1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
|
<?xml version="1.0" encoding="UTF-8"?>
<protocol name="agl_shell">
<copyright>
Copyright © 2019, 2022 Collabora, Ltd.
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>
<interface name="agl_shell" version="3">
<description summary="user interface for Automotive Grade Linux platform">
Starting with version 2 of the protocol, the client is required to wait
for the 'bound_ok' or 'bound_fail' events in order to proceed further.
In case the client gets a 'bound_fail' event then it should consider that
there's another client already bound to the agl_shell protocol.
A client that receives a 'bound_ok' event should consider that there's
no other client already bound to the interface and can proceed further.
If the client uses an older version of the protocol it will receive
automatically an error and the compositor will terminate the connection,
if there's another client already bound the interface.
If the client receives the 'bound_fail' event and attempts to use the
interface further it will receive an error and the compositor will
terminate the connection. After the 'bound_fail' event was received the
client should call the destructor, which has been added with version 2
of the protocol. The client is free to try at a later point in time to
see if it will receive the 'bound_ok' event, but there's no explicit way
of finding out when that event will be delivered.
It is assumed that it can infer that information through other
means/other channels.
</description>
<request name="destroy" type="destructor" since="2">
<description summary="destroys the factory object">
</description>
</request>
<enum name="error">
<entry name="invalid_argument" value="0"/>
<entry name="background_exists" value="1"/>
<entry name="panel_exists" value="2"/>
<entry name="split_failed" value="3" since="3"/>
</enum>
<enum name="edge">
<entry name="top" value="0"/>
<entry name="bottom" value="1"/>
<entry name="left" value="2"/>
<entry name="right" value="3"/>
</enum>
<enum name="tile_orientation" since="3">
<entry name="none" value="0"/>
<entry name="top" value="1"/>
<entry name="bottom" value="2"/>
<entry name="left" value="3"/>
<entry name="right" value="4"/>
</enum>
<request name="ready">
<description summary="client is ready to be shown">
Tell the server that this client is ready to be shown. The server
will delay presentation during start-up until all shell clients are
ready to be shown, and will display a black screen instead.
This gives the client an opportunity to set up and configure several
surfaces into a coherent interface.
The client that binds to this interface must send this request, otherwise
they may stall the compositor unnecessarily.
If this request is called after the compositor has already finished
start-up, no operation is performed.
</description>
</request>
<request name="set_background">
<description summary="set surface as output's background">
Set the surface to act as the background of an output. After this
request, the server will immediately send a configure event with
the dimensions the client should use to cover the entire output.
The surface must have a "desktop" surface role, as supported by
libweston-desktop.
Only a single surface may be the background for any output. If a
background surface already exists, a protocol error is raised.
</description>
<arg name="surface" type="object" interface="wl_surface"/>
<arg name="output" type="object" interface="wl_output"/>
</request>
<request name="set_panel">
<description summary="set surface as panel">
Set the surface to act as a panel of an output. The 'edge' argument
says what edge of the output the surface will be anchored to.
After this request, the server will send a configure event with the
corresponding width/height that the client should use, and 0 for the
other dimension. E.g. if the edge is 'top', the width will be the
output's width, and the height will be 0.
The surface must have a "desktop" surface role, as supported by
libweston-desktop.
The compositor will take the panel's window geometry into account when
positioning other windows, so the panels are not covered.
XXX: What happens if e.g. both top and left are used at the same time?
Who gets to have the corner?
Only a single surface may be the panel for an output's edge. If a
surface already exists on an edge, a protocol error is raised.
</description>
<arg name="surface" type="object" interface="wl_surface"/>
<arg name="output" type="object" interface="wl_output"/>
<arg name="edge" type="uint" enum="edge"/>
</request>
<request name="activate_app">
<description summary="make client current window">
Ask the compositor to make a toplevel to become the current/focused
window for window management purposes.
See xdg_toplevel.set_app_id from the xdg-shell protocol for a
description of app_id.
If multiple toplevels have the same app_id, the result is unspecified.
XXX: Do we need feedback to say it didn't work? (e.g. client does
not exist)
</description>
<arg name="app_id" type="string"/>
<arg name="output" type="object" interface="wl_output"/>
</request>
<event name="bound_ok" since="2">
<description summary="event sent if binding was ok">
Informs the client that it was able to bind the agl_shell
interface succesfully. Clients are required to wait for this
event before continuing further.
</description>
</event>
<event name="bound_fail" since="2">
<description summary="event sent if binding was nok">
Informs the client that binding to the agl_shell interface was
unsuccesfull. Clients are required to wait for this event for
continuing further.
</description>
</event>
<request name="set_app_split" since="3">
<description summary="set the application split">
Request changes the application from the original mode (whatever that
might be) to a split, tiled orientation mode defined in the orientation
enum. Clients need to implement resizing (meaing handle xdg-shell
configure events) for this to work correctly.
This request only handles a single level of tiling for practical
reasons: to keep implementation simple and straight forward. The
compositor will ignore requests if there are already two windows
present, and will terminate the client's connection with a protocol
violation, if it detects more than one tiling level.
If there's no app_id with the supplied name it will be added to a
pending list in order to be applied when an application id gets started.
Applications can use this approch if they want to be started in a
tiled orientantion position, before creating the xdg-shell toplevel role.
A none orientation type would make the window go back to the original
maximized mode. If two windows are side by side, returning one of them
back the original mode would mean the other one will be made hidden
and the one doing the request for the none orientation will become
the currently active window. A further activation, using activate_app
request for the other window would make that one active.
Closing the window in the tiled orientation state implies that either
the background surface will displayed, or in case there was another
applications being shown at that time, will make that application be
returned to the original, maximized state.
The tiled orientation could be applied independently of each other,
such that a client can transition from one tiled orientation to
another. Note that any other window already present would literally
take the opposite orientation with the one currently being changed. So
tiled orientation modification automatically implies a tile orientation
for any other application already present/active at that time.
In case there's already a client active at that time, it will be
attributed automatically the opposite tiled orientation, such that two
concurrent applications can be displayed at the same time.
The orientation tiles can not be combined, and only state at a time
can be active. Only horizontal and vertical tiling is possible.
Input focus is being delivered to the last started/activated window, such
that users can cycle between that one or the other, assumes there's
another window in the first place.
See xdg_toplevel.set_app_id from the xdg-shell protocol for a
description of app_id.
</description>
<arg name="app_id" type="string"/>
<arg name="orientation" type="uint" enum="tile_orientation"/>
<arg name="output" type="object" interface="wl_output"/>
</request>
</interface>
</protocol>
|
