Skip to content

Fixing the Sauce Labs tests and some old browser compatibility #2884

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 17 commits into from
Oct 22, 2020
Merged

Fixing the Sauce Labs tests and some old browser compatibility #2884

merged 17 commits into from
Oct 22, 2020

Conversation

jgonggrijp
Copy link
Collaborator

The good:

  • Updated the selection of browser versions to include in Sauce Labs testing.
  • Made the _.throttle and _.debounce tests less likely to randomly fail.
  • Added fixes and workarounds for _.isEqual, _.isDataView, _.isMap, _.isWeakMap and _.isSet in older browsers that have partial support for these types.
  • Made tests more aware of browser support so that Firefox 11 doesn't attempt to run the tests that involve DataView.
  • Overall: Travis CI + Sauce Labs should run more reliable again, hence be more informative.

The bad:

  • Testing of IE 8, Safari 7 and Firefox 4 is no longer possible.
  • Three new internal modules just for workarounds.
  • The workaround for _.isMap, _.isWeakMap and _.isSet in IE 11 is quite large.
  • Minified+gzipped weight increases by 230 bytes.

@jgonggrijp
Update the selection of browser versions in Sauce Labs (jashkenas#2879)
f236594
@jgonggrijp
Drop an old outcomment from the .travis.yml (jashkenas#2857)
96432bf
@jgonggrijp
Run coveralls only in Node 10 (jashkenas#2857)
d3dd874
@jgonggrijp
Add a fallback for byteOffset in toDataView (jashkenas#2879)
2223939 
Some older browsers break if the byteOffset argument is undefined.
@jgonggrijp
Use typed array, not DataView as comparison view in _.isEqual (jashke…
b16b623 
…nas#2879)

Firefox 11 supports typed arrays but not DataView.

@sktguha FYI. This would change the implementation of your
optimization(s) a bit, but I think the possibility is still there.
@jgonggrijp
Make DataView-depending tests conditional on its existence (jashkenas…
21ff8ed 
…#2879)

Some older browsers support typed arrays but not DataView.
@jgonggrijp
Add checks to diagnose and work around IE/edge test errors (jashkenas…
eea411e 
…#2879)
@jgonggrijp
Address a timing issue in the debounce asap test (jashkenas#2879)
8fe0c99
@jgonggrijp
Work around incorrect string tags in IE 10 - Edge 13 (jashkenas#2879)
376e03f
@jgonggrijp
Remove the diagnostic checks from eea411e (jashkenas#2879)
0061801 
This partially reverts 26356a98d760cb4a57c138a25.
@jgonggrijp
Add new diagnostic checks for still failing tests in IE 11 (jashkenas…
0fffc03 
…#2879)
@jgonggrijp
Take non-enumerability of Map/WeakMap/Set methods into account (jashk…
59494c9 
…enas#2879)
@jgonggrijp
Update diagnostics from 0fffc03 to match 59494c9 (jashkenas#2879)
7380637
@jgonggrijp
Fix an oversight that caused _.isWeakMap(new Map) == true (jashkenas#…
74b7d3a 
…2879)
@jgonggrijp
Remove the diagnostics from 0fffc03+73806370 again (jashkenas#2879)
5e3f191
@jgonggrijp
Add 4ms slack to _.throttle tests sensitive to setTimeout imprecision
3bc482d 
These tests fail relatively often, randomly, though more in some
engines than in others. The random failures are a combination of a
race condition and the imprecision of the global setTimeout function.

This measure shouldn't be taken lightly. It can only be applied if it
does not affect the timing of the throttled (or debounced) function
invocations. In the current patch, only the time at which the result
is read out is slightly delayed so that the readout is less likely to
be too early.

Part of jashkenas#2879, also related to jashkenas#2802.
@jgonggrijp
Make the new IE fallback code a bit more compact (jashkenas#2879)
dce5d42
@jgonggrijp jgonggrijp requested a review from jashkenas October 11, 2020 22:55
@coveralls
Copy link

Coverage Status

Coverage decreased (-1.1%) to 95.16% when pulling dce5d42 on jgonggrijp:saucelabs-fix into d9741b3 on jashkenas:master.

@jgonggrijp
Copy link
Collaborator Author

Coverage decreases because coveralls runs in an engine that doesn't need the workaround code.

@jgonggrijp
Copy link
Collaborator Author

Successful build here: https://travis-ci.org/github/jashkenas/underscore/builds/734840508

@jashkenas
Copy link
Owner

Sorry for the slowness!

@jgonggrijp
Copy link
Collaborator Author

No problem, thanks for reviewing!

@jgonggrijp jgonggrijp merged commit 1964cdb into jashkenas:master Oct 22, 2020
@jgonggrijp jgonggrijp deleted the saucelabs-fix branch October 22, 2020 20:42
jgonggrijp referenced this pull request in jgonggrijp/underscore Oct 22, 2020
@jgonggrijp
Use typed array,not DataView as comparison view in _.isEqual (jashken…
b8bb87b 
…as#2879)

@sktguha FYI. This would change the implementation of your
optimization(s) a bit, but I think the possibility is still there.
@jgonggrijp jgonggrijp mentioned this pull request Oct 24, 2020
Merged
jgonggrijp added a commit that referenced this pull request Oct 24, 2020
@jgonggrijp
Include #2884 in the change log
605028e
@jgonggrijp jgonggrijp mentioned this pull request Nov 5, 2020
Open
@jgonggrijp jgonggrijp mentioned this pull request Nov 25, 2020
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jashkenas jashkenas jashkenas approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@jgonggrijp @coveralls @jashkenas