[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: ncurses-5.9-20111022.patch.gz
From: |
Thomas Dickey |
Subject: |
Re: ncurses-5.9-20111022.patch.gz |
Date: |
Sun, 23 Oct 2011 11:11:21 -0400 (EDT) |
On Sun, 23 Oct 2011, Damien Guibouret wrote:
Hello,
I take a look at this last patch and I have a strange feeling on _mouse_mask2
use: in get_mouse and _nc_mouse_parse functions, everywhere you used
_mouse_mask2 I would had kept _mouse_mask and everywhere you kept
I made a copy, because _mouse_mask is returned to the caller in the
mousemask() function - the last parmeter. The caller should only get back
what it provided in a previous call.
(I noticed the problem a week ago while testing dialog - essentially just
run the password script in its samples directory, and click on the
text-field while doing tabs to switch the button focus - some of the
clicks were lost).
_mouse_mask, I would had used _mouse_mask2 (the temporary set mask shall be
used for the click/doubleclick/tripleclick merge but not to return event to
user) or I misunderstand something.
And the
} else if (!ValidEvent(next)) {
continue;
maybe - in debugging I did add that, but it may be redundant.
you add at line 1254 is theorically not needed (the first invalid event that
is encountered in loop is eventp and it is treated by previous condition that
will end loop at end of this iteration) (or if you encounter a case where
next is invalid, it is the previous loop that needs to be fixed).
That gives something as attached file.
Regards,
Damien
--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net