qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Gerd Hoffmann <kraxel@redhat.com>
To: qemu-devel@nongnu.org
Cc: libvir-list@redhat.com, Peter Maydell <peter.maydell@linaro.org>,
	Gerd Hoffmann <kraxel@redhat.com>,
	Akihiko Odaki <akihiko.odaki@gmail.com>
Subject: [PULL 11/11] ui/cocoa: Comment about modifier key input quirks
Date: Tue, 16 Mar 2021 06:38:13 +0100	[thread overview]
Message-ID: <20210316053813.1719442-12-kraxel@redhat.com> (raw)
In-Reply-To: <20210316053813.1719442-1-kraxel@redhat.com>

From: Akihiko Odaki <akihiko.odaki@gmail.com>

Based-on: <20210310042348.21931-1-akihiko.odaki@gmail.com>
Signed-off-by: Akihiko Odaki <akihiko.odaki@gmail.com>
Message-Id: <20210312133212.3131-1-akihiko.odaki@gmail.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
 ui/cocoa.m | 38 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

diff --git a/ui/cocoa.m b/ui/cocoa.m
index 9da0e884b712..37e1fb52eb4d 100644
--- a/ui/cocoa.m
+++ b/ui/cocoa.m
@@ -690,7 +690,43 @@ QemuCocoaView *cocoaView;
     NSPoint p = [self screenLocationOfEvent:event];
     NSUInteger modifiers = [event modifierFlags];
 
-    // emulate caps lock keydown and keyup
+    /*
+     * Check -[NSEvent modifierFlags] here.
+     *
+     * There is a NSEventType for an event notifying the change of
+     * -[NSEvent modifierFlags], NSEventTypeFlagsChanged but these operations
+     * are performed for any events because a modifier state may change while
+     * the application is inactive (i.e. no events fire) and we don't want to
+     * wait for another modifier state change to detect such a change.
+     *
+     * NSEventModifierFlagCapsLock requires a special treatment. The other flags
+     * are handled in similar manners.
+     *
+     * NSEventModifierFlagCapsLock
+     * ---------------------------
+     *
+     * If CapsLock state is changed, "up" and "down" events will be fired in
+     * sequence, effectively updates CapsLock state on the guest.
+     *
+     * The other flags
+     * ---------------
+     *
+     * If a flag is not set, fire "up" events for all keys which correspond to
+     * the flag. Note that "down" events are not fired here because the flags
+     * checked here do not tell what exact keys are down.
+     *
+     * If one of the keys corresponding to a flag is down, we rely on
+     * -[NSEvent keyCode] of an event whose -[NSEvent type] is
+     * NSEventTypeFlagsChanged to know the exact key which is down, which has
+     * the following two downsides:
+     * - It does not work when the application is inactive as described above.
+     * - It malfactions *after* the modifier state is changed while the
+     *   application is inactive. It is because -[NSEvent keyCode] does not tell
+     *   if the key is up or down, and requires to infer the current state from
+     *   the previous state. It is still possible to fix such a malfanction by
+     *   completely leaving your hands from the keyboard, which hopefully makes
+     *   this implementation usable enough.
+     */
     if (!!(modifiers & NSEventModifierFlagCapsLock) !=
         qkbd_state_modifier_get(kbd, QKBD_MOD_CAPSLOCK)) {
         qkbd_state_key_event(kbd, Q_KEY_CODE_CAPS_LOCK, true);
-- 
2.29.2



  parent reply	other threads:[~2021-03-16  5:49 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-16  5:38 [PULL 00/11] Ui 20210316 patches Gerd Hoffmann
2021-03-16  5:38 ` [PULL 01/11] ui: introduce "password-secret" option for VNC servers Gerd Hoffmann
2021-03-16  5:38 ` [PULL 02/11] ui: introduce "password-secret" option for SPICE server Gerd Hoffmann
2021-03-16  5:38 ` [PULL 03/11] ui: deprecate "password" " Gerd Hoffmann
2021-03-16  5:38 ` [PULL 04/11] opengl: Do not convert format with glTexImage2D on OpenGL ES Gerd Hoffmann
2021-03-16  5:38 ` [PULL 05/11] ui/cocoa: Do not exit immediately after shutdown Gerd Hoffmann
2021-03-16  5:38 ` [PULL 06/11] ui: add more trace points for VNC client/server messages Gerd Hoffmann
2021-03-16  5:38 ` [PULL 07/11] ui: avoid sending framebuffer updates outside client desktop bounds Gerd Hoffmann
2021-03-16  5:38 ` [PULL 08/11] ui: use client width/height in WMVi message Gerd Hoffmann
2021-03-16  5:38 ` [PULL 09/11] ui: honour the actual guest display dimensions without rounding Gerd Hoffmann
2021-03-16  5:38 ` [PULL 10/11] ui: fold qemu_alloc_display in only caller Gerd Hoffmann
2021-03-16  5:38 ` Gerd Hoffmann [this message]
2021-03-17 14:59 ` [PULL 00/11] Ui 20210316 patches Peter Maydell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210316053813.1719442-12-kraxel@redhat.com \
    --to=kraxel@redhat.com \
    --cc=akihiko.odaki@gmail.com \
    --cc=libvir-list@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).