* Return an error seperately from Response
* Remove BuildErrorResponse
* Drop Response.Error from model/client4.go
* Migrate require.Nil checks
* Migrate require.NotNil checks
* More manual fixes
* Move error check out of CheckOKStatus and CheckCreatedStatus
* Move error check out of CheckForbiddenStatus
* Move error check out of CheckUnauthorizedStatus
* Move error check out of CheckNotFoundStatus
* Move error check out of CheckBadRequestStatus
* Move error check out of CheckNotImplementedStatus and CheckRequestEntityTooLargeStatus
* Move error check out of CheckInternalErrorStatus
* Move error check out of CheckServiceUnavailableStatus
* Remove error check from checkHTTPStatus
* Remove remaining references to Response.Error
* Check previously unchecked errors
* Manually fix compile and linter errors
* Return error in CreateWebSocket methods
* Return error instead of *AppError in DoApi methods
* Manually fix bad replacments
* Conistently return Response and error
* Use err instead of seperate bool return value to indicate success
* Reduce ussage of model.AppError in web/oauth_test.go
* Remove client4.Must
* Check error in buf.ReadFrom
* Fix failing tests
* format using `goimports -local github.com/mattermost/mattermost-server/v5 -w`
* added goimports lint check to .golangci.yml
* format using `goimports -local github.com/mattermost/mattermost-server/v5 -w` for a corner case
* make app-layers, *-mocks and store-layers for ci check
Co-authored-by: Mahmudul Haque <mahmudulhaque@protonmail.com>
Co-authored-by: Mattermod <mattermod@users.noreply.github.com>
* MM-27149: optimize initBasic
Mostly, all tests just needed the user initialization part and not
the channel and group creation. So we move the user initialization inside
the Setup call. This avoids unnecessary DB calls which take around 250-300ms
on average.
And we make the login requests concurrently to shave off a few more ms.
According to my tests, the 2 login calls take 140 ms on average, which
shaves off 70ms.
So approximately, we shave off 350ms per test. And there are 114 occurences
of these. So around 39 seconds.
* make initlogin only for Setup/SetupEnterprise
Co-authored-by: Mattermod <mattermod@users.noreply.github.com>
* Consistent license message for all the go files
* Fixing the last set of unconsistencies with the license headers
* Addressing PR review comments
* Fixing busy.go and busy_test.go license header
* api4: fix TestGetUsersNotInTeam assertions
This test was relying on data from a previous test run. With the data cleared before each test, the assertions much match reality.
* *testlib: always InitSystemAdmin
Some tests implicitly relied on the basic user having system
administrator privileges because it was the first user created as such.
Eliminate `InitSystemAdmin` and explicitly create the system admin user
instead to avoid this ambiguity going forward.
* *testlib: drop all tables before each test
* api4: split up TestChannelDelete to avoid duplicate InitBasic
* api4: teardown in TestResetPassword, for when this test comes back
* invalidate cache on DropAllTables
This is necessary since the test store persists across tests.
* disable parallel tests
While tests within a package must be explicitly parallelized using `t.Parallel()`, tests across packages are run in parallel by default. This causes problems given that the tests all currently share the same database instance.
Unfortunately, this also means that running the tests is much slower, but we can return to this later.
* don't use global app for api / api4 tests
* put sleep back. we're gonna have to do some goroutine wrangling
* fix oauth test config assumptions
* jobs package, i'm comin' for you next
* app test fix
* try increasing sleep a little
