weston/protocol/weston-test.xml

145 lines
5.6 KiB
XML
Raw Normal View History

<?xml version="1.0" encoding="UTF-8"?>
<protocol name="weston_test">
<copyright>
Copyright © 2012 Intel Corporation
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="weston_test" version="1">
<description summary="weston internal testing">
Internal testing facilities for the weston compositor.
It can't be stressed enough that these should never ever be used
outside of running weston's tests. The weston-test.so module should
never be installed.
These requests may allow clients to do very bad things.
</description>
<request name="move_surface">
<arg name="surface" type="object" interface="wl_surface"/>
<arg name="x" type="int"/>
<arg name="y" type="int"/>
</request>
<request name="move_pointer">
<arg name="tv_sec_hi" type="uint"/>
<arg name="tv_sec_lo" type="uint"/>
<arg name="tv_nsec" type="uint"/>
<arg name="x" type="int"/>
<arg name="y" type="int"/>
</request>
<request name="send_button">
<arg name="tv_sec_hi" type="uint"/>
<arg name="tv_sec_lo" type="uint"/>
<arg name="tv_nsec" type="uint"/>
<arg name="button" type="int"/>
<arg name="state" type="uint"/>
</request>
<request name="send_axis">
<arg name="tv_sec_hi" type="uint"/>
<arg name="tv_sec_lo" type="uint"/>
<arg name="tv_nsec" type="uint"/>
<arg name="axis" type="uint"/>
<arg name="value" type="fixed"/>
</request>
<request name="activate_surface">
<arg name="surface" type="object" interface="wl_surface" allow-null="true"/>
</request>
<request name="send_key">
<arg name="tv_sec_hi" type="uint"/>
<arg name="tv_sec_lo" type="uint"/>
<arg name="tv_nsec" type="uint"/>
<arg name="key" type="uint"/>
<arg name="state" type="uint"/>
</request>
<request name="device_release">
<arg name="device" type="string"/>
</request>
<request name="device_add">
<arg name="device" type="string"/>
</request>
<event name="pointer_position">
<arg name="x" type="fixed"/>
<arg name="y" type="fixed"/>
</event>
<request name="capture_screenshot">
<description summary="records current screen image">
Records an image of what is currently displayed on a given
display output, returning the image as an event.
</description>
<arg name="output" type="object" interface="wl_output"
summary="output to capture from"/>
<arg name="buffer" type="object" interface="wl_buffer"
summary="buffer for returning screenshots to the test client"/>
</request>
<event name="capture_screenshot_done">
<description summary="screenshot capture is done">
The capture_screenshot_done signal is sent when a screenshot has been copied into the
provided buffer.
</description>
</event>
<request name="send_touch">
<arg name="tv_sec_hi" type="uint"/>
<arg name="tv_sec_lo" type="uint"/>
<arg name="tv_nsec" type="uint"/>
<arg name="touch_id" type="int"/>
<arg name="x" type="fixed"/>
<arg name="y" type="fixed"/>
<arg name="touch_type" type="uint"/>
</request>
</interface>
tests: ivi_layout test infrastructure Testing the ivi_layout API requires two things: - the tests must be written as a controller module to access the API - the tests need a helper client to create some objects that can then be managed via the API This patch adds all the infrastructure and two different kinds of example tests. Internal ivi-shell (ivi_layout) API tests are listed as ivi-*.la files in TESTS in Makefile.am. Weston-tests-env detects these, and runs Weston with ivi-shell, and loads the given module as a controller module, not as a normal plugin. The test controller module ivi-*.la will launch a helper client. For ivi-layout-test.la the helper client is ivi-layout.ivi. The helper client uses the weston-test-runner framework to fork and exec each TEST with a fresh connection to the compositor. The actual test is triggered by the weston_test_runner protocol interface, a new addition to weston-test.xml. The helper client uses weston_test_runner to trigger a test, and the server side of the interface is implemented by the test controller module (ivi-layout-test.la). The server side of weston_test_runner uses the same trick as weston-test-runner.h to gather a list of defined tests. A test is defined with the RUNNER_TEST macro. If a test defined by RUNNER_TEST succeeds, an event is sent to the helper client that it can continue (or exit). If a test fails, a fatal protocol error is sent to the helper client. Once the helper client has iterated over all of its tests, it signals the batch success/failure via process exit code. That is cought in the test controller module, and forwarded as Weston's exit code. In summary: each ivi_layout test is a combination of a client side helper/setup and server side actual tests. v2: Load weston-test.so, because create_client() needs it. v3: add a comment about IVI_TEST_SURFACE_ID_BASE. v4: Rebased to upstream weston-tests-env changes. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk> Reviewed-by: Derek Foreman <derekf@osg.samsung.com> (v2)
2015-03-25 13:50:31 +03:00
<interface name="weston_test_runner" version="1">
<description summary="weston internal testing">
This is a global singleton interface for Weston internal tests.
This interface allows a test client to trigger compositor-side
test procedures. This is useful for cases, where the actual tests
are in compositor plugins, but they also require the presence of
a particular client.
This interface is implemented by the compositor plugins containing
the testing code.
A test client starts a test with the "run" request. It must not send
another "run" request until receiving the "finished" event. If the
compositor-side test succeeds, the "finished" event is sent. If the
compositor-side test fails, the compositor should send the protocol
error "test_failed", but it may also exit with an error (e.g. SEGV).
Unknown test name will raise "unknown_test" protocol error.
</description>
<enum name="error">
<entry name="test_failed" value="0" summary="compositor test failed"/>
<entry name="unknown_test" value="1" summary="unrecognized test name"/>
</enum>
<request name="destroy" type="destructor"/>
<request name="run">
<arg name="test_name" type="string"/>
</request>
<event name="finished"/>
</interface>
</protocol>