| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
It does not appear to use anything from QtNetwork.
[ChangeLog][Potentially source-incompatible changes] QtNfc
no longer depends on QtNetwork.
Change-Id: I23a71684fee760b091579e2f57eaf62d90a4743a
Reviewed-by: Ievgenii Meshcheriakov <ievgenii.meshcheriakov@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
|
|
|
|
|
|
|
|
|
| |
The localAdapter variable is a leftover from the BlueZ4 support.
The remoteDevicePath variable was never used.
Pick-to: 6.5
Change-Id: I93a691627467700cc01d2ee372a5d8e661083cd4
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Now when we have the fine-grained public QBluetoothPermission API, we
can use it to check for permissions in the Android implementation.
Adapt QtBluetooth examples to use only the minimal required set of
permissions.
As a drive-by: fix a couple of qCWarning() strings.
Task-number: QTBUG-109964
Change-Id: I55c67bcae27926e76bf9c1a8a9367af2b58883ff
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
This doesn't seem to be necessary anymore. Maybe it was at some point,
Removing them, even without moving it to qtbase doesn't cause any
problems either.
Pick-to: 6.5
Task-number: QTBUG-109394
Change-Id: Id9a62f23aba355e2e37f70a374aa20c48ce4ef83
Reviewed-by: Alexey Edelev <alexey.edelev@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
Expect it to be done on the application side. The library only checks
for the needed permissions.
Task-number: QTBUG-109964
Change-Id: Ibb6b756ddea13c14a4c096d200418923e9ccd69c
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Peripheral applications don't work properly due to permission-related
initialization changes.
The underlying CBPeripheralManager was instantiated only when
startAdvertising() was called, which can be too late; the peripheral
manager is needed already earlier if the application calls
addService().
Calling addService() before startAdvertising() is not mandatory
though, and hence the lazy initialization needs to be attempted in
both functions.
Amends ae1a1f5efce291f613a13757ff6f744fcca2d2ce
Task-number: QTBUG-112215
Change-Id: I7771a2e1bd951093e33030546e5da69538c8dcd1
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Namely: fromCBUUID and toCBUUID, similar to fromNSUUID and toNSUUID that
we already have.
[ChangeLog][QtBluetooth] Add CoreBluetooth-specific conversion to
QBluetoothUuid class, to get CBUUID out of QBluetoothUuid and convert
CBUUID to QBluetoothUuid.
Fixes: QTBUG-112303
Change-Id: I4ca0b13395b4346f7c830a0e042a16f976998a33
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
|
|
|
|
|
|
|
|
| |
Checks are already done at base or the public class
Pick-to: 6.5
Change-Id: Idf5ae08cb34fa1293de8f270c8980799866a8b08
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
When building with Unity, Windows Bluetooth symbols from WIN32/COM and
more modern WinRT APIs, e.g., AsyncStatus, start to confuse the compiler
as related source files start to be pooled together.
Pick-to: 6.5
Task-number: QTBUG-109394
Change-Id: I3d59f09caca3e56b6961fb6ffcdf8233369313d2
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The deviceDiscovered() signal can be emitted more than once for the
same device, so implement a check before adding a device.
This commit amends a1f43b9dcd722527d52ceb0c7a138bd2994c1a7b
Task-number: QTBUG-111972
Pick-to: 6.5
Change-Id: Iff0a2b301d6d05090745ef97cedf5fab4532da7e
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some code clean-ups:
* Use qt_standard_project_setup() and PRIVATE linking in CMake
* Port away from QScopedPointer and use std::unique_ptr instead
Documentation updates:
* Add Connectivity category
* Mention the new BlueZ D-Bus backend in the documentation
* Small wording improvements
* Provide more links to Qt classes and enums
Task-number: QTBUG-111972
Pick-to: 6.5 6.5.0
Change-Id: Id2a0ef48d3ce82f1784dd4b51f94236731730432
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously we were doing full service discovery every time when the
BluetoothLEAdvertisementWatcher reports a new advertisement.
This results in a very high CPU usage during BTLE device discovery
on some configurations.
In practice such approach might not be needed, because most BTLE
devices repeatedly advertise the same data.
