Chris Povirk

Test IterableSubject methods on objects that don't implement hashCode()

2

Such objects are at least a little evil, but they exist in one or two Google-internal projects. If we currently support this, then we should continue to do so, or...

Write a comparison to Kluent, AssertK, Strikt, Atrium, Kotest, other Kotlin assertion libraries?

13

They have various Kotlin-specific advantages. It would be good to understand what those are. Maybe someday we can provide some of them ourselves (though we haven't followed through on our...

Document how to write a Truth extension under Kotlin

8

It's pretty simple (and pretty similar to Java), but it's always nice for users to have an example. \[edit: I now have [a draft example](https://github.com/google/truth/issues/661#issuecomment-1452472936) below.\] We could also consider...

Externally publish JUnit->Truth migrations

1

RE: https://twitter.com/foobarjet/status/1148700099331219456 We have these internally (in `TruthExhaustiveShortcuts` and scattered through the classes in the `truth` subpackage), but we haven't published them for external users. I believe that they all...

Kotlin: Support `assertThat(foo) { isNotNull() isNotEqualTo(bar) ... }`

32

It's already possible to write nearly that by using `apply`: ``` assertThat(foo).apply { isNotNull() isNotEqualTo(bar) } ``` Might it be worth a shortcut? Anecdotes welcome.

Document that isEqualTo is not for testing equals(), etc.

16

Our principle is: When you call `assertThat(foo)`, you're not testing any of the methods of foo. Rather, you're testing the result of some previous call: UserId id = employeeService.manager("cpovirk"); assertThat(id).isEqualTo(...

AssertJ's [`AbstractIterableAssert`](https://joel-costigliola.github.io/assertj/core/api/org/assertj/core/api/AbstractIterableAssert.html) is large. We don't have an equivalent to every last assertion (and this is [sometimes intentional](https://truth.dev/comparison#assertion-count)), but for cases in which we _do_ have an equivalent, we could...

Consider Kotlin extension-method "overloads" to StandardSubjectBuilder.that

1

Everywhere we expose a `Subject.Factory` (or `CustomSubjectBuilder.Factory`), we could expose a `that` method. Such methods would save users of `assertWithMessage`, `check`, `expect`, etc. (maybe including `ExpectFailure` in the `Subject`'s own...

If matches(x) fails but containsMatch(x) would have passed, suggest containsMatch in the failure message

1

Similar in spirit to https://github.com/google/truth/issues/783. ```java assertThat("foo bar baz").matches("ba[rz]"); ``` ``` Did you mean to call containsMatch() instead of match()? ```

As noted in https://github.com/google/truth/pull/789#issuecomment-732213007, it's _very_ frequently misused. Users who want that behavior would likely be better served by having to opt into it more explicitly with a call like...