BREAKING CHANGE:
The `blur` and `focus` event fire synchronously, also during DOM operations
that remove elements. This lead to errors as the Angular model was not
in a consistent state. See this [fiddle](http://jsfiddle.net/fq1dq5yb/) for a demo.
This change executes the expression of those events using
`scope.$evalAsync` if an `$apply` is in progress, otherwise
keeps the old behavior.
Fixes#4979Fixes#5945Closes#8803Closes#6910Closes#5402
We run unit tests in “strict” mode and thus can’t monkey-patch `window.location` nor `window.location.reload`. In order to avoid full page reload, we could pass location as argument, or another level of indirection, something like this:
```js
var ourGlobalFunkyLocation = window.location;
function reloadWithDebugInfo() {
window.name = 'NG_ENABLE_DEBUG_INFO!' + window.name;
ourGlobalFunkyLocation.reload();
}
// in the test
ourGlobalFunkyLocation = {
reload: function() {}
};
reloadWithDebugInfo();
ourGlobalFunkyLocation = window.location;
```
I don’t think any of these make sense, just so that we can test setting `window.name`. If the `reloadWithDebugInfo` function was more complicated, I would do it.
I don’t think it’s worthy to confuse production code with extra logic which purpose was only to make testing possible.
The compiler adds scope information (`ng-scope` CSS class and `$scope` data property) to elements
when the are bound to the scope. This is mostly to aid debugging tools such as Batarang. In
production this should be unnecesary and adds a performance penalty.
In the bench/apps/largetable-bp this change caused an improvement of ~100ms (7%).
This can be now disabled by calling `$compileProvider.debugInfoEnabled(false)`
in a module `config` block:
```
someModule.config(['$compileProvider', function($compileProvider) {
$compileProvider.debugInfoEnabled(false);
}]);
```
In the bench/apps/largetable-bp benchmark this change, with debug info disabled,
improved by ~120ms, that is ~10%.
Measuring the "create" phase, 25 loops, mean time ~1200ms -> ~1080ms.
The compiler and ngBind directives add binding information (`ng-binding`
CSS class and `$binding` data property) to elements when they are bound to
the scope. This is only to aid testing and debugging for tools such as
Protractor and Batarang. In production this is unnecessary and add a
performance penalty.
This can be now disabled by calling `$compileProvider.debugInfoEnabled(false)`
in a module `config` block:
```
someModule.config(['$compileProvider', function($compileProvider) {
$compileProvider.debugInfoEnabled(false);
}]);
```
In the bench/apps/largetable-bp benchmark this change, with debug info disabled,
improved by ~140ms, that is 10%.
Measuring the "create" phase, 25 loops, mean time ~1340ms -> ~1200ms.
We were storing the whole `interpolationFn` in the `$binding` data on
elements but this function was bringing a lot of closure variables with it
and so was consuming unwanted amounts of memory.
Now we are only storing the parsed interpolation expressions from the
binding (i.e. the values of `interpolationFn.expressions`).
BREAKING CHANGE:
The value of `$binding` data property on an element is always an array now
and the expressions do not include the curly braces `{{ ... }}`.
Prior to this fix when an Angular application is bootstrapped it would only
place an animation guard to prevent animations from running when the application
starts for the first two digest cycles. However, if any controllers or directives,
that are executed upon boostrap, trigger any remote code to be downloaded (via $http)
then the guard does not put that into consideration. This fix now properly addresses
that circumstance and removes the guard once all outbound HTTP requests are complete
when an Angular application is bootstrapped.
Closes#8275Closes#5262
This handy service is designed to download and cache template contents
and to throw an error when a template request fails.
BREAKING CHANGE
Angular will now throw a $compile minErr each a template fails to download
for ngView, directives and ngMessage template requests. This changes the former
behavior of silently ignoring failed HTTP requests--or when the template itself
is empty. Please ensure that all directive, ngView and ngMessage code now properly
addresses this scenario. NgInclude is uneffected from this change.
When multiple responses are received within a short window from each other, it can be wasteful to
perform full dirty-checking cycles for each individual response. In order to prevent this, it is
now possible to coalesce calls to $apply for responses which occur close together.
This behaviour is opt-in, and the default is disabled, in order to avoid breaking tests or
applications.
In order to activate coalesced apply in tests or in an application, simply perform the following
steps during configuration.
angular.module('myFancyApp', []).
config(function($httpProvider) {
$httpProvider.useApplyAsync(true);
});
OR:
angular.mock.module(function($httpProvider) {
$httpProvider.useApplyAsync(true);
});
Closes#8736Closes#7634Closes#5297
It is now possible to queue up multiple expressions to be evaluated in a single digest using
$applyAsync. The asynchronous expressions will be evaluated either 1) the next time $apply or
$rootScope.$digest is called, or 2) after after the queue flushing scheduled for the next turn
occurs (roughly ~10ms depending on browser and application).
