... Format arguments are substituted into the translation of the msgid, which
may not be known at the time the format arguments are captured (because locale
may change). This allows TranslatableString with arguments to be constructed
at static initialization time.
There is also a special "verbatim" or null context which makes no translations
of msgids.
There is not yet any use of other contexts besides default or null.
Problem:
NumericTextCtrls act on numeric keys, even when modifier keys are pressed.
This shouldn't be the case, and has the knock on effect that shortcuts like ctrl + 1, don't work when a NumericTextCtrl is the focus.
Fix:
Check that there are no modifier keys pressed.
... Just one low-level function for each that simply sets; just one high-level
function for each that also pushes notifications; and to make that work,
toolbars need some rewriting to avoid recursion when the notifications they
post come back to them.
... and break its compile dependency on CommandManager.h by letting it install
callbacks.
This also removes Objective-C mixed code from CommmandManager.
This also eliminates four inclusions of Project.h!
Capture handler state is also global, not per project, though the
CommandManager's callbacks still do depend on the active project.
... Unnecessary because transitively included.
But each .cpp file still includes its own .h file near the top to ensure
that it compiles indenendently, even if it is reincluded transitively later.
Problem:
Using the Narrator screen reader on Windows 10, there are a couple of problems:
1. When using the left/right arrow keys, Narrator reads the current field, rather than the digit which is now the focus.
2. Using up/down arrow keys, Narrator is silent.
Fixes:
1. The existing code assumes that NumericTextCtrlAx::GetName() is called only once after left/right arrow is pressed. However, Narrator causes this function to be called more than once. Solution: handle the case where the function is called, and neither the focus or the digits have changed, and use a cached value of the name.
2. If the focus has not changed, then after a focus event, Narrator does not read the name, even if the name has changed. Solution: add a name change event. (The focus event has been retained to keep Window-Eyes happy until we stop supporting it.)
Note:
One of the focus events has been removed from NumericTextCtrl::SetFieldFocus(), as it no longer appears to be necessary.
... which will make it easier to change the types of those containers to
std::vectors of other string-like classes
for wxString,
IsEmpty => empty
Clear => clear
Alloc => reserve
for wxArrayString,
Count => size
GetCount => size
IsEmpty => empty
Add => push_back
Clear => clear
Empty => clear
Sort => std::sort (only with default comparator)
SetCount => resize
Last => back
Item => operator []
Alloc => reserve
All backing bitmaps (not just the one in track panel) now are set to 24 bits.
Big thanks to David Bailes for tracking down the root cause of slow BitBlts, and the comments at https://trac.wxwidgets.org/ticket/14403 which led to the fix.
In testing the bitmaps, I also fixed the sizer errors reported for Export.cpp that wxWidgets now reports as ASSERTS when running in debug builds.
Modification to commit 7ab4b8c. The first non-zero digit only becomes the focus on the first visit to the control. On subsequent visits, the focus is the previous focus. Example use case: modifying start and end times by a tenth of a second. The user can move between the controls without loosing his "place".
The main change in wx accessibility is this:
7dab555f71 (diff-04f5191d86f95b1c4d5d9c979da65878)
However wxWindowAccessible has not been updated to take into account of that change. In particular wxWindowAccessible::GetParent() was always wrong, but it was consistent with the rest of the framework. Now it's wrong and inconsistent. This function should return an object with role window, and which has the same name.
The fix is to introduce class WindowAccessible, which is effectively our own version of wxWindowAccessible. This class does not override GetParent(), and so just relies on a standard accessible object to to the right thing in wxIAccessible::get_accParent() (which is does). This class also allows us to have our own version of GetName(), which allows us to set the accessibility names of buttons.
These changes will break the accessibility of Audacity if it is built with wxWidgets 3.0.X. If this is a problem, then there could be some #if stuff in WindowAccessible.h to turn the WindowAccessible class into one which simply inherits from wxWindowAccessible, and doesn't override anything.
Defaulted move constructors and assignments introduced at 080dd34 are not
essential but might give a bit more efficiency in std::vector on non-Windows.
Previous compilation fix was incorrect! Excpliticly defined move assignments
should not have had no-effect bodies!
Problem: On Windows, if you set the "length" radio button in the selection bar, then tab round to it, this sets the "end" radio button.
This was caused by commit 7e49dc4.
The fix is to restore the previous code just for Windows.
... A non-narrowing conversion out to long long is a necessity, but the
conversions to float and double are simply conveniences.
Conversion from floating is explicit, to avoid unintended consequences with
arithmetic operators, when later sampleCount ceases to be an alias for an
integral type.
Some conversions are not made explicit, where I expect to change the type of
the variable later to have mere size_t width.
Caused by optimisation of NumericTextCtrl::ValueToControls(). It now only updates on a change, and because mValueStr was being updated without calling value to controls, the string was not being identified as having changed.
... I believe this list of four places is exhaustive.
There are many, many more safe narrowings that I examined.
This resulted from changing the definition of sampleCount in my builds so that
narrowing conversions failed to compile without some fixes, and I examined and
fixed every place.
The rest of that work is not yet shared.
