kwin/xkb.cpp

507 lines
16 KiB
C++
Raw Normal View History

/********************************************************************
KWin - the KDE window manager
This file is part of the KDE project.
Copyright (C) 2013, 2016, 2017 Martin Gräßlin <mgraesslin@kde.org>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*********************************************************************/
#include "xkb.h"
#include "keyboard_input.h"
#include "utils.h"
#include "wayland_server.h"
// frameworks
#include <KConfigGroup>
#include <KKeyServer>
// KWayland
#include <KWayland/Server/seat_interface.h>
// Qt
#include <QTemporaryFile>
// xkbcommon
#include <xkbcommon/xkbcommon.h>
#include <xkbcommon/xkbcommon-compose.h>
#include <xkbcommon/xkbcommon-keysyms.h>
// system
#include <sys/mman.h>
#include <unistd.h>
Q_LOGGING_CATEGORY(KWIN_XKB, "kwin_xkbcommon", QtCriticalMsg)
namespace KWin
{
static void xkbLogHandler(xkb_context *context, xkb_log_level priority, const char *format, va_list args)
{
Q_UNUSED(context)
char buf[1024];
if (std::vsnprintf(buf, 1023, format, args) <= 0) {
return;
}
switch (priority) {
case XKB_LOG_LEVEL_DEBUG:
qCDebug(KWIN_XKB) << "XKB:" << buf;
break;
case XKB_LOG_LEVEL_INFO:
qCInfo(KWIN_XKB) << "XKB:" << buf;
break;
case XKB_LOG_LEVEL_WARNING:
qCWarning(KWIN_XKB) << "XKB:" << buf;
break;
case XKB_LOG_LEVEL_ERROR:
case XKB_LOG_LEVEL_CRITICAL:
default:
qCCritical(KWIN_XKB) << "XKB:" << buf;
break;
}
}
Xkb::Xkb(InputRedirection *input)
: m_input(input)
, m_context(xkb_context_new(static_cast<xkb_context_flags>(0)))
, m_keymap(NULL)
, m_state(NULL)
, m_shiftModifier(0)
, m_capsModifier(0)
, m_controlModifier(0)
, m_altModifier(0)
, m_metaModifier(0)
, m_numLock(0)
, m_capsLock(0)
, m_scrollLock(0)
, m_modifiers(Qt::NoModifier)
, m_consumedModifiers(Qt::NoModifier)
, m_keysym(XKB_KEY_NoSymbol)
, m_leds()
{
qRegisterMetaType<KWin::Xkb::LEDs>();
if (!m_context) {
qCDebug(KWIN_XKB) << "Could not create xkb context";
} else {
xkb_context_set_log_level(m_context, XKB_LOG_LEVEL_DEBUG);
xkb_context_set_log_fn(m_context, &xkbLogHandler);
// get locale as described in xkbcommon doc
// cannot use QLocale as it drops the modifier part
QByteArray locale = qgetenv("LC_ALL");
if (locale.isEmpty()) {
locale = qgetenv("LC_CTYPE");
}
if (locale.isEmpty()) {
locale = qgetenv("LANG");
}
if (locale.isEmpty()) {
locale = QByteArrayLiteral("C");
}
m_compose.table = xkb_compose_table_new_from_locale(m_context, locale.constData(), XKB_COMPOSE_COMPILE_NO_FLAGS);
if (m_compose.table) {
m_compose.state = xkb_compose_state_new(m_compose.table, XKB_COMPOSE_STATE_NO_FLAGS);
}
}
}
Xkb::~Xkb()
{
xkb_compose_state_unref(m_compose.state);
xkb_compose_table_unref(m_compose.table);
xkb_state_unref(m_state);
xkb_keymap_unref(m_keymap);
xkb_context_unref(m_context);
}
void Xkb::reconfigure()
{
if (!m_context) {
return;
}
xkb_keymap *keymap = nullptr;
if (!qEnvironmentVariableIsSet("KWIN_XKB_DEFAULT_KEYMAP")) {
keymap = loadKeymapFromConfig();
}
if (!keymap) {
qCDebug(KWIN_XKB) << "Could not create xkb keymap from configuration";
keymap = loadDefaultKeymap();
}
if (keymap) {
updateKeymap(keymap);
} else {
qCDebug(KWIN_XKB) << "Could not create default xkb keymap";
}
}
xkb_keymap *Xkb::loadKeymapFromConfig()
{
// load config
if (!m_config) {
return nullptr;
}
const KConfigGroup config = m_config->group("Layout");
const QByteArray model = config.readEntry("Model", "pc104").toLocal8Bit();
const QByteArray layout = config.readEntry("LayoutList", "").toLocal8Bit();
const QByteArray options = config.readEntry("Options", "").toLocal8Bit();
xkb_rule_names ruleNames = {
.rules = nullptr,
.model = model.constData(),
.layout = layout.constData(),
.variant = nullptr,
.options = options.constData()
};