With this patch we do a detailed service discovery only when a new
device is added, or when the list of advertised services has changed.
Note that we still need to handle all the incoming advertisements,
because we want to report changes in other parameters, such as RSSI.
Fixes: QTBUG-111116
Pick-to: 6.5 6.5.0 6.2
Change-Id: I0863739a2fb21d04e9a728adc68e3f197f1966e3
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The example contains a lot of rather complicated code, which is not
directly related to the demonstrated Bluetooth features.
Remove it in favor of BtChat example, which demonstrates the same
Bluetooth features in a much more compact and understandable way.
Task-number: QTBUG-111972
Pick-to: 6.5 6.5.0
Change-Id: Iccb2598fa6e88816697286b2252858f35a635ee5
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The example mostly shows the usage of
QBluetooth{Device,Service}DiscoveryAgent classes, which is also
illustrated by other examples. Move it, because it's not fully
functional on mobile platforms (because it's widget based).
Do not remove it completely, because it might still be useful
for doing Classic device scan.
Also implement some improvements while on it:
* fix CMakeLists.txt by using qt_standard_project_setup() and PRIVATE
linking
* fix memory leak in DeviceDiscoveryDialog
* rework the code to use Qt parent-child model instead of manually
deleting objects where possible
* fix includes
* fix forward declarations
Task-number: QTBUG-111972
Pick-to: 6.5 6.5.0
Change-Id: Ie4bc9e25ccdda6d5f5de2f57528df349c71cdc12
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We only _check_ if permissions were granted, leaving it up to an application
to _request_ permissions. If permission status is not 'Granted', we bail
out ealy setting MissionPermissionsError. QLowEnergyController::init is
now a no-op for Darwin, because it's OK to create a peripheral/central and no need
to set any error yet. Autotests require minor adjustments - they were
already passing if BT is off, as it is on CI (checking Bluetooth local
device and its status), but if BT is somewhere on, tests can try to scan or
connect not having permissions granted - for this we adjust the tests,
using permission API.
[ChangeLog][Important Behavior Changes][QtBluetooth][Darwin] Do not request
permissions implicitly, only check them and leave it to applications to request
permissions explicitly.
Task-number: QTBUG-109964
Change-Id: I95c04744e979614ffb6d992da2e279e86b272679
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The old implementation didn't take care of escaping the XML-specific
characters and didn't handle non-printable characters.
This patch makes use of QUrl class to properly %-encode the input data.
The QUrl::toEncoded() method %-encodes all XML-specific characters
except '&', so we need to manually replace it with "&" before
adding the url to the generated XML.
Escaping special XML characters potentially allows Qt Bluetooth to
handle more URLs received from sdpscanner, because QXmlStreamReader
discards attributes with unescaped special characters, so previously
part of the URLs could be silently skipped.
For other potential sdpscanner users this change shouldn't make much
difference, because they should anyway parse the returned XML documents
according to XML standard.
%-encoding of URLs potentially changes the way the URL looks for the
user, but not for the software that should handle the URLs, so this
change is also safe.
[ChangeLog][Qt Bluetooth][sdpscanner] sdpscanner now %-encodes the
URLs and escapes all XML-specific characters in them before adding
the result to the generated XML output.
Fixes: QTBUG-111369
Pick-to: 6.5 6.4 6.4.3 6.2 5.15
Change-Id: I6de080fef7689ef96fe5e5e26c62a3c48ebc45b7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QByteArray::resize() treats all negative parameters as a request for
a zero length. So the code
text.resize(text.indexOf('\0'));
can completely erase the text if there is no '\0' in it.
Fix it by explicitly checking the return value of QByteArray::indexOf().
Pick-to: 6.5 6.4 6.2 5.15
Change-Id: Idc42bf4b96a9be5b007916263d6cf1e831b96c07
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Do not use the fixed-size temporary buffer, instead just parse the
data as a QByteArray.
Grepping through BlueZ sources, I could find only several usages of
SDP_URL_STR{8,16,32}, and all of them suggest that the url is simply
a NULL-terminated string (see [0], [1], [2]).
However, the older BlueZ sources suggest that the url can be not
NULL-terminated as well (see [3]).
