WM_CLOSE was not processed at all for native windows wrapped by
wxNativeContainerWindow because we don't handle it ourselves at wxWindow level
but still mark it as processed in order to prevent DefWindowProc() from
destroying the window. Unfortunately this also prevented the original handler
for this message in the native window from being called.
Calling just the original handler and not the wxWidgets one is not ideal
neither but is much better as it allows to e.g. close MFC frames wrapped in
wxNativeContainerWindow whereas before this didn't work at all as WM_CLOSE was
completely ignored.
Also call the original handler for WM_DESTROY to avoid similar potential
problems with this message, even if it doesn't seem to create any with MFC.
Extend the mfc sample to show how a wxPanel can be embedded into the existing
CFrameWnd.
Comparing wchar_t with int doesn't compile under OpenVMS because of
%CXX-E-AMBIGUOUSOPRFUN, more than one operator ">=" matches these operands:
built-in operator "arithmetic >= arithmetic"
function "operator>=(wchar_t, const wxUniChar &)"
operand types are: wchar_t >= enum <unnamed>
Try to work around this by explicitly casting wchar_t to int.
Don't use wxGROW and wxALIGN_CENTRE together in a wxBoxSizer, this doesn't
make sense as the latter is just ignored.
Also use wxEXPAND instead of wxGROW for consistency with the flags elsewhere
in the same file.
Closes#17063.
Horizontal alignment of bitmap associated with wxPGChoice item drawn in wxPGComboBox (in wxPropertyGrid::OnComboItemPaint) needs to be adjusted by 1 pixel (as it is already done for text label) in order to get the bitmap in the same position when wxPGComboBox is selected.
The number of frames between the code containing the assert and the code
generating the stack trace is not the same under different platforms and so
hardcoding 8 for it in wxAppTraitsBase::GetAssertStackTrace() worked for wxMSW
but not e.g. wxGTK.
Instead, just ignore all frames up to and including the one for wxOnAssert()
itself. This makes the code work correctly on all platforms and it also won't
need to be modified whenever any extra functions are added/removed
(wxGTK-specific code in utilsgtk.cpp used wrong number of frames too, even
though it was presumably correct once before).
UTF8FromUTF16() now only NUL-terminates the string if there is enough space in
it for the trailing NUL, so pass the correct length of the buffer, including
the last byte reserved for this NUL to this function.
Also allocate one byte less in wxCharBuffer, it was adding 1 extra byte
unnecessarily.
See #16776.
Non-blocking sockets can't work in worker threads without additional locking
as they generate events that can be dispatched from the main thread after the
socket object, created in the worker thread, is already destroyed, so don't
even attempt to use them if wxProtocol object is created from non-main thread.
Also simplify the code by removing the calls to SetFlags(), Notify() and
{Save,Restore}State() and simply put the socket from the beginning in
blocking, wait all mode that it needs to be in.
This, with the fixes in the previous commit, allows wxHTTP and wxFTP to work
from worker threads too.
Test using wxHTTP from a worker thread in the socket client sample.
Closes#17031.
Events are not needed for this kind of sockets, using wxSOCKET_BLOCK is
supposed to ensure that calling socket IO operations blocks until the bytes
are read/written without dispatching any events.
See #17031.
Implement Freeze() by blocking the GtkWindow "expose-event"/"draw" signal
instead. Since the introduction of client-side windows in GTK+ 2.18,
gdk_window_freeze_updates() is unuseable because the impl_window (and thus the
update_freeze_count) for a given GdkWindow can change unpredictably. See #16795
Calling SetMinSize() on wxMDIChildFrame in wxMSW prevented SetMaxSize() from
working as wxMDIChildFrame::HandleGetMinMaxInfo() didn't take the max size
into account and prevented the execution of the base class version of the same
method from taking place if min size was set.
Fix this simply by always delegating to the base class version after using
DefMDIChildProc() to compute the default max size values.
Closes#17029.
Definitions of non-standard LLONG_MAX, LLONG_MIN and ULLONG_MAX constants are
excluded from MinGW limits.h when compiling in "strict ANSI" mode, so define
them ourselves.
Don't ignore errors (this resulted in warnings in optimized builds because
variable "hr" containing the error code to was assigned but never used) and
don't leak memory in the (admittedly unlikely) case an error really occurs.
Also don't duplicate the code for creating a one element SAFEARRAY<VARIANT>,
extract it into a helper function.
Don't cast NMHDR::code field to int, this cast was added in the previous
millennium, probably to work around the wrong definition of NM_DBLCLK in some
ancient MinGW headers, but nowadays all MinGW distributions (tested with
MinGW 4.8.1, MinGW-64 4.9.1 and TDM-GCC 4.9.2) define it correctly and so
using this cast results in a warning -- just remove it to get rid of it.
These options enable "strict ANSI" mode in MinGW which omits declarations of
POSIX functions from the standard headers. To allow the library and, possibly
even more importantly, the user code including our headers, to compile with
these options, declare the functions that we need ourselves.
This might appear to go against the spirit of "strict ANSI" mode, but the only
alternative would be to not use such functions at all and silently cripple the
library when -std=c++NN is used, compared to -std=g++NN case, and this doesn't
seem appealing neither.
Closes#16984.
Fixes the bug which could be observed if string containing more then one flag entry was provided to wxPGProperty::SetFlagsFromString. In this case property flags were not set properly (were not set all) due to the error in splitting the string into tokens (whole string instead of its individual token was taken to compare).
Lookup table used to map between flag values and their names is redesigned in order to reduce the size and to make indexing simpler. There are no longer NULL entries in the table and every flag name is identified explicitly by the corresponding flag value and not by the index of the name entry in the table. Thanks to this the relation between flag value and flag name is more clear and the loops in wxPGProperty::GetFlagsAsString/SetFlagsAsString can be simplified.
