You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I frequently use byebug to create breakpoints where I can fiddle with the correct capybara command while writing/editing tests. It seems that when I do this, my test will freeze. It seems to be inconsistent. Sometimes it happens, other times it doesn't - sorry for the lack of reproducible case. I've seen this happens with the headless option set to either true or false.
Here's what I do:
Add byebug somewhere in my test suite
run the test with: ./bin/rails test path/to/test.rb
watch browser open, test run and wait for breakpoint
attempt some finder at the byebug promt (for example: find('#some_ele'))
test freezes(?) - the command doesn't return. Nothing happens in the browser.
I cannot do anything at this point. crtl-c doesn't stop the running test, using kill <pid> doesn't work either. I have to resort to kill -9 <pid>
As stated above, I still don't have a concrete reproducible test case. In some test runs, using the same breakpoint and the same finder command it works. Other times it does not. I'd say it happens about 50% of the time.
The text was updated successfully, but these errors were encountered:
perhaps unsurprising, but worth mentioning, when I kill -9 the process, the browser (chrome) is left open. I also cmd-q that before attempting a second run.
I frequently use byebug to create breakpoints where I can fiddle with the correct capybara command while writing/editing tests. It seems that when I do this, my test will freeze. It seems to be inconsistent. Sometimes it happens, other times it doesn't - sorry for the lack of reproducible case. I've seen this happens with the
headless
option set to either true or false.Here's what I do:
byebug
somewhere in my test suite./bin/rails test path/to/test.rb
I cannot do anything at this point. crtl-c doesn't stop the running test, using
kill <pid>
doesn't work either. I have to resort tokill -9 <pid>
As stated above, I still don't have a concrete reproducible test case. In some test runs, using the same breakpoint and the same finder command it works. Other times it does not. I'd say it happens about 50% of the time.
The text was updated successfully, but these errors were encountered: