[WIP] (refactor) Use Java-version-independent adapter for internal API #429

JordanMartinez · 2017-09-04T21:07:44Z

There's a number of issues with the current state of this PR only because I'm not yet familiar with how to set up some things with Java 9 or integrate it into this project's somewhat complex build system.

Tests don't work correctly due to not having access to things
Javadoc warns of bad javadoc (using & instead & and > instead of >)
CI files need to be updated

At least that was the issues it raised. I changed something in the build file and now it doesn't work due to

* What went wrong:
Execution failed for task ':testfx-internal-java9:javadoc'.
> Could not write to file 
    'TestFX/subprojects/testfx-internal-java9/build/tmp/javadoc/javadoc.options'.

JordanMartinez · 2017-09-04T21:10:28Z

@brcolow Here's your comparison to #425

brcolow · 2017-09-05T17:15:24Z

Just trying to see if we can get the build to work, can you try changing the _JAVA_OPTIONS for .travis.yml for the JDK9 build to:

- _JAVA_OPTIONS="--add-opens=java.base/java.util=ALL-UNNAMED --add-exports=javafx.graphics/com.sun.glass.ui=ALL-UNNAMED"

Also the Window iterator order returned by the Java 9 interface implementation should be reversed. That should make the build pass, but we will see.

JordanMartinez · 2017-09-05T20:03:28Z

can you try changing the _JAVA_OPTIONS for .travis.yml for the JDK9 build to:

It's already that. I rebased this on top of the latest commit you pushed that had the change.

Looks like this comment shows how to fix a related issue. I'm not sure if that will be the same for us or not. Thoughts?

Also the Window iterator order returned by the Java 9 interface implementation should be reversed.

Ah, didn't know that.

JordanMartinez · 2017-09-05T20:25:27Z

This is what is written in the javadoc.options file

-Xdoclint:none '-quiet'
-author 
-charSet 'UTF-8'
-d '/home/jordan/Git Projects/TestFX/subprojects/testfx-internal-java9/build/docs/javadoc'
-doctitle 'TestFX Internal Java 9 4.0.7-alpha API'
-encoding 'UTF-8'
-footer 'Copyright &copy; 2013-2016 The TestFX Contributors. All rights reserved.'

JordanMartinez · 2017-09-05T20:57:52Z

Turns out the javadoc task was failing because the given URL to Java 9 javadoc links weren't correct. Once these were updated, it would build.... until it got to the test classes ;-) So, I added read access and it works now. However, there are still the 2 flaky WindowFinderImplTest that prevented me from adding read access to JUnit 4/5. (This may be due to the windows list not being reversed yet)

brcolow · 2017-09-06T00:09:51Z

We must change the logging verbosity in ./ci/script.sh:

# Increase build log verbosity for JDK 9.
if [ "$TRAVIS_JDK_VERSION" == "oraclejdk9" ]; then
  ./gradlew build --debug --stacktrace
else

Just remove the conditional and have them both build with --info.

brcolow · 2017-09-06T00:12:52Z

Not really related to this PR but just noticed it because of you posting the contents of javadoc.options, might as well change the javadoc.footer in gradle.properties:

To 2017. It would be nice to use a templating feature so that the current year is always used, but that's probably more trouble than it's worth.

JordanMartinez · 2017-09-06T01:04:37Z

It built, but it's failing on 2 tests (same thing happened on my on local machine). If I reverse the iterator, via

List<Window> windows = Window.getWindows();
Collections.reverse(windows);
return windows.iterator();

then 19 tests fail. So I don't think the reverse part is the issue, but correct me if I'm wrong.

JordanMartinez · 2017-09-06T01:06:11Z

First failure

org.testfx.service.finder.impl.WindowFinderImplTest > window_windowIndex FAILED
    java.lang.AssertionError: 
    Expected: is <javafx.stage.Stage@23359ada>
         but: was <javafx.stage.Stage@372bce1d>
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:8)
        at org.testfx.service.finder.impl.WindowFinderImplTest.window_windowIndex(WindowFinderImplTest.java:162)

Second failure

org.testfx.service.finder.impl.WindowFinderImplTest > targetWindow_windowIndex FAILED
    java.lang.AssertionError: 
    Expected: is <javafx.stage.Stage@23359ada>
         but: was <javafx.stage.Stage@372bce1d>
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:8)
        at org.testfx.service.finder.impl.WindowFinderImplTest.targetWindow_windowIndex(WindowFinderImplTest.java:137)

