends AbstractTaskListener implements Serializable, Closeable {
+public class StreamTaskListener implements TaskListener, Closeable {


That is true, but AbstractTaskListener was only ever a convenience base class, so there was never a legitimate reason to do so to begin with.

oleg-nenashev · 2017-11-06T21:29:03Z

No strong opinion since I do not see any usages which will be affected by the change.
Would be great to get more votes from @jenkinsci/code-reviewers in order to elaborate whether it's fine to break binary compatibility for such cleanup.

jglick · 2017-11-13T14:58:27Z

I do not see any usages which will be affected by the change.

And I would be happy to break any such usage if there were one, because it would be ridiculous.

oleg-nenashev

Weak 🐝 since I cannot find any usages of the API to be broken. @reviewbybees done, I'd guess

oleg-nenashev · 2017-11-19T12:25:58Z

formally it's ready to go. YOLO.

jtnord · 2018-03-09T11:34:43Z

noting that this caused a breakage as the class serialVersionUID are different in the AbstractTaskListener - so Serialized formats are no longer the compatable.
The follow up #3154 did not restore this :-(

jglick · 2018-03-09T19:42:43Z

@jtnord Normally that would be irrelevant since master → agent connections will use identical bytecode. Would only matter for other uses of Remoting involving diachronic serialization. Probably fixable by adding a serialVersionUID to AbstractTaskListener matching the original value.

daniel-beck · 2018-03-10T21:21:14Z

@jtnord Do you have an issue reference for that?

jtnord · 2018-03-12T11:02:46Z

@daniel-beck JENKINS-50124

…nkins#3122.

Implement default methods in TaskListener and BuildListener.

9ed6b01

jglick requested a review from svanoort October 31, 2017 23:08

jglick mentioned this pull request Oct 31, 2017

[JEP-210] Log handling rewrite jenkinsci/workflow-support-plugin#15

Merged

6 tasks

svanoort reviewed Nov 1, 2017

View reviewed changes

svanoort suggested changes Nov 1, 2017

View reviewed changes

jglick commented Nov 2, 2017

View reviewed changes

jglick added the needs-more-reviews Complex change, which would benefit from more eyes label Nov 2, 2017

jglick requested a review from oleg-nenashev November 6, 2017 20:57

jglick requested a review from svanoort November 13, 2017 18:36

svanoort approved these changes Nov 13, 2017

View reviewed changes

oleg-nenashev approved these changes Nov 14, 2017

View reviewed changes

oleg-nenashev added ready-for-merge The PR is ready to go, and it will be merged soon if there is no negative feedback and removed needs-more-reviews Complex change, which would benefit from more eyes labels Nov 14, 2017

oleg-nenashev merged commit 5b9ad64 into jenkinsci:master Nov 19, 2017

oleg-nenashev mentioned this pull request Nov 19, 2017

[JENKINS-47896] - Introduce SerializableOnlyOverRemoting in classes. #3144

Closed

4 tasks

jglick deleted the TaskListener branch November 20, 2017 21:45

oleg-nenashev mentioned this pull request Nov 21, 2017

[JENKINS-48116] - Restore AbstractTaskListener binary compatibility in the core. #3154

Merged

4 tasks

jglick added a commit to jglick/workflow-support-plugin that referenced this pull request Jun 11, 2018

Deleting LessAbstractTaskListener as it is superseded by jenkinsci/je…

54f2094

…nkins#3122.

jglick added a commit to jglick/workflow-durable-task-step-plugin that referenced this pull request Jun 11, 2018

Deleting LessAbstractTaskListener as it is superseded by jenkinsci/je…

34125ba

…nkins#3122.

jglick mentioned this pull request Jun 26, 2018

[JEP-206] Define API for gathering command output in a local encoding jenkinsci/durable-task-plugin#61

Merged

orgi pushed a commit to orgi/workflow-durable-task-step-plugin that referenced this pull request Nov 13, 2018

Deleting LessAbstractTaskListener as it is superseded by jenkinsci/je…

c52fe3b

…nkins#3122.

jglick mentioned this pull request Jan 27, 2022

[JEP-7] Comment updates after removing JRuby support #6224

Merged

6 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implement default methods in TaskListener and BuildListener #3122

Implement default methods in TaskListener and BuildListener #3122

jglick commented Oct 31, 2017

ghost commented Oct 31, 2017

svanoort Nov 1, 2017

jglick Nov 2, 2017

svanoort Nov 1, 2017

jglick Nov 2, 2017

oleg-nenashev Nov 3, 2017

jglick Nov 6, 2017

svanoort left a comment

jglick left a comment

jglick Nov 2, 2017

jglick Nov 2, 2017

oleg-nenashev commented Nov 6, 2017

jglick commented Nov 13, 2017

oleg-nenashev left a comment

oleg-nenashev commented Nov 19, 2017

jtnord commented Mar 9, 2018

jglick commented Mar 9, 2018

daniel-beck commented Mar 10, 2018

jtnord commented Mar 12, 2018 •

edited

Loading

Implement default methods in TaskListener and BuildListener #3122

Implement default methods in TaskListener and BuildListener #3122

Conversation

jglick commented Oct 31, 2017

ghost commented Oct 31, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

svanoort left a comment

Choose a reason for hiding this comment

jglick left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

oleg-nenashev commented Nov 6, 2017

jglick commented Nov 13, 2017

oleg-nenashev left a comment

Choose a reason for hiding this comment

oleg-nenashev commented Nov 19, 2017

jtnord commented Mar 9, 2018

jglick commented Mar 9, 2018

daniel-beck commented Mar 10, 2018

jtnord commented Mar 12, 2018 • edited Loading

jtnord commented Mar 12, 2018 •

edited

Loading