Add new type: confusing and type: use-case labels to the triage readme as well
as clarify that freq and severity are only required for type: bug/fix
PR Close#38081
Some issue reports don't really fall into any of the current buckets that count
towards triage level 2: bug/fix, feature, or refactor. Some reports are:
* working as intended but confusing - the labels might be 'type: confusing', 'comp: docs', 'comp: router'
* generally working as originally designed but a use-case could be argued for a different implementation.
This type of report is a little hard to triage; it may be neither a bug, nor feature but more of a
'type: use-case'. These may eventually turn into a bug/fix or feature, but can't necessarily be
put in those buckets immediately.
PR Close#38081
HTML is very lenient when it comes to closing elements, so Angular's parser has
rules that specify which elements are implicitly closed when closing a tag.
The parser keeps track of the nesting of tag names using a stack and parsing
a closing tag will pop as many elements off the stack as possible, provided
that the elements can be implicitly closed.
For example, consider the following templates:
- `<div><br></div>`, the `<br>` is implicitly closed when parsing `</div>`,
because `<br>` is a void element.
- `<div><p></div>`, the `<p>` is implicitly closed when parsing `</div>`,
as `<p>` is allowed to be closed by the closing of its parent element.
- `<ul><li>A <li>B</ul>`, the first `<li>` is implicitly closed when parsing
the second `<li>`, whereas the second `<li>` would be implicitly closed when
parsing the `</ul>`.
In all the cases above the parsed structure would be correct, however the source
span of the closing `</div>` would incorrectly be assigned to the element that
is implicitly closed. The problem was that closing an element would associate
the source span with the element at the top of the stack, however this may not
be the element that is actually being closed if some elements would be
implicitly closed.
This commit fixes the issue by assigning the end source span with the element
on the stack that is actually being closed. Any implicitly closed elements that
are popped off the stack will not be assigned an end source span, as the
implicit closing implies that no ending element is present.
Note that there is a difference between self-closed elements such as `<input/>`
and implicitly closed elements such as `<input>`. The former does have an end
source span (identical to its start source span) whereas the latter does not.
Fixes#36118
Resolves FW-2004
PR Close#38126
Adds a note to the provider docs that users shouldn't mutate an array that
is returned from a `multi` provider, because it can cause unforeseen
consequences in other parts of the app.
Closes#37481.
PR Close#37645
Docs state that `strictInjectionParameters` is true by default in `ng new`, however this is not the case in `10.0.1`. It is only set when `--strict` is provided. Clarified that the `--strict` flag is required.
`strictTemplates` does not mention anything about `--strict`, so I included a similar point that it is `true` when a new project is generated with `--strict`.
PR Close#37982
We currently use 16 bits to store information about nodes in a view.
The 16 bits give us 65536 entries in the array, but the problem is that while
the number is large, it can be reached by ~4300 directive instances with host
bindings which could realistically happen is a very large view, as seen in #37876.
Once we hit the limit, we end up overflowing which eventually leads to a runtime error.
These changes bump to using 20 bits which gives us around 1048576 entries in
the array or 16 times more than the current amount which could still technically
be reached, but is much less likely and the user may start hitting browser limitations
by that point.
I picked the 20 bit number since it gives us enough buffer over the 16 bit one,
while not being as massive as a 24 bit or 32 bit.
I've also added a dev mode assertion so it's easier to track down if it happens
again in the future.
Fixes#37876.
PR Close#38014
Fix two issues that affected displaying of SVG icons in IE11:
1. All SVG icons except for one appeared empty. This was related how the
CustomIconRegistry re-used the same <div> element to create all
SVG elements.
2. The GitHub and Twitter buttons next to the search bar were not sized
properly.
Fixes#37847
PR Close#38046
This commit adds a script to build @angular/language-service
locally so that it can be consumed by the Angular extension for
local development.
PR Close#38103
We recently reworked our `ng_rollup_bundle` rule to no longer output
ESM5 and to optimize applications properly (previously applications were
not optimized properly due to incorrect build optimizer setup).
This change meant that a lot of symbols have been removed from the
golden correctly. See: fd65958b88
Unfortunately though, a few symbols have been accidentally removed
because they are now part of the bundle as ES2015 classes which the
symbol extractor does not pick up. This commit fixes the symbol
extractor to capture ES2015 classes. We also update the golden to
reflect this change.
PR Close#38093
Previously, the i18n message extractor just quietly ignored messages that
it extracted that had the same id. It can be helpful to identify these
to track down messages that have the same id but different message text.
Now the messages are checked for duplicate ids with different message text.
Any that are found can be reported based on the new `--duplicateMessageHandling`
command line option (or `duplicateMessageHandling` API options property).
* "ignore" - no action is taken
* "warning" - a diagnostic warning is written to the logger
* "error" - the extractor throws an error and exits
Fixes#38077
PR Close#38082
Currently the Ivy language service bundle is [10MB](
https://unpkg.com/browse/@angular/language-service@10.0.4/bundles/) because we
accidentally included typescript in the bundle.
With this change, the bundle size goes down to 1.6MB, which is even smaller
than the View Engine bundle (1.8MB).
```bash
$ yarn bazel build //packages/language-service/bundles:ivy
$ ls -lh dist/bin/packages/language-service/bundles/ivy.umd.js
1.6M Jul 15 15:49 dist/bin/packages/language-service/bundles/ivy.umd.js
```
PR Close#38088
`ls_rollup_bundle` is no longer needed since we could invoke `ng_rollup_bundle`
directly.
Background: language service runs rollup to produce a single file to reduce
startup time in the editor. However, due to the need to load dynamic versions
of typescript at runtime (think the case where users can change typescript
version in their editor), we hack the "banner" to export a CommonJS default function,
so that we could dynamically load the typescript module provided at runtime via AMD
and use it throughout the implementation.
PR Close#38086
* Add petebacondarwin to public-api, size-tracking, and circular-dependencies
* Add mhevery, josephperrott, and jelbourn to code-ownership
PR Close#37994
Currently we read lifecycle hooks eagerly during `ɵɵdefineComponent`.
The result is that it is not possible to do any sort of meta-programing
such as mixins or adding lifecycle hooks using custom decorators since
any such code executes after `ɵɵdefineComponent` has extracted the
lifecycle hooks from the prototype. Additionally the behavior is
inconsistent between AOT and JIT mode. In JIT mode overriding lifecycle
hooks is possible because the whole `ɵɵdefineComponent` is placed in
getter which is executed lazily. This is because JIT mode must compile a
template which can be specified as `templateURL` and those we are
waiting for its resolution.
- `+` `ɵɵdefineComponent` becomes smaller as it no longer needs to copy
lifecycle hooks from prototype to `ComponentDef`
- `-` `ɵɵNgOnChangesFeature` feature is now always included with the
codebase as it is no longer tree shakable.
Previously we have read lifecycle hooks from prototype in the
`ɵɵdefineComponent` so that lifecycle hook access would be monomorphic.
This decision was made before we had `T*` data structures. By not
reading the lifecycle hooks we are moving the megamorhic read form
`ɵɵdefineComponent` to instructions. However, the reads happen on
`firstTemplatePass` only and are subsequently cached in the `T*` data
structures. The result is that the overall performance should be same
(or slightly better as the intermediate `ComponentDef` has been
removed.)
- [ ] Remove `ɵɵNgOnChangesFeature` from compiler. (It will no longer
be a feature.)
- [ ] Discuss the future of `Features` as they hinder meta-programing.
Fix#30497
PR Close#35464
Fixes the following issues related to how we validate properties during JIT:
- The invalid property warning was printing `null` as the node name
for `ng-content`. The problem is that when generating a template from
`ng-content` we weren't capturing the node name.
- We weren't running property validation on `ng-container` at all.
This used to be supported on ViewEngine and seems like an oversight.
In the process of making these changes, I found and cleaned up a
few places where we were passing in `LView` unnecessarily.
PR Close#37773
The function was removed by default in Bazel 0.27.
It is still accessible with the flag `--incompatible_new_actions_api`
(which is set in Google code base), but the flag will be deleted very soon.
This change should be a no-op for Bazel users. The change was tested in
Google (cl/318277076) and should be safe as well.
