[Webkit-unassigned] [Bug 86690] Rename test_expectations.txt to TestExpectations

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=86690





--- Comment #10 from Ryosuke Niwa <rniwa at webkit.org>  2012-05-16 22:39:55 PST ---
(In reply to comment #9)
> Is the confusion you're seeking to avoid the use of "expect" in both names? Or is it that "test expectations" sounds too much like "text expectation"? Can someone use these names in an example sentence or two to better illustrate the confusion? 

"Test expectations of my-test.html have changed since r12345, and we need to rebaseline them."  Here, the speaker is referring to expected results as test expectations.

> For example, I often ask someone "what's in the test expectations" or say "I will add that to the expectations file"; it seems like the meaning of both of those is reasonably clear, but I could see how the first one could be misheard.

The problem is that people have used test expectations to mean expected results before nrwt existed.

> Maybe something like "suppressions" is better, since we often talk about suppressing an error?

TestSuppressions? or FailureSuppressions? I like that!

> On a different note, I would like to note that there is a clear connection between the name of the file and the name of the classes that process the file in the code today (e.g., TestExpectaions{Parser,Serializer,Model}). If we change the one, we should probably change the others as well. To that end, I am somewhat leery of "result" insofar as that is already badly overloaded in the NRWT code (e.g., TestResult, ResultSummary).

Yeah, I dislike CurrentResults for that reason. "current results" sound as if we're talking about the current result (i.e. -expected.* files in the sense of results we currently expect) of tests.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.