- Support serializing class properties instead of just dataclass fields
- filter_string is now a property that's always up-to-date
- Account display names matching the current SidePane filter are shown
Chnaging delegate heights is a terrible idea that leads to graphic bugs.
Handle filtering (both room and accounts) and collapsing from a function
that sets the JsonListModel source.
- Remove Python.loadingAccounts/willLoadAccounts
- Fix HPage SwipeView bug that caused its inner SidePane to steal focus
- Use overshoot for animation
- Get rid of sidepane flickering
- Set default HNumberAnimation easing.type to OutQuad, specify InOutQuad
for HCheckBox
- Cleaned up code
- Sizes are standardized
- New HCheckBox component
- Replace RememberAccount page by a checkbox on the SignIn page
- The page can now be flicked if it's not tall enough to display
everything
- HNumberAnimation now has an overshoot property that will set the right
easing type, overshoot and adapt the animation factor/duration.
- HImage can now be colorized
Instead of passing all sorts of events for the JS to handle and manually
add to different data models, we now handle everything we can in Python.
For any change, the python models send a sync event with their
contents (no more than 4 times per second) to JS, and the QSyncable
library's JsonListModel takes care of converting it to a QML ListModel
and sending the appropriate signals.
The SortFilterProxyModel library is not used anymore, the only case
where we need to filter/sort something now is when the user interacts
with the "Filter rooms" or "Filter members" fields. These cases are
handled by a simple JS function.
We now keep separated room and timeline models for different accounts,
the previous approach of sharing all the data we could between accounts
created a lot of complications (local echoes, decrypted messages
replacing others, etc).
The users's own account profile changes are now hidden in the timeline.
On startup, if all events for a room were only own profile changes, more
events will be loaded.
Any kind of image format supported by Qt is now handled by the
pyotherside image provider, instead of just PNG/JPG.
SVGs which previously caused errors are supported as well.
The typing members bar paddings/margins are fixed.
The behavior of the avatar/"upload a profile picture" overlay is fixed.
Config files read from disk are now cached (TODO: make them reloadable
again).
Pylint is not used anymore because of all its annoying false warnings
and lack of understanding for dataclasses, it is replaced by flake8 with
a custom config and various plugins.
Debug mode is now considered on if the program was compiled with
the right option, instead of taking an argument from CLI.
When on, C++ will set a flag in the Window QML component.
The loading screen is now unloaded after the UI is ready, where
previously it just stayed in the background invisible and wasted CPU.
The overall refactoring and improvements make us now able to handle
rooms with thousand of members and no lazy-loading, where previously
everything would freeze and simply scrolling up to load past events
in any room would block the UI for a few seconds.
- No more translatable, content_type, show_name_line attrs for
TimelineEventReceived.
Since they are UI concerns, they are handled directly in QML.
- Refactor the EventDelegate and get rid of errors when new items
are added to the timeline
- Messages, events and emotes all combine correctly.
- No more 28px wide avatars for events, to make them uniform with
messages.
http://es6-features.org/#PropertyShorthand
Instead of doing {"foo": foo, "bar": bar, ...}, we can just do
{foo, bar} now.
The function parameters of EventHandlers have all been renamed to
camelCase to make use of this, as the JS style conventions intend.
Other functions will follow in a later commit.
- max_instances renamed to max_running
- consider_args parameter:
if True, $max_running of this function with the same arguments can be
running, else:
$max_running of this function, no matter the arguments, can be running
- discard_if_max_running:
if True and there are already the maximum possible number of running
functions running, cancel this task (previous default behavior), else:
Wait for a spot to be free before running
- New ListItem implemented using a metaclass, which makes defining new
ListItem subclasses much cleaner and shorter
- ListModel functions taking an index now accept either an int
(list index) or str (value of a main key)
- getWhere() gone, simply use get(a_main_key_value) now
- updateOrAppendWhere replaced by update()/upsert()
- Standardized capitalization for variables and file names everywhere in
QML and JS, get rid of mixed camelCase/snakeCase,
use camelCase like everywhere in Qt
- ListModel items are now stored and returned as real QObjects with
PyQt properties and signals.
This makes dynamic property binding a lot easier and eliminates the need
for many hacks.
- New update(), updateOrAppendWhere() methods and roles property
for ListModel
- RoomHeader now properly updates when the room title or topic changes
- Add Backend.pdb(), to make it easier to start the debugger from QML
- Move HTTP connect/disconnect logic to networkManager
- If a talk fails due to socket error, HTTP transport error or
nio bad response that might change, retry every 2s until success
- Clean up some leftover debug prints
For any name not found in rooms data, rely on new
nio.HttpClient.get_displayname() function to get and cache it,
e.g. for our own name if no room is joined and past events from users
who left the room.
@futurize now returns PyQtFuture objects, wrapper for the
concurrent.futures.Future objects that can be used from QML,
to ensure name retrieval does not block the GUI.
- Trigger when room is shown if there's not enough messages to fill the
list height
- Trigger whenever user is scrolling before a certain point, instead of
when dragging is released/scrolling stopped and the top edge is hit
- Prevent multiple load requests at same time
- Keep a set of fully loaded rooms, don't request anymore history
if a room is fully loaded
