Formed in 2009, the Archive Team (not to be confused with the archive.org Archive-It Team) is a rogue archivist collective dedicated to saving copies of rapidly dying or deleted websites for the sake of history and digital heritage. The group is 100% composed of volunteers and interested parties, and has expanded into a large amount of related projects for saving online and digital history.
Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.
Sign upIterableSubject constructor documentation for check method #634
Comments
|
Right. That's for implementations of |
|
(I did figure that someone would call me on this incomplete doc eventually, so I, too, am interested to hear about what brought you here. Thanks for filing the issue.) |
|
This came out in GoogleCloudPlatform/cloud-opensource-java#954 I was doing a quick check if a list had changed between versions. To detect the change, I compared the sizes of the lists. I didn't compare the contents because the list was long and I didn't want to reproduce and keep up to date the several hundred elements of the list in the test code. Not perfect, I know, but this allows us to update the test to the new version without updating the previously published version. If, and only if, the size of the list has changed then I want to see exactly what changed. I was trying to use IterableSubject to compare the two lists to give me a nicely formatted diff. I was not using Truth.assertThat(list).containsExactly() because I can't update that without publishing a new canonical version, and I can't publish a new canonical version until the tests pass. Hence the simple size comparison. |
|
I still don't fully understand why It seems to me like whether to use an explicit set of expected elements vs. one read from some other source (as the test is currently doing) is orthogonal to which mechanism it uses to compare the actual set to the expected set. |
|
You pointed out the reason a few comments ago. If I take out the check for the size I can't update the test so it passes when the local copy differs from the last published copy. |
|
I think that's a separate issue. Even if you leave the size check in, you can't update the test so it passes when the local copy intentionally adds an element. And if the local copy unintentionally differs but the size is the same, the size check means you'll have a false negative (the test will pass but it shouldn't). |
The API doc for the constructor https://truth.dev/api/1.0/com/google/common/truth/IterableSubject.html#IterableSubject-com.google.common.truth.FailureMetadata-java.lang.Iterable- says "Constructor for use by subclasses. If you want to create an instance of this class itself, call check(...).that(actual)."
However, the check method it links to and the others I can see are protected so I can't call them either.