JogAmp IRC Chat Logs

#jogamp @ irc.freenode.net - 20151009 05:05:16 (UTC)

20151009 05:05:16 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20151008050516.html
20151009 05:05:16 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20151009050516.html
20151009 05:05:42 * i-make-robots (~i-make-ro@anon) Quit ()
20151009 05:38:07 <xranby_f22> good morning!.. wow testing went from 90 something minutes down to 35 minutes to run all tests on my machine after this nights changes!
20151009 05:44:53 <xranby_f22> <testcase classname="com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT" name="testScreenModeChange01" time="2.681">
20151009 05:44:53 <xranby_f22> <error type="java.lang.NullPointerException">java.lang.NullPointerException
20151009 05:44:53 <xranby_f22> at com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT.testScreenModeChange01(TestScreenMode00cNEWT.java:137)
20151009 05:45:00 <xranby_f22> final List<MonitorMode> allMonitorModes = screen.getMonitorModes();
20151009 05:45:00 <xranby_f22> Assert.assertTrue(allMonitorModes.size()>0);
20151009 05:45:14 <xranby_f22> allMonitorModes is null?
20151009 05:46:33 * xranby_f22 scratches head
20151009 05:48:10 <xranby_f22> <testcase classname="com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasSWTAWT" name="testHoldOneButtonAndPressAnother" time="0.739">
20151009 05:48:10 <xranby_f22> <error type="java.lang.NullPointerException">java.lang.NullPointerException
20151009 05:48:10 <xranby_f22> at com.jogamp.nativewindow.swt.SWTAccessor.invoke(SWTAccessor.java:497)
20151009 05:50:00 <xranby_f22> public static void invoke(final org.eclipse.swt.widgets.Display display, final boolean wait, final Runnable runnable) {
20151009 05:50:00 <xranby_f22> if( display.isDisposed() || Thread.currentThread() == display.getThread() ) {
20151009 05:50:00 <xranby_f22>
20151009 05:50:08 <xranby_f22> display is null?
20151009 05:50:15 * xranby_f22 scratches head
20151009 05:54:28 <xranby_f22> those two above are the only NullPointerExceptions seen when testing on my machine
20151009 05:54:56 <xranby_f22> [familjen@localhost results]$ grep -r NullPointerException -A 2 *
20151009 05:54:56 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasSWTAWT.xml: <error type="java.lang.NullPointerException">java.lang.NullPointerException
20151009 05:54:56 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasSWTAWT.xml- at com.jogamp.nativewindow.swt.SWTAccessor.invoke(SWTAccessor.java:497)
20151009 05:54:56 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasSWTAWT.xml- at com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasSWTAWT.eventDispatchImpl(TestNewtEventModifiersNewtCanvasSWTAWT.java:75)
20151009 05:54:56 <xranby_f22> --
20151009 05:54:56 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT.xml: <error type="java.lang.NullPointerException">java.lang.NullPointerException
20151009 05:54:56 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT.xml- at com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT.testScreenModeChange01(TestScreenMode00cNEWT.java:137)
20151009 05:54:57 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT.xml- at com.jogamp.newt.util.MainThread$UserApp.run(MainThread.java:207)
20151009 05:58:29 <xranby_f22> ~~~
20151009 05:58:52 <xranby_f22> Good news, no *zero* newt.parenting Failures seen on my machine :D
20151009 06:04:35 <xranby_f22> ok so what is the output from TestScreenMode00cNEWT
20151009 06:05:17 <xranby_f22> ++++ TestCase.setUp: com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT - testScreenModeChange01 - Test not supported on this platform.
20151009 06:05:17 <xranby_f22> Test.0: Window bounds : [ 0 / 0 0 x 0 ] [wu] within [ 0 / 0 0 x 0 ] [wu]
20151009 06:05:17 <xranby_f22> Test.0: Window size : 0 x 0 [pixels]
20151009 06:05:17 <xranby_f22> Test.0: Screen viewport : [ 0 / 0 0 x 0 ] [pixels]
20151009 06:05:30 <xranby_f22> Test not supported on this platform. <-- huh?
20151009 06:05:37 <xranby_f22> ok maybe a false test then?