To be on a safe side, we provide an implementation that handles both
cases correctly.
[0]: https://github.com/bluez/bluez/blob/9be85f867856195e16c9b94b605f65f6389eda33/lib/sdp.c#L465
[1]: https://github.com/bluez/bluez/blob/9be85f867856195e16c9b94b605f65f6389eda33/src/sdp-xml.c#L351
[2]: https://github.com/bluez/bluez/blob/9be85f867856195e16c9b94b605f65f6389eda33/tools/sdptool.c#L517
[3]: https://android.googlesource.com/platform/external/bluetooth/bluez/+/master/src/sdp-xml.c#324
Fixes: QTBUG-111242
Pick-to: 6.5 6.4 6.2 5.15
Change-Id: I22f9521582863fb316dd0b2c49a78928b80a6078
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Only kept for in removed_api.cpp for binary compatibility.
[ChangeLog][QtBluetooth] The undocumented quint128 was removed from
QtBluetooth. This affects the QBluetoothUuid constructor and the
toUInt128() method. QUuid has support for an integer quint128 on certain
64-bit platforms (all except MSVC) with a constructor and a getter of
the same name, so this change is mostly source-compatible, so long as
the application code doesn't attempt to treat this type as a structure.
For cross-platform support, use QUuid::Id128Bytes.
Change-Id: Id8e48e8f498c4a029619fffd1728c34931ac525e
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
| |
Change-Id: Id8e48e8f498c4a029619fffd1728c8facac724fd
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
| |
This will simplify the operation when QUuid::toBytes supports selecting
the endianness.
Change-Id: Id8e48e8f498c4a029619fffd1728c8e666dbc087
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
minimumSize() and toUInt16() need to perform the same check, so do it
only once. And make one function call the other, to avoid repeating
oneself.
Change-Id: Id8e48e8f498c4a029619fffd1728c8f415f36045
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is now supported by the QUuid class directly instead of being an
invention of QBluetoothUuid on top. Additionally, as a trivial structure
of 16 bytes in size, it should be passed by value, not by const-ref.
The BluezUint128 type is used where the 128 bits of data aren't a UUID (like
the Connection Signature Resolving Key).
Change-Id: Id8e48e8f498c4a029619fffd1728c1af9149a561
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
iOS needs some time after invalidating a session before a new session
can be started. Otherwise the NFC dialog of iOS will not show up.
For restarting a session inside the iOS NearfieldManager, this was
already solved with a timeout of 2 seconds.
This commit fixes the case, that a user of the Nearfieldmanager
restarts a session manually too fast.
Pick-to: 6.5 6.4 6.3 6.2 5.15
Change-Id: Ic91ad225a9cab13ba92523f33a19f44af68575a0
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
The versions were too old to begin with (minimum is 23) and perhaps
better use the default to begin with.
Pick-to: 6.5
Change-Id: Ie5c502001c4c4c59da091e7fbf899e4a5234034c
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
The overarching theme for these API deprecations is to
improve memory safety. This patch replaces the descriptor
related APIs.
Pick-to: 6.5
Fixes: QTBUG-107180
Change-Id: Ief098d67d2601e06a2f29fa38d9b481bbda605f9
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The overarching theme for these API deprecations is to
improve memory safety. This patch replaces the characteristic
related APIs.
The list of deprecated APIs, addressed by this and the followup commit:
* BluetoothGatt::writeCharacteristic()
* BluetoothGatt::writeDescriptor()
* BluetoothGattCharacteristic::getValue(), no replacement
* BluetoothGattCharacteristic::setValue(), no replacement
* BluetoothGattDescriptor::getValue(), no replacement
* BluetoothGattDescriptor::setValue(), no replacement
* BluetoothGattCallback::onCharacteristicChanged()
* BluetoothGattCallback::onCharacteristicRead()
* BluetoothGattCallback::onDescriptorRead()
* BluetoothGattServer::notifyCharacteristicChanged()
Pick-to: 6.5
Task-number: QTBUG-107180
Change-Id: I054db07301038ceb56b8af8471d71309d61ae777
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Reproducing the issue reliably has proven to be near impossible,
but the stack trace can be seen every now and then, and it does
not seem to cause any functional harm.
