Clean up output for Hardware CLI #43
Conversation
fixtures/conch-v3/datacenter.yaml
Outdated
| @@ -18,15 +18,15 @@ interactions: | |||
| Cache-Control: | |||
| - no-cache | |||
| Date: | |||
| - Mon, 13 Jan 2020 23:16:20 GMT | |||
| - Sun, 26 Jan 2020 19:04:25 GMT | |||
There was a problem hiding this comment.
This is just churn and committing these is surely not needed.
what are you doing to generate these changes? when I make test, nothing is altered in the repository.
There was a problem hiding this comment.
To generate these files simply remove fixtures/conch-v3/*.yaml and they'll be recreated on the next call to go test (which make test calls) by attempting to hit whatever Conch API endpoint is configured in your shell. I create them locally by hitting a fresh dev server with an empty database.
I've opened #44 to track replacing this system with something less ugly. Sungo specifically warned me that Lane had tried something like this and it sucked, and it does … I'm just not sure how I want to replace everything. Half re-implementing the API using Gock isn't really more appealing … yet.
There was a problem hiding this comment.
Also the most recent rebase reduces the number of fixtures changed to just the ones directly impacted.
perigrin
force-pushed
the
36-add-hardware-cli
branch
from
502840b
to
57d4e96
Compare
perigrin
force-pushed
the
36-add-hardware-cli
branch
from
57d4e96
to
41331a6
Compare
This PR cleans up the output for the Hardware CLI. I had inadvertently not implemented stringification for the hardware structs in my rush to get something functional out the door.
This also adds tests for the Hardware CLI output, and implements a proof of concept for testing the other CLI outputs going forward.
Finally I cleaned up a couple places that
go vetwas complaining about in the code, and started migrating tests to usestretchr/testifybecause it's output is easier for debugging test failures than my hand rolled output.