JordanMartinez · 2017-09-06T01:25:42Z

Actually, the reverse thing fails to build because reverse can't be called on an unmodifiable collection. However, once that's fixed, the above two tests still fail:

First failure on my machine

org.testfx.service.finder.impl.WindowFinderImplTest > window_windowIndex STANDARD_OUT
    Last target window is: null
    windows = [
        javafx.stage.Stage@16fdccba, 
        javafx.stage.Stage@50e43cf2, 
        javafx.stage.Stage@66ac14f0, 
        javafx.stage.Stage@daa41e4, 
        javafx.stage.Stage@380c429b, 
        javafx.stage.Stage@497eeac4
    ]

org.testfx.service.finder.impl.WindowFinderImplTest > window_windowIndex FAILED
    java.lang.AssertionError: 
    Expected: is <javafx.stage.Stage@66ac14f0>
         but: was <javafx.stage.Stage@50e43cf2>
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:8)
        at org.testfx.service.finder.impl.WindowFinderImplTest.window_windowIndex(WindowFinderImplTest.java:162)

Second failure on my machine

org.testfx.service.finder.impl.WindowFinderImplTest > targetWindow_windowIndex STANDARD_OUT
    Last target window is: null
    windows = [
        javafx.stage.Stage@16fdccba, 
        javafx.stage.Stage@50e43cf2, 
        javafx.stage.Stage@66ac14f0, 
        javafx.stage.Stage@daa41e4, 
        javafx.stage.Stage@380c429b, 
        javafx.stage.Stage@497eeac4
    ]

org.testfx.service.finder.impl.WindowFinderImplTest > targetWindow_windowIndex FAILED
    java.lang.AssertionError: 
    Expected: is <javafx.stage.Stage@66ac14f0>
         but: was <javafx.stage.Stage@50e43cf2>
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:8)
        at org.testfx.service.finder.impl.WindowFinderImplTest.targetWindow_windowIndex(WindowFinderImplTest.java:137)

brcolow · 2017-09-06T01:58:17Z

Sorry I was mis-remembering a bit. To get those tests to pass I changed:

     @SuppressWarnings("deprecation")
    private List<Window> fetchWindowsInQueue() {
        List<Window> windows = Lists.newArrayList(Window.impl_getWindows());
        return ImmutableList.copyOf(Lists.reverse(windows));
    }

To:

 private List<Window> fetchWindowsInQueue() {
        return ImmutableList.copyOf(Lists.newArrayList(Window.getWindows()));
    }

For the JDK 9 implementation. In other words I removed reversal.

JordanMartinez · 2017-09-06T02:04:28Z

Still fails:

org.testfx.service.finder.impl.WindowFinderImplTest > window_windowIndex STANDARD_OUT
    Last target window is: null
    windows = [javafx.stage.Stage@347b170d, javafx.stage.Stage@6fe8dcec, javafx.stage.Stage@29e55e60, javafx.stage.Stage@3ff5778b, javafx.stage.Stage@42f630af, javafx.stage.Stage@6b508ba1]

org.testfx.service.finder.impl.WindowFinderImplTest > window_windowIndex FAILED
    java.lang.AssertionError: 
    Expected: is <javafx.stage.Stage@29e55e60>
         but: was <javafx.stage.Stage@6fe8dcec>
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:8)
        at org.testfx.service.finder.impl.WindowFinderImplTest.window_windowIndex(WindowFinderImplTest.java:162)

org.testfx.service.finder.impl.WindowFinderImplTest > targetWindow_window PASSED

org.testfx.service.finder.impl.WindowFinderImplTest > window_scene PASSED

org.testfx.service.finder.impl.WindowFinderImplTest > targetWindow_windowIndex STANDARD_OUT
    Last target window is: null
    windows = [javafx.stage.Stage@347b170d, javafx.stage.Stage@6fe8dcec, javafx.stage.Stage@29e55e60, javafx.stage.Stage@3ff5778b, javafx.stage.Stage@42f630af, javafx.stage.Stage@6b508ba1]

org.testfx.service.finder.impl.WindowFinderImplTest > targetWindow_windowIndex FAILED
    java.lang.AssertionError: 
    Expected: is <javafx.stage.Stage@29e55e60>
         but: was <javafx.stage.Stage@6fe8dcec>
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:8)
        at org.testfx.service.finder.impl.WindowFinderImplTest.targetWindow_windowIndex(WindowFinderImplTest.java:137)

