[PATCH wayland-protocols v3] Introduce pointer locking and confinement protocol
Jonas Ådahl
jadahl at gmail.com
Tue Jan 12 22:43:30 PST 2016
On Wed, Jan 13, 2016 at 01:59:21PM +0800, Jonas Ådahl wrote:
> On Wed, Jan 13, 2016 at 02:44:29PM +1000, Peter Hutterer wrote:
> > On Wed, Jan 13, 2016 at 10:14:50AM +0800, Jonas Ådahl wrote:
> > > This patch introduces a new protocol for locking and confining a
> > > pointer. It consists of a new global object with two requests; one for
> > > locking the surface to a position, one for confining the pointer to a
> > > given region.
> > >
> > > Signed-off-by: Jonas Ådahl <jadahl at gmail.com>
> > > Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net>
> > > Reviewed-by: Derek Foreman <derekf at osg.samsung.com>
> > > ---
> > >
> > > Changes since v2:
> > >
> > > Added a "lifetime" enum which is passed to the locking/confining requests. It
> > > is used to specify whether the constraints should be oneshot or reoccurring.
> > > Oneshot and reoccurring both has race conditions when they are deactivated, and
> > > this enables the client to choose what race condition it prefers.
> > >
> > > The oneshot race condition is that client may not be able to re-request the
> > > constraint in time, and as a result will loose the constraint where it should
> > > not have.
> > >
> > > The reoccurring race condition is that the client may not be able to destroy
> > > the constraint in time, and as a result the compositor will re-constrain the
> > > pointer when it shouldn't.
> > >
> > > Whether each of these race conditions are preferrable depends on the
> > > application using them, so lets give the client the option to choose.
> > >
> > >
> > > Another change is that the factory requests now take a wl_pointer instead of a
> > > wl_seat.
> > >
> > >
> > > Jonas
> > >
> > >
> > > Makefile.am | 1 +
> > > unstable/pointer-constraints/README | 4 +
> > > .../pointer-constraints-unstable-v1.xml | 341 +++++++++++++++++++++
> > > 3 files changed, 346 insertions(+)
> > > create mode 100644 unstable/pointer-constraints/README
> > > create mode 100644 unstable/pointer-constraints/pointer-constraints-unstable-v1.xml
> > >
> > > diff --git a/Makefile.am b/Makefile.am
> > > index e3b60ad..44041a6 100644
> > > --- a/Makefile.am
> > > +++ b/Makefile.am
> > > @@ -6,6 +6,7 @@ unstable_protocols = \
> > > unstable/input-method/input-method-unstable-v1.xml \
> > > unstable/xdg-shell/xdg-shell-unstable-v5.xml \
> > > unstable/relative-pointer/relative-pointer-unstable-v1.xml \
> > > + unstable/pointer-constraints/pointer-constraints-unstable-v1.xml \
> > > $(NULL)
> > >
> > > nobase_dist_pkgdata_DATA = \
> > > diff --git a/unstable/pointer-constraints/README b/unstable/pointer-constraints/README
> > > new file mode 100644
> > > index 0000000..8a242f8
> > > --- /dev/null
> > > +++ b/unstable/pointer-constraints/README
> > > @@ -0,0 +1,4 @@
> > > +Pointer constraints protocol
> > > +
> > > +Maintainers:
> > > +Jonas Ådahl <jadahl at gmail.com>
> > > diff --git a/unstable/pointer-constraints/pointer-constraints-unstable-v1.xml b/unstable/pointer-constraints/pointer-constraints-unstable-v1.xml
> > > new file mode 100644
> > > index 0000000..607a538
> > > --- /dev/null
> > > +++ b/unstable/pointer-constraints/pointer-constraints-unstable-v1.xml
> > > @@ -0,0 +1,341 @@
> > > +<?xml version="1.0" encoding="UTF-8"?>
> > > +<protocol name="pointer_constraints_unstable_v1">
> > > +
> > > + <copyright>
> > > + Copyright © 2014 Jonas Ådahl
> > > + Copyright © 2015 Red Hat Inc.