This commit introduces a 2nd validation queue called `$asyncValidators`. Each time a value
is processed by the validation pipeline, if all synchronous `$validators` succeed, the value
is then passed through the `$asyncValidators` validation queue. These validators should return
a promise. Rejection of a validation promise indicates a failed validation.
With this commit, ngModel will now handle parsing first and then validation
afterwards once the parsing is successful. If any parser along the way returns
`undefined` then ngModel will break the chain of parsing and register a
a parser error represented by the type of input that is being collected
(e.g. number, date, datetime, url, etc...). If a parser fails for a standard
text input field then an error of `parse` will be placed on `model.$error`.
BREAKING CHANGE
Any parser code from before that returned an `undefined` value
(or nothing at all) will now cause a parser failure. When this occurs
none of the validators present in `$validators` will run until the parser
error is gone.
The HTML5 spec allows to use seconds for `input[time]` and `input[datetime-local]`,
even though they are not displayed by all browsers.
Related to #8447.
Angular used to always use the browser timezone when parsing
`input[date]`, `input[time]`, … The timezone can now be changed
to `UTC` via `ngModelOptions`.
Closes#8447.
BREAKING CHANGE:
According to the HTML5 spec `input[time]` should create dates
based on the year 1970 (used to be based on the year 1900).
Related to #8447.
Angular used to always use the browser timezone for
`dateFilter`. An additional parameter was added to allow to use
`UTC` timezone instead.
Related to #8447.
In some cases, the type of Error thrown by minErr is meaningful, such as in $q where a TypeError
is sometimes required. This fix allows providing an error constructor as the second argument to
minErr, which will be used to construct the error that gets returned by the factory function.
When transition-delay and animation-delay were used to drive the staggering
animation the result was unpredictable at times due to the browser not being
able to register the generated delay styles in time. This caused a hard to
track down bug that didn't have a solid solution when styles were being used.
This fix ensures that stagger delays are handled by the $timeout service.
Closes#7228Closes#7547Closes#8297Closes#8547
BREAKING CHANGE
If any stagger code consisted of having BOTH transition staggers and delay staggers
together then that will not work the same way. Angular will now instead choose
the highest stagger delay value and set the timeout to wait for that before
applying the active CSS class.
The $animate service (both the service inside of ng and ngAnimate) now
makes use of promises instead of callback functions.
BREAKING CHANGE
Both the API for the cancallation method and the done callback for
$animate animations is different. Instead of using a callback function
for each of the $animate animation methods, a promise is used instead.
```js
//before
$animate.enter(element, container, null, callbackFn);
//after
$animate.enter(element, container).then(callbackFn);
```
The animation can now be cancelled via `$animate.cancel(promise)`.
```js
//before
var cancelFn = $animate.enter(element, container);
cancelFn(); //cancels the animation
//after
var promise = $animate.enter(element, container);
$animate.cancel(promise); //cancels the animation
```
All class-based animation methods (addClass, removeClass and setClass) on $animate
are now processed after the next digest occurs. This fix prevents any sequencing
errors from occuring from excessive calls to $animate.addClass, $animate.remoteClass
or $animate.setClass.
BREAKING CHANGE
$animate.addClass, $animate.removeClass and $animate.setClass will no longer start the animation
right after being called in the directive code. The animation will only commence once a digest
has passed. This means that all animation-related testing code requires an extra digest to kick
off the animation.
```js
//before this fix
$animate.addClass(element, 'super');
expect(element).toHaveClass('super');
//now
$animate.addClass(element, 'super');
$rootScope.$digest();
expect(element).toHaveClass('super');
```
$animate will also tally the amount of times classes are added and removed and only animate
the left over classes once the digest kicks in. This means that for any directive code that
adds and removes the same CSS class on the same element then this may result in no animation
being triggered at all.
```js
$animate.addClass(element, 'klass');
$animate.removeClass(element, 'klass');
$rootScope.$digest();
//nothing happens...
```
Via transclusion, svg elements can occur outside an `<svg>` container in an
Angular template but are put into an `<svg>` container through compilation
and linking.
E.g.
Given that `svg-container` is a transcluding directive with
the following template:
```
<svg ng-transclude></svg>
```
The following markup creates a `<circle>` inside of an `<svg>` element
during runtime:
```
<svg-container>
<circle></circle>
</svg-container>
```
However, this produces non working `<circle>` elements, as svg elements
need to be created inside of an `<svg>` element.
This change detects for most cases the correct namespace of transcluded content
and recreates that content in the correct `<svg>` container
when needed during compilation. For special cases it adds an addition argument
to `$transclude` that allows to specify the future parent node of elements
that will be cloned and attached using the `cloneAttachFn`.
Related to #8494Closes#8716
Also corrects the tests for MathML that use `directive.templateNamespace`.
BREAKING CHANGE (within 1.3.0-beta): `directive.type` was renamed to `directive.templateNamespace`
The property name `type` was too general.
