This commit adds a server level setting which controls whether the setting
can be set to anonymous user groups. We only allow it in the tests for
now because the UI can only handle named user groups.
This commit adds a new helper function to create or update
a UserGroup object for a setting. We could have used existing
update_or_create_user_group_for_setting but that also validates
user IDs and subgroup IDs which we can skip in tests.
This commit updates code, majorly in tests, to use
setting values from enums instead of directly using
the constants defined in Realm.
We still have those constants defined Realm as they
are used in a couple of places where the same code
is used for different settings. These will be
handled later.
This commit adds support to pass object containing both old and new
values of the can_mention_group setting, as well as detailed API
documentation for this part of the API system.
Co-authored-by: Tim Abbott <tabbott@zulip.com>
Co-authored-by: Greg PRice <greg@zulip.com>
This commit removes name, description, is_system_group and
can_mention_group fields from UserGroup model and rename
them in NamedUserGroup model.
Fixes#29554.
This commit adds get_recursive_strict_subgroups function
which returns all the subgroups but not includes the user
group passed to the function.
We also update the test to check subgroups of named user
groups using the get_recursive_strict_subgroups function.
This is fine as we already test the get_recursive_subgroups
function.
There order of group ids doesn't matter here and thus the
compared values can have the ids in different order and test
should still pass. So, using `set` for comparing unordered
lists seems like the right fix here.
This commit moves constants for system group names to a new
"SystemGroups" class so that we can use these group names
in multiple classes in models.py without worrying about the
order of defining them.
Python evaluates function parameter defaults at definition time, not
call time. This function wouldn’t work with other realms anyway.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
I add a bunch of cute helper methods to make
the test a bit more readable.
And then I make sure to get clean objects,
which precludes the need for our callback
functions to refresh the user objects.
And finally I make sure that our validation
functions don't cause any round trips (assuming
we have fetched objects using a standard
Zulip helper, which example_user ensures.)
**Background**
User groups are expected to comply with the DAG constraint for the
many-to-many inter-group membership. The check for this constraint has
to be performed recursively so that we can find all direct and indirect
subgroups of the user group to be added.
This kind of check is vulnerable to phantom reads which is possible at
the default read committed isolation level because we cannot guarantee
that the check is still valid when we are adding the subgroups to the
user group.
**Solution**
To avoid having another transaction concurrently update one of the
to-be-subgroup after the recursive check is done, and before the subgroup
is added, we use SELECT FOR UPDATE to lock the user group rows.
The lock needs to be acquired before a group membership change is about
to occur before any check has been conducted.
Suppose that we are adding subgroup B to supergroup A, the locking protocol
is specified as follows:
1. Acquire a lock for B and all its direct and indirect subgroups.
2. Acquire a lock for A.
For the removal of user groups, we acquire a lock for the user group to
be removed with all its direct and indirect subgroups. This is the special
case A=B, which is still complaint with the protocol.
**Error handling**
We currently rely on Postgres' deadlock detection to abort transactions
and show an error for the users. In the future, we might need some
recovery mechanism or at least better error handling.
**Notes**
An important note is that we need to reuse the recursive CTE query that
finds the direct and indirect subgroups when applying the lock on the
rows. And the lock needs to be acquired the same way for the addition and
removal of direct subgroups.
User membership change (as opposed to user group membership) is not
affected. Read-only queries aren't either. The locks only protect
critical regions where the user group dependency graph might violate
the DAG constraint, where users are not participating.
**Testing**
We implement a transaction test case targeting some typical scenarios
when an internal server error is expected to happen (this means that the
user group view makes the correct decision to abort the transaction when
something goes wrong with locks).
To achieve this, we add a development view intended only for unit tests.
It has a global BARRIER that can be shared across threads, so that we
can synchronize them to consistently reproduce certain potential race
conditions prevented by the database locks.
The transaction test case lanuches pairs of threads initiating possibly
conflicting requests at the same time. The tests are set up such that exactly N
of them are expected to succeed with a certain error message (while we don't
know each one).
**Security notes**
get_recursive_subgroups_for_groups will no longer fetch user groups from
other realms. As a result, trying to add/remove a subgroup from another
realm results in a UserGroup not found error response.
