2 tests fail with cvs HEAD. The second one looks reasonably important.
==== bind-13.13 Tk_BindEvent procedure: invalid key detail FAILED
==== Contents of test case:
setup
bind .b.f <Key> "lappend x Key%K"
bind .b.f <KeyRelease> "lappend x Release%K"
set x {}
event gen .b.f <Key> -keycode 0
event gen .b.f <KeyRelease> -keycode 0
set x
---- Result was:
Keya Releasea
---- Result should have been (exact matching):
Key?? Release??
==== bind-13.13 FAILED
==== bind-15.7 MatchPatterns procedure, ignoring type mismatches
FAILED
==== Contents of test case:
setup
bind .b.f <Double-1> {set x 1}
set x 0
event gen .b.f <Button-1>
event gen .b.f <Key-Shift_L>
event gen .b.f <ButtonRelease-1>
event gen .b.f <Button-1>
event gen .b.f <ButtonRelease-1>
set x
---- Result was:
0
---- Result should have been (exact matching):
1
==== bind-15.7 FAILED
Logged In: YES
user_id=143885
About the first test (bind-13.13): This tests that keycode
0 is invalid as required on X11. But in Aqua, 0 is actually
valid and the test result is correct. The test should use
some other value instead, e.g. -1, which shoulkd be invalid
on all supported platforms.