return xkb_keymap_new_from_names(m_context, &ruleNames, XKB_KEYMAP_COMPILE_NO_FLAGS);
}
xkb_keymap *Xkb::loadDefaultKeymap()
{
return xkb_keymap_new_from_names(m_context, nullptr, XKB_KEYMAP_COMPILE_NO_FLAGS);
}
void Xkb::installKeymap(int fd, uint32_t size)
{
if (!m_context) {
return;
}
char *map = reinterpret_cast<char*>(mmap(NULL, size, PROT_READ, MAP_SHARED, fd, 0));
if (map == MAP_FAILED) {
return;
}
xkb_keymap *keymap = xkb_keymap_new_from_string(m_context, map, XKB_KEYMAP_FORMAT_TEXT_V1, XKB_MAP_COMPILE_PLACEHOLDER);
munmap(map, size);
if (!keymap) {
qCDebug(KWIN_XKB) << "Could not map keymap from file";
return;
}
updateKeymap(keymap);
}
void Xkb::updateKeymap(xkb_keymap *keymap)
{
Q_ASSERT(keymap);
xkb_state *state = xkb_state_new(keymap);
if (!state) {
qCDebug(KWIN_XKB) << "Could not create XKB state";
xkb_keymap_unref(keymap);
return;
}
// now release the old ones
xkb_state_unref(m_state);
xkb_keymap_unref(m_keymap);
m_keymap = keymap;
m_state = state;
m_shiftModifier = xkb_keymap_mod_get_index(m_keymap, XKB_MOD_NAME_SHIFT);
m_capsModifier = xkb_keymap_mod_get_index(m_keymap, XKB_MOD_NAME_CAPS);
m_controlModifier = xkb_keymap_mod_get_index(m_keymap, XKB_MOD_NAME_CTRL);
m_altModifier = xkb_keymap_mod_get_index(m_keymap, XKB_MOD_NAME_ALT);
m_metaModifier = xkb_keymap_mod_get_index(m_keymap, XKB_MOD_NAME_LOGO);
m_numLock = xkb_keymap_led_get_index(m_keymap, XKB_LED_NAME_NUM);
m_capsLock = xkb_keymap_led_get_index(m_keymap, XKB_LED_NAME_CAPS);
m_scrollLock = xkb_keymap_led_get_index(m_keymap, XKB_LED_NAME_SCROLL);
m_currentLayout = xkb_state_serialize_layout(m_state, XKB_STATE_LAYOUT_EFFECTIVE);
Handle modifier updates in the same sequence as Wayland does Summary: Consider the case that capslock gets pressed and released. In the case of Weston we have a sequence of: 1. Key press event 2. Modifier changed event 3. Key release event 4. Modifier changed event KWin however used to send the events in the following sequence: 1. Modifier changed event (on key press) 2. Key press event 3. Modifier changed event (on key release) 4. Key release event It looks like Xwayland is not able to properly process the sequence sent by KWin. And in fact KWin's sequence is wrong as it sends a state which does not match. We report that the caps lock is pressed in the modifiers prior to the application getting informed about the key press of caps lock. This change aligns KWin's implementation to the behavior of Weston. The main difference is that when modifiers change Xkb internally caches the serialized modifier states. And KeyboardInputRedirection just forwards the modifiers to KWayland::Server::SeatInterface once the processing has finished. SeatInterface ignores the forwarding if no states changes, so it is fine to do it that way. BUG: 377155 Test Plan: Not yet tested with an affected Xwayland as I only have 1.18 and the problem started with 1.19. But verified the sequence of events with WAYLAND_DEBUG and caps lock stil working in QtWayland clients and Xwayland 1.18 Reviewers: #kwin, #plasma Subscribers: plasma-devel, kwin Tags: #kwin Differential Revision: https://phabricator.kde.org/D5452
2017-04-14 22:01:20 +03:00
m_modifierState.depressed = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_DEPRESSED));
m_modifierState.latched = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_LATCHED));
m_modifierState.locked = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_LOCKED));
createKeymapFile();
