During a brief period while a custom emoji is being uploaded, it could
be visible to clients even though it was still in the process of being
uploaded, an operation that can fail.
In certain cases, we call `RealmEmoji.save()` before the filename
becomes available. This result in getting invalid urls generated and
flushed. Normally we call it again shortly after, making it harder to
trigger this bug.
Fixes#22552.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
We do not allow changing enable_spectator_access to True using API on
limited plan realms. Frontend changes have been done previously.
This is a follow-up of #22179.
It is apparently possible to have a mention of a user who is not (or
no longer?) in the `users.bson` table.
Skip such mention for the purposes of Zulip import; there's nothing
better for us to do.
This is likely an error somewhere in rocketchat's MongoDB "eventual
consistency," but there is no problem with skipping the chunks at this
step.
In the one case where this was observed so far, the upload-id was not
referenced in any message -- if it is referenced and has chunks, but
has no metadata, we will fail later, at that reference.
This `mimetype` parameter was introduced in c4fa29a and its last
usage removed in 5bab2a3. This parameter was undocumented in the
OpenAPI endpoint documentation for `/user_uploads`, therefore
there shouldn't be client implementations that rely on it's
presence.
Removes the `request.GET` call for the `mimetype` parameter and
replaces it by getting the `content_type` value from the file,
which is an instance of Django's `UploadedFile` class and stores
that file metadata as a property.
If that returns `None` or an empty string, then we try to guess
the `content_type` from the filename, which is the same as the
previous behaviour when `mimetype` was `None` (which we assume
has been true since it's usage was removed; see above).
If unable to guess the `content_type` from the filename, we now
fallback to "application/octet-stream", instead of an empty string
or `None` value.
Also, removes the specific test written for having `mimetype` as
a url parameter in the request, and replaces it with a test that
covers when we try to guess `content_type` from the filename.
Updates `json_change_settings` so that the default value for the `email`,
`full_name`, `new_password` and `old_password` parameters is `None` instead
of an empty string, which also makes the type annotation `Optional[str]`.
Also, updates tests for email and full name changes to include an empty
string as one of the tested invalid values.
Because Django's ContentType objects are, by default, created lazily
when an actual object is created that will use them, this migration
would fail on any server that actually had RealmReactivationStatus
objects already, and had not yet created the ContentType for them.
ContentType objects are very simple:
zulip=> select * from django_content_type where model = 'realmreactivationstatus';
id | app_label | model
----+-----------+-------------------------
85 | zerver | realmreactivationstatus
So we can simply patch this by using get_or_create.
* Remove likely useless use of `controls` Bootstrap class.
* Remove outer control-group elements, which just added 10px bottom margin, and
were always used with .no-margin, which removed that margin :).
* Rename message-edit-timer-control-group to message-edit-timer, since
it isn't a control-group.
* Remove Bootstrap `form-horizontal` class and !important override
required by it.
* Standardize on 10px bottom margins; this fixes subtle bugs in some
sizes.
Previously, we had you select where to send notifications before
specifying which messages you're moving, which seems like the wrong
decision hierarchy.
Fixes#19973.
The stream list left sidebar currently has 3 sections:
* Pinned (+ Muted pinned streams)
* Active (+ Muted active streams)
* Inactive streams
Previously, these sections were separated by horizontal lines, which
did not provide an easy way to discern why there were sections. We add
labels to these section dividers to help with this.
Additionally, within each section, we now sort all muted streams to
the bottom, so that they general minimal clutter.
Fixes#19812.
In compose box, the "x" button tooltip text changed to
"Cancel compose and save draft" except when unsent
message length is short(<3).
Also in help(?) > keyboard shortcuts, text for `Esc`
changed to "Cancel compose and save draft".
The help center page updated with the above changes.
Fixes#21599.
Following c31ab1bcb5, the keyboard
navigation in Recent Topics has been buggy, because the number of
columns with actionable elements now varies with whether the topic
has any unreads.
Fix the keyboard dead reckoning logic to understand that there's a
different number of columns depending on whether the unread count is
present.
This does not fully make the experience nice, but it's enough to tide
us over for now.
Fixes: #21654.
This is preperatory commit for #21654.
We extract the logic for arrow key navigation when focused on table to
separate functions so it is easy to add more functionality in future
and will not clutter the switch block.
Fixes a part of: #21654
In future commits, it will become possible to have a non-null
`compose_state.topic()` while in private message view, because
we'll be keeping that state for switching between the stream
and private message views. See #21853 for further context.
We don't want to warn about a resolved topic unless the topic
is actively visible in the compose box.