> > > +
> > > + Permission is hereby granted, free of charge, to any person obtaining a
> > > + copy of this software and associated documentation files (the "Software"),
> > > + to deal in the Software without restriction, including without limitation
> > > + the rights to use, copy, modify, merge, publish, distribute, sublicense,
> > > + and/or sell copies of the Software, and to permit persons to whom the
> > > + Software is furnished to do so, subject to the following conditions:
> > > +
> > > + The above copyright notice and this permission notice (including the next
> > > + paragraph) shall be included in all copies or substantial portions of the
> > > + Software.
> > > +
> > > + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> > > + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> > > + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> > > + THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> > > + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> > > + FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
> > > + DEALINGS IN THE SOFTWARE.
> > > + </copyright>
> > > +
> > > + <description summary="Protocol for constraining pointer motions">
> > > + This protocol specifies a set of interfaces used for adding constraints to
> > > + the motion of a pointer. Possible constraints include confining pointer
> > > + motions to a given region, or locking it to its current position.
> > > +
> > > + In order to contrain the pointer, a client must first bind the global
> > > + interface "wp_pointer_constraints" which, if a compositor supports pointer
> > > + constraints, is exposed by the registry. Using the bound global object, the
> > > + client uses the request that corresponds to the type of constraint it wants
> > > + to make. See wp_pointer_constraints for more details.
> > > +
> > > + Warning! The protocol described in this file is experimental and backward
> > > + incompatible changes may be made. Backward compatible changes may be added
> > > + together with the corresponding interface version bump. Backward
> > > + incompatible changes are done by bumping the version number in the protocol
> > > + and interface names and resetting the interface version. Once the protocol
> > > + is to be declared stable, the 'z' prefix and the version number in the
> > > + protocol and interface names are removed and the interface version number is
> > > + reset.
> > > + </description>
> > > +
> > > + <interface name="zwp_pointer_constraints_v1" version="1">
> > > + <description summary="constrain the movement of a pointer">
> > > + The global interface exposing pointer constraining functionality. It
> > > + exposes two requests; lock_pointer for locking the pointer to its
> > > + position, and confine_pointer for locking the pointer to a region.
> > > +
> > > + The lock_pointer and confine_pointer requests create the objects
> > > + wp_locked_pointer and wp_confined_pointer respectively, and the client can
> > > + use these objects to interact with the lock.
> > > +
> > > + For any surface, only one lock or confinement per seat may be active at
> > > + any time. If a lock or confinement is requested when another lock or
> > > + confinement is active or requested on that surface and seat, an
> > > + 'already_constrained' error will be raised.
> >
> > should this read as "pointer" instead of seat now?
>
> This paragraph is now changed to:
>
> For any surface, only one lock or confinement may be active across all
> wl_pointer objects of the same seat. If a lock or confinement is requested
> when another lock or confinement is active or requested on the same surface
> and with any of the wl_pointer objects of the same seat, an
> 'already_constrained' error will be raised.
>
> >
> > > + </description>
> > > +
> > > + <enum name="error">
> > > + <description summary="wp_pointer_constraints error values">
> > > + These errors can be emitted in response to wp_pointer_constraints
> > > + requests.
> > > + </description>
> > > + <entry name="already_constrained" value="1"
> > > + summary="pointer constraint already requested on that surface"/>
> > > + </enum>
> > > +
> > > + <enum name="lifetime">
> > > + <description summary="constraint lifetime">
> > > + These values represent different lifetime semantics. They are passed
> > > + as argument to the factory requests to specify how the constraint
> > > + lifetimes should be managed.
> > > + </description>
> > > + <entry name="oneshot" value="1">
> > > + <description summary="the pointer constraint is defunct once deactivated">
> > > + A oneshot pointer constraint will never re-activate once it has been
> > > + deactivated. See the corresponding deactivation event
> > > + (wp_locked_pointer.unlocked and wp_confined_pointer.unconfined) for
> > > + details.
> > > + </description>
> > > + </entry>
> > > + <entry name="reoccurring" value="2">
> >
> > drop the 'o', this should be "recurring", goes for all uses of the word
> > below.
>
> Hmm. I guess so. What I can see about the difference between reoccurring
> and recurring is that recurring means occurring again at a regular
> interval, while reoccurring means maybe occurring again at an irregular
> interval. So is recurring really correct?
>
> >
> > > + <description summary="the pointer constraint is defunct once deactivated">
> >
> > and this should be a bit different to the oneshot :)
>
> Changed to "the pointer constraint is may reactivate"
>
> >
> > > + A reoccurring pointer constraint may again re-activate once it has
> > > + been deactivated. See the corresponding deactivation event
> >
> > if deactivate doesn't get a -, reactivate shouldn't either.
>
> Fixed locally.
>
> >
> > > + (wp_locked_pointer.unlocked and wp_confined_pointer.unconfined) for
> > > + details.
> > > + </description>
> > > + </entry>
> > > + </enum>
> > > +
> > > + <request name="destroy" type="destructor">
> > > + <description summary="destroy the pointer constraints manager object">
> > > + Used by the client to notify the server that it will no longer use this
> > > + pointer constraints object.
> > > + </description>
> > > + </request>
> > > +
> > > + <request name="lock_pointer">
> > > + <description summary="lock pointer to a position">
> > > + The lock_pointer request lets the client request to disable movements of
> > > + the virtual pointer (i.e. the cursor), effectively locking the pointer
> > > + to a position. This request may not take effect immediately; in the
> > > + future, when the compositor deems implementation-specific constraints
> > > + are satisfied, the pointer lock will be activated and the compositor
> > > + sends a locked event.
> > > +
> > > + The protocol provides no guarantee that the constraints are ever
> > > + satisfied, and does not require the compositor to send an error if the
> > > + constraints cannot ever be satisfied. It is thus possible to request a
> > > + lock that will never activate.
> > > +
> > > + There may not be another lock of any kind requested or active on the
> > > + surface for the seat of the given pointer when requesting a lock. If
> > > + there is, an error will be raised. See general pointer lock
> > > + documentation for more details.
> >
> > needs a seat/pointer rewording
>
> This paragraph is now:
>
> There may not be another pointer constraint of any kind requested or
> active on the surface for any of the wl_pointer objects of the seat of
> the passed pointer when requesting a lock. If there is, an error will be
> raised. See general pointer lock documentation for more details.
>
> >
> > > +
> > > + The intersection of the region passed with this request and the input
> > > + region of the surface is used to determine where the pointer must be
> > > + in order for the lock to activate. It is up to the compositor whether to
> > > + warp the pointer or require some kind of user interaction for the lock
> > > + to activate. If the region is null the surface input region is used.
> > > +
> > > + A surface may receive pointer focus without the lock being activated.
> > > +
> > > + The request will create a new object wp_locked_pointer which is used to
> >
> > s/will create/creates/
>
> Fixed locally.
>
> >
> > > + interact with the lock as well as receive updates about its state. See
> > > + the the description of wp_locked_pointer for further information.
> > > +
> > > + Note that while a pointer is locked, the wl_pointer objects of the
> > > + corresponding seat will not emit any wl_pointer.motion events, but
> > > + relative motion events will still be emitted via wp_relative_pointer
> > > + objects of the same seat. wl_pointer.axis and wl_pointer.button events
> > > + are unaffected.
> > > + </description>
> > > +
> > > + <arg name="id" type="new_id" interface="zwp_locked_pointer_v1"/>
> > > + <arg name="surface" type="object" interface="wl_surface"
> > > + summary="surface to lock pointer to"/>
> > > + <arg name="pointer" type="object" interface="wl_pointer"
> > > + summary="the pointer that should be locked"/>
> > > + <arg name="region" type="object" interface="wl_region" allow-null="true"
> > > + summary="region of surface"/>
> > > + <arg name="lifetime" type="uint" summary="lock lifetime"/>
> > > + </request>
> > > +
> > > + <request name="confine_pointer">
> > > + <description summary="confine pointer to a region">
> > > + The confine_pointer request lets the client request to confine the
> > > + pointer cursor to a given region. This request may not take effect
> > > + immediately; in the future, when the compositor deems implementation-
> > > + specific constraints are satisfied, the pointer confinement will be
> > > + activated and the compositor sends a confined event.
> > > +
> > > + The intersection of the region passed with this request and the input
> > > + region of the surface is used to determine where the pointer must be
> > > + in order for the confinement to activate. It is up to the compositor
> > > + whether to warp the pointer or require some kind of user interaction for
> > > + the confinement to activate. If the region is null the surface input
> > > + region is used.
> > > +
> > > + The request will create a new object wp_confined_pointer which is used
> > > + to interact with the confinement as well as receive updates about its
> > > + state. See the the description of wp_confined_pointer for further
> > > + information.