20151009 06:07:34 <xranby_f22> if( !manualTest || NativeWindowFactory.TYPE_X11 != NativeWindowFactory.getNativeWindowType(true) ) {
20151009 06:07:34 <xranby_f22> setTestSupported(false);
20151009 06:12:59 <xranby_f22> list of test FailingAssertions is also impressively small
20151009 06:15:18 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.xml: <failure message="expected:<false> but was:<true>" type="junit.framework.AssertionFailedError">junit.framework.AssertionFailedError: expected:<false> but was:<true>
20151009 06:15:18 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.xml- at com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.testDisplayCreate02Impl(TestDisplayLifecycle02NEWT.java:299)
20151009 06:15:18 <xranby_f22> TEST-com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.xml- at com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.testDisplayCreate02(TestDisplayLifecycle02NEWT.java:357)
20151009 06:15:41 <xranby_f22> // just make the Window invisible
20151009 06:15:41 <xranby_f22> window1.setVisible(false);
20151009 06:15:41 <xranby_f22> Assert.assertEquals(true,window1.isNativeValid());
20151009 06:15:41 <xranby_f22> Assert.assertEquals(false,window1.isVisible());
20151009 06:16:02 <xranby_f22> thats the only one.. after running all tests
20151009 06:17:19 <xranby_f22> evil...
20151009 06:17:49 <xranby_f22> why do all newt.parenting tests pass while this one in newt fail?
20151009 06:25:32 * xranby_f22 scratches head
20151009 06:25:43 <xranby_f22> so i can see three tests that needs to look into
20151009 06:26:24 <xranby_f22> com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT
20151009 06:26:35 <xranby_f22> com.jogamp.opengl.test.junit.newt.mm.TestScreenMode00cNEWT
20151009 06:26:40 <xranby_f22> com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasSWTAWT
20151009 06:27:53 <xranby_f22> can someone else than me reproduce the above exceptions and asserts?
20151009 06:29:40 <xranby_f22> sgothel: reading that you added a new test flag... (01:55:58) sgothel: src/newt/native/X11Common.h: enable the VERBOSE_ON <- if you encounter an issue
20151009 06:29:52 <xranby_f22> will do!
20151009 06:30:39 * netlike (~netlike@anon) has joined #jogamp
20151009 06:35:35 <xranby_f22> https://jogamp.org/chuck/job/jogl/lastCompletedBuild/testReport/ <- jogl buildbots do not reproduce the things i see on my machine which is of course bigsuccess!
20151009 06:37:32 <xranby_f22> i will soon start wearing a thinfoil hat blaiming all errors i see on the geomagnetic storm that is all time high right now
20151009 06:40:28 <netlike> I think tinfoil would actually not block anything, but actually increase reception of said alien/storm interference? haha =)
20151009 06:43:08 <xranby_f22> great lucid dreams maybe :)
20151009 06:43:52 <xranby_f22> http://www.aurora-service.eu/aurora-forecast/ <- i live in scandinavia..
20151009 06:45:28 <netlike> nice website, very cool. I live in South Africa.. its so hot here currently its hard to focus on anything..
20151009 06:45:30 <xranby_f22> this is data from the ACE spacecraft http://www.aurora-service.eu/aurora-school/nasas-ace-spacecraft/
20151009 06:47:48 <xranby_f22> netlike: oh, i have a question, do you recall people playing a lot of music by the artist Sixto Rodriguez on vinyl where you are at?
20151009 06:48:20 <netlike> so do you actually see the aurora at night or am I miss understanding you xranby_f22? because that would be awesome! its 35ºC today.. with no wind and no chance of rain..
20151009 06:48:50 <netlike> no I don't know Sixto, maybe a little before my time
20151009 06:49:13 <xranby_f22> i looked out but no at the longitude i am at right now i could not see anything.. people living in norway and iceland are more lucky
20151009 06:49:39 <xranby_f22> err .. latitude i mean
20151009 06:49:48 * elect (~GBarbieri@anon) has joined #jogamp
20151009 06:50:07 <xranby_f22> some people in UK have been able to capture photos from this years nothern lights
20151009 06:50:18 <xranby_f22> due to the storms being larger than usual
20151009 06:51:23 <xranby_f22> https://www.facebook.com/photo.php?fbid=10153737068378982&set=a.10153737068083982.1073741948.787463981&type=3&theater <- i dont know if you can see this.. but this picture was taken by one of the openjdk developers living in UK
20151009 06:52:20 <xranby_f22> like, his camera was able to capture it while his eyes was not
20151009 06:53:02 <netlike> nope that page seems to have expired. anyway let me get back to work, chat laterz xranby_f22! have a good day!
