Skip to content

Conversation

@lg2de
Copy link
Contributor

@lg2de lg2de commented May 8, 2020

@lg2de lg2de marked this pull request as ready for review June 5, 2020 21:17
@lg2de lg2de changed the title [WIP] do not remove existing synchronization context on async assertions keep synchronization context of executing thread when asserting asynchronous operations Jun 5, 2020
@lg2de
Copy link
Contributor Author

lg2de commented Jun 6, 2020

Hey @dennisdoomen and @jnyrup , the PR is ready for review.
The failed test is IMHO not related to my changed, but a common timing issue.

Copy link
Member

@jnyrup jnyrup left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nothing strikes out to me, but as mentioned in Slack discussions this is not my strongest topic.
I assume we have the complementary async tests for the removed sync tests.

If I e.g. have existing code that for whatever reason mixes sync and async code, can this PR make it deadlock on a CI-build, where it would previously have passed?

@dennisdoomen dennisdoomen changed the title keep synchronization context of executing thread when asserting asynchronous operations Keep synchronization context of executing thread when asserting asynchronous operations Jul 1, 2020
@dennisdoomen dennisdoomen requested a review from jnyrup July 1, 2020 05:18
@jnyrup
Copy link
Member

jnyrup commented Jul 20, 2020

Nothing strikes out to me, but as mentioned in Slack discussions this is not my strongest topic.
I assume we have the complementary async tests for the removed sync tests.

If I e.g. have existing code that for whatever reason mixes sync and async code, can this PR make it deadlock on a CI-build, where it would previously have passed?

@lg2de Do you have any comments on this case?
I'd like to avoid causing hanging tests due to bad, yet "working" user code.

@lg2de
Copy link
Contributor Author

lg2de commented Jul 20, 2020

Do you have any comments on this case?
I'd like to avoid causing hanging tests due to bad, yet "working" user code.

I thought, we had already discussed this.
Yes, it may happen that "wrong code" will run into deadlock.
But it is not the scope of FluentAssertion to work-around issues of user code.

@jnyrup
Copy link
Member

jnyrup commented Jul 20, 2020

We might already have discussed it, in which case it simply has slipped my mind.
Thanks for re-stating it.

@dennisdoomen
Copy link
Member

Shall we squash this PR or do you plan to clean-up the commits so we can keep some history?

@lg2de
Copy link
Contributor Author

lg2de commented Jul 20, 2020

All the recent cleanups I collected into single (force-pushed) "cleanup commit". In total we have 5 commits in this PR. I hope this is ok. It shows the development steps.

@dennisdoomen dennisdoomen merged commit 4396c8f into fluentassertions:develop Jul 20, 2020
@dennisdoomen
Copy link
Member

dennisdoomen commented Jul 20, 2020

Now I'm wondering whether we should ship this as an alpha version...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants