Skip to main content

Notice: This Wiki is now read only and edits are no longer possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.

Jump to: navigation, search

Difference between revisions of "JUnit4Migration"

(Eclipse projects are currently ''not required'' to convert tests)
(Move the note to proper and less intrusive real content to the page.)
 
(One intermediate revision by the same user not shown)
Line 1: Line 1:
{{important|Note|The following is one user's personal opinion. JDT will not add support for JUnit 5 if that would mean dropping support for JUnit 3 tests, and Eclipse projects are currently '''not required''' to waste time converting their tests. Converting tests is not trivial, and any contributions need to be verified carefully. See e.g. {{bug|487095}}}}
 
 
 
With JUnit 5 alpha being already available (http://repo1.maven.org/maven2/org/junit/) it seems like it will drop support for JUnit 3 APIs. Once JUnit 5 comes to JDT it would be nice not to have 3 different style of junit tests around. This work will have to eventually be done to keep up with the newer JUnit releases.
 
With JUnit 5 alpha being already available (http://repo1.maven.org/maven2/org/junit/) it seems like it will drop support for JUnit 3 APIs. Once JUnit 5 comes to JDT it would be nice not to have 3 different style of junit tests around. This work will have to eventually be done to keep up with the newer JUnit releases.
  
Also https://junit-team.github.io/junit5/#running-tests does not even mention the possibility of running JUnit 3 tests.
+
Also http://junit.org/junit5/docs/current/user-guide/#running-tests does not even mention the possibility of running JUnit 3 tests.
  
 
The APIs are cleaner in most cases and most developers are more used to 4 API rather than 3, and that is always good for the project in terms of potential contributors.
 
The APIs are cleaner in most cases and most developers are more used to 4 API rather than 3, and that is always good for the project in terms of potential contributors.
  
To summarize the goal is: make sure for all (as much as we can) projects to be prepared when JUnit 5 comes and it kills support for JUnit 3 tests.
+
JDT will not add support for JUnit 5 if that would mean dropping support for JUnit 3 tests, and Eclipse projects are '''currently not required''' to convert their tests. Converting tests may not be trivial for some tests, and any contributions need to be verified carefully. See {{bug|487095}} for examples of troubles faced. Green tests doesn't mean the conversion was done properly.
 +
 
 +
Having as many projects prepared for JUnit 5 release (without JUnit 3 support) would ensure easier migration when JUnit 3 might no longer be an option for the project.

Latest revision as of 08:24, 28 April 2016

With JUnit 5 alpha being already available (http://repo1.maven.org/maven2/org/junit/) it seems like it will drop support for JUnit 3 APIs. Once JUnit 5 comes to JDT it would be nice not to have 3 different style of junit tests around. This work will have to eventually be done to keep up with the newer JUnit releases.

Also http://junit.org/junit5/docs/current/user-guide/#running-tests does not even mention the possibility of running JUnit 3 tests.

The APIs are cleaner in most cases and most developers are more used to 4 API rather than 3, and that is always good for the project in terms of potential contributors.

JDT will not add support for JUnit 5 if that would mean dropping support for JUnit 3 tests, and Eclipse projects are currently not required to convert their tests. Converting tests may not be trivial for some tests, and any contributions need to be verified carefully. See bug 487095 for examples of troubles faced. Green tests doesn't mean the conversion was done properly.

Having as many projects prepared for JUnit 5 release (without JUnit 3 support) would ensure easier migration when JUnit 3 might no longer be an option for the project.

Back to the top