Change of default behavior of When

[Nir Lisker]

Hi all,

The class javafx.beans.binding.When is using eager evaluation of its
arguments [1]. Kevin and I discussed the idea of changing this default
behavior to the more intuitive lazy evaluation as done in a simple ternary
expression. This is a potentially breaking change for current code that
(for some reason) relies on both values being evaluated regardless of the
condition's value. This behavior is not documented.

If anyone has any reservations, please voice them over the next few days.

Thanks,
Nir

[1] https://bugs.openjdk.java.net/browse/JDK-8089579

Re: [11] JDK-8204621: Upgrade MarlinFX to 0.9.2

[Kevin Rushforth]

I'm giving a +1 on the implementation changes. I scanned the webrev and 
didn't see anything out of place. I compared the diffs of the FX Marlin 
0.9.2 with the Java2D 0.9.1 changeset, and there were a few more diffs 
than I might have expoected, but nothing jumped out of me as a problem. 
Also, I've tested it pretty well on all three platforms.


The overall +1 is pending the fixes needed for the test: at least the 
copyright header and shortening or disabling the test.


-- Kevin


On 6/29/2018 11:25 AM, Kevin Rushforth wrote:
One more thing about the test. All of the OpenJFX unit tests should 
have GPL v2 + Classpath Exception (this differs from the JDK).


-- Kevin


On 6/29/2018 10:23 AM, Kevin Rushforth wrote:


I'll plan to review the code today if possible. This will need one 
more reviewer, so maybe Phil can also review it, since he reviewed 
the Java2D patch?


As for my comments on the test:

Finally I think this test should be manually run only if Marlin 
renderer is modified.

How to do that ? use @Ignore or specific tags ?


As a slight variation of this: How about running a small number (say, 
200 or 250) by default, but adding a flag to run more? Alternatively, 
you could use a flag to enable it, but since you would need to do 
something extra (provide a flag or modify the test) to run it, we 
might as well get at least some testing all the time. Unless you 
really think there is no value in doing this.


I deliberately set all these Marlin clip (runtime + always 
subdivider) / curve quality settings (quads / cubics thresholds) to 
be sure of the concrete Marlin setup as quality thresholds are 
sensitive to such values.


As a best practice, tests should generally be run using the same 
settings as are used in production. Other than to verify how it 
behaves when you change these settings, I don't see the value in 
testing the system running in a mode that no application will ever 
see. I may be missing some point here.


-- Kevin


On 6/25/2018 9:01 AM, Laurent Bourgès wrote:

Kevin,

Here are my comments below:

2018-06-16 1:47 GMT+02:00 Kevin Rushforth 
mailto:kevin.rushfo...@oracle.com>>:


    I tested this on all three platforms and the updated rasterizer
    looks good.

    I spot checked the code changes, but didn't get time to do a
    complete review yet. I was mostly looking for diffs between the
    Java2D version which was already reviewed, and this one.

    I do have a couple comments on the new ClipShapeTest (which looks
    like a nice accuracy test, btw).

    1. The test runs for way too long (about 20x too long) to include
    in our normal test runs. By default the entire class file (all
    three tests) needs to take < 5 minutes and 2 minutes would be
    better. I measured the time on 4 machines that I have and found
    that if you cut the number of iterations down from 5000 to 250 it
    will be just about the right run time. Then you can set the
    timeout to 120 seconds (the slowest test on the slowest of my
    machines took about 48 seconds, so a 2 minute timeout should be
    plenty).


I agree this test is very long but it is the only mean I found to 
test all possible stroke combinations and test enough shapes (5000) 
to detect bugs.
I wondered if using mask directly (via ShapeUtils.getMaskData()) 
would become faster but it will never run below the 2 minutes 
threshold in total.


Finally I think this test should be manually run only if Marlin 
renderer is modified.

How to do that ? use @Ignore or specific tags ?


    2. Can you explain the reason for setting the following?

     206 // disable static clipping setting:
     207 System.setProperty("prism.marlin.clip", "false");
     208
    System.setProperty("prism.marlin.clip.runtime.enable", "true");
     209
     210 // enable subdivider:
     211 System.setProperty("prism.marlin.clip.subdivider",
    "true");
     212
     213 // disable min length check: always subdivide curves
    at clip edges
     214 System.setProperty("prism.marlin.clip.subdivider.minLength",
    "-1");
     215
     216 // If any curve, increase curve accuracy:
     217 // curve length max error:
     218 System.setProperty("prism.marlin.curve_len_err", "1e-4");
     219
     220 // cubic min/max error:
     221 System.setProperty("prism.marlin.cubic_dec_d2", "1e-3");
     222 System.setProperty("prism.marlin.cubic_inc_d1",
    "1e-4"); // or disabled ~ 1e-6
     223
     224 // quad max error:
     225 System.setProperty("prism.marlin.quad_dec_d2", "5e-4");

    It seems better to test with the default parameters (i.e., it
    makes a better regression test that way).


