Make sure that we actually lazily initialize our storage layers.
With this patch applied, storage layers (history, logins, bookmarks) will be initialized when first
accessed. We will no longer block GeckoEngine init, for example, on waiting for the logins storage
to initialize (which needs to access the costly securePrefStorage).
Similarly, BackgroundServices init will no longer require initialized instances of the storage
components - references to their "lazy wrappers" will suffice.
In practice, this change changes when our storage layers are initialized in the following ways.
Currently, we will initialize everything on startup. This includes loading our megazord, as well.
With this change, init path depends on if the user is signed-into FxA or not.
If user is not an FxA user:
- on startup, none of the storage layers are initialized
- history storage will be initialized once, whenever:
- first non-customTab page is loaded (access to the HistoryDelegate)
- first interaction with the awesomebar
- history UI is accessed
- bookmarks storage will be initialized once, whenever:
- something is bookmarked, or we need to figure out if something's bookmarked
- bookmarks UI is accessed
- logins storage will be initialized once, whenever:
- first page is loaded with a login/password fields that can be autofilled
- (or some other interaction by GV with the autofill/loginStorage delegates)
- logins UI is accessed
- all of these storages will be initialized if the user logs into FxA and starts syncing data
- except, if a storage is not chosen to be synced, it will not be initialized
If user is an FxA user:
- on startup, none of the storage layers are initialized
- sometime shortly after startup is complete, when a sync worker runs in the background, all storage
layers that are enabled to sync will be initialized.
This change also means that we delay loading the megazord until first access (as described above).
The a-c side of this work is in https://github.com/mozilla-mobile/android-components/pull/6128
This switches Fenix to use `SyncableLoginsStorage`, which caches a connection internally
on first access, and doesn't expose any lock/unlock APIs at the public boundary.
Workaround for issue described in:
https://github.com/mozilla-mobile/android-components/issues/5989
For debug builds it is unnecessary to use the actual location provider since those builds
do not have an API key configured. With that patch we replace the location provider with
a dummy implementation in debug builds.
This new event will be sent when the user has successfully migrated from Fennec
to Fenix.
This event will only be sent to Leanplum and not to the other telemetry
services like Glean or Adjust.
Co-authored-by: ValentinTimisica <valentin.timisica@softvision.ro>
- The "Add to Firefox Home" browser menu item adds a top site to the top site storage.
- Refactors the FenixSnackbar from BaseBrowserFragment into BrowserToolbarController
since there are multiple menu items that need to show a FenixSnackbar.
- Adds metrics for the new browser menu item.
This removes Fretboard. The goal is to reduce cold startup costs associated with loading the experiments on the main thread. We currently have two experiments frameworks in use and should only require one.
Cache the list of installed browsers. Calling `Browsers.all`
the application directly redundantly recalculates the list.
Accessing the list of installed browsers through this cache
will reduce that overhead.
* For #4977 - Support opening Fennec pinned website shortcuts in Fenix
Fennec's pinned website shortcuts are set to open the BrowserApp activity.
So we need a new activity alias to actually catch such Intents. Otherwise they
would open "org.mozilla.firefox/.App" without any way to inform that this is
the result of the user clicking on a pinned shortcut.
For actually checking if the newly received Intent is of a Fennec pinned
shortcut we introduce a new FennecBookmarkShortcutsIntentProcessor which will
prepare the Intent to open the shortcut's URL in a new tab.
* For #4977 - Don't keep IntentReceiverActivity on the back stack
For successive Fennec pinned shortcuts to create a new IntentReceiverActivity
and be processed as normal we need to not keep this as our task root.
* For #4977 - Test the FennecBookmarkShortcutsIntentProcessor
* For #7352: integrate highlightable browser menu changes
* For 7352: invalidate menu when reader mode availability changes
* For 7352: removed highlight from reader mode appearance per UX
* For #5334: added private custom tab processor
* For #5334 - Fixes up IntentReceiverActivity for handling intents
* For 5334: update styling for private custom tabbs
* For 5334: update tests to account for new processors
Note that two are still failing. These appear to be true failures, and will be corrected in a later commit.
* For 5334: fixes bug introduced by changes to IntentReceiverActivity
RCA: intent className and extra were previously set based on which processors matched, not which successfully processed. This patch reintroduces that behavior.
* For 5334: add tests for custom tabs processing
* For #5694 & #6054: Adds preference screen for toolbar
* For #5694: Adds changing toolbar position functionality
* No issue: Updates telemetry links to actually work lol 😬
* For #6054: Adds toolbar position to core ping
* For #5577 - Adds button to add a new search engine
* For #5577 - Adds custom engine store
* For #5577 - Creates a custom SearchEngineProvider
* For #5577 - Gives the ability to delete search engines
* For #5577 - Adds the UI to add a custom search engine
* For #5577 - Adds form to create a custom search engine
* For #5577 - Adds the ability to add a custom search engine
* For #5577 - Adds the ability to delete custom search engines
* For #5577 - Selects the first element on the add custom search engine screen
* For #5577 - Prevents adding a search engine that already exists
* For #5577 - Styles the add search engine preference
* For #5577 - Makes the name check case-insensitive
* For #5577 - Fix bug where home screen doesnt see new search engines
* For #5577 - Moves Search URL validation to its own type
* For #5577 - Fixes linting errors
* For #5577 - Adds the ability to edit a custom search engine
* For #5577 - Allows the user to edit a serach engine even when it is the last item in the list
* For #5577 - Adds an undo snackbar when deleting a search engine
* For #5577 - Moves all of the strings to be translated
* For #5577 - Fixes bug when deleting your default search engine
* For #5577 - Puts adding search engines behind a feature flag
* For #5577 - Navigate to custom search engine SUMO article when tapping learn more
* For #5577 - Fixes nits
* For #5577 - Uses concept-fetch to validate search string
* For #5577 - Adds string resources for the cannot reach error state
* For #4844: add test cases for url elision
* For 4844: implement toShortUrl to pass test cases
* For 4844: update plumbing to use toShortUrl
* For 4844: adds/handles suggested url elision test case
* For #4281: small ToolbarMenu refactor
This makes it easier to see how items are ordered in the menuItems list
* For 4281: add QAB buttons to menu
* For 4281: removed menu back button per mocks
I double checked with UX, and we'll be relying on the hardware back button for its functionality
* For 4281: add content descriptions for bookmarking
* For 4281: updated BrowserToolbarController for new functionality
* For 4281: provided simple dependencies to browser controller
More complex changes will be in a following commit, for review readability
* For 4281: move toolbar controller dependencies up to BaseBrowserFragment
The functionality they control is being moved into the toolbar menu, which is shared by both normal tabs and custom ones
* For 4281: removed (now unused) code related to QAB
* For 4281: fix test compilation after QAB removal
Tests still need to be expanded to include added functionality
* For 4281: updated menu to show if url is bookmarked
This sloppy workaround is required because TwoStateButton requires that `isInPrimaryState` be a synchronous call, and checking whether or not the current site is bookmarked is quite slow (10-50 MS, in my tests). After days of work and many attempted solutions, this was the least abhorrent among them.
https://github.com/mozilla-mobile/android-components/issues/4915 was opened against AC to evaluate potentially supporting async `isInPrimaryState` functions.
https://github.com/mozilla-mobile/fenix/issues/6370 was opened against Fenix to investigate the unexpectedly slow call to `BookmarkStorage`.
* For 4281: update reader mode switch
* For 4281: selectively show/hide menu items
* For 4281: add reader mode appearance
* For 4281: update bookmark button when it is clicked
* For 4281: removed unused QAB code
* For 4281: removed QAB robot, updated UI tests
* For 4281: removed QuickActionSheet metrics
Since this behavior now lives in the toolbar, it is tracked via Event.BrowserMenuItemTapped
* For 4281: fixed lint errors
* For 4281: add new strings for buttons added to menu
This is necessary because the location change (from QAB to toolbar menu) could affect the grammar in some languages
* For 4281: remove outdated TODOs
* For 4281: removed QAB container
* For 4281: removed back button reference from UI test
This button no longer exists
* For 4821: Fixes a visual defect (extra padding on top of toolbar)
* For 4281: update copy on reader mode
* For 4281: fixed review nits
This reverts commit ed239ec250.
