This commit adds an endpoint `/mobile_push/e2ee/test_notification`
to send an end-to-end encrypted test push notification to the user's
selected mobile device or all of their mobile devices.
If the client has passed `simplified_presence_events` as true
in the `client_capabilities` parameter of the `POST /register`
request, then the server will send `presence` events with the
`presences` field, which has the user presence data in the
modern API format. When that client capability is false, the
`presence` event will be unchanged and sent with the user
presence data in the legacy format.
This commit includes the following changes:
- Add an administrator setting to customize the Welcome Bot
message when sending an invitation.
- Add an API endpoint to test the customized Welcome Bot message
by sending a copy of the message to the administrator.
Fixes#27663.
Co-authored-by: Akarsh Jain <akarsh.jain.790@gmail.com>
We do not test relative links here since relative links were only
used in the old help pages. We should probably remove the logic
for relative links after we have done the official cutover to the
starlight help center: #35654.
Help center is going to cutover to using starlight. We won't have that
help center running during our backend tests for now, that will be done
after cutover. Remove help related opengraph tests. We have replaced
instances from api_docs wherever possible, but for most of the cases, an
equivalent was not found in api_docs.
There is no api_docs file that starts with `{settings_tab` or an
admonition or `{start_tabs}` and thus we need to remove those tests.
test_settings_tab was not testing what it said it was testing,
deactivate-your-account.md did not start with `{settings_tab` anymore.
That was changed in 584124bb45.
We are going to move the help center to starlight soon and the tests
won't be able to access that help center until we solve things in a
followup issue. We move some of the urls being tested from /help to /api
to not loose coverage on those tests.
The test cases are copied from ReorderCustomProfileFieldTest since we
are imitating the reordering mechanism from custom profile fields to
channel folders.
UTF-8 is far and wide the dominant encoding, so we assume that
encoding if we have more than 66% confidence -- this allows it to be
guessed even for short files. ISO-8859-1 caps at 73% confidence, so
we assume that if it's the best option and it hits that max.
Otherwise, we still require 90% confidence.
This commit updates the data that gets encrypted to be
the same on both android and iOS.
The data and its format is almost the same as what we send
as FCM payload to android clients with no E2EE support,
changes are:
For send push notification payload:
* 'realm_id`, 'server', 'sender_email', and 'realm_uri' fields
don't exist in the new payload.
* 'event' field renamed to 'type'
* 'stream' and 'stream_id' fields renamed to 'channel_name'
and 'channel_id' respectively.
* The value of 'recipient_type' will be 'channel' & 'direct'
instead of 'stream' & 'private' respectively.
* 'zulip_message_id' field renamed to 'message_id'
For remove push notification payload:
* 'realm_id`, 'server', and 'realm_uri' fields don't exist
in the new payload.
* 'event' field renamed to 'type'
* 'zulip_message_ids' field renamed to 'message_ids' and it's
value will be a JSON array instead of a string.
In the existing iOS client, we have no code of our own involved
in constructing the notifications in the UI, and instead we
leave it to the iOS SDK to do so.
Since, for clients with E2EE support the data is going to be
interpreted by our own code, not by the iOS SDK - we are free
to keep the same data and format.
Co-authored-by: Tim Abbott <tabbott@zulip.com>
This commit removes the 'zulip_message_id' field from the FCM
payload for remove push notification.
Clients that use `zulip_message_ids` instead have been out for
years and we no longer support older client versions that don't.
Fixes#20028.
There's no reason to have a special `RealmCreationKey` class - the
`Confirmation` system already does this job.
This is somewhat complicated by the need to write a migration for
`RealmCreationKey`->`Confirmation` for pre-existing, valid objects, to
avoid breaking realm creation links that haven't been used yet.
Clears up some naming confusion - `REALM_CREATION_LINK_VALIDITY_DAYS` is
a setting which describes the validaty period of RealmCreationKey, not
of Confirmation with REALM_CREATION type. Now that the latter is called
NEW_REALM_USER_REGISTRATION, there should be no confusion about this.
This commit updates code to require unique names for only
non-archived folders in a realm, which means a folder can
use a name which is being used by an archived folder.
The order of token values returned by the database query isn't
guaranteed, so 'assertEqual' fails in a flaky way.
This commit uses 'assertCountEqual' to fix that.
This reverts 1862c3d9ab.
The "time" field was added in the APNs payload to log
"Remote queuing latency".
The client doesn't use this field. Ideally, we should have
removed the field from the payload before sending it to APNs.
We no longer log "Remote queuing latency", so it's safe to
remove it from the APNs payload.
Only FCM payload will have this field as it is used by android
clients.
The function was used to reformat apns payload sent
by servers older than 613d093d7.
Since we no longer need to worry about servers older
than 1.6.0, it's safe to remove that function.
This commit adds "max_channel_folder_name" and
"max_channel_folder_description" fields to response.
Also updated webapp to use these fields instead of hardcoding
the values.
This commit replaces the `PUSH_NOTIFICATION_REDACT_CONTENT` server
setting with `require_e2ee_push_notifications` realm setting.
If `require_e2ee_push_notifications` set to True:
* Older clients: Content redacted
* Updated clients: Encrypted content
If `require_e2ee_push_notifications` set to False:
* Older clients: Content NOT redacted
* Updated clients: Encrypted content
Note: Older clients refers to clients that don't support E2EE.
Fixes part of #35370.
This commit adds a realm setting:
"Require end-to-end encryption for push notification content".
This commit just involves backend & UI changes to add the setting,
the impact of this setting is in the next commit.
Fixes part of #35370.
Previously, when there were two topics -- A and B, and a user
follows topic A, then when messages of topic B are merged into
topic A then topic A loses its follow status.
This is fixed by updating the algorithm for getting the visibility
policy after merge to be followed in the above case.
Fixes#35226
This commit adds a test and updates a few existing tests to
cover more cases related to send push notifications.
* We no longer mock the 'send_push_notifications_legacy' function
while testing 'send_push_notifications' codepath and vice-versa.
This makes the tests more realistic as both functions gets called
in 'handle_push_notification'.
This covers the case when only old clients (which don't support
E2EE) exists for a user. Or only updated clients (which supports
E2EE) exist.
* Adds a test 'test_both_old_and_new_client_coexists' for the case
when a user has both type of clients at an instant i.e. they have
updated a few devices only.
In 'send_push_notifications_legacy', when a user has no registered
devices:
* `uses_notification_bouncer()`=True: we log "Skipping..." and return.
* `uses_notification_bouncer()`=False: we make some function calls,
which effectively does nothing.
It's better to have a common check (and log) early in the codepath.
This commit makes that change.
Earlier, we were passing a map `device_id_to_encrypted_data`
and http headers as separate fields to bouncer.
The downside of that approach is it restricts the bouncer to
process only one type of notice i.e. either notification for
a new message or removal of sent notification, because it
used to receive a fixed priority and push_type for all the
entries in the map.
Also, using map restricts the bouncer to receive only one
request per device_id. Server can't send multiple notices
to a device in a single call to bouncer.
Currently, the server isn't modelled in a way to make a
single call to the bouncer with:
* Both send-notification & remove-notification request data.
* Multiple send-notification request data to the same device.
This commit replaces the old protocol of sending data with
a list of objects where each object has the required data
for bouncer to send it to FCM or APNs.
This makes things a lot flexible and opens possibility for
server to batch requests in a different way if we'd like to.
