Uh oh!
There was an error while loading. Please reload this page.
Consolidate Compiler ErrorMessage tests under one suite. - #9525
Conversation
KevinRansom
left a comment
There was a problem hiding this comment.
Minor tweaks, then it's good to go , thanks
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
smoothdeveloper
commented
Jun 22, 2020
In my experience, tests asserting error messages are easier to maintain when they use the baseline files. Adjusting the test code when they are using I've added
fsharp/tests/FSharp.Test.Utilities/CompilerAssert.fs Lines 449 to 478 in d53716c This would also lend well into something where we cover all actual error messages the compiler can produce (which isn't the case, we've seen a regression on match on DU cases), which can also be useful for future work on compiler documentation (description of each error message with samples and explanation to fix). |
baronfel
commented
Jun 22, 2020
regarding the last point about compiler error message documentation I 100% took samples and phrasing from the baseline tests because they were such an easy to read resource. |
Uh oh!
There was an error while loading. Please reload this page.
…ricLiteralTests.fs Co-authored-by: Thorsten Reichert <ThorstenReichert@users.noreply.github.com>
vzarytovskii
commented
Jun 23, 2020
It makes sense to me, I wouldn't do it as a part of this PR thought, this one is just moving tests from older suites to a new one. |
smoothdeveloper
commented
Jun 24, 2020
@vzarytovskii sorry I didn't catch the tests I mentioned were just moved. What I say makes most sense for new tests around error / warnings, or at the time change to the actual message makes the test red. |
* Consolidate ErrorMessages compiler tests under component tests suite * Fix newlines in the assertion * Cleaned up test list; Make RIDs conditional as well Co-authored-by: Thorsten Reichert <ThorstenReichert@users.noreply.github.com>
As part of #7075