I deliberately set all these Marlin clip (runtime + always 
subdivider) / curve quality settings (quads / cubics thresholds) to 
be sure of the concrete Marlin setup as quality thresholds are 
sensitive to such values.


The ClipShapeTest is dedicated to test the clipper (+ subdivider) 
part of 

Re: [11] JDK-8204621: Upgrade MarlinFX to 0.9.2

[Kevin Rushforth]

One more thing about the test. All of the OpenJFX unit tests should have 
GPL v2 + Classpath Exception (this differs from the JDK).


-- Kevin


On 6/29/2018 10:23 AM, Kevin Rushforth wrote:


I'll plan to review the code today if possible. This will need one 
more reviewer, so maybe Phil can also review it, since he reviewed the 
Java2D patch?


As for my comments on the test:

Finally I think this test should be manually run only if Marlin 
renderer is modified.

How to do that ? use @Ignore or specific tags ?


As a slight variation of this: How about running a small number (say, 
200 or 250) by default, but adding a flag to run more? Alternatively, 
you could use a flag to enable it, but since you would need to do 
something extra (provide a flag or modify the test) to run it, we 
might as well get at least some testing all the time. Unless you 
really think there is no value in doing this.


I deliberately set all these Marlin clip (runtime + always 
subdivider) / curve quality settings (quads / cubics thresholds) to 
be sure of the concrete Marlin setup as quality thresholds are 
sensitive to such values.


As a best practice, tests should generally be run using the same 
settings as are used in production. Other than to verify how it 
behaves when you change these settings, I don't see the value in 
testing the system running in a mode that no application will ever 
see. I may be missing some point here.


-- Kevin


On 6/25/2018 9:01 AM, Laurent Bourgès wrote:

Kevin,

Here are my comments below:

2018-06-16 1:47 GMT+02:00 Kevin Rushforth >:


    I tested this on all three platforms and the updated rasterizer
    looks good.

    I spot checked the code changes, but didn't get time to do a
    complete review yet. I was mostly looking for diffs between the
    Java2D version which was already reviewed, and this one.

    I do have a couple comments on the new ClipShapeTest (which looks
    like a nice accuracy test, btw).

    1. The test runs for way too long (about 20x too long) to include
    in our normal test runs. By default the entire class file (all
    three tests) needs to take < 5 minutes and 2 minutes would be
    better. I measured the time on 4 machines that I have and found
    that if you cut the number of iterations down from 5000 to 250 it
    will be just about the right run time. Then you can set the
    timeout to 120 seconds (the slowest test on the slowest of my
    machines took about 48 seconds, so a 2 minute timeout should be
    plenty).


I agree this test is very long but it is the only mean I found to 
test all possible stroke combinations and test enough shapes (5000) 
to detect bugs.
I wondered if using mask directly (via ShapeUtils.getMaskData()) 
would become faster but it will never run below the 2 minutes 
threshold in total.


Finally I think this test should be manually run only if Marlin 
renderer is modified.

How to do that ? use @Ignore or specific tags ?


    2. Can you explain the reason for setting the following?

     206 // disable static clipping setting:
     207 System.setProperty("prism.marlin.clip", "false");
     208
    System.setProperty("prism.marlin.clip.runtime.enable", "true");
     209
     210 // enable subdivider:
     211 System.setProperty("prism.marlin.clip.subdivider",
    "true");
     212
     213 // disable min length check: always subdivide curves
    at clip edges
     214 System.setProperty("prism.marlin.clip.subdivider.minLength",
    "-1");
     215
     216 // If any curve, increase curve accuracy:
     217 // curve length max error:
     218 System.setProperty("prism.marlin.curve_len_err", "1e-4");
     219
     220 // cubic min/max error:
     221 System.setProperty("prism.marlin.cubic_dec_d2", 
"1e-3");

     222 System.setProperty("prism.marlin.cubic_inc_d1",
    "1e-4"); // or disabled ~ 1e-6
     223
     224 // quad max error:
     225 System.setProperty("prism.marlin.quad_dec_d2", "5e-4");

    It seems better to test with the default parameters (i.e., it
    makes a better regression test that way).


I deliberately set all these Marlin clip (runtime + always 
subdivider) / curve quality settings (quads / cubics thresholds) to 
be sure of the concrete Marlin setup as quality thresholds are 
sensitive to such values.


The ClipShapeTest is dedicated to test the clipper (+ subdivider) 
part of the Marlin renderer.




    3. Related to that, I think you should eliminate the following (I
    don't recommend running functional tests with this set), although
    since you don't do any animation, it probably doesn't matter.