> > > + </description>
> > > +
> > > + <arg name="id" type="new_id" interface="zwp_confined_pointer_v1"/>
> > > + <arg name="surface" type="object" interface="wl_surface"
> > > + summary="surface to lock pointer to"/>
> > > + <arg name="pointer" type="object" interface="wl_pointer"
> > > + summary="the pointer that should be confined"/>
> > > + <arg name="region" type="object" interface="wl_region" allow-null="true"
> > > + summary="region of surface"/>
> > > + <arg name="lifetime" type="uint" summary="confinement lifetime"/>
> > > + </request>
> > > + </interface>
> > > +
> > > + <interface name="zwp_locked_pointer_v1" version="1">
> > > + <description summary="receive relative pointer motion events">
> > > + The wp_locked_pointer interface represents a locked pointer state.
> > > +
> > > + While the lock of this object is active, the wl_pointer objects of the
> > > + associated seat will not emit any wl_pointer.motion events.
> > > +
> > > + This object will send the event 'locked' when the lock is activated.
> > > + Whenever the lock is activated, it is guaranteed that the locked surface
> > > + will already have received pointer focus and that the pointer will be
> > > + within the region passed to the request creating this object.
> > > +
> > > + To unlock the pointer, send the destroy request. This will also destroy
> > > + the wp_locked_pointer object.
> > > +
> > > + If the compositor decides to unlock the pointer the unlocked event is
> > > + sent. The wp_locked_pointer object is at this point defunct and should be
> > > + destroyed.
> >
> > needs updates for the lifetime, or just a reference to the unlocked event
> > documentation.
>
> Fixed by just referencing .unlocked.
>
> >
> > > +
> > > + When unlocking, the compositor may warp the cursor position to the set
> > > + cursor position hint. If it does, it will not result in any relative
> > > + motion events emitted via wp_relative_pointer.
> > > +
> > > + If the surface the lock was requested on is destroyed and the lock is not
> > > + yet activated, the wp_locked_pointer object is now defunct and must be
> > > + destroyed.
> > > + </description>
> > > +
> > > + <request name="destroy" type="destructor">
> > > + <description summary="destroy the locked pointer object">
> > > + Destroy the locked pointer object. If applicable, the compositor will
> > > + unlock the pointer.
> > > + </description>
> > > + </request>
> > > +
> > > + <request name="set_cursor_position_hint">
> > > + <description summary="set the pointer cursor position hint">
> > > + Set the cursor position hint relative to the top left corner of the
> > > + surface.
> > > +
> > > + If the client is drawing its own cursor, it should update the position
> > > + hint to the position of its own cursor. A compositor may use this
> > > + information to warp the pointer upon unlock in order to avoid pointer
> > > + jumps.
> > > +
> > > + The cursor position hint is double buffered. The new hint will only take
> > > + effect when the associated surface gets it pending state applied. See
> > > + wl_surface.commit for details.
> > > + </description>
> > > +
> > > + <arg name="surface_x" type="fixed"
> > > + summary="x coordinate in surface-relative coordinates"/>
> > > + <arg name="surface_y" type="fixed"
> > > + summary="y coordinate in surface-relative coordinates"/>
> > > + </request>
> > > +
> > > + <request name="set_region">
> > > + <description summary="set a new lock region">
> > > + Set a new region used to lock the pointer.
> > > +
> > > + The new lock region is double-buffered. The new lock region will
> > > + only take effect when the associated surface gets its pending state
> > > + applied. See wl_surface.commit for details.
> > > +
> > > + For details about the lock region, see wp_locked_pointer.
> > > + </description>
> > > +
> > > + <arg name="region" type="object" interface="wl_region" allow-null="true"
> > > + summary="region of surface"/>
> > > + </request>
> > > +
> > > + <event name="locked">
> > > + <description summary="lock activation event">
> > > + Notification that the pointer lock of the seat's pointer is activated.
> > > + </description>
> > > + </event>
> > > +
> > > + <event name="unlocked">
> > > + <description summary="lock deactivation event">
> > > + Notification that the pointer lock of the seat's pointer is no longer
> > > + active. If this is a oneshot pointer lock (see
> > > + wp_pointer_constraints.lifetime) this object is now defunct and should
> > > + be destroyed. If this is a reoccurring pointer lock (see
> > > + wp_pointer_constraints.lifetime) this the pointer lock may again
> > > + reactivate in the future.
