[
https://issues.apache.org/jira/browse/CALCITE-2457?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16573556#comment-16573556
]
Andrei Sereda commented on CALCITE-2457:
----------------------------------------
{quote}
Is that IDE-friendly?
{quote}
Yes you get {{Click to see the difference}} link. Same one as with standard
JUnit.
Regarding failure messages, consider the following assertion :
{code:java}
assertThat(Collections.singletonMap("key1", "val2")).containsEntry("key2",
"val2");
// error
// Not true that <{key1=val2}> contains entry <key2=val2>. However, the
following keys are mapped to <val2>: [key1]
{code}
Here I don't need to manually debug and trace the variables to understand why
"expected true got false". Failure message already includes all necessary debug
information (this saves time). Obviously you can do it manually (with string
concat) but it will not be as informational.
h3. Kotlin
Interesting proposal. This will definitely make writing DSLs easier. But
introducing new language is contentious topic, much more than introducing new
library. Obviously CALCITE-2458 has to be decided before assertions.
But, since I'm not a calcite dev, I will remain neutral.
> Upgrade to JUnit 5
> ------------------
>
> Key: CALCITE-2457
> URL: https://issues.apache.org/jira/browse/CALCITE-2457
> Project: Calcite
> Issue Type: Improvement
> Reporter: Vladimir Sitnikov
> Assignee: Julian Hyde
> Priority: Major
>
> JUnit 5 brings multiple useful features so tests are easier to read and write.
> Is there something that blocks upgrading to JUnit 5?
> By upgrade I mean bumping up the dependency version and creating new tests
> with JUnit 5 features.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