Handle modifier updates in the same sequence as Wayland does Summary: Consider the case that capslock gets pressed and released. In the case of Weston we have a sequence of: 1. Key press event 2. Modifier changed event 3. Key release event 4. Modifier changed event KWin however used to send the events in the following sequence: 1. Modifier changed event (on key press) 2. Key press event 3. Modifier changed event (on key release) 4. Key release event It looks like Xwayland is not able to properly process the sequence sent by KWin. And in fact KWin's sequence is wrong as it sends a state which does not match. We report that the caps lock is pressed in the modifiers prior to the application getting informed about the key press of caps lock. This change aligns KWin's implementation to the behavior of Weston. The main difference is that when modifiers change Xkb internally caches the serialized modifier states. And KeyboardInputRedirection just forwards the modifiers to KWayland::Server::SeatInterface once the processing has finished. SeatInterface ignores the forwarding if no states changes, so it is fine to do it that way. BUG: 377155 Test Plan: Not yet tested with an affected Xwayland as I only have 1.18 and the problem started with 1.19. But verified the sequence of events with WAYLAND_DEBUG and caps lock stil working in QtWayland clients and Xwayland 1.18 Reviewers: #kwin, #plasma Subscribers: plasma-devel, kwin Tags: #kwin Differential Revision: https://phabricator.kde.org/D5452
2017-04-14 22:01:20 +03:00
forwardModifiers();
}
void Xkb::createKeymapFile()
{
if (!waylandServer()) {
return;
}
// TODO: uninstall keymap on server?
if (!m_keymap) {
return;
}
ScopedCPointer<char> keymapString(xkb_keymap_get_as_string(m_keymap, XKB_KEYMAP_FORMAT_TEXT_V1));
if (keymapString.isNull()) {
return;
}
const uint size = qstrlen(keymapString.data()) + 1;
QTemporaryFile *tmp = new QTemporaryFile(m_input);
if (!tmp->open()) {
delete tmp;
return;
}
unlink(tmp->fileName().toUtf8().constData());
if (!tmp->resize(size)) {
delete tmp;
return;
}
uchar *address = tmp->map(0, size);
if (!address) {
return;
}
if (qstrncpy(reinterpret_cast<char*>(address), keymapString.data(), size) == nullptr) {
delete tmp;
return;
}
waylandServer()->seat()->setKeymap(tmp->handle(), size);
}
void Xkb::updateModifiers(uint32_t modsDepressed, uint32_t modsLatched, uint32_t modsLocked, uint32_t group)
{
if (!m_keymap || !m_state) {
return;
}
xkb_state_update_mask(m_state, modsDepressed, modsLatched, modsLocked, 0, 0, group);
updateModifiers();
Handle modifier updates in the same sequence as Wayland does Summary: Consider the case that capslock gets pressed and released. In the case of Weston we have a sequence of: 1. Key press event 2. Modifier changed event 3. Key release event 4. Modifier changed event KWin however used to send the events in the following sequence: 1. Modifier changed event (on key press) 2. Key press event 3. Modifier changed event (on key release) 4. Key release event It looks like Xwayland is not able to properly process the sequence sent by KWin. And in fact KWin's sequence is wrong as it sends a state which does not match. We report that the caps lock is pressed in the modifiers prior to the application getting informed about the key press of caps lock. This change aligns KWin's implementation to the behavior of Weston. The main difference is that when modifiers change Xkb internally caches the serialized modifier states. And KeyboardInputRedirection just forwards the modifiers to KWayland::Server::SeatInterface once the processing has finished. SeatInterface ignores the forwarding if no states changes, so it is fine to do it that way. BUG: 377155 Test Plan: Not yet tested with an affected Xwayland as I only have 1.18 and the problem started with 1.19. But verified the sequence of events with WAYLAND_DEBUG and caps lock stil working in QtWayland clients and Xwayland 1.18 Reviewers: #kwin, #plasma Subscribers: plasma-devel, kwin Tags: #kwin Differential Revision: https://phabricator.kde.org/D5452
2017-04-14 22:01:20 +03:00