The strong suspicion on what happens is that in some places of the
code we create a temporary QBluetoothLocalDevice object to check
for example address(). This happens for example when creating a
QLowEnergyController object, which is one of the scenarios I recall
seeing the unhandled exception.
The localdevice creation triggers broadcast receiver registration,
and the ~immediate destruction triggers unregistration. Reading the
address() works as it does not rely on any broadcast events.
It seems plausible that in some cases the registration is not yet
fully complete on Android side when the unregistration is ~immediately
called.
Pick-to: 6.5
Fixes: QTBUG-106938
Change-Id: Iac734f17fb9d2a1879fca39d9b2839cb530e5df5
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
| |
Remove clashing string constants.
Pick-to: 6.5
Task-number: QTBUG-109394
Change-Id: I44c89cbc58fcc2e2ae33d23acd541eaba1a5a288
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Because we include <winrt/base.h>, use the factory-cache-cleaner
registration developed in
qtbase/ffb9dee1b0954e4d4f9e9791175609a80ecafc31 to make QtBluetooth
contribute its part to fixing QTBUG-103611.
Including the qfactorycacheregistration_p.h header from another header
is explicitly supported (inline variable).
Pick-to: 6.5 6.4 6.2
Task-number: QTBUG-103611
Change-Id: I1f21a7001a3eea88ddc5ba647c72c3dd22c69374
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
values
Android backend did not update the serivice's characteristic and
descriptor values after executing writeCharacteristic/writeDescriptor.
As a result, the connected central saw the correct updated value, while
local request still returned the old value.
This patch fixes it by properly updating the local values when the write
operations complete successfully.
Pick-to: 6.5
Change-Id: Ie09299b6c72bbf92ab6c824c3ca23f0136f0457e
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The Bluez DBus peripheral role support cleared the remoteDevice so that
the QLowEnergyController::remoteAddress() will return empty after
disconnection.
However the resetController() is also called in client role
connectToDeviceHelper(), and by reseting the remoteDevice
address there, the controller won't find the remote device path and
connect will fail.
Instead clear the remoteDevice address in places specific to the
peripheral role.
Amends: e7499c2cca615eaff3bd4c9ffe5f72d7112055cf
Pick-to: 6.5
Change-Id: I53a0de157873e95b9e832f67a24d902277f233a1
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
paired devices
Commit 0b60ca266f0fe27053a58eff3dbd903e3a1678ca introduced Close()
calls for GattDeviceService objects. IIRC, it was done to avoid
potential GattCommunicationStatus_AccessDenied errors, which could
otherwise happen when trying to re-acquire the non-closed service.
This commit does the proper cleanup when the device is disconnected.
However there is one corner-case: when we try to connect to a paired
device which is turned off, Windows provides its services (using some
cached data), but the service details discovery fails.
The details discovery is executed in a background thread using a set
of async calls. If the user disconnects from the device during this
discovery, the GattDeviceService::Close() call for the respective
service will block until the async operation fails (which takes
a couple of seconds on Windows 11). As a result, the UI thread freezes
for this time.
This patch is an attempt to avoid it by shifting the Close() call to
the background helper thread in such cases. If there is an error during
details discovery, or if it just takes too long, and the user decides
to disconnect from the device, the service will be closed directly
in the helper thread.
However, if the details discovery is completed successfully, the
GattDeviceService object can still be re-used for other operations,
and it will not be deleted from cache until the device is disconnected.
Fixes: QTBUG-108461
Pick-to: 6.5 6.4 6.2
Change-Id: I8ebe945130808ed7bd8852a76bba84c73651f5a5
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The QWinRTLowEnergyServiceHandler helper class is used to query low
energy service details. It works in a background thread and performs
multiple async operations to get the results.
As the class works in a background thread, we use
QWinRTFunctions::await() to wait for the results of async operations.
This allows to simplify the code.
In practice the user might want to disconnect from the device while
service details discovery is performed. Previously the background
thread was still performing all the requests.
This patch handles the already-existing
QLowEnergyControllerPrivateWinRT::abortConnection() signal and
provides the exit condition to all QWinRTFunctions::await() calls,
so that the background thread does not do unnecessary work.
