This bug was introduced by my commit
002aab309055b7feeeeeb25c1835b6abe6646729
"fdo#46808, convert sdb::tools::ConnectionTools to new style"
Change-Id: Ib10f97bbe46d20db663047d2f35ec4850df839fa
This reverts commit 6c61b20a8d4a6dcac28801cde82a211fb7e30654. As discussed at
<http://lists.freedesktop.org/archives/libreoffice/2013-May/052449.html> "Re:
fdo#46808, Convert awt::UnoControlDialogModel to new style problem" why the odd
change in 2e2a4827ce6708f0e8677dba9cc92e1479a44086 "scripting: get
CreateUnoDialog() work again" appears to fix things again:
The problem is that the implementation of the css.awt.UnoControlDialogModel
involves UNO aggregation
(IMPL_CREATE_INSTANCE_WITH_GEOMETRY(UnoControlDialogModel) in
toolkit/soruce/helper/registerservices.cxx creating a
OGeometryControlModel<UnoControlDialogModel> instance that aggregates a
UnoControlDialogModel instance). That means that queryInterface can return a
reference to something that is technically a different object, and that's
what's happening here, and explains why calling setPropertyValue in two
different ways on what logically appears to be a single object can end up
calling two different implementations (of two different physical objects).
(UNO aggregation is known to be broken and should not be used. Nevertheless,
there's still code that does---code that is a horrible mess and hard to clean
up.)
That all this worked as intended in the past is just sheer luck, but any
way of substantially touching it is asking for trouble. I'm going to
revert 6c61b20a8d4a6dcac28801cde82a211fb7e30654 again.
I wasn't able to revert without also reverting
be50ad28f5bbdaeff527f646481ce263843c2401 "fdo#46808, Convert
awt::XUnoControlDialog to new style," as the two were tightly dependant. Also
reverts all the follow-up fixes cb4b6dde8fda2a5848e11063028bf44d72f85431
"-Werror,-Wuninitialized" (sans the const-ness fix in
UpdateHandler::insertControlModel), 697a007c61b9cabceb9767fad87cd5822b300452
"Fix exception specifications," 2ce6828bbbf6ba181bb2276adeec279e74151ef6 "fix
awt::UnoControlModelDialog crash," and 2e2a4827ce6708f0e8677dba9cc92e1479a44086
"scripting: get CreateUnoDialog() work again."
Conflicts:
basctl/source/dlged/dlged.cxx
filter/source/t602/t602filter.cxx
xmlscript/test/imexp.cxx
Change-Id: I5d133468062f3ca36300db52fbd699be1ac72998
To be honest, I do not userstand the code much. Julien pointed out that
the setParentTitle and onTypeSelected were newer called. The condition
looked like an optimisation and it helped to remove it.
It would be great to put it back. It might be enough to hand the
intial values.
Change-Id: Id06cc9c63a9e578cb3c698c8526851fec71ef9da
and pass it to the parser and PredicateInput constructors.
This makes the whole story consistent; before system locale settings were already manually passed to parseNodeToPredicateStr, which led to some things being parsed as en_US and others as system locale.
Change-Id: Ib9571b10d79183571e8ab3f79660b41594dc2d1c
...so move it from osl/mutex.hxx to its own comphelper/solarmutex.hxx. It looks
like a newbie mistake that 59e7685d8d812ee8773f57475cbe3aa2a0bdfc81 "Create an
abstract interface to be used to implement a SolarMutex" put it here in the
first place.
I do not consider this an incompatible change really, as no external URE client
code should have used SolarMutex anyway.
(Also included some clean up, like removing unused
{Clearable,Resettable}SolarGuard, and spelling out SolarGuard in the few places
it is used.)
Change-Id: I121ffb5b7cefbc19e88b5405e5a85ffc895be852
Let foo.odb be a database file that has a macro that connects to the
Database on "Open Document" event (and needs to prompt user for
user/password).
There was a race condition between two actions:
1) the asynchronous treatment of "OnFirstControllerConnected" in dbaui::OApplicationController,
which tries to get dbaui::OApplicationController's mutex
2) the StarBasic macro calling dbaui::OApplicationController::connect
which needs to display a dialog (to get username and password),
and thus puts that dialog in the main thread's event queue
and waits for it ... with dbaui::OApplicationController's mutex held
Now, if "1)" is before "2)" in the event queue of the the main thread,
*but* "1)" is executed *after* "2)" has taken the lock, there is a deadlock.
Fix:
1) Make OnFirstControllerConnected synchronous.
Make sure (by taking mutex in dbaui::OApplicationController::attachFrame, its ancestor in the call graph)
that nothing else will happen with the OApplicationController as long as it is not finished.
---> it does not need to take mutex itself anymore
This avoids the "order in the asynchronous events" dependency.
2) Change dbaui::OApplicationController::ensureConnection to do the user prompting
WITHOUT HOLDING the mutex, and use the mutex "only" to protect actually assigning
the connection to m_xDataSourceConnection.
Theoretically, in some race condition, we could connect twice and then discard one connection <shrug>.
ensureConnection will never return the discarded connection, though.
(I think I got that right with respect to http://www.cs.umd.edu/~pugh/java/memoryModel/DoubleCheckedLocking.html)
This keeps it from locking on another condition while holding the mutex.
Change-Id: Iab1bbec5d5df12bb89d027d43e498c78c92ffc32
Reviewed-on: https://gerrit.libreoffice.org/3310
Reviewed-by: David Tardon <dtardon@redhat.com>
Tested-by: David Tardon <dtardon@redhat.com>
Just use SAL_WARN, building up the message using chained << operators instead
of constructing it into a OStringBuffer (or OUStringBuffer) and then passing
that to OSL_FAIL.
Or should we have a real assert() here?
Change-Id: I2c3f64ff2b29a3d81ba444d98c2a38ef49e6d7af
