Past npm versions required the `--save` flag to save anything in `package.json`
when installing packages but newer ones do this by default. Pass `--no-save` to
restore the original behavior in the `release-test.js` script.
Ref gh-1966
jQuery UI 1.12.1 was released using an outdated (at the time of its release)
version of the download.jqueryui.com package; the latest ones wouldn't generate
relevant JS files.
This commit accounts for the changes so that the latest version can be used.
Closes gh-1965
Add a new option named onUpdateDatepicker that allows a custom callback
to be provided. If provided, the callback is called at the end of
$.datepicker._updateDatepicker.
Instead of using enormous HTML strings, various elements are now constructed
using jQuery APIs. This makes it more obvious user-provided data is used
correctly.
Fixes#15284
Closes gh-1953
In IE in jQuery 3.4+ a sequence:
```js
$( inputNode ).trigger( "focus" ).trigger( "blur" ).trigger( "focus" );
```
doesn't end up with a focused input. However, in this test we only want to check
that the last focused input receives the focus back when `_focusTabbable()` is
called which in reality doesn't happen so quickly so let's avoid the issue by
waiting a bit.
Ref jquery/jquery#4856
Closes gh-1951
Some bad pages are being detected as Catalan randomly which makes the build
fail. This is reproducible both locally & on Travis. PR gh-1949 added a new
regex to account for this error but it didn't escape parens properly so it's
not matching the problematic error message.
Ref gh-1949
Closes gh-1950
Some of the APIs return fractional values in newer jQueries, making comparisons
sometimes not being 100% accurate. Allow some delta.
This is similar to what was already done in
98b539171b but a few cases affecting IE and/or
Edge Legacy were missed.
Closes gh-1947
Some bad pages are being detected as Catalan randomly which makes the build
fail. This is reproducible both locally & on Travis. This commit adds a new
regex to account for this error.
Closes gh-1949
Focus re-triggering in jQuery 3.4/3.5 makes the original element
have its focus event propagated last, breaking the re-targeting.
Trigger focus in a delay in addition if needed to avoid the issue.
This fixes the "interaction between overlay and other dialogs" core dialog
test when tested against jQuery 3.4/3.5.
Closes gh-1946
Ref jquery/jquery#4382
jQuery >=3.4.0 uses a special focus/blur handler pair needed to fix various
issues with checkboxes/radio buttons as well as being able to pass data in focus
triggers. This leaves extra focus & blur events if any of these events were ever
listened to at a particular element.
We've started skipping these handlers in the `domEqual` assertion in gh-1930 but
we missed a case where an event is triggered before any handler is attached -
jQuery >=3.4.0 attaches then an extra noop listener just to force the code path
to go through the setup code before the trigger happens. We now skip this extra
handler as well.
This fixes a test failure in "dialog: methods" destroy tests.
Closes gh-1945
Ref jquery/jquery#4496
Ref gh-1930
Fix for `options === true` when using jQuery UI under `use strict`,
which throws:
```
Uncaught TypeError: Cannot create property 'complete' on boolean 'true'
```
on line:
```js
options.complete = callback;
```
Closes gh-1931
jQuery >=3.2 doesn't include scrollbars in `.height()`, this commit switches
it to `.innerHeight()` which does so in jQuery >=3.3. In jQuery 3.2 it doesn't
either so include scrollbars in innerHeight, add it back.
Using `.innerHeight()` instead of `.height()` should be fine as menu doesn't
define padding styles.
Closes gh-1929
jQuery >=3.4.0 uses a special focus/blur handler pair
needed to fix various issues with checkboxes/radio buttons
as well as being able to pass data in focus triggers.
However, this leaves dummy focus & blur events if any of these
events were ever listened to at a particular element. There's not
a lot UI can do to fix this so we now just skip these handlers for
data comparisons in tests.
Ref jquery/jquery#4496
Closes gh-1930
Co-authored-by: Richard Gibson <richard.gibson@gmail.com>
jQuery 3.2 & newer have a different animation logic and the animated elements
width/height differ from the starting ones even at the beginning of the
animation. The point of the assertions checking that they're identical was
to ensure bug #5245 is fixed; that issue manifested by a big jump to half the
element size. To test for that, it's enough to check that the first obtained
values are close to the original ones.
This makes effects tests pass in all supported jQuery versions.
Closes gh-1928
Ref #5245
This helps fix issues that make tooltip tests sometimes fail when run against
jQuery 3.2 or newer due to timing differences.
Details:
* Add the `moduleAfterEach` function ensuring no timers are running.
* Attach this function via `common.testWidget`.
* Attach this function to most test suites.
* Add a tooltip test helper cleaning up leftover timers.
* Rename legacy `setup`/`teardown` hooks to `beforeEach`/`afterEach`.
Closes gh-1920