We also implement subgroup-specific checks in has_user_group_access to
keep permission managing in a single place. Do note that the API
currently don't have a way to violate that check because we are only
checking the realm ID now.
The most expensive thing for adding user groups is sending
all the notification messages, but we at least want to make
sure that the basic stuff runs in constant time.
Earlier the API endpoints related to user_group accepts and returns a
field `can_mention_group_id` which represents the ID
of user_group whose members can mention the group.
This commit renames this field to `can_mention_group`.
This commit removes "@" from name of role-based system groups
since we have added a restricion on having user group names
starting with "@" in the previous commit as they look odd in
mention syntax.
We also add a migration in this commit to update the name of
role-based system groups in existing realms to remove "@"
from the name. This migration also updates the names of
non-system user groups by removing the invalid prefixes
from their names and if there is a group already with that
name, we insted name the group as "group:{group_id}".
Fixes#26148.
We do not allow user group names to start with "@", "role:",
"user:", "stream:" and "channel:".
Group names starting with "@" look odd in mentions and
"role:", "user:" and "stream:" prefixes are reserved for
system groups which will be used in the new groups-based
permission model. We do not allow "channel:" prefix for
now just to be safe in a case where we use it instead of
"stream:" prefix for stream based groups in future.
Fixes part of #26148.
Previously we had database level restriction on length of
user group names. Now we add the same restriction to API
level as well, so we can return a better error response.
This commit changes the code in test_user_groups.py to use
check_add_user_group function to create user groups instead
of directly using django ORM to make sure that settings
would be set to the correct defaults in further commits.
This verifies that updates of the user group name/description are
correctly done by doing additional queries. This also empathsizes on
checking that the state before and after API calls are indeed different.
We extract the checks needed for user membership changes into a method,
verifying that the members of the user group are matching the expected
values exactly.
The user group depedency graph should always be a DAG.
This commit adds code to make sure we keep the graph DAG
while adding subgroups to a user group.
Fixes#25913.
After this commit a notification message is sent to users if they are
added to user_groups by someone else or they are removed from user_groups
by someone else.
Fixes#23642.
This commit adds code to create a "Nobody" system user group
to realms which will be used in settings to represent "Nobody"
option.
We also add a migration to add this group to existing realms.
"check_add_user_group" is a safer helper function than
"create_user_group" to use when creating user_groups. It does
error handling and notify the client with the appropriate event.
Note that the populate_db command still uses "create_user_group"
because we do not need to enqueue events at that point.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Since this function creates a new user group into the database,
it is more appropriate to have it not as a generic "lib" function
but as an "action".
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
View that handled `PATCH user_groups/<int:user_group_id>` required
both name and description parameters to be passed. Due to this
clients had to pass values for both these parameters even if
one of them was changed.
To resolve this name description parameters to
`PATCH user_groups/<int:user_group_id>` are made optional.
Previously, test cases or clients accessing /json/ views using HTTP
Basic Auth would be accepted, while we intended to only allow clients
authenticated with a session cookie to access these views.
This adds a check on the accessed path to avoid this possibility.
It seems unlikely that any API clients clients were taking advantage
of this unintended quirk; so we're not going to bother documenting
this bug fix as an API change. In any case, it should be trivial for
anyone affected to consult the documentation and then switch their
/json/foo URL to a correct /api/v1/foo URL.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Although our POST /messages handler accepts the ‘to’ parameter with or
without JSON encoding, there are two problems with passing it as an
unencoded string.
Firstly, you’d fail to send a message to a stream named ‘true’ or
‘false’ or ‘null’ or ‘2022’, as the JSON interpretation is prioritized
over the plain string interpretation.
Secondly, and more importantly for our tests, it violates our OpenAPI
schema, which requires the parameter to be JSON-encoded. This is
because OpenAPI has no concept of a parameter that’s “optionally
JSON-encoded”, nor should it: such a parameter cannot be unambiguously
decoded for the reason above.
Our version of openapi-core doesn’t currently detect this schema
violation, but after the next upgrade it will.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
We now use EVERYONE_GROUP_NAME instead of writing
the actual group name at multiple places, so that we
can have all the group names coded at one place only.
Vector73
Sahil Batra
Anders Kaseorg
Alex Vandiver
Aman Agrawal
Steve Howell
Zixuan James Li
Ujjawal Modi
m-e-l-u-h-a-n