This enables VoiceOver user to see the document view as a single
element and interact with it (i.e. inspect its subelements) only when
needed by using the VO-Shift-down shortcut (and stop interacting with
it using VO-Shift-up shortcut). This makes the element structure more
hierarchical where at the top level are only major user interface
components, so that it's easier for orientation. It also prevents
unrelated UI elements to be sequential as VoiceOver by default orders
elements by the visual order (top-down, left-right) rather then
the order in the AXChildren array. For example in this case,
the bottom status bar of the document window gets separated from
the bottom toolbar of the file browser, unlike the situation before
grouping when both bars were adjacent to the VoiceOver user.
This fixes semantics of the button and thus also its
accessibility. The macro recording button represents state - more
specifically a boolean state, so it has to be a NSToggleButton (or
its specialization). Then the button is presented to the VoiceOver
user as an AXCheckbox - that is a checkbox with on and off value -
which is exactly what the button's function is. Examples of this in
system apps is e.g. the "Play" button in QuickTime Player, which is
also an AXCheckbox ("Pause" is then simply an unchecked "Play"
checkbox).
VoiceOver reads an AXValue for a pop-up button - that is the selected
item. But button's title does not set an AXValue, as it probably is
not a menu item. So change the code set to the title by setting a
single menu item with that title.
This fixes accessibility of the pop-up when accessing its value before
it is first popped up. It makes the code consistent with setGrammarName:
which works fine (that's how I discovered how to fix it).
When enabled the height of the document is increased by the height of the view port, which means the last lines of the document are not anchored to the bottom of the view port but can e.g. be centered.
Closes#513.
We now try to find a scrollable view under the mouse, and if any scrolling was done, we reject the gesture as a “go back/forward” event.
Additionally we reject gestures with a duration of more than 0.2 seconds.
Maybe fixes#854.
Ideally we would be able to remove arbitrary callbacks rather than the last one added, but for this we would need to introduce some sort of handle (the callback itself is a block so not suitable as handle). The push/pop API for callbacks is sufficient for our use-case though.
This is so that tests can use the exact same executable to create the repository, as is used to obtain status (svn repositories created with newer executables are often not backwards compatible).
This was previously public so that we could write tests for it, but since the test runner is linked with the object files, it can access symbols that has visibility set to hidden.
If we closed a window with a document SCM callback for the current document, but that document was untitled (so it wouldn’t be cleared during shutdown), we would keep ‘self’ retained.
Normally we enable these on application launch when we receive an ‘application did become active’ notification, for tests though, this notification is lacking, but we do want it enabled for tests, and while we could do a constructor in the tests to enable it, it seems better to just enable it by default.
One issue is that if TextMate is launched “in the background” then it will still do SCM status gathering. Though long-term I may actually want to always have it enabled, even when TextMate is not active.
By using the menu item’s indentation level property it is possible to “type to select” items, which previously would not work for items that had leading em-space as a prefix.
Closes#806.