forwardModifiers();
}
void Xkb::updateKey(uint32_t key, InputRedirection::KeyboardKeyState state)
{
if (!m_keymap || !m_state) {
return;
}
xkb_state_update_key(m_state, key + 8, static_cast<xkb_key_direction>(state));
if (state == InputRedirection::KeyboardKeyPressed) {
const auto sym = toKeysym(key);
if (m_compose.state && xkb_compose_state_feed(m_compose.state, sym) == XKB_COMPOSE_FEED_ACCEPTED) {
switch (xkb_compose_state_get_status(m_compose.state)) {
case XKB_COMPOSE_NOTHING:
m_keysym = sym;
break;
case XKB_COMPOSE_COMPOSED:
m_keysym = xkb_compose_state_get_one_sym(m_compose.state);
break;
default:
m_keysym = XKB_KEY_NoSymbol;
break;
}
} else {
m_keysym = sym;
}
}
updateModifiers();
updateConsumedModifiers(key);
}
void Xkb::updateModifiers()
{
Qt::KeyboardModifiers mods = Qt::NoModifier;
if (xkb_state_mod_index_is_active(m_state, m_shiftModifier, XKB_STATE_MODS_EFFECTIVE) == 1 ||
xkb_state_mod_index_is_active(m_state, m_capsModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::ShiftModifier;
}
if (xkb_state_mod_index_is_active(m_state, m_altModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::AltModifier;
}
if (xkb_state_mod_index_is_active(m_state, m_controlModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::ControlModifier;
}
if (xkb_state_mod_index_is_active(m_state, m_metaModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::MetaModifier;
}
m_modifiers = mods;
// update LEDs
LEDs leds;
if (xkb_state_led_index_is_active(m_state, m_numLock) == 1) {
leds = leds | LED::NumLock;
}
if (xkb_state_led_index_is_active(m_state, m_capsLock) == 1) {
leds = leds | LED::CapsLock;
}
if (xkb_state_led_index_is_active(m_state, m_scrollLock) == 1) {
leds = leds | LED::ScrollLock;
}
if (m_leds != leds) {
m_leds = leds;
emit m_input->keyboard()->ledsChanged(m_leds);
}
Handle modifier updates in the same sequence as Wayland does Summary: Consider the case that capslock gets pressed and released. In the case of Weston we have a sequence of: 1. Key press event 2. Modifier changed event 3. Key release event 4. Modifier changed event KWin however used to send the events in the following sequence: 1. Modifier changed event (on key press) 2. Key press event 3. Modifier changed event (on key release) 4. Key release event It looks like Xwayland is not able to properly process the sequence sent by KWin. And in fact KWin's sequence is wrong as it sends a state which does not match. We report that the caps lock is pressed in the modifiers prior to the application getting informed about the key press of caps lock. This change aligns KWin's implementation to the behavior of Weston. The main difference is that when modifiers change Xkb internally caches the serialized modifier states. And KeyboardInputRedirection just forwards the modifiers to KWayland::Server::SeatInterface once the processing has finished. SeatInterface ignores the forwarding if no states changes, so it is fine to do it that way. BUG: 377155 Test Plan: Not yet tested with an affected Xwayland as I only have 1.18 and the problem started with 1.19. But verified the sequence of events with WAYLAND_DEBUG and caps lock stil working in QtWayland clients and Xwayland 1.18 Reviewers: #kwin, #plasma Subscribers: plasma-devel, kwin Tags: #kwin Differential Revision: https://phabricator.kde.org/D5452
2017-04-14 22:01:20 +03:00
m_currentLayout = xkb_state_serialize_layout(m_state, XKB_STATE_LAYOUT_EFFECTIVE);
m_modifierState.depressed = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_DEPRESSED));
m_modifierState.latched = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_LATCHED));
m_modifierState.locked = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_LOCKED));
}
void Xkb::forwardModifiers()
{
if (!waylandServer()) {