Now the following characters are allowed before @-mentions and stream
references (starting with #) for proper rendering - {, [, /.
This commit makes the markdown rendering consistent with autocomplete
(anything that is autocompleted is also rendered properly).
Now the following characters are allowed before @-mentions, stream
references (starting with #) and emojis (starting with :) - space, (,
{, [, ", ', /, <.
Earlier only the opening brace type characters and space was allowed.
Initializing a dictionary from an iterable requires the each item to be
a tuple containg a key and a value. `mypy_django_plugin` cannot infer
the number of items in an queryset with annotated values, so we have to
explicitly unpack each row with a dictionary comprehension here.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
With the new signature of has_request_variables, we can now use
`HttpResponseBase` as the return type of the decorated function.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Decorators like `require_server_admin_api` turns user_profile into a
positional-only parameter, requiring the callers to stop passing it as a
keyword argument.
Functions like `get_chart_data` that gets decorated by both
`require_non_guest_user` and `has_request_variables` now have accurate
type annotation during type checking, with the first two parameters
turned into positional-only, and thus the change in
`analytics.views.stats`.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
This makes `has_request_variables` more generic, in the sense of the return
value, and also makes it more accurate, in the sense of requiring the
first parameter of the decorated function to be `HttpRequest`, and
preserving the function signature without using `cast`.
This affects some callers of `has_request_variables` or the callers of its
decoratedfunctions in the following manners:
- Decorated non-view functions called directly in other functions cannot
use `request` as a keyword argument. Becasue `Concatenate` turns the
concatenated parameters (`request: HttpRequest` in this case) into
positional-only parameters. Callers of `get_chart_data` are thus
refactored.
- Functions to be decorated that accept variadic keyword arguments must
define `request: HttpRequest` as positional-only. Mypy in strict mode
rejects such functions otherwise because it is possible for the caller to
pass a keyword argument that has the same name as `request` for `**kwargs`.
No defining `request: HttpRequest` as positional-only breaks type safety
because function with positional-or-keyword parameters cannot be considered
a subtype of a function with the same parameters in which some of them are
positional-only.
Consider `f(x: int, /, **kwargs: object) -> int` and `g(x: int,
**kwargs: object) -> int`. `f(12, x="asd")` is valid but `g(12, x="asd")`
is not.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
This adds an assertion ensuring the type of `store` before accessing the
`cache_key` attribute that does not exist in the base class. Also note
that `.decode` returns `Dict[str, Any]` instead of a `str`.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
This allows this component to follow existing design patterns, rather
than being its own unusual element with various quirks.
The implementation is approximately the same as before I just migrated
"Edit custom profile field" form into modal, like "Add new custom
profile field" form modal.
Fixes: #21634
This commit is a preparatory commit of "custom_profile_fields: Edit
custom profile field form into modal."
The object returned by get_profile_field_info, which contained jQuery
objects for the profile-field-row and profile-field-form elements, was
just a source of unnecessary indirection, since none of the functions
consuming it accessed the row.
This lets us remove this fairly ugly user interface widget, which was
inconsistent with the rest of the settings UI.
The implementation is approximately the same as before I just migrated
"Add new custom profile field" form into a modal, status update
notifications about these forms will be displayed inside their modal,
and made some little design changes as discussed on CZO.
Fixes part of: #21634
This is a preparatory commit for commit "Edit custom profile
field form into modal."
Above mentioned commit will use class input-group instead of
control-group so here we are already changing it to input-group.
There is no control-group class to hide/show in custom profile fields
list instead there is input-group class, kind of little typo I guess
from this 9e343f1 commit.
We now send a new user_topic event while muting and unmuting topics.
fetch_initial_state_data now returns an additional user_topics array to
the client that will maintain the user-topic relationship data.
This will support any future addition of new features to modify the
relationship between a user-topic pair.
This commit adds the relevent backend code and schema for the new
event.
Updates documentation to include information about user presence
objects with `aggregated` key (instead of the user's email) where
appropriate.
Also, cleans up spelling, grammar and formatting errors in the
descriptive text for these objects / endpoints.
`render_table` calls itself recursively when it finds nested
`additionalProperties` (i.e. nested objects) in response schema,
to render their properties.
This fixes `render_table` to call `render_desc` along with
calling itself, to render the description of the nested
`additionalProperties` as well.
Kartik Srivastava
Tim Abbott
Zixuan James Li
Sahil Batra
Alex Vandiver
luisaadanttas
Ganesh Pawar
Lauryn Menard
anurastogiji
madrix01
Julia Bichler
Yogesh
evykassirer
N-Shar-ma
Anders Kaseorg