Re: Proposal For Inclusion of Robot and ParametersImpl in the Public API

[Michael Ennen]

Kevin,

You mentioned in your initial response that Jemmy would need to updated
with any changes
that result from moving Robot into the JavaFX public API. However, from
what I can tell
from a cursory glance, Jemmy seems to use "java.awt.Robot" and not the
JavaFX Glass
Robot API and thus no changes to Jemmy would be required after moving Robot
to the
JavaFX public API.

Am I missing something?

Thanks, happy holidays (not expecting a response for a while :) ).



On Fri, Dec 22, 2017 at 10:51 PM, Michael Ennen 
wrote:

> I didn't realize "_1" is an escape character in JNI method naming and is
> only
> needed when the method has a "_" in its' name. Nevermind!
>
> On Fri, Dec 22, 2017 at 2:57 PM, Michael Ennen 
> wrote:
>
>> I have made quite a bit of good progress I think:
>>
>> https://github.com/brcolow/openjfx/commit/d92ef24101cf32dfc0
>> 7b21bdcb575528d8a58eaf
>>
>> The public API is starting to look much cleaner:
>>
>> https://github.com/brcolow/openjfx/blob/master/modules/javaf
>> x.graphics/src/main/java/javafx/scene/robot/Robot.java
>>
>> Still a lot to do.
>>
>> Currently I am running into a JNI UnsatisfiedLinkError for GtkRobot but I
>> am unsure why:
>>
>>   java.lang.UnsatisfiedLinkError: com.sun.glass.ui.gtk.GtkRobot.
>> getScreenCapture([I)V
>> at 
>> javafx.graphics/com.sun.glass.ui.gtk.GtkRobot.getScreenCapture(Native
>> Method)
>> at javafx.graphics/com.sun.glass.ui.gtk.GtkRobot.getPixelColor(
>> GtkRobot.java:93)
>>
>> I have the following Java method in GtkRobot.java:
>>
>> @Override native protected void getScreenCapture(int x, int y, int width,
>> int height, int[] data);
>>
>> Which is overriding the abstract method in Robot.java:
>>
>> protected abstract void getScreenCapture(int x, int y, int width, int
>> height, int[] data);
>>
>> and it is implemented in GlassRobot.cpp thusly:
>>
>> /*
>>  * Class: com_sun_glass_ui_gtk_GtkRobot
>>  * Method:getScreenCapture
>>  * Signature: ([I)V
>>  */
>> JNIEXPORT void JNICALL Java_com_sun_glass_ui_gtk_GtkR
>> obot_1getScreenCapture
>>   (JNIEnv * env, jobject obj, jint x, jint y, jint width, jint height,
>> jintArray data)
>> {
>> (void)obj;
>>
>> GdkPixbuf *screenshot, *tmp;
>> GdkWindow *root_window = gdk_get_default_root_window();
>>
>> tmp = glass_pixbuf_from_window(root_window, x, y, width, height);
>> screenshot = gdk_pixbuf_add_alpha(tmp, FALSE, 0, 0, 0);
>> g_object_unref(tmp);
>>
>> jint *pixels = (jint *)convert_BGRA_to_RGBA((int*)g
>> dk_pixbuf_get_pixels(screenshot), width * 4, height);
>> env->SetIntArrayRegion(data, 0, height * width, pixels);
>> g_free(pixels);
>>
>> g_object_unref(screenshot);
>> }
>>
>> Have I somehow messed up the signature? As you can see I removed one of
>> the "_" prefixes
>> to "1getScreenCapture" (which seemed to work for the other Robots) as
>> necessary because
>> the native method is no longer itself prefixed with "_".
>>
>> Thanks again.
>>
>> On Wed, Dec 20, 2017 at 3:05 PM, Kevin Rushforth <
>> kevin.rushfo...@oracle.com> wrote:
>>
>>> Sure, no problem. One quick comment is that a common way to solve this
>>> is by delegating to an implementation class, which would then be
>>> sub-classes.
>>>
>>>
>>> -- Kevin
>>>
>>>
>>> Michael Ennen wrote:
>>>
>>> I am not trying to be a burden here. I understand that you may not have
>>> time to hand-hold
>>> to this degree. I will try and make progress, sorry for the follow up
>>> question.
>>>
>>> On Wed, Dec 20, 2017 at 2:08 PM, Michael Ennen 
>>> wrote:
>>>
 How can Robot call into the implementation when it is a super class of
 the implementations?

 On Wed, Dec 20, 2017 at 2:04 PM, Kevin Rushforth <
 kevin.rushfo...@oracle.com> wrote:

>
>
> Michael Ennen wrote:
>
> I have a question about how to proceed with the Robot code.
>
> The base abstract Robot class is: https://github.com/brcolow
> /openjfx/blob/master/modules/javafx.graphics/src/main/java/j
> avafx/scene/robot/Robot.java
>
> As you can see for each method, such as "getMouseX()" there is a "_"
> prefixed method
> which is abstract and a non-prefixed method:
>
> protected abstract int _getMouseX();
>
> public int getMouseX() {
> Application.checkEventThread();
> return _getMouseX();
> }
>
> I have copied this from the private Robot API.
>
> Is there a better way to do this? Would this pass review?
>
>
> Yes there are better ways to do this. No it would not pass review,
> since this would be leaking implementation into the public API.
>
> Rather than copying the public / protected methods from the internal
> package, it probably makes more sense to start with what a Robot API 
> should
> look like and then have that call into the implementation (suitably
> modified so it better matches the public 

REMINDER: Deadline for FX fixes for JDK 10 RDP1

[Kevin Rushforth]

As a reminder, the deadline for getting FX fixes in during RDP1 of JDK 
10 is Monday, January 15, 2018 (end of the day Pacific time) [1]. To be 
more precise, fixes need to be reviewed, committed, and pushed to 
openjfx/10-dev/rt no later than 23:59 Pacific Time on that day.