Currently if a reserved word occurs anywhere within the aliasAs identifier, we throw. This CL fixes
this behaviour by allowing these identifiers, since they are technically perfectly valid.
Closes#8729
Make angular.equals() Date comparison NaN-aware to prevent infinite digest errors when a dealy watched
date has an invalid value.
Closes#8650Closes#8715
allOrNothing interpolation is now used for ng-attr-*, under all circumstances. This prevents
uninitialized attributes from being added to the DOM with invalid values which cause errors
to be shown.
BREAKING CHANGE:
Now, ng-attr-* will never add the attribute to the DOM if any of the interpolated expressions
evaluate to `undefined`.
To work around this, initialize values which are intended to be the empty string with the
empty string:
For example, given the following markup:
<div ng-attr-style="border-radius: {{value}}{{units}}"></div>
If $scope.value is `4`, and $scope.units is undefined, the resulting markup is unchanged:
<div ng-attr-style="border-radius: {{value}}{{units}}"></div>
However, if $scope.units is `""`, then the resulting markup is updated:
<div ng-attr-style="border-radius: {{value}}{{units}}" style="border-radius: 4"></div>
Closes#8376Closes#8399
Do not trim input[type=password] values
BREAKING CHANGE:
Previously, input[type=password] would trim values by default, and would require an explicit ng-trim="false"
to disable the trimming behaviour. After this CL, ng-trim no longer effects input[type=password], and will
never trim the password value.
Closes#8250Closes#8230
Ensure that aliasAs expressions are valid simple identifiers. These are still assigned to $scope in the same way
that they were previously, however now you won't accidentally create a property named "filtered.collection".
This change additionally restricts identifiers to prevent the use of certain ECMAScript reserved words ("null",
"undefined", "this" --- should probably add "super", "try", "catch" and "finally" there too), as well as certain
properties used by $scope or ngRepeat, including $parent, $index, $even, $odd, $first, $middle, or $last.
Closes#8438Closes#8440
It is now possible to ask the $compiler's isolate scope property machinery to bind isolate
scope properties to a controller rather than scope itself. This feature requires the use of
controllerAs, so that the controller-bound properties may still be referenced from binding
expressions in views.
The current syntax is to prefix the scope name with a '@', like so:
scope: {
"myData": "=someData",
"myString": "@someInterpolation",
"myExpr": "&someExpr"
},
controllerAs: "someCtrl",
bindtoController: true
The putting of properties within the context of the controller will only occur if
controllerAs is used for an isolate scope with the `bindToController` property of the
directive definition object set to `true`.
Closes#7635Closes#7645
The Promises A+ 1.1 spec introduces new constraints that would cause $q to fail,
particularly specs 2.3.1 and 2.3.3.
Newly satisfied requirements:
* "then" functions that return the same fulfilled/rejected promise
will fail with a TypeError
* Support for edge cases where "then" is a value other than function
Full 1.1 spec: https://github.com/promises-aplus/promises-spec/tree/1.1.0
This commit also modifies the adapter to use "resolve" method instead of "fulfill"
The $sanitize service was returning an empty string to the error page
because the input was usually a single html tag (sometimes it could be
`document`). This fix replaces angle brackets with html entities.
Closes#8683
Array.prototype.forEach will not invoke the callback function if the properety is not present in the
object. Because of this, we have the illusion of not iterating over non-added properties in a sparse
array.
From ECMAScript:
9. Repeat while k < len
a. Let Pk be ToString(k).
b. Let kPresent be HasProperty(O, Pk).
c. ReturnIfAbrupt(kPresent).
d. If kPresent is true, then
i. Let kValue be Get(O, Pk)
... (steps for invoking the function and aborting if it throws)
Closes#8510Closes#8522Closes#8525
Currently, legacy browsers get to use a clever scheme for resolving relative URIs in html5Mode,
and resolve the URI relative to $location.path().
Currently, $location.path() can be '/' under certain circumstances, which means that when we
split $location.path() on '/' and later join by '/' after adding another path component,
we end up with '//pathComponent'. $$rewrite fails to deal with this correctly, and effectively
the $location is never changed from the root path.
This CL corrects this by ensuring that the duplicate '/' situation does not occur when resolving
relative URIs.
Closes#8684
.context is a deprecated jQuery api still being used by at least live() queries, so
we need to keep it in up to date during replacement.
Because of the if check, we can be sure that we replace the context only when jQuery is being
used and the context property is set to the element being replaced.
Closes#8253Closes#7900
This reverts commit 0d608d041f.
The commits caused more breaking changes at Google than initially expected and since its
benefit is small, so it's not worth keeping.
Change jqLite's implementation of wrap() to clone the wrapNode before
wrapping the target element in it.
Match jQuery's wrap() behavior and prevent accidentally attaching
target element to the DOM as a side effect.
Closes#3860Closes#4194