JordanMartinez · 2017-09-06T02:10:40Z

Also, since targetWindow is null in the following code:

private List<Window> orderWindowsByProximityTo(Window targetWindow,
                                                   List<Window> windows) {
        List<Window> copy = new ArrayList<>(windows);
        copy.sort(Comparator.comparingInt(w -> calculateWindowProximityTo(targetWindow, w)));
        return Collections.unmodifiableList(copy);
    }

    private int calculateWindowProximityTo(Window targetWindow,
                                           Window window) {
        if (window == targetWindow) {
            return 0;
        }
        if (isOwnerOf(window, targetWindow)) {
            return 1;
        }
        return 2;
    }

it will return 2, not 0 or 1. Is that supposed to happen? I thought comparator returns -1, 0, or 1, not 2.

brcolow · 2017-09-06T02:29:29Z

It is okay for Comparator to return any negative, zero, or positive value as long as it is an equivalence relation which would take some thought to investigate if our code is one or not. I will investigate the test failures in more detail tomorrow.

brcolow · 2017-09-06T02:35:40Z

subprojects/testfx-core/src/main/java/org/testfx/service/finder/impl/WindowFinderImpl.java

@@ -138,7 +140,7 @@ public Window window(Node node) {

    @SuppressWarnings("deprecation")
    private List<Window> fetchWindowsInQueue() {
-        List<Window> windows = Lists.newArrayList(Window.impl_getWindows());
+        List<Window> windows = Lists.newArrayList(getWindows());
        return ImmutableList.copyOf(Lists.reverse(windows));


Remove the reversal here, and let the JDK 8 implementation return it in reverse, and the JDK 9 implementation not reversed.

Ok, looks like this is another method to add to JavaVersionAdapter.

Also, when I did remove the reversal and ran the tests solo, I believe they worked fine. When I run it in context of an entire gradle build, that's when it fails. So, it might be more than just that going on here.

brcolow · 2017-09-06T02:37:19Z

subprojects/testfx-core/src/main/java/org/testfx/util/NodeQueryUtils.java

@@ -250,7 +252,7 @@ private static boolean matchesNodeMatcher(Node node,

    @SuppressWarnings("deprecation")
    private static boolean isNodeVisible(Node node) {
-        if (!node.isVisible() || !node.impl_isTreeVisible()) {
+        if (isNotVisible(node)) {


I did the same thing in my PR, don't really know what the implications are of removing the isTreeVisible() condition but there is no way to check that in Java 9 so I don't think it should be relied upon for anything. Would maybe be a candidate for the JavaFX devs to move into the public API but that is outside our control.

I'd guess it only matters if one is using a Tree-related control, a component which I'm not sure we test for in TestFX or not. For all we know, they could have implemented whatever code that was into the Tree node itself... Otherwise, yes, outside of our control

JordanMartinez · 2017-09-07T04:36:15Z

When I looked at the output, there were 6 windows stored in Window.getWindows(). There's only 5 in the test class, so I knew the extra window, which comes at index 1, was coming from another test. I added a change listener to Window.getWindows() and printed all windows whenever a new one was added.

Turns out BoundsQueryUtilsTest doesn't clean up its stage after its tests run. The build has progressed further, but it still fails due to check style running on the internal9 subproject. Error message below:

> Task :testfx-internal-java9:checkstyleMain
Putting task artifact state for task ':testfx-internal-java9:checkstyleMain' into context took 0.01 secs.
Executing task ':testfx-internal-java9:checkstyleMain' (up-to-date check took 0.005 secs) due to:
  No history is available.
[ant:checkstyle] Running Checkstyle 7.6.1 on 2 files

:testfx-internal-java9:checkstyleMain (Thread[Task worker for ':' Thread 3,5,main]) completed. Took 0.128 secs.

FAILURE: Build failed with an exception.

* What went wrong:
Execution failed for task ':testfx-internal-java9:checkstyleMain'.
> Unable to process files: [
    TestFX/subprojects/testfx-internal-java9/src/main/java/
        module-info.java,
    TestFX/subprojects/testfx-internal-java9/src/main/java/
        org/testfx/service/adapter/JavaVersionAdapter.java
]

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

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

Prevents the stage from later interfering with other future tests

JordanMartinez · 2017-09-07T05:13:06Z

K, now it gets to JUnit 5 and fails with:

  JUnit Jupiter:ApplicationStartTest:should_click_on_button()
    MethodSource [className = 'org.testfx.cases.acceptance.ApplicationStartTest', methodName = 'should_click_on_button', methodParameterTypes = '']
    => java.lang.AssertionError: 
Expected: Node has text "clicked!"
     but: was <Button@5af7e405[styleClass=button]'click me!'>

Test run finished after 2808 ms
[         5 containers found      ]
[         1 containers skipped    ]
[         4 containers started    ]
[         0 containers aborted    ]
[         4 containers successful ]
[         0 containers failed     ]
[        10 tests found           ]
[         4 tests skipped         ]
[         6 tests started         ]
[         0 tests aborted         ]
[         5 tests successful      ]
[         1 tests failed          ]

:testfx-junit5:junitPlatformTest (Thread[Task worker for ':' Thread 5,5,main]) completed. Took 3.895 secs.

