| Age | Commit message (Collapse) | Author |
|---|
|
Previously Q_NAMESPACE did not require its own moc, somehow blending
into others; now it does.
|
|
QChar now accepts more types for construction, and that unraveled
concatenation of a Type/SecondaryType character with a QString.
To fix it, give the compiler a hint by casting to the enum's underlying
type (which also nicely documents that we _actually_ switch from enum
to character type).
|
|
Now that QMetaType introspects into types, it reveals hidden problems
(which is very nice of it).
|
|
|
|
Waiting for the Multimedia arrival in Qt 6.2.
|
|
This fixes reliance on QIODevice being magically available for
std::unique_ptr<> by indirect inclusion. Since Qt 6 this inclusion no
more happens, time to #include <QIODevice> explicitly.
|
|
|
|
erase_if is now also provided by Qt; doing pretty much the same thing,
the Qt implementation only returns the number of removed entries instead
of returning a collection of them, however. Worth admitting at this
point that the function in connection.cpp has never had the semantics
of STL's erase_if() and doesn't quite have the semantics of remove_if()
either; but at least it's closer to remove_if().
|
|
See https://www.qt.io/blog/whats-new-in-qmetatype-qvariant#qmetatype-knows-your-properties-and-methods-types
|
|
|
|
This is an adjustment to the earlier fix of #471: if a join is
immediately followed by a leave (e.g. from another client/bot - you
can't do it programmatically from libQuotient) the sync may bring the
room already in the Leave state; therefore `joinRoom` should not
impose the state but rather ask `provideRoom` to create a `Join` room -
just as it's designed when passed an empty `joinState`.
|
|
Add libquotient.kdev4 to .gitignore
|
|
|
|
An alternative implementation of #463 (and thanks to Carl for spotting
the original problem).
|
|
GitHub images no more have GCC 8.
|
|
Make it possible to load the user metadata
|
|
In the normal case there is always a room that is associated with an
user. So it is in most of the cases, possible to load the metadata
(display name and avatar url) with the help of the room.
In some cases, it is not possible. For example, when opening an user
matrix link pointing to an user and not to a room. In this case, we need
to load the metadata independly of the room, since the user is not
linked to a room.
|
|
Fix broken links in README
|
|
|
|
Add public method to determine if we can change the user password
|
|
Co-authored-by: Alexey Rusakov <Kitsune-Ral@users.sf.net>
|
|
|
|
Adds ability remove the avatar
|
|
Not possible previously
|
|
(cherry picked from commit b25785d294669f2bab7dcd1e3cd1fba61991fe46)
|
|
As per the latest iteration of MSC2312, room/, user/ and event/ are
only supported for parsing and replication but not for emitting from
Matrix identifiers.
(cherry picked from commit 86f24d1ecf300b82b3a7253b81a2c392669d2c2b)
|
|
Load user info (display name + avatar) for the local user.
|
|
Co-authored-by: Alexey Rusakov <Kitsune-Ral@users.sf.net>
|
|
|
|
Fix rich edits (transmit)
|
|
Fix rich replies json format (transmit)
|
|
If someone still needs it - fix it; otherwise building with qmake
will be dropped in 0.7
|
|
The new formatted_body was not included into new content on edit
due to badly constructed json.
|
|
With this patch it looks like:
"m.relates_to": {
"m.in_reply_to": {
"event_id": "$another:event.com"
}
}
instead of:
"m.relates_to": {
"event_id": "$another:event.com",
"rel_type": "m.in_reply_to"
},
So it fits the specification by now.
https://matrix.org/docs/spec/client_server/r0.6.1#rich-replies
|
|
|
|
Maybe it's not even that bad, given that an effort is taken to recover
from the internal member list corruption.
|
|
For some reason the installed one doesn't find QtTest dynamic library.
|
|
|
|
Shared libraries are a bit of a chore to handle, maybe another time.
|
|
|
|
That way InstallQuotest feature is also tested. Also fix the Valgrind
suppression file path.
Also: use cmake arguments instead of pushd/popd dance
|
|
|
|
- The feature summary is only generated at the end of the configuration.
- InstallQuotest feature is defined in quotest/CMakeLists.txt now,
and therefore is only available if quotest is getting built
(i.e., if BUILD_TESTING is on).
- API generation configuration code merged from two places into one.
|
|
The current test may fail with "undefined reference" errors discussed at
https://stackoverflow.com/questions/14198972/undefined-symbols-for-qcompare.
|
|
Otherwise CMake's automoc complains that it doesn't see a moc-able class
definition in the file that includes "*.moc".
|
|
Now that CTest is included, quotest target should build (and install)
automatically together with the library.
|
|
It's a missing line from the recent autotests introduction. Without it
CMake doesn't introduce BUILD_TESTING option, and if you explicitly
pass it, fails to configure the project.
|
|
|
|
|
|
This is needed for a few cases like the account list in NeoChat or the
account switcher. In this cases we don't have a room binded to the user
and can't fetch the real display name and avatar.
|
