Fixes#7632.
If Caps Lock was on, rules that did not specify Caps Lock state were
not matched, for example:
```
+ [K_SLASH] > 'foo'
```
The truth table for matches is:
Event | Rule | Result
---------|-------------|-----------
NCAPS | 0 | match
NCAPS | CAPS | no match
NCAPS | NCAPS | match
CAPS | 0 | match
CAPS | CAPS | match
CAPS | NCAPS | no match
(Internally, Keyman for Mac currently maps 0 to NCAPS on the event for
simplicity.)
The fix is to ignore the Event Caps Lock state if the rule specifies
neither `CAPS` nor `NCAPS`. If the rule specifies either of these
state masks, then the Caps component of the state mask must match.
Fixes#6364.
Removes `&layerChanged` and adds `&newLayer` and `&oldLayer` system
stores with the additional nuances around values as described in #6364.
Fixes#6168.
This does not any support for changing any system store values, but does
ensure that the statement will be correctly handled when emitting
text.
Relates to #3621.
Add support for `&layerChanged` system store. This store is set to `1`
before a `begin postKeystroke`, if the keystroke it follows resulted in
a layer change, either programatically through a keyboard rule, or
through a `nextlayer` property of the touched key.
Fixes#5316.
Keyman for macOS did not recognise the updated version number 0x0E00,
which is version 14.0. There are no changes to the file format which
impact Keyman for macOS -- the only change is support for notany and
context for Keyman Engine for Web.
Fixes#2601.
This PR adds support for the 102nd key as found on European layouts. It
does not however reshape the Enter key and move the BKSLASH key, as that
is a much bigger job; functionally this makes it possible to press the
102nd key, which was the biggest gap.
I wasn't going to do this, but realised that crash reporting was going
to be a problem on mac because it still used crashlytics/fabric. Turned
out to be a good exercise to get me back into macOS world, and fairly
straightforward to get right (helps that the other platforms broke the
ice for me).
Fixes#2375.
For applications that do not provide context to Keyman, Keyman was losing
the context buffer whenever a modifier key was pressed or released. This
change modifies the cache reset to occur only when the Command modifier is
pressed.
I attempted unsuccessfully to find an actual failing case so I could add a unit test for this. Even though I couldn't, I agree that it is the correct logic, so I'm making the change as Marc suggested.
