-
Notifications
You must be signed in to change notification settings - Fork 139
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
[CI only] Streamlined/fixed sanitize/coverage matrices #790
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## fix_test_ssl_verify_hostname #790 +/- ##
================================================================
+ Coverage 68.85% 70.70% +1.85%
================================================================
Files 49 47 -2
Lines 15219 15133 -86
Branches 3133 3088 -45
================================================================
+ Hits 10479 10700 +221
+ Misses 1693 1450 -243
+ Partials 3047 2983 -64 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, but wondering if we need all those code coverage runs?
|
||
sanitize-no-host-verify: | ||
name: "Sanitize address (NO force_host_verification)" | ||
coverage-NO-verify_host: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we have a single coverage run (basically with the defaults of the library that has TLS/enfore_hostname_verification/etc..)?
Or is it that we would get a much lower score (for all code that is in the "#else" of those "#ifdef") without these new runs?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Right. What prompted this was the refactoring of deliveries - the thread pool refactor code generated a lot of missing coverage comments since it was not ran with coverage.
My thinking: the coverage overhead is relatively small (relative to sanitize) so it's ok to run a bigger matrix. For the same reason, the coverage runs are probably sufficient to assert functional correctness (no need to dial down the parameters, for example). We still run the no-coverage tests for the default build config.
FWIW, I failed to package them into a single compact matrix statement, but the current job set is probably more intention revealing anyway then the matrix would have been.
…OFF (#788) * Fixed test SSLVerifyHostName when build NATS_FORCE_HOST_VERIFICATION=OFF This test would fail unless we force host verification, so adapt test to take into consideration the expected result based on the build environment variable. Signed-off-by: Ivan Kozlovic <[email protected]> * Added a GHA job for NATS_FORCE_HOST_VERIFICATION * fixed default: ON * job: no-host-sanitize * job names * [CI only] Streamlined/fixed sanitize/coverage matrices (#790) --------- Signed-off-by: Ivan Kozlovic <[email protected]> Co-authored-by: Lev Brouk <[email protected]> Co-authored-by: Lev <[email protected]>
Streamlined and fixed the test matrix for coverage, sanitize, and environment variable/commpile flag configurations.