FAILURE: Build failed with an exception.

* What went wrong:
Execution failed for task ':testfx-junit5:junitPlatformTest'.
> Process 'command '/usr/lib/jvm/java-9-oracle/bin/java'' finished with non-zero exit value 1

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

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

When I ran it separately, it worked fine, so I think there's an issue somewhere else that causes the problem.

JordanMartinez · 2017-09-07T05:15:28Z

When I ran the full build, it seemed like there is one test in the JUnit5 tests where the mouse is supposed to move towards a button and click on it, but that did not happen. Even the fired events below show that its clicking on the stack pane, not the button:

Failures (1):
  JUnit Jupiter:ApplicationStartTest:should_click_on_button()
    MethodSource [className = 'org.testfx.cases.acceptance.ApplicationStartTest', methodName = 'should_click_on_button', methodParameterTypes = '']
    => java.lang.AssertionError: 
Expected: Node has text "clicked!"
     but: was <Button@20be827a[styleClass=button]'click me!'>

   Context:
      Test image saved at:
         TestFX/subprojects/testfx-junit5/testfx-test - 0.png
      Keys pressed at test failure:
      Mouse Buttons pressed at test failure:
      Fired events since test began:
         WindowEvent [source = javafx.stage.Stage@5db6b9cd, target = javafx.stage.Stage@5db6b9cd, eventType = WINDOW_SHOWING, consumed = false]
         WindowEvent [source = javafx.stage.Stage@5db6b9cd, target = javafx.stage.Stage@5db6b9cd, eventType = WINDOW_SHOWN, consumed = false]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = javafx.scene.Scene@210ab13f, eventType = MOUSE_ENTERED_TARGET, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = NONE, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_ENTERED_TARGET, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = NONE, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_MOVED, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = NONE, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_MOVED, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = NONE, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_MOVED, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = NONE, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_PRESSED, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = PRIMARY, primaryButtonDown, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_RELEASED, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = PRIMARY, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]
         MouseEvent [source = javafx.stage.Stage@5db6b9cd, target = StackPane@58f672b1[styleClass=root], eventType = MOUSE_CLICKED, consumed = false, x = 49.0, y = 27.0, z = 0.0, button = PRIMARY, pickResult = PickResult [node = StackPane@58f672b1[styleClass=root], point = Point3D [x = 49.0, y = 27.0, z = 0.0], distance = 186.60254037844388]

Test run finished after 3060 ms
[         5 containers found      ]
[         1 containers skipped    ]
[         4 containers started    ]
[         0 containers aborted    ]
[         4 containers successful ]
[         0 containers failed     ]
[        10 tests found           ]
[         4 tests skipped         ]
[         6 tests started         ]
[         0 tests aborted         ]
[         5 tests successful      ]
[         1 tests failed          ]


:testfx-junit5:junitPlatformTest (Thread[Task worker for ':' Thread 2,5,main]) completed. Took 3.897 secs.

JordanMartinez · 2017-09-08T02:54:48Z

Everything passed except the flaky test on one of the Windows' builds. What else needs to be done to clean this up? How should I exclude module-info.java from the check style task?

JordanMartinez · 2017-09-08T20:39:43Z

I think this should be merged once the non-flaky tests pass. The findbugs task doesn't appear to work with Java 9 yet (maybe it requires a different version?). Since it needs further investigation and since the build files should be cleaned up later, it's outside the scope of this PR.

brcolow · 2017-09-09T06:55:22Z

Great work @JordanMartinez - thanks so much.

JordanMartinez · 2017-09-09T18:56:46Z

Woot!

JordanMartinez · 2017-10-31T21:13:50Z

Just FYI. Someone extended the experimental Java 9 plugin for Gradle and started work on what they call Chainsaw.

