-
Notifications
You must be signed in to change notification settings - Fork 4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Introduce CTest-based test harness #80
Conversation
Codecov Report
@@ Coverage Diff @@
## develop #80 +/- ##
=======================================
Coverage 96.2% 96.2%
=======================================
Files 34 34
Lines 2667 2667
Branches 470 470
=======================================
+ Hits 2566 2567 +1
+ Misses 77 76 -1
Partials 24 24
Continue to review full report at Codecov.
|
b75b310
to
212f1f4
Compare
Enable no languages by default
This is a temporary state of affairs, meant to ease the transition to using CTest for all testing.
As I recently found out the hard way, `empty.to` has been (since commit 52c5c50) even emptier than null.to, and the difference shows up in test coverage, specifically in tld.o, which requires at least one record in each object.
Generating these failure cases with `dd` places an undue burden on the build environment. Since we are moving toward a CMake build environment, we may soon be able to build without a GNU-like environment.
The wine-run tests will have a different value for argv[0], so we check for "Usage:" in the output instead of forcing an exact textual match.
CMake 3.19 is the minimum, and 3.20 is the latest we have tested.
Fixes #50.
The approach in this PR is to iteratively peel away pieces of the Makefile-based test harness and replace them with CTest tests while maintaining non-decreasing code coverage.