When I try to run the tests for this repo locally, sometimes my test run looks like this:
Randomized with seed 7379
................................................................................................*..................................mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
WARNING: option --ssl-verify-server-cert is disabled, because of an insecure passwordless login.
Pending: (Failures listed here are expected and do not affect your suite's status)
1) Hanami::CLI::Commands::App::Generate::Action generate for app, without hanami view bundled allows to specify MIME Type for template
# Temporarily skipped with xit
# ./spec/unit/hanami/cli/commands/app/generate/action_spec.rb:252
Finished in 8.57 seconds (files took 0.85412 seconds to load)
132 examples, 0 failures, 1 pending
Other times like this:
Randomized with seed 6645
...................mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
WARNING: option --ssl-verify-server-cert is disabled, because of an insecure passwordless login.
Finished in 4.46 seconds (files took 0.84088 seconds to load)
20 examples, 0 failures
Either way, the whole suite is pretty much never completed. It looks like something (probably around mysql tests) kills the test process abruptly, but in a way that the RSpec runner thinks that all the test are run. I'm not sure how to start debugging it. Running with doc format does not reveal any further hints.
Example:
Randomized with seed 15437
Hanami::CLI::Server
#call
delegates config as config
delegates debug as debug
delegates port as Port
delegates warn as warn
delegates host as Host
Hanami::CLI::Commands::App::DB::Structure::Load
mysql
mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
WARNING: option --ssl-verify-server-cert is disabled, because of an insecure passwordless login.
Finished in 0.32933 seconds (files took 1.25 seconds to load)
6 examples, 0 failures
Another example (truncated)
Randomized with seed 49833
bin/hanami
console
defaulting to pry when it's present
starts pry console
irb
starts irb console (PENDING: Temporarily skipped with xit)
[...]
Hanami::CLI::Commands::App::DB::Structure::Dump
postgres
prints errors for any dumps that fail and exits with non-zero status
dumps the structure for the app db when given --app
dumps the structure for each db, including schema_migrations
dumps the structure for a slice db when given --slice
sqlite
prints errors for any dumps that fail and exits with non-zero status
dumps the structure for each db, including schema_migrations
dumps the structure for the app db when given --app
dumps the structure for a slice db when given --slice
slice with gateways
dumps the structure for for a slice's gateway when given --slice and --gateway
dumps the structure for all the slice's gateways when given --slice
app with gateways
dumps the structure for all the app's gateways when given --app
dumps the structure for for a slice's gateway when given --app and --gateway
mysql
mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
WARNING: option --ssl-verify-server-cert is disabled, because of an insecure passwordless login.
Pending: (Failures listed here are expected and do not affect your suite's status)
1) bin/hanami console irb starts irb console
# Temporarily skipped with xit
# ./spec/integration/run_spec.rb:59
Finished in 9.42 seconds (files took 0.55687 seconds to load)
47 examples, 0 failures, 1 pending
And a somewhat related question:
Could setup for running the tests be documented somewhere in the repository, perhaps the README? I find myself reverse-engineering Github Actions setup every time, just to figure out how to run against different databases and perhaps I'm doing something wrong in that regard.
The command I used to run above is:
MYSQL_BASE_URL=mysql2://[email protected]/hanami_cli_test POSTGRES_BASE_URL=postgres://postgres:password@localhost:5432/hanami_cli_test bundle exec rake spec
(tried replacing with bundle exec rspec but the outcome is the same)
When I try to run the tests for this repo locally, sometimes my test run looks like this:
Other times like this:
Either way, the whole suite is pretty much never completed. It looks like something (probably around mysql tests) kills the test process abruptly, but in a way that the RSpec runner thinks that all the test are run. I'm not sure how to start debugging it. Running with
docformat does not reveal any further hints.Example:
Another example (truncated)
And a somewhat related question:
Could setup for running the tests be documented somewhere in the repository, perhaps the README? I find myself reverse-engineering Github Actions setup every time, just to figure out how to run against different databases and perhaps I'm doing something wrong in that regard.
The command I used to run above is:
(tried replacing with
bundle exec rspecbut the outcome is the same)