I note that January 15 is a US public holiday, so for practical 
purposes, you will need to have your fixes reviewed by Friday, January 
12 in order to push before the deadline. Unless you have a very simple 
fix, please don't send your code review on January 12 and expect that it 
will be reviewed in time to meet the deadline.


RDP2 formally starts on January 18, 2018, but given the FX integration 
schedule, we will be under RDP2 rules starting 00:00 on Tuesday, January 
16, 2018. See the jdk10 project page [2] for all JDK 10 dates.


-- Kevin

[1] 
http://mail.openjdk.java.net/pipermail/openjfx-dev/2017-December/021149.html


[2] http://openjdk.java.net/projects/jdk/10/

Re: Updated OpenJFX build instructions

[Kevin Rushforth]

It's quite possible that Microsoft changed something in their 
installation of Visual Studio. We are relying on the VC/Auxiliary/Build 
directory being created as part of the (default) Visual Studio 
installation so we can find vcvars32.bat. You might look in your 
installation and see whether that file exists somewhere else. If it 
does, then point VS150COMNTOOLS to the directory containing that bat file.


I won't have any more time to look at it this until after the new 
year...maybe someone else on the list can help?


-- Kevin


Nir Lisker wrote:
I did set them, but... if I set them in bash they disappear the next 
time I launch it, which made me think they are not being set properly 
for some reason even though they appear in `export -p` immediately 
after. Launching bash as admin didn't do anything (I thought write 
permissions issue).


So I went to the /home/user/.bash_profile file and added the export 
declarations there. Now they appear on `export -p` properly every 
launch. I also did `gradle --stop` and `rm -rf build`, but `gradle 
clean` gives the same error.


Otherwise, I also noticed that the folder VS150COMNTOOLS is pointing 
to does not exist. I downloaded today VS 2017 Community version 15.5.2 
(which is what I set for MSVC_VER). This is as close as it gets:


---
Nir@Nir-Desktop /cygdrive/c/Program Files (x86)/Microsoft Visual 
Studio/2017/Community

$ dir
Common7  Licenses  MSBuild  Team\ Tools  Xml
---

(no VC/Auxiliary/Build)

Is that directory supposed to be created by some process, I needed to 
select some installation package for VS, or did something change from 
version 14.x to 15.x?


Nir

On Sat, Dec 23, 2017 at 4:14 AM, Kevin Rushforth 
> wrote:


Thanks for the feedback...I'll note it.

As for the build failure, did you install Visual Studio 2017 and
set the two env variables (VS150COMNTOOLS and MSVC_VER) to point
to it? If so, then it should determine WINSDK_DIR without anything
else needed.

-- Kevin



Nir Lisker wrote:

Thanks Kevin,

I'm going through the process now on Win 10.

A few things to note:
- In the Platform Prerequisites/cygwin it says to make sure
mercurial package is installed. Later, under Common
Prerequisites/Mercurial it says "you can also install Mercurial
as a cygwin package". It's not clear then if it's needed or
optional and if it replaces or complements the other hg toolings.
- The link to Tortoise should be https://tortoisehg.bitbucket.io
 (not .org).
- Might be trivial but I would say it's worth noting that all
commands listed there are to be ran in cygwin.
- The hg clone command probably needs to have a note added about
the destination folder.

The problems starts after navigating to the /rt directory and
executing 'gradle tasks' or `gradle projects`:

$ gradle tasks
:buildSrc:generateGrammarSource
A problem was found with the configuration of task
':buildSrc:generateGrammarSource'. Registering invalid inputs and
outputs via TaskInputs and TaskOutputs methods has been
deprecated and is scheduled to be removed in Gradle 5.0.
 - Directory 'D:\rt\buildSrc\src\main\antlr' specified for
property '$1' does not exist.
:buildSrc:generateGrammarSource UP-TO-DATE
:buildSrc:compileJava NO-SOURCE
:buildSrc:compileGroovy UP-TO-DATE
:buildSrc:processResources NO-SOURCE
:buildSrc:classes UP-TO-DATE
:buildSrc:jar UP-TO-DATE
:buildSrc:assemble UP-TO-DATE
:buildSrc:compileTestJava NO-SOURCE
:buildSrc:compileTestGroovy NO-SOURCE
:buildSrc:processTestResources NO-SOURCE
:buildSrc:testClasses UP-TO-DATE
:buildSrc:test NO-SOURCE
:buildSrc:check UP-TO-DATE
:buildSrc:build UP-TO-DATE

FAILURE: Build failed with an exception.

* Where:
Script 'D:\rt\buildSrc\win.gradle' line: 93

* What went wrong:
A problem occurred evaluating script.
> FAIL: WINSDK_DIR not defined

* Try:
Run with --stacktrace option to get the stack trace. Run with
--info or --debug option to get more log output.

* Get more help at https://help.gradle.org

BUILD FAILED in 2s

---

I did not install DirectX SDK. Also the output of the above
commands in the instructions page seems out of date compared to
the contents of the directory... or I messed up.

Any ideas?

 


I did a first pass over the build instructions on the OpenJFX
Wiki [1].
I think I cleaned up the worst of the inaccuracies, and added
some
information that will make it easier to build.

I'm not the best person to see whether anything important is
missing,
though. Someone less familiar with the build should look it
over and let
me know what I've missed. I won't be able to get back to this
until
after the