fix: bind infinite loop check to clock instance #398
Conversation
|
Pushed a commit that's almost there - |
| checkIsNearInfiniteLimit(clock, i); | ||
| clock.next(); |
There was a problem hiding this comment.
moving this after the check fixes all but the setInterval tests
SimenB
changed the title
|
@fatso83 I don't know how to deal with One quick fix is of course to handle |
|
Pushed that latest thing - not ideal, but I don't know how to solve it properly |
|
This looks fine, but TBH I am just too tired to grok the details of this right now, so I cannot really say something very useful about the |
SimenB
force-pushed
the
reset-is-near-on-uninstall
branch
from
ffcc06a
to
7411b8b
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #398 +/- ##
=======================================
Coverage 94.10% 94.10%
=======================================
Files 1 1
Lines 611 611
=======================================
Hits 575 575
Misses 36 36
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
SimenB
force-pushed
the
reset-is-near-on-uninstall
branch
from
7411b8b
to
72adb63
Compare
| @@ -5036,6 +5036,7 @@ describe("loop limit stack trace", function () { | |||
| }); | |||
| }); | |||
|
|
|||
| // This doesn't really work since we're unable to add an error to all running intervals | |||
There was a problem hiding this comment.
Not totally sure what "this" refers to really. The tests or the setInteval function or what.
There was a problem hiding this comment.
tracking of callbacks triggered by setInterval
|
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
|
Is this going somewhere? Can I help? Not totally sure what to do between your last commend and Benji's approval. |
Purpose (TL;DR) - mandatory
Failing test from #375 (comment).
By not resetting, the counter is not tied to a specific clock but rather then entire
withGlobal.I'll push a second commit moving the attribute from living in the
withGlobalclosure to living on theclockobject, which I think is more correct