Closed mirekdlugosz closed 2 years ago
Merging #370 (50b5210) into master (f4bb8a7) will not change coverage. The diff coverage is
n/a
.
@@ Coverage Diff @@
## master #370 +/- ##
=======================================
Coverage 79.82% 79.82%
=======================================
Files 5 5
Lines 228 228
=======================================
Hits 182 182
Misses 46 46
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update f4bb8a7...50b5210. Read the comment docs.
Thanks for another great contribution. LGTM!
This PR includes / builds upon #365 . Story is similar to #369 - I can include this in #365 , or we can close #365 and continue discussion here, or I can rebase this once #365 is merged.
This is parallel to #369 and one should nicely rebase on top of another. I didn't include one in another to indicate that this is largely independent work.
This introduces proper end-to-end UI tests. For each source type (Network, Satellite and vCenter), it executes following path:
Created data is later removed using API, so we should be able to run test repeatedly without leaving too much rubbish behind.
Input data for forms may be specified inside function. I have done this for Network source and credential, based on data available on OpenStack instance. I did not find reference values for Satellite and vCenter, so these were left out. When running tests, use
-k network
to only run test that should pass.Report assertions are relatively weak - I only check that file can be extracted, that all files inside have non-zero size and that their SHA256 sums match what is specified in report file. We probably should focus more on reports content, but (1) reports are the same no matter the interface, and we already have some reports tests in CLI and API; (2) report helpers I have found are not easy to use, and at this point I am not in position to fully refactor them.