return;
}
Handle modifier updates in the same sequence as Wayland does Summary: Consider the case that capslock gets pressed and released. In the case of Weston we have a sequence of: 1. Key press event 2. Modifier changed event 3. Key release event 4. Modifier changed event KWin however used to send the events in the following sequence: 1. Modifier changed event (on key press) 2. Key press event 3. Modifier changed event (on key release) 4. Key release event It looks like Xwayland is not able to properly process the sequence sent by KWin. And in fact KWin's sequence is wrong as it sends a state which does not match. We report that the caps lock is pressed in the modifiers prior to the application getting informed about the key press of caps lock. This change aligns KWin's implementation to the behavior of Weston. The main difference is that when modifiers change Xkb internally caches the serialized modifier states. And KeyboardInputRedirection just forwards the modifiers to KWayland::Server::SeatInterface once the processing has finished. SeatInterface ignores the forwarding if no states changes, so it is fine to do it that way. BUG: 377155 Test Plan: Not yet tested with an affected Xwayland as I only have 1.18 and the problem started with 1.19. But verified the sequence of events with WAYLAND_DEBUG and caps lock stil working in QtWayland clients and Xwayland 1.18 Reviewers: #kwin, #plasma Subscribers: plasma-devel, kwin Tags: #kwin Differential Revision: https://phabricator.kde.org/D5452
2017-04-14 22:01:20 +03:00
waylandServer()->seat()->updateKeyboardModifiers(m_modifierState.depressed,
m_modifierState.latched,
m_modifierState.locked,
m_currentLayout);
}
QString Xkb::layoutName() const
{
return layoutName(m_currentLayout);
}
QString Xkb::layoutName(xkb_layout_index_t layout) const
{
if (!m_keymap) {
return QString{};
}
return QString::fromLocal8Bit(xkb_keymap_layout_get_name(m_keymap, layout));
}
QMap<xkb_layout_index_t, QString> Xkb::layoutNames() const
{
QMap<xkb_layout_index_t, QString> layouts;
const auto size = m_keymap ? xkb_keymap_num_layouts(m_keymap) : 0u;
for (xkb_layout_index_t i = 0; i < size; i++) {
layouts.insert(i, layoutName(i));
}
return layouts;
}
void Xkb::updateConsumedModifiers(uint32_t key)
{
Qt::KeyboardModifiers mods = Qt::NoModifier;
if (xkb_state_mod_index_is_consumed2(m_state, key + 8, m_shiftModifier, XKB_CONSUMED_MODE_GTK) == 1) {
mods |= Qt::ShiftModifier;
}
if (xkb_state_mod_index_is_consumed2(m_state, key + 8, m_altModifier, XKB_CONSUMED_MODE_GTK) == 1) {
mods |= Qt::AltModifier;
}
if (xkb_state_mod_index_is_consumed2(m_state, key + 8, m_controlModifier, XKB_CONSUMED_MODE_GTK) == 1) {
mods |= Qt::ControlModifier;
}
if (xkb_state_mod_index_is_consumed2(m_state, key + 8, m_metaModifier, XKB_CONSUMED_MODE_GTK) == 1) {
mods |= Qt::MetaModifier;
}
m_consumedModifiers = mods;
}
Qt::KeyboardModifiers Xkb::modifiersRelevantForGlobalShortcuts() const
{
if (!m_state) {
return Qt::NoModifier;
}
Qt::KeyboardModifiers mods = Qt::NoModifier;
if (xkb_state_mod_index_is_active(m_state, m_shiftModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::ShiftModifier;
}
if (xkb_state_mod_index_is_active(m_state, m_altModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::AltModifier;
}
if (xkb_state_mod_index_is_active(m_state, m_controlModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::ControlModifier;
}
if (xkb_state_mod_index_is_active(m_state, m_metaModifier, XKB_STATE_MODS_EFFECTIVE) == 1) {
mods |= Qt::MetaModifier;
}
Qt::KeyboardModifiers consumedMods = m_consumedModifiers;
if ((mods & Qt::ShiftModifier) && (consumedMods == Qt::ShiftModifier)) {
// test whether current keysym is a letter
// in that case the shift should be removed from the consumed modifiers again