We're choosing to revert LeanPlum push messaging integration since we believe
that this has introduced a crash in Firebase push messaging. To verify
this assumption we're reverting this change, and when confirmed, we can
re-introduce Leamplum in a safer manner.
See #6250 and #6251 for related info.
* For #4596: move code from CollectionCreationComponent to CollectionCreationStore
Other than adding comments, no changes were made. The code will be updated in a following commit. This is in order to make the commit diff more readable.
* For 4596: update CollectionCreateStore to libstate
* For 4596: copied CollectionCreationUIView into CollectionCreationView
Otherwise, no code was changed. The next commit will update this code. This is in order to make the commit diff more readable.
* For 4596: update CollectionCreationView to LibState
Note that the minimal changes possible to enable migration were made. Refactoring will happen in a later commit.
* For 4596: updated CollectionCreationTabListAdapter to work with the new View
* For 4596: updated SaveCollectionListAdapter to work with the new View
* For 4596: implemented CollectionCreationController
For now, it has an identical interface to the interactor. In a later commit several of its responsibilities will be moved around, some to the interactor and some to the reducer
* For 4596: copied over previous reducer code
No other changes were made. The code will be updated in the following commit. This is done to make changes more readable for the reviewer
* For 4596: update reducer code param names
Otherwise, no changes at this time
* For 4596: add arguments to CreateCollectionFragment in nav_graph
These will be used to replace the current CreateCollectionViewModel, which shares data between fragments in a way that doesn't fit within our architecture.
* For 4596: pass arguments to collection via transaction instead of VM
The VM will be removed in a later commit
* For 4596: update BrowserToolbarController to share state to collection via its Direction
* For 4596: removed CreateCollectionViewModel
* For 4596: test tab retrieval in CreateCollectionFragment
* For 4596: fix crashing CreateCollectionFragmentTest
* For 4596: removed classes create collection classes used by old architecture
* For 4596: collection interactor rename + kdoc
* For 4596: moved collection interactor interface
* For 4596: renamed CreateCollectionFragment
All related classes followed the pattern of CollectionCreationX
* For 4596: kdoc CollectionCreationController
There's no effective difference between these calls and their interactor equivalent, so I linked to them
* For 4596: fix bug that caused rename to not work
* For 4596: removed unused collection actions
These were unused before the LibState refactor
* For 4596: kdoc StepChanged
* For 4596: removed todos about moving logic to the reducer
saveTabsToCollection: this could be moved, but that would involve creating a new action. SaveCollectionStep should probably be refactored out, so adding this layer of indirection seemed counterproductive
handleBackPress: needs to be able to call dismiss(). The reducer doesn't (and shouldn't) be able to do that, so this needs to live here
stepBack: called by handleBackPress. See above
* For 4596: wrote tests for CollectionCreationController#stepback
* For 4596: fixed tests broken by changes to collections
* For 4596: small readability refactor for CollectionController#stepBack
No change to functionality (see tests)
* For 4596: broke apart CollectionView#update
There's probably a lot more that could be done here, but smaller changes were made to reduce scope
* For 4596: remove unnecessary todos
It looks like we don't follow the suggested pattern in this project
* For 4596: test CollectionCreationController#normalSessionSize
* For 4596: updated naming in CollectionCreationController per review
5775: Closes#5233, don't display private-browsing-shortcut creation items if shortcut already exists r=NotWoods a=bswe
Closes issue #5233:
Added doesPrivateBrowsingPinnedShortcutExist() method to the PrivateShortcutCreateManager class to check for the existence of the shortcut. Code only works on devices with API 25 or higher, so a check for that condition was also included in the new method.
Calls to the new method were added to the SettingsFragment and HomeFragment classes to determine whether the settings menu item and the recommendation dialog should be displayed.
**Quality**: The pre-push test failed to start on my windows 10 OS, but this code has been thoroughly tested manually on simulators and physical devices.
**Tests**: This code only effects UI appearance so not sure how to test for that and no tests were written.
**Screenshots**: This code does not add anything new to the UI, it only controls whether to display existing items.
Co-authored-by: bswe <bbollenbacher@mindspring.com>
5908: Closes#5876: Removed @ObsoleteCoroutinesApi in tests r=sblatz a=shldhll
### Pull Request checklist
<!-- Before submitting the PR, please address each item -->
- [x] **Quality**: This PR builds and passes detekt/ktlint checks (A pre-push hook is recommended)
- [ ] **Tests**: This PR includes thorough tests or an explanation of why it does not
- [ ] **Screenshots**: This PR includes screenshots or GIFs of the changes made or an explanation of why it does not
- [ ] **Accessibility**: The code in this PR follows [accessibility best practices](https://github.com/mozilla-mobile/shared-docs/blob/master/android/accessibility_guide.md) or does not include any user facing features
### After merge
- [ ] **Milestone**: Make sure issues finished by this pull request are added to the [milestone](https://github.com/mozilla-mobile/fenix/milestones) of the version currently in development.
### To download an APK when reviewing a PR:
1. click on Show All Checks,
2. click Details next to "Taskcluster (pull_request)" after it appears and then finishes with a green checkmark,
3. click on the "Fenix - assemble" task, then click "Run Artifacts".
4. the APK links should be on the left side of the screen, named for each CPU architecture
Co-authored-by: shldhll <iam.shldhll@gmail.com>
5633: Add fxa/sync integration tests r=csadilek a=isabelrios
Pull Request checklist
<!-- Before submitting the PR, please address each item -->
- [x] **Quality**: This PR builds and passes detekt/ktlint checks (A pre-push hook is recommended)
- [x] **Tests**: This PR includes thorough tests or an explanation of why it does not
- [-] **Screenshots**: This PR includes screenshots or GIFs of the changes made or an explanation of why it does not
- [-] **Accessibility**: The code in this PR follows [accessibility best practices](https://github.com/mozilla-mobile/shared-docs/blob/master/android/accessibility_guide.md) or does not include any user facing features
This PR tries to add new tests, sync integration tests, to check the sync process Desktop<->Fenix, first for Bookmarks and in the future for more.
Co-authored-by: Isabel Rios <isabelrios@mackbookirios.home>
Co-authored-by: isabelrios <isabelrios@gmail.com>
This patch includes:
- WebChannels support enabled by default, with ability to disable it via remote flag
- expanded FxA telemetry (closes#4971)
Co-authored-by: Arturo Mejia <arturomejiamarmol@gmail.com>
* Upgrade to Mozilla Android components 15.0.0-SNAPSHOT
* Update findinpage integration to use browser-state
* Use new UseCases APIs after download migration.
* For #5092 - Show a Snackbar with retry option if sharing to devices fails
ShareController will contain all the business logic for checking the result
status of the `share to device` operations. When this fails it will show a
snackbar which also offer the possibility to retry the just failed operation.
To allow this even in the event the user has closed the share fragment we'll
use a GlobalScope's coroutine.
Refactored out the TabsSharedCallback from ShareFragment because otherwise we
would have neede to sent through that the just failed operation. After the
refactor the ShareController is solely responsable for showing the right
snackbar and handling the retry actions.
* For #5092 - Refactor ShareControllerTest
* For #5092: Adds color theming of snackbars
This introduces test coverage, using the Glean SDK
testing API, for the metrics that are set at startup
by Fenix in the GleanMetricsService.
This additional adds a basic test for the translation
of the `app_opened` event.
This patch includes:
- WebChannels support enabled by default, with ability to disable it via remote flag
- expanded FxA telemetry (closes#4971)
Co-authored-by: Arturo Mejia <arturomejiamarmol@gmail.com>
There is a "unused" lint suppression in place with the comment
"Referenced from XML". I found no such usage.
It's documentation says that this Behavior will always position FindInPageBar
above BrowserToolbar but the current code ensures
BrowserToolbar.visibility == GONE when FindInPageBar.visibility == VISIBLE
so there's no need for such behavior.
FindInPageFeature is used inside the app as a LifecycleAwareFeature and as such
it receives the onStart / onStop lifecycle calls.
The onStart() lifecycle call would not get passed to the feature but in
onStop() FindInPageFeature will detach it's Presenter and Interactor so when
the user comes back to the screen she could not interact anymore with the
feature.
To fix this we'll propagate LifecycleAwareFeature's onStart() to the inner
feature which is to rebind it's Presenter and Interactor in onStart().
This behavior is common to all the other features so all of them who implement
LifecycleAwareFeature will now get the onStart() call also.
Grisha Kruglov