> > > + </description>
> > > + </event>
> > > + </interface>
> > > +
> > > + <interface name="zwp_confined_pointer_v1" version="1">
> > > + <description summary="confined pointer object">
> > > + The wp_confined_pointer interface represents a confined pointer state.
> > > +
> > > + This object will send the event 'confined' when the confinement is
> > > + activated. Whenever the confinement is activated, it is guaranteed that
> > > + the surface the pointer is confined to will already have received pointer
> > > + focus and that the pointer will be within the region passed to the request
> > > + creating this object. It is up to the compositor to decide whether this
> > > + requires some user interaction and if the pointer will warp to within the
> > > + passed region if outside.
> > > +
> > > + To unconfine the pointer, send the destroy request. This will also destroy
> > > + the wp_confined_pointer object.
> > > +
> > > + If the compositor decides to unconfine the pointer the unconfined event is
> > > + sent. The wp_confined_pointer object is at this point defunct and should
> > > + be destroyed.
> > > + </description>
> > > +
> > > + <request name="destroy" type="destructor">
> > > + <description summary="destroy the confined pointer object">
> > > + Destroy the confined pointer object. If applicable, the compositor will
> > > + unconfine the pointer.
> > > + </description>
> > > + </request>
> > > +
> > > + <request name="set_region">
> > > + <description summary="set a new confine region">
> > > + Set a new region used to confine the pointer.
> > > +
> > > + The new confine region is double-buffered. The new confine region will
> > > + only take effect when the associated surface gets its pending state
> > > + applied. See wl_surface.commit for details.
> > > +
> > > + If the confinement is active when the new confinement region is applied
> > > + and the pointer ends up outside of newly applied region, the pointer is
> > > + warped to a position within the new confinement region. If warped, a
> > > + wl_pointer.motion event will be emitted, but no
> > > + wp_relative_pointer.relative_motion event.
> >
> > is there a case where the compositor will break the confinement if it's not
> > happy with the new region? If so, we should add a comment here to state
> > that.
>
> Hmm. I suppose a 0x0 region would be hard to make any sense of. I guess we can
> add a blurb about that an effective 0x0 region (intersection between set region
> and input region is 0x0) would result in an unconfined event.
>
> Something like:
>
> If the effective confinement region (the intersection of the applied
> confinement region and the input region of the corresponding wl_surface)
> is empty, an active confinement would be deactivated.
>
> maybe?
After IRC discussion, this paragraph and the one above was changed to:
If the confinement is active when the new confinement region is applied
and the pointer ends up outside of newly applied region, the pointer may
warped to a position within the new confinement region. If warped, a
wl_pointer.motion event will be emitted, but no
wp_relative_pointer.relative_motion event.
The compositor may also, instead of using the new region, unconfine the
pointer.
to make it clear that the compositor may choose to ignore any new region and
just unconfine the pointer if it wants to.
Jonas
>
> >
> > my rev-by still stands.
>
> Thanks!
>
>
> Jonas
>
> >
> > Cheers,
> > Peter
> >
> > > +
> > > + For details about the confine region, see wp_confined_pointer.
> > > + </description>
> > > +
> > > + <arg name="region" type="object" interface="wl_region" allow-null="true"
> > > + summary="region of surface"/>
> > > + </request>
> > > +
> > > + <event name="confined">
> > > + <description summary="pointer confined">
> > > + Notification that the pointer confinement of the seat's pointer is
> > > + activated.
> > > + </description>
> > > + </event>
> > > +
> > > + <event name="unconfined">
> > > + <description summary="pointer unconfined">
> > > + Notification that the pointer confinement of the seat's pointer is no
> > > + longer active. If this is a oneshot pointer confinement (see
> > > + wp_pointer_constraints.lifetime) this object is now defunct and should
> > > + be destroyed. If this is a reoccurring pointer confinement (see
> > > + wp_pointer_constraints.lifetime) this the pointer confinement may again
> > > + reactivate in the future.
> > > + </description>
> > > + </event>
> > > + </interface>
> > > +
> > > +</protocol>
> > > --
> > > 2.4.3
> > >
More information about the wayland-devel
mailing list