Re: RFR: 8213573: MouseLocationOnScreenTest fails intermittently [v2]
On Fri, 23 Oct 2020 13:57:46 GMT, Pankaj Bansal wrote: >> The test test.robot.javafx.scene.MouseLocationOnScreenTest fails >> intermittently on Windows and Mac system though it fails on Windows more >> often than Mac. >> >> The test uses robot to move the mouse at different positions on screen using >> mouseMove and then validates the mouse coordinates by robot.getMouseX/Y. If >> the coordinates returned from Robot.getMouseX/Y are same as mouse >> coordinates passed to Robot.mouseMove, the test passes, else the test fails. >> >> The issue here is that there is no wait/delay between the calls to mouseMove >> and getMouseX/Y. When Robot.mouseMove is used to move mouse to a coordinate, >> robot in turn sends the event to native system. When Robot.getMouseX/Y is >> called to get the current mouse coordinates, robot again uses native >> functions to get the mouse location. In current test, it is possible that >> the Robot.mouseMove is called with particular x,y coordinates and which in >> turn has send the events to native. As there is no delay/wait after calling >> Robot.mouseMove, When the Robot.getMouseX/Y is called, it is possible that >> the earlier event sent to native has not been processed yet. So the mouse >> has not been moved and Robot.getMouseX/Y may return older mouse coordinates. >> This results in the test to fail. >> >> In short, the Robot.mouseMove is an asynchronous call, so it should be used >> in same way. A follow up bug will be filled to update the specification to >> reflect this. The test is assuming it to be synchronous and results in >> intermittent failures. >> >> The fix is to add small delay after calling Robot.mouseMove. Along with >> this, now as the test will take more time, so the timeout time for test is >> increased to 120s. Also, I have divided the calls to edge(...) among more >> Util.runAndWait calls as there is a timeout limit for Util.runAndWait which >> is set to 10s. >> >> I tested this on Windows 10, Ubuntu 18.04, Ubuntu 20.04, Oracle Linux 8.2, >> Mac 10.13. For me this test is passing in less than 50 seconds on all >> platforms. Following command can be used to run the test >> gradle --continue --info -PFULL_TEST=true -PUSE_ROBOT=true >> :systemTest:cleanTest :systemTests:test --tests >> test.robot.javafx.scene.MouseLocationOnScreenTest > > Pankaj Bansal has updated the pull request incrementally with one additional > commit since the last revision: > > Fixing review comments Marked as reviewed by kcr (Lead). - PR: https://git.openjdk.java.net/jfx/pull/333
Re: RFR: 8213573: MouseLocationOnScreenTest fails intermittently [v2]
On Fri, 23 Oct 2020 13:33:35 GMT, Kevin Rushforth wrote:
>> Pankaj Bansal has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Fixing review comments
>
> tests/system/src/test/java/test/robot/javafx/scene/MouseLocationOnScreenTest.java
> line 138:
>
>> 136: for (int y = y1; y <= y2; y++) {
>> 137: robot.mouseMove(x, y);
>> 138: delay(delayTime);
>
> Minor: you can use `Util.sleep` here instead of creating your own method.
Fixed
-
PR: https://git.openjdk.java.net/jfx/pull/333
Re: RFR: 8213573: MouseLocationOnScreenTest fails intermittently [v2]
> The test test.robot.javafx.scene.MouseLocationOnScreenTest fails > intermittently on Windows and Mac system though it fails on Windows more > often than Mac. > > The test uses robot to move the mouse at different positions on screen using > mouseMove and then validates the mouse coordinates by robot.getMouseX/Y. If > the coordinates returned from Robot.getMouseX/Y are same as mouse coordinates > passed to Robot.mouseMove, the test passes, else the test fails. > > The issue here is that there is no wait/delay between the calls to mouseMove > and getMouseX/Y. When Robot.mouseMove is used to move mouse to a coordinate, > robot in turn sends the event to native system. When Robot.getMouseX/Y is > called to get the current mouse coordinates, robot again uses native > functions to get the mouse location. In current test, it is possible that the > Robot.mouseMove is called with particular x,y coordinates and which in turn > has send the events to native. As there is no delay/wait after calling > Robot.mouseMove, When the Robot.getMouseX/Y is called, it is possible that > the earlier event sent to native has not been processed yet. So the mouse has > not been moved and Robot.getMouseX/Y may return older mouse coordinates. This > results in the test to fail. > > In short, the Robot.mouseMove is an asynchronous call, so it should be used > in same way. A follow up bug will be filled to update the specification to > reflect this. The test is assuming it to be synchronous and results in > intermittent failures. > > The fix is to add small delay after calling Robot.mouseMove. Along with this, > now as the test will take more time, so the timeout time for test is > increased to 120s. Also, I have divided the calls to edge(...) among more > Util.runAndWait calls as there is a timeout limit for Util.runAndWait which > is set to 10s. > > I tested this on Windows 10, Ubuntu 18.04, Ubuntu 20.04, Oracle Linux 8.2, > Mac 10.13. For me this test is passing in less than 50 seconds on all > platforms. Following command can be used to run the test > gradle --continue --info -PFULL_TEST=true -PUSE_ROBOT=true > :systemTest:cleanTest :systemTests:test --tests > test.robot.javafx.scene.MouseLocationOnScreenTest Pankaj Bansal has updated the pull request incrementally with one additional commit since the last revision: Fixing review comments - Changes: - all: https://git.openjdk.java.net/jfx/pull/333/files - new: https://git.openjdk.java.net/jfx/pull/333/files/4992e528..adc62f56 Webrevs: - full: https://webrevs.openjdk.java.net/?repo=jfx=333=01 - incr: https://webrevs.openjdk.java.net/?repo=jfx=333=00-01 Stats: 12 lines in 1 file changed: 0 ins; 8 del; 4 mod Patch: https://git.openjdk.java.net/jfx/pull/333.diff Fetch: git fetch https://git.openjdk.java.net/jfx pull/333/head:pull/333 PR: https://git.openjdk.java.net/jfx/pull/333
