github.com/benbjohnson/clock #35
Labels
needs investigation
we need to investigate some things about this project
Comments
theckman
added
the
needs investigation
yrashk
added a commit
to yrashk/bpxe
that referenced
this issue
May 21, 2021
For the time being, all timer events are being sent manually in tests. This is not enough for a functioning BPMN engine. Solution: introduce `timer` package This is something to be used by the engine initiate timer events. It's not yet integrated into the engine. Speaking of the mock time implementation, and picking https://github.com/benbjohnson/clock. It is known to have issues and is not actively maintained, here's a short overview: gofrs/help-requests#35 There are other alternatives, and I tried a few of them, but for some reason they don't seem to work for the tests I wrote: https://github.com/derision-test/glock https://github.com/microsoft/clock https://github.com/tilinna/clock I haven't tried the approach laid out in https://github.com/cabify/timex yet. My current hypothesis is that it doesn't really matter which one will be picked as long as it facilitates proper testing, as they all delegate to `time` package for non-mocking use cases. That being said, the tests of `timer` package are a bit on the slow side, and that's a nuisance.
yrashk
added a commit
to yrashk/bpxe
that referenced
this issue
May 22, 2021
For the time being, all timer events are being sent manually in tests. This is not enough for a functioning BPMN engine. Solution: introduce `timer` package This is something to be used by the engine initiate timer events. It's not yet integrated into the engine. Speaking of the mock time implementation, and picking https://github.com/benbjohnson/clock. It is known to have issues and is not actively maintained, here's a short overview: gofrs/help-requests#35 There are other alternatives, and I tried a few of them, but for some reason they don't seem to work for the tests I wrote: https://github.com/derision-test/glock https://github.com/microsoft/clock https://github.com/tilinna/clock I haven't tried the approach laid out in https://github.com/cabify/timex yet. My current hypothesis is that it doesn't really matter which one will be picked as long as it facilitates proper testing, as they all delegate to `time` package for non-mocking use cases. That being said, the tests of `timer` package are a bit on the slow side, and that's a nuisance.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
With subsequent forks:
None.
github.com/benbjohnson/clock: 2016/12/16github.com/facebookgo/clock: 2015/04/09github.com/andres-erbsen/clock: 2016/05/26All of the listed repos have a lack of activity, and some outstanding issues / PRs. There is no visible request for assistance.
Yes, each project has its share of bugs. Some ones to note:
It seems like a small API that offers an interesting solution to a problem quite a few may need to solve. I've ran in to similar pains with testing, and would have definitely tried to see if this package made my tests easier to manage.
There was a small Slack conversation about this starting here: https://gophers.slack.com/archives/CBP4N9BEU/p1552522641030300
The text was updated successfully, but these errors were encountered: