* Add an end-to-end test for stickers
* More logs on login
* Wait for spinners to go away
* Factor out spinner waiting as it seems useful
* Move stickers to the end
* More waiting
* When all else fails... add sleeps
* Waiting for the server picker to appear seems to work..?
* Typos
Co-authored-by: J. Ryan Stinnett <jryans@gmail.com>
* remove commented code from registration usecase
Co-authored-by: J. Ryan Stinnett <jryans@gmail.com>
* Add a new flag pseudonymousAnalyticsOptIn replacing analyticsOptIn, stored at account level, so people only need to opt in once.
* Show a toast in login to users that have analyticsOptIn set but not yet pseudonymousAnalyticsOptIn prompting them confirm the new method is okay. Update the copy of the existing opt-in toast. Don't notify users that previously opted out.
* Update the copy in settings
* Add a new learn more dialog
* Support a new config flag analyticsOwner which is used in these toasts when explaining which entity the data is sent to ("Help improve %(analyticsOwner)"). If unset, display brand. This allows deployments whose brand differs from the receiver of the analytics to explain the situation to their users (e.g. AcmeCorp badges their app, but explains the data is sent to Element, not them)
* The new opt-in and flags are only used when posthog is configured; prior to that there are no changes to UX or tracking behaviour.
* Move user avatar to Space panel
* Add room list header for 'Home' or 'Space Name' to room list
Add existing Space context menus to room list header
* Re-add pending room join spinner
* Iterate RoomListHeader plus context menu
* Iterate space context menu
* Iterate room list + interactions
* Move DND to new iA model
* Replace composer custom status management with usermenu one
* Cull Quick Actions
* Iterate minimized room list state
* delint
* Merge the RoomListNumResults into the RoomListHeader
* Make the search shortcut prompt semi-bold
* Iterate RoomListHeader based on design review
* Iterate UserMenu based on feedback
* Add name to expanded spacepanel usermenu button
* i18n
* Make room sub list aux button components more generic
* Change left panel explore button to only refer to room directory
* Iterate RoomListHeader
* Fix custom user status input field width in Chrome
* Bring back Notification settings button
* delint
* i18n
* post-merge fix
* iterate pr
* Remove unused state
* update copy
* Apply suggestions from PR review
* delint
* Update invite iconography
* Iterate Space context menu to match Figma
* Fix chevron alignment
* Fix edge case for RoomListHeader on metaspaces
* Wire up general rageshake-driven feedback mechanism
* Add IA1.1 info toast
* add missing alt attribute
* delint
* delint
* tweak ia toast priority
* e2e test account for new toast
* autofocus feedback field and remove old subheading
* tweak copy
* Iterate space panel colours to match Figma
* Iterate PR
* delint
* Fix feedback submission with object setting values
* iterate based on review
* Tweak colours and update splash image
* Tweaks based on review
* Remove room list prompt, made redundant by the big fat `+`
* Fix edge cases around User Menu positioning and dnd
* Add missing import, bad merge?
* Update aria label in e2e test
* Fix room list space rooms context menu explore button behaviour
* Tweak copy
* Revert order of options in the UserMenu
* Tweak copy
* i18n
Fixes https://github.com/vector-im/riot-web/issues/13226
This isn't the fastest route, but it is a predictable route for the happy path we probably want to test. For example, Alice will already be staring at the DM and could easily accept the verification there, but we probably want to make sure that the toast is present and does the right thing. Similarly, neither of them need to verify that there's green shields everywhere, they should be implied, however an explicit check follows a real user's gaze.
Because the tests are run directly by node, we have to use the CommonJS module syntax. We could run the thing through babel, but then we just have another babel.
Windows instructions are from experience and may not be optimized.
We use `export default` begrudgingly here. Ideally we'd use just `export`, though this entire SDK expects things to be exported as a default. Instead of breaking everything, we'll sacrifice our export pattern for a smaller diff - a later commit can always do the default export -> regular export conversion.
