| Age | Commit message (Collapse) | Author | 
|---|
|  | It's great that IE11 wants to be compatible enough that it doesn't want
to be special cased and treated differently.
However, as long as one has to have a different code path for IE than
for the other supported browsers, we still need to detect and special
case it.  For instance, our URL parsing code still needs the same
workaround the we used for IE10.  We still see the same Access denied /
TypeError exceptions when setting certain values.  FYI, Angular doesn't
generally blindly test for IE – we also check the version number.
Thanks to modern.ie for the free IE11 test VM.
Closes #3682 | 
|  | Nothing would prevent a user from accidentally calling angular.bootstrap on an element that had already been bootstrapped. If this was done, odd behavior could manifest in an application, causing different scopes to update the same DOM, and causing debugger confusion.
This fix adds a check inside of angular.bootstrap to check if the passed-in element already has an injector, and if so, will throw an error. | 
|  | angular.copy previously copied RegExp as an empty object. Change detects
RegExp instance and clones into new RegExp. This change is based on a previous
fix to allow Date to be copied.
Closes #3473
Closes #3474 | 
|  | Change the implementation of isArrayLike to use one heavily based on the
implementation in jQuery in order to correctly detect array-like
objects, that way functionality like ngRepeat works as expected. | 
|  |  | 
|  | - ngAnimate directive is gone and was replaced with class based animations/transitions
- support for triggering animations on css class additions and removals
- done callback was added to all animation apis
- $animation and $animator where merged into a single $animate service with api:
  - $animate.enter(element, parent, after, done);
  - $animate.leave(element, done);
  - $animate.move(element, parent, after, done);
  - $animate.addClass(element, className, done);
  - $animate.removeClass(element, className, done);
BREAKING CHANGE: too many things changed, we'll write up a separate doc with migration instructions | 
|  | angular.equals was returning inconsistent values for the comparison between
{} and []:
    angular.equals({}, []) // true
    angular.equals([], {}]) // false
Since these object are not of the same type, they should not be considered
equivalent. | 
|  | ngScenario expects an ngApp directive to be used, and doesn't work for
manually bootstrapped apps. The failure mode is to hang on navigation.
Trying to make this wont-fix bug less obscure by documenting it.
Eventually Protractor will replace ngScenario and fix this. | 
|  | Regular expression objects didn't used to be considered to be equal when using
'angular.equals'. Dirty checking therefore failed to recognize a
property modification.
Closes #2685 | 
|  | Return early in `angular.toJson` if the object to be stringified is `undefined`.
IE8 stringifies `undefined` to `'undefined'` whereas other browsers return
`undefined`. This normalizes behavior and passes currently broken unit tests
in IE8. | 
|  | Closes #2890 | 
|  |  | 
|  | Merely testing for object[key] will give incorrect results on keys
defined in Object.prototype.
Note: IE8 is generally broken in this regard since `for...in` never returns
certain property keys even if they are defined directly on the object.
See #2141 - partially merges this PR | 
|  |  | 
|  | parseKeyValue and toKeyValue can now handle duplicate values in the query.
```
?x=1&x=2 <-> {x:[1,2]}
```
The algorithm looks like:
    1)parseKeyValue looks for presence of obj[key]
    2)detects and replaces obj[key] with [obj[key],val]
    3)then pushes more duplicates if necessary
    4)toKeyValue decodes array correctly
    5)(not changed)$location.search({param: 'key'}) still replaces if necessary
    6)(not changed)$location.search({param: ['key1', 'key2']}) sets the url with duplicates
BREAKING CHANGE: Before this change:
- `parseKeyValue` only took the last key overwriting all the previous keys;
- `toKeyValue` joined the keys together in a comma delimited string.
This was deemed buggy behavior. If your server relied on this behavior
then either the server should be fixed or a simple serialization of
the array should be done on the client before passing it to $location. | 
|  | In Angular.toJson, any properties with a leading '$' character will be
stripped from the resulting string since angular uses this notation
internally for services.  There have been complaints of not knowing
about this functionality until it breaks within their code. | 
|  |  | 
|  |  | 
|  |  | 
|  | jQuery switched to a completely new event binding implementation as of
1.7.0, centering around on/off methods instead of previous bind/unbind.
This patch makes jqLite match this implementation while still supporting
previous bind/unbind methods. | 
|  |  | 
|  |  | 
|  |  | 
|  | - add toThrowNg matcher | 
|  | When real jQuery is present, Angular monkey patch it to fire `$destroy` event.
This commit fixes two issues in the jQuery patch:
- passing a selector to the $.fn.remove method (only fire `$destroy` on the matched elements)
- using `$.fn.html` without parameters as a getter (do not fire `$destroy`) | 
|  |  | 
|  |  | 
|  | Copying the $$hashKey as part of copy/extend operations makes little
sense since hashkey is used primarily as an object id, especially in
the context of the ngRepeat directive. This change maintains the
existing $$hashKey of an object that is being copied into (likewise for
extend).
It is not uncommon to take an item in a collection, copy it,
and then append it to the collection. By copying the $$hashKey, this
leads to duplicate object errors with the current ngRepeat.
Closes #1875 | 
|  |  | 
|  |  | 
|  |  | 
|  | Use this directive to auto-bootstrap **an** application. | 
|  |  | 
|  |  | 
|  | Remove fromCharCode function as it was used only in two inner
functions in the code, and its functionality is achieved in several
other places by using String.fromCharCode
Breaks fromCharCode closure function, String.fromCharCode should be
used instead | 
|  | This features enables tools like Batarang and test runners to
hook into angular's bootstrap process and sneak in more modules
into the DI registry which can replace or augment DI services for
the purpose of instrumentation or mocking out heavy dependencies.
If window.name contains prefix NG_DEFER_BOOTSTRAP! when
angular.bootstrap is called, the bootstrap process will be paused
until angular.resumeBootstrap is called.
angular.resumeBootstrap takes an optional array of modules that
should be added to the original list of modules that the app was
about to be bootstrapped with. | 
|  | Fixes issue in encodeUriQuery used by $http and $resource that
treats null as a string and replaces the characters "null" with "+". | 
|  |  | 
|  |  | 
|  | This reverts commit 28273b7f1ef52e46d5bc23c41bc7a1492cf23014o.
slice(0) doesn't perform deep copy of the array so its unsuitable
for our purposes. | 
|  |  | 
|  | slice(0) is way faster on most browsers | 
|  | Should handle JQLite, jQuery, NodeList and other objects like arrays
but not other generic objects or instances of user defined types
with length property.
Closes #1840 | 
|  |  | 
|  | Update src/Angular.js
removed redundant 'the' from nextUid()'s ngdoc | 
|  | Remove var Error = window.Error
window.Error is a read-only property in Apps Script.
Igor says, "we should just delete that line instead. I think it was
misko's attempt to get better closure minification, but it turns out
that it's actually hurting us after gzip (I verified it)." | 
|  | in 5ae63fd3 the comparison was made consistent but strict, so that
angular.equals({}, {foo: undefined}) // always returns false
this turns out to cause issues for data that is being roundtripped via network
and serialized via JSON because JSON.stringify serializes {foo: undefined} as {}.
Since angular.equals() behaved like this before the 5ae63fd3 in 50% of the cases,
changing the behavior in this way should not introduce any significant issues.
Closes #1648 | 
|  | previously:
a = {};
b = {x:undefined};
angular.equals(a, b) == angular.equals(b, a) // returns false.
both should return false because these objects are not equal.
unlike in implemented in #1695 the comparison should be stricter
and return false not true. if we need to relax this in the future
we can always do so.
Closes #1648 | 
|  |  | 
|  |  |