Needs manual testing in the common case that the
intended crash does not occur.
Change-Id: I99de4f466df9e3a470a6bcb98b819d15da3a7067
Reviewed-by: hjk <hjk@qt.io>
no-one ever queried this state since S60 support was removed in
ae23d50576 (2012).
Change-Id: I3e05d90bb43514b4c326e124834cf9c5e89a0168
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
since fa6d0f1231, DEPLOYMENT is aliased to INSTALLS, so we would have to
actually look for .sources in entries listed in that variable, which we
didn't. apparently, nobody noticed, among other things possibly because
the qt4 variant already supports .files in later versions.
also, our actual deployment code doesn't use .sources, either.
Change-Id: I990240716817118047fc9aa97eff55305fcf7eca
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
We relied on the CppModelManager to tell us whether a project was reparsed
after a kit change. While this worked, it was not guaranteed that the project
is really finished (and ready for e.g. building) after pushing new ProjectInfos
to the CppModelManager.
Rely on the projects telling when they are finished with parsing. This is more
accurate and future-proof.
The introduced signals in Project and SessionManager are (at the moment)
only for tests.
Change-Id: I1b368ec4585ffa8755eb28fac6d187cce31243ee
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
We fallback to create types from source, if other means failed.
This can be required if the meta info is incomplete/incorrect when
a C++ type is mocked.
Change-Id: I50d45a6f9d7eba3483da10d9e1d7d6644d9aeb96
Reviewed-by: Tim Jenssen <tim.jenssen@qt.io>
In examples and tutorials pages.
Since the pages are unloaded and loaded again when the tab changes, they
loose their internal state and require the search text to be kept as
state in the model (which makes sense anyhow).
Task-number: QTCREATORBUG-15901
Change-Id: I4fb05ca47f8337008c51e17cc95962a5e9e67fef
Reviewed-by: Thomas Hartmann <Thomas.Hartmann@theqtcompany.com>
The performance counters are broken on most (all?) i.MX6 boards, not
only on the ones from Boundary Devices.
Task-number: QTBUG-56769
Change-Id: I9e0f291be7e00df3d9d1b98c7414dfd37e265b39
Reviewed-by: Kari Hormi <kari.hormi@qt.io>
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
QTCREATORBUG-17190 is fixed by this since initialization is fixed as a side effect;-)
Task-number: QTCREATORBUG-17190
Change-Id: I9d1f81d6cca26fcdd9ba8a49ee66e05607d6cae1
Reviewed-by: Orgad Shaneh <orgads@gmail.com>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Move the instructions for changing the default colors
to the end and add a section title.
Change-Id: I837e4d25a686d1bf7e9288a438f6321a9ae9d7ed
Reviewed-by: Jarek Kobus <jaroslaw.kobus@qt.io>
You can now view unsaved changes in the current file
or all open files.
Added a subtitle and moved the image to the new section.
Change-Id: I8135dc7c9e2edc60b88d84a241f6d959ba590464
Reviewed-by: Jarek Kobus <jaroslaw.kobus@qt.io>
Update screenshot and move numbers around, because the time
ruler was not described before.
Change-Id: I6e13269349b847249ff6b1a9d9c8062ed55bc7f6
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
This is apparently necessary to establish policy before setting up
things.
Task-number: QTCREATORBUG-17197
Change-Id: I8b12e752e05e65b2275c0a97cf1b727f60c88cd0
Reviewed-by: Tim Jenssen <tim.jenssen@qt.io>
Otherwise C/C++ with the same ABI are detected as duplicates.
Change-Id: I99090b4cc776047b1ecb958242d0b4f5ecec5f0a
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
To make sure CDB is really installed on a Windows Kit, we need to assure the
'inc' subdirectory is present. Before this patch, if, for instance, the
Windows 8.1 kit is present (including a Debuggers subdirectory), but without
CDB, the first test under "32-bit" qmake will yield a false positive,
preventing the following 32-bit kits to be tested, then to fail under the
64-bit qmake test because the 'inc' directory is inexistent, finally
bringing about an empty CDB_PATH variable.
Change-Id: I1d7cf3350f4a63a556ef106faa5b3f0593dcf8c8
Reviewed-by: Andrew Knight <andrew.knight@intopalo.com>
Reviewed-by: Brett Stottlemyer <bstottle@ford.com>
Reviewed-by: David Schulz <david.schulz@qt.io>
Inside a Connections item JavaScript blocks should be allowed.
Change-Id: Ia6a08fc575a72980bd53a32249bf302b7a426266
Reviewed-by: Marco Benelli <marco.benelli@qt.io>
When rescanning imports we generally get a new version of the parent
document. When replacing the cached documents we want to release the
old ones, not the new ones. Failing to do so leads to types
disappearing from the QML code model and to leaking memory.
Task-number: QTCREATORBUG-17175
Change-Id: I3994444ac0a6cd87f9d9d0b47ab3d6015660e416
Reviewed-by: Thomas Hartmann <Thomas.Hartmann@theqtcompany.com>
We don't know which protocol QtCreator will choose to connect to the
QML debug server. Currently QtCreator uses whatever "localhost"
resolves to.
Change-Id: Id41fb54e5eb975581d382767bdd125fbb9801f4f
Task-number: QTCREATORBUG-17141
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Apparently a failed connection counts as "pending" for a while before
it is finally removed. The QML profiler aggressively tries to connect
with different timeouts and multiple sockets, so quite often there is
such a failed connection which then prevents new connections from
getting established.
We already enforce the limit of only one successful connection via the
SingleRelayServer subclass.
Change-Id: I441e2ed898d28f04359a9976d5b25569b8447f3f
Task-number: QTCREATORBUG-17141
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
You can now visualize memory consumption and allocations
of functions in addition to their time consumption.
Change-Id: I1c36019a5e3a161332a956a4b2d4aa833ea781dc
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
The index was handled wrongly.
Change-Id: I8a98c7a16e32798a9b7662c6c5c1683d248580da
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
Reviewed-by: Tim Jenssen <tim.jenssen@qt.io>
This is where the "Waiting for connection ..." message usually appears.
In general this mechanism of first replacing the port in the message
and then parsing the message to figure out that the iostool (not the
application) is actually listening is very confusing and backwards,
but fixing this is material for the master branch.
Change-Id: I73d3c5a34482403d275e6ea7a9ad59996121b02e
Task-number: QTCREATORBUG-17141
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Editing subcomponents should only be possible for
graphical items.
Qt Creator was crashing in case the node was something else.
Change-Id: Ia0e286c666b0b485da0817678ca2048ace260c9b
Reviewed-by: Tim Jenssen <tim.jenssen@qt.io>
This should fix kits complaining about wrongly set C compilers and
ABI incompatibility between C and C++ compilers.
Task-number: QTCREATORBUG-17165
Task-number: QTCREATORBUG-17166
Change-Id: Ia002490b471e0f5306c3a76b27158869920452ed
Reviewed-by: BogDan Vatra <bogdan@kdab.com>
Nothing ever fills that column, so do not bother to show it.
Change-Id: I5787aa72ee5c414fbf01ca01eed3aea20b6b7201
Reviewed-by: Tim Jenssen <tim.jenssen@qt.io>
