Merge the functionality of CollapseStateProxyModel into MessageFilterModel there is no need for a whole separate model the filters can be combined trivialy.
In some cases (where we don't get a thumbnail) we load the entire image,
which might be large. Even when we get a thumbnail, the smoothing we
apply doesn't get the best results and the only other option is
mipmapping which is more memory intensive - it stores progressively
smaller images but keeps all of them in memory.
Now the ImageDelegate sets the sourceSize, which helps reduce the memory
footprint (since Qt will throw away whatever parts of the image we don't
need) and makes the images look smoother when scaled. However,
AnimatedImages do not support this so this change introduces a new
property to tell if the image could possibly be animated. It will
erroneously pick up formats that could support animation (e.g. WebP) but
other image formats will still benefit so it's a net positive. The
ImageDelegate will load the correct image component depending on this
property.
When auto-completing rooms, there may be some that do not have
canonical addresses (e.g. 1-to-1 chats). These can't be linked anyway,
so skip them and any other unlinkable entries.
- Split the list into sections for "this devices", "verified devices", "unverified devices", and "devices without encryption support"
- Sort the lists by last activity
Add missing #pragma once + missing include
* speeds up incremental builds as changes to a header will not always
need the full mocs_compilation.cpp for all the target's headers rebuild,
while having a moc file sourced into a source file only adds minor
extra costs, due to small own code and the used headers usually
already covered by the source file, being for the same class/struct
* seems to not slow down clean builds, due to empty mocs_compilation.cpp
resulting in those quickly processed, while the minor extra cost of the
sourced moc files does not outweigh that in summary.
Measured times actually improved by some percent points.
(ideally CMake would just skip empty mocs_compilation.cpp & its object
file one day)
* enables compiler to see all methods of a class in same compilation unit
to do some sanity checks
* potentially more inlining in general, due to more in the compilation unit
* allows to keep using more forward declarations in the header, as with the
moc code being sourced into the cpp file there definitions can be ensured
and often are already for the needs of the normal class methods
This is a significant rework of the handling of push rules. Rather than using a lot of boilerplate code for the default models `KeywordNotificationModel` has been converted to `PushRuleModel` and now handles all push rules.
The new model has the following features:
- Handles all push rules
- Has special handling for the names of default keywords (i.e. it still gives the same text as previously for showing in the settings menus)
- Push rules for blocking individuals or room overrides are still there but hidden so will be available for developer tools (to follow)
- Room specific keywords are now supported.
The notification settings pages have also been refactored to take advantage of the new models. Each section is now just a repeater with a filter for the rules that it should contain. The push rule delegate has now been cleaned up and uses required properties.
Implements network/neochat#574
Create a media model for all the media message in the timeline and then setup `NeoChatMaximizeComponent` so that it can use the media model to scroll through all loaded images and video in the current room.
Depends upon libraries/kirigami-addons!105
FEATURE: 467411
data() being called with an invalid model index seems to happen during
normal operations here, with something in the proxy model chain not
catching that case (due to a row being filtered out presumably?).
This check worked for accounts that already had custom emojis, but if it
never had any added, m_images was never filled and the model would crash
when adding more.