(refactor) Use Java version independent adapter for internal API

3d16d77

JordanMartinez changed the title ~~[WIP] (refactor) Use Java version independent adapter for internal API~~ [WIP] (refactor) Use Java-version-independent adapter for internal API Sep 4, 2017

JordanMartinez added 2 commits September 5, 2017 13:52

(refactor) Use Java-version-specific javadoc links

5a16778

(refactor) Add module access info to test classes

50ce3a4

JordanMartinez force-pushed the jdk8and9 branch 2 times, most recently from 5eee926 to 50ce3a4 Compare September 5, 2017 21:04

(refactor) Use ./gradlew build --info for both Java versions

3f1662b

brcolow reviewed Sep 6, 2017

View reviewed changes

JordanMartinez mentioned this pull request Sep 6, 2017

Deflake flaky tests #411

Closed

(fix) Clean up stage by closing it to prevent conflict with other tests

f944f56

JordanMartinez added 3 commits September 6, 2017 21:40

(refactor) exclude check style on internal java 9 subproject

39c29d4

(refactor) Add license header to module

8b81875

(refactor) Close rule's stage by default when stop is called

21e7046

Prevents the stage from later interfering with other future tests

JordanMartinez added 2 commits September 6, 2017 22:23

(refactor) Reverse window list for Java 8, don't for Java 9

f5114f5

(refactor) Include debug info

0422388

JordanMartinez force-pushed the jdk8and9 branch from dca20df to 0422388 Compare September 7, 2017 05:25

brcolow mentioned this pull request Sep 8, 2017

[WIP] WIP Try splitting off jdk9 sub-project. #425

Closed

(refactor) Run checkstyle on Java 9 builds by ignoring module-info.java

6cb3cb8

brcolow merged commit 2fb6690 into TestFX:master Sep 9, 2017

JordanMartinez deleted the jdk8and9 branch September 9, 2017 18:56

JordanMartinez mentioned this pull request Sep 26, 2017

Allow RichtTextFX to access Java9 hidden methods FXMisc/RichTextFX#593

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP] (refactor) Use Java-version-independent adapter for internal API #429

[WIP] (refactor) Use Java-version-independent adapter for internal API #429

JordanMartinez commented Sep 4, 2017 •

edited

JordanMartinez commented Sep 4, 2017 •

edited

brcolow commented Sep 5, 2017

JordanMartinez commented Sep 5, 2017

JordanMartinez commented Sep 5, 2017

JordanMartinez commented Sep 5, 2017 •

edited

brcolow commented Sep 6, 2017

brcolow commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017 •

edited

brcolow commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

brcolow commented Sep 6, 2017 •

edited

brcolow Sep 6, 2017

JordanMartinez Sep 6, 2017

brcolow Sep 6, 2017

JordanMartinez Sep 6, 2017

JordanMartinez commented Sep 7, 2017

JordanMartinez commented Sep 7, 2017

JordanMartinez commented Sep 7, 2017

JordanMartinez commented Sep 8, 2017 •

edited

JordanMartinez commented Sep 8, 2017

brcolow commented Sep 9, 2017

JordanMartinez commented Sep 9, 2017

JordanMartinez commented Oct 31, 2017

[WIP] (refactor) Use Java-version-independent adapter for internal API #429

[WIP] (refactor) Use Java-version-independent adapter for internal API #429

Conversation

JordanMartinez commented Sep 4, 2017 • edited

JordanMartinez commented Sep 4, 2017 • edited

brcolow commented Sep 5, 2017

JordanMartinez commented Sep 5, 2017

JordanMartinez commented Sep 5, 2017

JordanMartinez commented Sep 5, 2017 • edited

brcolow commented Sep 6, 2017

brcolow commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017 • edited

brcolow commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

JordanMartinez commented Sep 6, 2017

brcolow commented Sep 6, 2017 • edited

brcolow Sep 6, 2017

Choose a reason for hiding this comment

JordanMartinez Sep 6, 2017

Choose a reason for hiding this comment

brcolow Sep 6, 2017

Choose a reason for hiding this comment

JordanMartinez Sep 6, 2017

Choose a reason for hiding this comment

JordanMartinez commented Sep 7, 2017

JordanMartinez commented Sep 7, 2017

JordanMartinez commented Sep 7, 2017

JordanMartinez commented Sep 8, 2017 • edited

JordanMartinez commented Sep 8, 2017

brcolow commented Sep 9, 2017

JordanMartinez commented Sep 9, 2017

JordanMartinez commented Oct 31, 2017

JordanMartinez commented Sep 4, 2017 •

edited

JordanMartinez commented Sep 4, 2017 •

edited

JordanMartinez commented Sep 5, 2017 •

edited

JordanMartinez commented Sep 6, 2017 •

edited

brcolow commented Sep 6, 2017 •

edited

JordanMartinez commented Sep 8, 2017 •

edited