| Age | Commit message (Collapse) | Author |
|---|
|
Quotient::function_traits<> did not support member functions in a proper
way (i.e. the way std::invoke_result<> treats them, with the function's
owning class represented as the first parameter). Now that I gained
the skill and understanding in function_traits<> somewhat wicked
machinery, I could properly support member functions. Overloads and
generic lambdas are not supported but maybe we'll get to those one day.
|
|
It has not much to do with the Visitor design pattern; also,
std::visit() has different conventions on the order of parameters.
|
|
|
|
There was a lot of excess redirection in fromJson() and toJson() with
most of JsonConverter<> specialisations being unnecessary boilerplate.
These have been replaced by overloads for toJson() and explicit
specialisations for fromJson() wherever possible without breaking
the conversion logic.
|
|
Proper linters recognise that the returned types are not primitive,
while people might still be confused a bit.
|
|
In both fixed cases the callee accepts a const reference, which makes
std::move() useless. Static analyzers apparently missed them because
the cases are inside a macro.
|
|
It's might look weird; but without making fromJson() a specialisation
it becomes an overload next to an implicit specialisation of
the template function defined just above, and then loses to that
specialisation because it (also) has the perfect match.
(would be great if the compiler shaded the implicit specialisation in
such cases - alas it's not how the standard works.)
|
|
|
|
Similar to contentPart() - apparently there are enough places across
the code that would benefit from it.
|
|
Given that QJsonObject only accepts QStrings in the list constructor,
the template is useless cruft.
|
|
|
|
There's a clash between Event::content() (a template function) and
RoomMessageEvent::content() (plain member). Out of these two, the name
more fits to the RME's member function - strictly speaking,
Event::content() retrieves a part of content, and so is renamed.
In addition, contentPart() defaults to QJsonValue now, which is pretty
intuitive (the function returns values from a JSON object) and allows
to implement more elaborate logic such as
if (const auto v = contentPart<>("key"_ls); v.isObject()) {
// foo
} else if (v.isString()) {
// bar
} else {
// boo
}
|
|
|
|
This is useful for implementing Spaces support, where all events of
type `m.space.child` are needed, and we don't know their state keys in
advance.
|
|
The problem is in Room::processStateEvent(): after
potentially-inserting-nullptr into currentState, pre-check failure
(that may occur on member and trigger events for now) leaves that
nullptr in the hash map. Basically anything that uses currentState
(e.g., Room::toJson) assumes that currentState has no nullptrs - which
leads to either an assertion failure, or nullptr dereferencing. The fix
removes the nullptr placeholder if the pre-checks failed.
|
|
Trying to test bits with Changes::testFlag(Change::Any) was a bad idea.
Along the way: made logging in setLastReadReceipt() refer to the actual
timeline item when possible.
|
|
Doesn't really help build times, instead breaking the build on older Qt.
|
|
The "original JSON" wording is misleading: the returned JSON can be and
is routinely edited as a part of event construction, redaction, editing.
Also, originalJson() name is misleading in that it returns a stringified
(in a very specific way) JSON and not an object. You have to call
fullJson() to get the object, and originalJsonObject(), confusingly,
returns exactly the same thing but as a value rather than as
a reference. The original intention of keeping originalJsonObject() was
to make it Q_INVOKABLE or use it as an accessor for a Q_PROPERTY.
unfortunately, this was never really practical as discussed in
the previous commit.
All that implies that clients have to handle passing event JSON to QML
themselves, in the form they prefer (as an object or a string). The
added complexity is negligible though; on the other hand, there's added
flexibility in, e.g., choosing a compact instead of default JSON layout
or even generate a highlighted JSON representation.
|
|
Makes compilation a tad lighter.
|
|
Q_GADGET is generally used to enable two things outside of QObject:
Q_PROPERTY/Q_INVOKABLE and Q_ENUM/Q_FLAG. While the latter can be used
in its own right in QML, the former requires Q_GADGET instances to be
passed to QML by value, which is not really possible with
uncopyable/unassignable classes. Bottom line is that Q_PROPERTY in
anything derived from Quotient::Event is not viable, making Q_GADGET
macro useless unless there's a Q_ENUM/Q_FLAG (as is the case with
RoomMessageEvent, e.g.).
|
|
|
|
|
|
|
|
|
|
|
|
This introduces a new API to count unread events that would allow to
obtain those unread and highlight counts since either the fully read
marker (Room::partiallyReadStats) or the last read receipt
(Room::unreadStats). Element uses the read receipt as the anchor
to count unread numbers, while Quaternion historically used the fully
read marker for that (with the pre-0.7 library sticking the two markers
to each other). From now on the meaning of "unread" in Quotient is
aligned with that of the spec and Element, and "partially read" means
events between the fully read marker and the local read receipt;
the design allows client authors to use either or both counting
strategies as they see fit. Respectively, Room::P::setFullyReadMarker()
updates partially-read statistics, while Room::P::setLastReadReceipt(),
when called on a local user, updates unread statistics.