PR Close#38080
As part of our CI checks, we ensure the a11y score on certain angular.io
pages do not fall below some thresholds.
This commit increases these thresholds based on our current scores to
ensure we do not regress below current values.
PR Close#37899
* Without this BUILD file we were seeing errors about the reference to
expand_template.bzl in ng_rollup_bundle.bzl because dev-infra/bazel
was not considered a package.
PR Close#38026
Adds Firefox as browser to `dev-infra/browsers` with RBE
compatibility. The default Firefox browser is not compatible similar to
the default Chromium version exposed by `rules_webtesting`.
The Angular Components repository will use this browser target as
it enables RBE support. Also it gives us more flexibility about
the Firefox version we test against. The version provided by
`rules_webtesting` is very old and most likely not frequently
updated (based on past experience).
PR Close#38029
Angular Console has been renamed and links no longer work. It has been decided to remove references to this third-party tool from the AIO documentation.
Closes#37604
PR Close#37608
In CLI v10 there was a move to use the new solution-style tsconfig
which became available in TS 3.9.
The result of this is that the standard tsconfig.json no longer contains
important information such as "paths" mappings, which ngcc might need to
correctly compute dependencies.
ngcc (and ngc and tsc) infer the path to tsconfig.json if not given an
explicit tsconfig file-path. But now that means it infers the solution
tsconfig rather than one that contains the useful information it used to
get.
This commit logs a warning in this case to inform the developer
that they might not have meant to load this tsconfig and offer
alternative options.
Fixes#36386
PR Close#38003
Remove an article from the `Data flow in forms` section of the forms overview guide. The use of `the` and `a` together
is not syntactically correct.
PR Close#37933
In #37957, parts of the testing guide were broken out into separate
guides. As part of that work, the `<live-example>` tags were also copied
to the new guides. These `<live-example>` tags did not specify the
targeted example project via the `name` attribute, thus they were
implicitly targeting the example with the same name as the guide they
were in. See the [Docs style guide][1] for more info.
However, there is only one example project (`testing/`) and all
`<live-example>` tags were supposed to target that. This worked fine on
the `testing.md` guide, but it broke on other guides (which tried to
target non-existing example projects based on their names).
This commit fixes it by explicitly specifying which example is targeted
by the `<live-example>` tags. It also removes the `embedded-style`
attribute that has no effect.
[1]: https://angular.io/guide/docs-style-guide#live-examplesFixes#38036
PR Close#38038
The current method of handling duplicate navigations caused by 'hashchange' and 'popstate' events for the same url change does not correctly handle cancelled navigations. Because `scheduleNavigation` is called in a `setTimeout` in the location change subscription, the duplicate navigations are not flushed at the same time. This means that if the initial navigation hits a guard that schedules a new navigation, the navigation for the duplicate event will not compare to the correct transition (because we inserted another navigation between the duplicates). See https://github.com/angular/angular/issues/16710#issuecomment-646919529Fixes#16710
PR Close#37674
With `rules_nodejs` v2.0.0 being in RC phase currently, we should
make sure that the package is compatible so that we can use it
in the components repo in combination with rules_nodejs v2.0.0.
In v2.0.0 of the NodeJS rules, Bazel workspaces are recommended
to no longer be symlinked under a separate repository. Instead,
bazel rules and targets should be accessed directly from the
user-selected NPM repository. Usually `@npm`, so that the import
changes to `@npm//@angular/dev-infra-private/<..>`.
PR Close#37968
Sonu Kapoor
Andrew Scott
JoostK
crisbeto
Douglas Parker
Misko Hevery
Saif
Keen Yee Liau
Paul Gschwendtner
Pete Bacon Darwin
Judy Bogart
Andrew Kushnir
Keen Yee Liau
Jeremy Elbourn
Ajit Singh
Emma Twersky
George Kalpakas
Laurent Le Brun
Wagner Maciel
Windvis
Sonu Kapoor
danranVm
Aristeidis Bampakos
David Martinez Barreiro
RN Lee
Samuel
ivanwonder
Shapovalov-Dmitry
Krzysztof Platis