// otherwise it would not be possible to trigger e.g. Shift+W as a shortcut
// see BUG: 370341
if (QChar(toQtKey(m_keysym)).isLetter()) {
consumedMods = Qt::KeyboardModifiers();
}
}
return mods & ~consumedMods;
}
xkb_keysym_t Xkb::toKeysym(uint32_t key)
{
if (!m_state) {
return XKB_KEY_NoSymbol;
}
return xkb_state_key_get_one_sym(m_state, key + 8);
}
QString Xkb::toString(xkb_keysym_t keysym)
{
if (!m_state || keysym == XKB_KEY_NoSymbol) {
return QString();
}
QByteArray byteArray(7, 0);
int ok = xkb_keysym_to_utf8(keysym, byteArray.data(), byteArray.size());
if (ok == -1 || ok == 0) {
return QString();
}
return QString::fromUtf8(byteArray.constData());
}
Qt::Key Xkb::toQtKey(xkb_keysym_t keysym) const
{
int key = Qt::Key_unknown;
KKeyServer::symXToKeyQt(keysym, &key);
if (key & Qt::KeyboardModifierMask) {
key &= ~Qt::KeyboardModifierMask;
}
return static_cast<Qt::Key>(key);
}
bool Xkb::shouldKeyRepeat(quint32 key) const
{
if (!m_keymap) {
return false;
}
return xkb_keymap_key_repeats(m_keymap, key + 8) != 0;
}
void Xkb::switchToNextLayout()
{
if (!m_keymap || !m_state) {
return;
}
const xkb_layout_index_t numLayouts = xkb_keymap_num_layouts(m_keymap);
const xkb_layout_index_t nextLayout = (xkb_state_serialize_layout(m_state, XKB_STATE_LAYOUT_EFFECTIVE) + 1) % numLayouts;
switchToLayout(nextLayout);
}
void Xkb::switchToPreviousLayout()
{
if (!m_keymap || !m_state) {
return;
}
const xkb_layout_index_t previousLayout = m_currentLayout == 0 ? numberOfLayouts() - 1 : m_currentLayout -1;
switchToLayout(previousLayout);
}
void Xkb::switchToLayout(xkb_layout_index_t layout)
{
if (!m_keymap || !m_state) {
return;
}
if (layout >= numberOfLayouts()) {
return;
}
const xkb_mod_mask_t depressed = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_DEPRESSED));
const xkb_mod_mask_t latched = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_LATCHED));
const xkb_mod_mask_t locked = xkb_state_serialize_mods(m_state, xkb_state_component(XKB_STATE_MODS_LOCKED));
xkb_state_update_mask(m_state, depressed, latched, locked, 0, 0, layout);
updateModifiers();
Handle modifier updates in the same sequence as Wayland does Summary: Consider the case that capslock gets pressed and released. In the case of Weston we have a sequence of: 1. Key press event 2. Modifier changed event 3. Key release event 4. Modifier changed event KWin however used to send the events in the following sequence: 1. Modifier changed event (on key press) 2. Key press event 3. Modifier changed event (on key release) 4. Key release event It looks like Xwayland is not able to properly process the sequence sent by KWin. And in fact KWin's sequence is wrong as it sends a state which does not match. We report that the caps lock is pressed in the modifiers prior to the application getting informed about the key press of caps lock. This change aligns KWin's implementation to the behavior of Weston. The main difference is that when modifiers change Xkb internally caches the serialized modifier states. And KeyboardInputRedirection just forwards the modifiers to KWayland::Server::SeatInterface once the processing has finished. SeatInterface ignores the forwarding if no states changes, so it is fine to do it that way. BUG: 377155 Test Plan: Not yet tested with an affected Xwayland as I only have 1.18 and the problem started with 1.19. But verified the sequence of events with WAYLAND_DEBUG and caps lock stil working in QtWayland clients and Xwayland 1.18 Reviewers: #kwin, #plasma Subscribers: plasma-devel, kwin Tags: #kwin Differential Revision: https://phabricator.kde.org/D5452
2017-04-14 22:01:20 +03:00
forwardModifiers();
}
quint32 Xkb::numberOfLayouts() const
{
if (!m_keymap) {
return 0;
}
return xkb_keymap_num_layouts(m_keymap);
}
}