Room::notificationCount() and Room::highlightCount() maintain their
previous meaning as the counters since the last read receipt;
Room::notificationCount() counts unread events locally, falling back
to the value from the above-mentioned key defined by MSC2654, and if
that is not there, further to `unread_notifications/notification_count`
defined in the current spec. Room::highlightCount(), however, is still
taken from the homeserver, not from Room::unreadStats().highlightCount.
|
|
This makes updating display name and emission of necessary signals
including Room::changed() more systematic when it has to occur outside
of updateData() flow - e.g. when loading all members.
|
|
Since MSC2654's unread count is counted from the m.read receipt, and
the course is to follow the spec's terminology and use "unread count"
for the number of notable events since m.read, this required to move
the existing number of notable events since m.fully_read to another
field, henceforth called partiallyReadCount. At the same time,
SyncData::notificationCount is dropped completely since MSC2654 claims
to supersede it.
Also: Room::resetNotificationCount() and Room::resetHighlightCount() are
deprecated, as these never worked properly overwriting values that can
be calculated or sourced from the server, only for these values to be
set back again the next time the room is updated from /sync.
|
|
Room::isEventNotable has been moved out from Room::Private and made
compliant with MSC2654.
The concept of Room::checkForNotifications is taken from Quaternion
where a method with the same name has been in QuaternionRoom for a long
time - however, actual body is a stub for now, always returning
{ Notification::None } (Quaternion's implementation is too crude to be
taken to the library). Now we really need a pushrules processor to fill
this method with something reasonably good. Internally the library now
calls checkForNotifications() on every event added to the timeline,
filling up the events-to-notifications map because it is anticipated
that calculation of notifications can be rather resource-intensive and
should only be done once for a given event.
Finally, Room::notificationsFor is an accessor into the mentioned map,
standing next to isEventNotable (but unlike isEventNotable, it's not
virtual; checkForNotifications is).
|
|
[skip ci]
|
|
|
|
By default enum class values are logged along with the qualifier; this
may or may not be desirable in a given setting. For JoinState(s) and
Membership(Mask) operator<< was overloaded to implicitly suppress
qualification; however, this is both overly sweeping and uses Qt's
internal API for the backend.
Instead, a new QDebug manipulator, terse(), is introduced, that does
the same as those operator<< overloads but on a per-invocation basis.
This makes it slightly more verbose to log enums but makes the QDebug
reconfiguration explicit and doesn't require to produce new overloads
every time a new enum ends up in logs. And it's built entirely on
the published Qt API, reusing the QDebugManip framework that Quotient
already has.
Also: operator<<(QDebug, QDebugManip) has been moved out of
the namespace to fix lookup issues when there's no prior
`using namespace Quotient`.
|
|
This enumeration sees very limited (if any) use outside Quotient; and
though this change will surely break code using it the fix is very
straightforward and quick.
|
|
|
|
|
|
# Conflicts:
# lib/room.cpp
|
|
To simplify retrieval of the local m.read receipt and the marker for it.
|
|
[skip ci]
|
|
This reduces the surface interacting with the User class that eventually
will be split into LocalUser (most part) and RoomMember (a tiny wrapper
around the member data in a given room, used almost everywhere in Room
where User currently is).
Also: dropped a log message when the new receipt is at or behind
the old one as it causes a lot of noise in the logs.
|
|
Port away from implicit 'this' captures in lambdas
|
|
Deprecated with C++20
|
|
|
|
Makes the Room::P::toJson() code more readable.
|
|
|
|
It didn't like using QT_IGNORE_DEPRECATIONS inside a statement.
|
|
|
|
|
|
|
|
A handy macro that introduces an enumerator with a respective
Q_DECL_DEPRECATED_X recommending the substitution.
|
|
These usually don't affect the room outlooks in the room list in any
way; so can be merged into OtherChange instead.
Also: OtherChanges synonym has been added, implying that there might
be more than one change behind a single "Other" flag.
|
