This prevented migration 0345
(517c2ed39d / #19696) from applying on
systems that were created after the refactoring that resulted in the
system bot realm potentially having null as its name.
(We've already confirmed that normal realms, created via
`do_create_realm`, shouldn't be able to have this unusual state).
This check was copied from upstream python-markdown's "safe mode"
before they removed that feature. The upstream history is that they
introduced this check in
2db5d1c8e4,
which was not a complete security check, and then added the
immediately following check (with an allowlist of schemes) in
0b4ffbb60e.
Their first, incomplete check provides no security benefit and makes
the code hard to reason about, so we remove it.
The 'update_global_notifications' type event is sent only for
existing settings and will not be sent for new settings, so we
should use notification_settings_legacy dict to check the type
of setting value in check_update_global_notifications instead
of notification_settings_types dict.
We still used notification_setting_types in copy_user_settings
function of create_user.py and in a test in test_event_system.py.
It is not required to do so since we have added all settings in
property_types already and we loop over property_types at both
these places which includes all settings.
This was likely initiall created with null=True in
5c5ffd6ea3 just because we didn't have a
plan for backfilling this field, but I verified that Zulip Cloud has
no realms without a name set, and that's the place most likely to have
any form of super-legacy nameless realms.
So we can clean up this aspect of the data model without a special
migration to do something with existing realms with name=None (which I
suspect would have resulted in a 500 anyway).
We can use this format to display text in plural form or not based
on a number. This helps translators easily translate text and
users get a better formatted text.
We are renaming stream to stream_name in formatted draft object
just to be more explicit and be clear that we are storing that
stream name in this variable.
This commit changes snapshot_message to store stream_id for
drafts along with stream names. The stream_id field is
undefined if draft is for empty or invalid stream name.
After this change:
- If draft has a valid stream_id stored and it maps to
a stream, then we display the stream name from the
obtained stream object.
- If draft.stream_id is undefined or doesn't map to a stream,
then we display the name stored in draft.stream, which can
be invalid (no stream of this name existed ever), can be
empty and can also be name of a deactivated stream.
This change helps us to show correct stream-name for drafts
in case of renaming a stream.
Fixes#15155.
On zulipchat.com we use SOCIAL_AUTH_SUBDOMAIN='auth', meaning
python-social-auth backend flows go through auth.zulipchat.com rather
than staying on the original subdomain.
We already test all the notification settinsg in
test_toggling_boolean_display_settings (which is
now renamed to test_toggling_boolean_user_settings)
as all settings are now moved to property_types and
we are merging other parts also to consider all the
settings under one category.
This commit adds a separate test for invalid values of user settings
and remove the existing code for it in test_change_user_setting.
This change will enable us to merge the tests for notification
settings to this because email_notification_batching_period_settings
has different invalid values than other integer values and we do the
same for realm settings also.
This commit adds `demo_organization_scheduled_deletion_date` to
the `realm` section of the `/register` response so that it is
available to clients when enabled.
This is a part of #19523.
This fixes a regression where one could end up deactivating all owners
of a realm when trying to synchronize LDAP with the `is_realm_admin`
flag configured in `AUTH_LDAP_USER_FLAGS_BY_GROUP`.
With tweaks by tabbott to add is_moderator as well.
Fixes#18677.
Along with the extraction, we do some simplifications of inserting
text in compose too. This same function can now be used in
compose formatting popover too.
We use "text-field-edit", which has good cross-browser `undo` /
`redo` support, to do the text replace for us instead of writing
that logic ourselves.
There is a bug when multiple message edit forms are opened at the
same time where undefined value of stream_id is sent to the server.
This happens because a global variable stream_widget is used to get
the id of stream selected in dropdown and value of stream_widget
variable keeps on changing when we open multiple message edit forms.
Thus, stream_widget can have the dropdown widget of already closed
edit form resulting in undefined value of stream id.
This commit changes the save_message_row_edit function to access
the dropdown element directly using message id instead of using
stream_widget.value() and thus we always use the correct dropdown
element to get the stream id.
We also move the stream_widget variable to be inside edit_message
function instead of being global variable for the module.
Fixes#19663.
Before this commit, the message or any draft is deleted as soon
as the compose box is closed. So, it removes that by removing
delete_active_drafts and instead this commit will add the deletion
process of drafts in reify_message_id that is called when a
message is successfully sent and received.
Now, see there are two types of messages, one that are locally
echoed and the second ones are that aren't locally echoed but
sent directly to server. This commit only saves the message in
draft if it is locally echoed as they are the only messages
that show message failed in message list. The non locally echoed
ones aren't remove from the compose box until they are
successfully sent. Now as the draft-id is stored in the message
data for locally echoed messages, as they are echoed from the
server, they are deleted using that draft-id.
This also adds node tests for echo reify_message_id for testing
this feature that this commit is adding.
Fixes#17697
Since 84742a0, all settings are sent in the `user_settings` dictionary
which were previously sent inline with other fields in /register
response.
In order to simplify the process of adding new personal settings, we
want to transition to a world where new settings only need to consider
the `property_types` object, and code that needs to reference the
legacy behavior interacts with an object with `legacy` in its name.
This way, contributors working on new settings don't need to think
about the legacy code paths at all.
See https://chat.zulip.org/#narrow/stream/378-api-design/topic/user.20settings.20response.20in.20.2Fregister
to understand this better.
Tim Abbott
rht
Sahil Batra
Aman Agrawal
Anders Kaseorg
sahil839
Wesley Aptekar-Cassels
Mateusz Mandera
Eeshan Garg
Shelly
Palash Raghuwanshi
Signior-X
Dinesh