Task-number: QTBUG-108461
Pick-to: 6.5 6.4 6.2
Change-Id: I1db05c906beb0078de8a00fd5e32d589306f6665
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
| |
Task-number: QTBUG-106533
Pick-to: 6.5
Change-Id: I5b09d882230ba6b996f5edc5ac4502d0bc004525
Reviewed-by: Amir Masoud Abdol <amir.abdol@qt.io>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
The various versions do not matter anymore as current Qt Version
are long past the mentioned versions.
Pick-to: 6.5 6.4 6.2
Change-Id: I285785eb64a82cacdd853380312e4810ec907c1c
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Juha Vuolle <juha.vuolle@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
Adjust variable names to be case-sensitive.
Pick-to: 6.5 6.4 6.2
Fixes: QTBUG-109315
Change-Id: I7d534fbfd7891740fbc8ac172b24a2b03832388b
Reviewed-by: Kai Köhne <kai.koehne@qt.io>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit adds error message for unauthorized writes. This may
occur if peripheral has set "authorize" flag for char/desc, and
the client device is not trusted.
In addition the commit sets a more appropriate DBus error for
invalid offset errors.
Task-number: QTBUG-107511
Pick-to: 6.5
Change-Id: I2e477dd77135e95100ef0cca814a6416df27c2b0
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
| |
These are Qt functions that BlueZ invokes over DBus, and in some
circumstances an error reply is needed.
Task-number: QTBUG-107511
Change-Id: If325621d12600990e86fe07681aabfba8adc9056
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
| |
Task-number: QTBUG-107511
Change-Id: Ia94262a29457489f8b7d5ffd2f5b50f943eb4b21
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
| |
[ChangeLog][QtBluetooth] Add support for Bluez DBus peripheral role
Fixes: QTBUG-107510
Change-Id: I1c26606ff0b01818c6f446147e005090582ba877
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
This is a preparatory commit for adding the Bluez DBus peripheral
role support.
Task-number: QTBUG-107510
Change-Id: I27860947c3f94fa0cc4d2af0ce90ab3c0f2287eb
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
| |
Fixes: QTBUG-107509
Change-Id: I6e26cbc0502cdc65eb6af20bfeb481f0b62ff386
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
A preparatory task for implementing Bluez DBus peripheral role support.
This commit introduces the bulk of the needed interfaces; further
manual adjustments are likely needed in the followup commits.
Fixes: QTBUG-107506
Change-Id: I5152002746db3b6ffc20981f34a34bbe13880b2f
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
|
| |
A preparatory commit for adding Bluez DBus peripheral role support.
Fixes: QTBUG-107508
Change-Id: Ifa2b12df9856011048a9bc311e5da96275700f5e
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
This fixes an issue where, upon reconfiguring, the package
would not be looked up again (a side-effect of an optimization
where we avoid looking up packages that have not been found).
This lead to a "force feature ON breaks condition" error.
Change-Id: I0be1bd6dd83186fe462d2dda9dbd93dfa372629a
Reviewed-by: Ievgenii Meshcheriakov <ievgenii.meshcheriakov@qt.io>
|
|
|
|
|
|
|
| |
QUuid can create ID128 strings on its own.
Change-Id: Id8e48e8f498c4a029619fffd1728938a1b70e4a0
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
|
|
|
|
|
|
|
|
|
| |
The snippets in the NFC overview page refer to the ndefeditor
example that is about to be completely replaced.
Task-number: QTBUG-103949
Change-Id: Ie0b42b36f82b58f680a5a355f79e04e0422e32c9
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|
|
|
|
|
|
|
|
|
| |
GPL-2.0 is deprecated in SPDX. This also fixes an issue with
qtattributionsscanner in latest qttools.
Pick-to: 6.4
Change-Id: I462b6c23ffaf1cbafcfd7feadc12f99ddbe797e5
Reviewed-by: Jörg Bornemann <joerg.bornemann@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
The row has been empty since its introduction and even if used, wouldn't
provide very useful information (other rows imply the support)
Pick-to: 6.2 6.4
Fixes: QTBUG-107196
Change-Id: I3e3ca4e6c446707772a24fc6b3ea000e4b889fdf
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
|