     227 System.setProperty("javafx.animation.fullspeed",
    "true"); // full speed


I will remove it and see if the overall test is not slower.
Is Platform.runLater() impacted by such setting (latency of FX thread 
-> Prism rendering thread ?) ?


Laurent



    On 6/8/2018 

Re: [11] JDK-8204621: Upgrade MarlinFX to 0.9.2

[Kevin Rushforth]

I'll plan to review the code today if possible. This will need one more 
reviewer, so maybe Phil can also review it, since he reviewed the Java2D 
patch?


As for my comments on the test:

Finally I think this test should be manually run only if Marlin 
renderer is modified.

How to do that ? use @Ignore or specific tags ?


As a slight variation of this: How about running a small number (say, 
200 or 250) by default, but adding a flag to run more? Alternatively, 
you could use a flag to enable it, but since you would need to do 
something extra (provide a flag or modify the test) to run it, we might 
as well get at least some testing all the time. Unless you really think 
there is no value in doing this.


I deliberately set all these Marlin clip (runtime + always subdivider) 
/ curve quality settings (quads / cubics thresholds) to be sure of the 
concrete Marlin setup as quality thresholds are sensitive to such values.


As a best practice, tests should generally be run using the same 
settings as are used in production. Other than to verify how it behaves 
when you change these settings, I don't see the value in testing the 
system running in a mode that no application will ever see. I may be 
missing some point here.


-- Kevin


On 6/25/2018 9:01 AM, Laurent Bourgès wrote:

Kevin,

Here are my comments below:

2018-06-16 1:47 GMT+02:00 Kevin Rushforth >:


I tested this on all three platforms and the updated rasterizer
looks good.

I spot checked the code changes, but didn't get time to do a
complete review yet. I was mostly looking for diffs between the
Java2D version which was already reviewed, and this one.

I do have a couple comments on the new ClipShapeTest (which looks
like a nice accuracy test, btw).

1. The test runs for way too long (about 20x too long) to include
in our normal test runs. By default the entire class file (all
three tests) needs to take < 5 minutes and 2 minutes would be
better. I measured the time on 4 machines that I have and found
that if you cut the number of iterations down from 5000 to 250 it
will be just about the right run time. Then you can set the
timeout to 120 seconds (the slowest test on the slowest of my
machines took about 48 seconds, so a 2 minute timeout should be
plenty).


I agree this test is very long but it is the only mean I found to test 
all possible stroke combinations and test enough shapes (5000) to 
detect bugs.
I wondered if using mask directly (via ShapeUtils.getMaskData()) would 
become faster but it will never run below the 2 minutes threshold in 
total.


Finally I think this test should be manually run only if Marlin 
renderer is modified.

How to do that ? use @Ignore or specific tags ?


2. Can you explain the reason for setting the following?

 206 // disable static clipping setting:
 207 System.setProperty("prism.marlin.clip", "false");
 208
System.setProperty("prism.marlin.clip.runtime.enable", "true");
 209
 210 // enable subdivider:
 211 System.setProperty("prism.marlin.clip.subdivider",
"true");
 212
 213 // disable min length check: always subdivide curves
at clip edges
 214 System.setProperty("prism.marlin.clip.subdivider.minLength",
"-1");
 215
 216 // If any curve, increase curve accuracy:
 217 // curve length max error:
 218 System.setProperty("prism.marlin.curve_len_err", "1e-4");
 219
 220 // cubic min/max error:
 221 System.setProperty("prism.marlin.cubic_dec_d2", "1e-3");
 222 System.setProperty("prism.marlin.cubic_inc_d1",
"1e-4"); // or disabled ~ 1e-6
 223
 224 // quad max error:
 225 System.setProperty("prism.marlin.quad_dec_d2", "5e-4");

It seems better to test with the default parameters (i.e., it
makes a better regression test that way).


I deliberately set all these Marlin clip (runtime + always subdivider) 
/ curve quality settings (quads / cubics thresholds) to be sure of the 
concrete Marlin setup as quality thresholds are sensitive to such values.


The ClipShapeTest is dedicated to test the clipper (+ subdivider) part 
of the Marlin renderer.




3. Related to that, I think you should eliminate the following (I
don't recommend running functional tests with this set), although
since you don't do any animation, it probably doesn't matter.

 227 System.setProperty("javafx.animation.fullspeed",
"true"); // full speed


I will remove it and see if the overall test is not slower.
Is Platform.runLater() impacted by such setting (latency of FX thread 
-> Prism rendering thread ?) ?


Laurent



On 6/8/2018 7:28 AM, Laurent Bourgès wrote:

Hi,

Please review this large patch to upgrade MarlinFX to 0.9.2 in
OpenJFX11:
JBS: