This commit adds a new ecosystem client ID to `ClientID.jsm`. It's
handled the same way as a regular telemetry client ID, in that it's a
random UUID that's persisted to disk, and reset together with the
client ID. Unlike the main client ID, we don't cache it in prefs.
Additionally:
* `setClientID` has been renamed to `setCanaryClientIDs`, and
automatically sets the main and ecosystem client IDs to the canary
(known) ID. It's no longer possible to set either client ID to a
specific value, but only test code did this.
* `resetClientID` has been renamed to `removeClientIDs`, and clears
both client IDs. It no longer returns the new ID; instead, consumers
should call `getClientID` or `getEcosystemClientID` after.
* The ecosystem client ID is included in the `deletion-ping`.
Differential Revision: https://phabricator.services.mozilla.com/D84233
This commit adds a new ecosystem client ID to `ClientID.jsm`. It's
handled the same way as a regular telemetry client ID, in that it's a
random UUID that's persisted to disk, and reset together with the
client ID. Unlike the main client ID, we don't cache it in prefs.
Additionally:
* `setClientID` has been renamed to `setCanaryClientIDs`, and
automatically sets the main and ecosystem client IDs to the canary
(known) ID. It's no longer possible to set either client ID to a
specific value, but only test code did this.
* `resetClientID` has been renamed to `removeClientIDs`, and clears
both client IDs. It no longer returns the new ID; instead, consumers
should call `getClientID` or `getEcosystemClientID` after.
* The ecosystem client ID is included in the `deletion-ping`.
Differential Revision: https://phabricator.services.mozilla.com/D84233
This is done because the test always suffers from shutdown leaks after the changes in Bug 1619953, which make the stream acquired during the test asynchronous.
Differential Revision: https://phabricator.services.mozilla.com/D81861
Also cleans up some test boilerplate:
1. avoid waitForFocus passing in urlbar tests
2. avoid useless window.gSomething
3. avoid useless UrlbarTestUtils imports
Differential Revision: https://phabricator.services.mozilla.com/D80082
It turns out that we sometimes do not get a `OnLocationChange` event from the
BrowserChild when navigating to an `about:blank` document from an initial
`about:blank`. We then would not have the `innerWindowID` set on the
`<xul:browser>`. Now we retrieve the `innerWindowId` from the `<browser>`'s
current window global.
Additionally, we were running into some timeouts due to intermittently not
receiving a `OnStateChange` event for `STATE_STOP`. Because we waiting for this
event, the test would time out sometimes.
Differential Revision: https://phabricator.services.mozilla.com/D79199
Creates a new pref, "...activity-stream.feeds.system.topsites" to control whether TopSitesFeed is enabled. The existing "...activity-stream-feeds.topsites" pref is kept to allow users to downgrade and to ensure that people who disabled Top Sites in the past will continue to not see Top Sites on the New Tab Page -- the difference being that Top Sites will now be generated for those users.
Differential Revision: https://phabricator.services.mozilla.com/D75603
The private search result is shown in Nightly but not otherwise. The test is
failing because it expects the private search result to be at index 1 and the
form history result to be at index 2, but the form history is actually at index 1.
The test just needs to get the right index. There's another part of the test
that clicks a remote suggestion, but it's not affected because the search
happens to match a couple of history results, so the private search result isn't
shown even on Nightly.
Differential Revision: https://phabricator.services.mozilla.com/D76585
Top-level content WindowGlobalParents now keep track of the site origins of the
documents in their document tree. When the WindowGlobalParent is torn down, the
maximum of the number of unique site origins is submitted for telemetry.
Differential Revision: https://phabricator.services.mozilla.com/D71493