20151009 06:53:26 <xranby_f22> netlike: Sixto Rodriguez is nicely portraited in the documentary movie Searching for Sugar Man http://www.imdb.com/title/tt2125608/
20151009 06:55:19 <elect> hi
20151009 06:56:19 <elect> rmk0, nothing, he is fine here right now :s
20151009 07:00:33 <xranby_f22> netlike: you too, have a good day!
20151009 07:10:27 * monsieur_max (~maxime@anon) has joined #jogamp
20151009 07:21:52 * doev (~doev@anon) Quit (Quit: Verlassend)
20151009 07:24:34 * eclesia (~husky@anon) has joined #jogamp
20151009 07:24:40 <eclesia> good morning
20151009 07:42:52 <xranby_f22> eclesia: good morning!
20151009 08:07:47 * elect (~GBarbieri@anon) Quit (Read error: Connection reset by peer)
20151009 08:08:49 * elect (~GBarbieri@anon) has joined #jogamp
20151009 08:35:12 <elect> https://www.preining.info/blog/2015/10/looking-at-the-facts-sarah-sharps-crusade/
20151009 08:55:57 <xranby_f22> I for one would be happy if people differented between blaming code and blaming people: These two replys to that thread is in my view good conclusions: https://www.preining.info/blog/2015/10/looking-at-the-facts-sarah-sharps-crusade/#comment-17206 https://www.preining.info/blog/2015/10/looking-at-the-facts-sarah-sharps-crusade/#comment-17209
20151009 08:59:17 <xranby_f22> meskarune 's comment is also very goodL https://www.preining.info/blog/2015/10/looking-at-the-facts-sarah-sharps-crusade/#comment-17221
20151009 09:06:32 <elect> but sarah failed hard, she didnt bringt a single proof about the toxic environment she talked about
20151009 09:06:44 <elect> that's the crucial point
20151009 09:14:03 <xranby_f22> do you think there is more proof needed than reading the LKML and seeing that maintainers there uses cursing as a way to lecture?
20151009 09:15:54 <xranby_f22> i get the impression that this debate goes on because ideoligical thinking like if free speech includes freedom to insult.
20151009 09:16:22 <xranby_f22> however I think it is a great loss every time a developer leaves only because people excersize such a privelige
20151009 09:17:47 <xranby_f22> and if a community excersize such a privelige to use harsh language in code reviews, then that is a toxic environment that will drive some developers away
20151009 09:19:25 <xranby_f22> the best way to fix it is to acknowledge that you have a choice to insult the code or the people who wrote the code and the outcome in the community is constructive or toxic depending on if the cursing is addressed towards code or people.
20151009 09:23:26 <guillaum1> hi. I also like the debate about intercultural differences. I agree that there are a big differences between country/civilisations in the way of communicating: how to say you agree or not / how to criticize work of someone else / humor (1st, 2nd degree) is totally different from my job experience.
20151009 09:24:49 <xranby_f22> my standard response will soon be: if you want to be provocative be like Ji Gong. Be kind hearted while you skip following the monastic rules.
20151009 09:27:14 <elect> if you refer to that one where Linus joke about Greg being a giant, it is not
20151009 09:27:49 <xranby_f22> elect: no i refer to that Linus at several occasions declare that he will curse at people
20151009 09:28:00 <xranby_f22> and do curse at people
20151009 09:29:36 <elect> well, we should see case by case
20151009 09:29:59 <elect> a dev in the link I pasted said for example there wasnt a single time when it was unjustified
20151009 09:30:47 <xranby_f22> well take this case for example.. here linus start cusing the comment, and ending up cursing the persion writing it by addressing the largest finnish curse to the "you" person http://staefcraeft.blogspot.se/2013/07/linus-torvalds-there-arent-enough-swear.html
20151009 09:31:18 <xranby_f22> that is what i mean that you have a choice between curing the code and comments and cursing a persion
20151009 09:32:42 <guillaum1> "There are huge differences between Finnish and Americal cultures which can lead to misunderstanding (true for all cultures). And this here is a perfect example: Finish culture is one of the mist direct cultures in the world. Finns don’ beat around the bush." https://www.preining.info/blog/2015/10/looking-at-the-facts-sarah-sharps-crusade/#comment-17216
20151009 09:33:08 <xranby_f22> i do not think it is justified ewven if the code is incorrect and the comment is misleadin
20151009 09:33:39 * bigpet (uid25664@anon) Quit (Ping timeout: 250 seconds)
20151009 09:39:04 <elect> Ok, Linus is not the person who'd define as the tollerance prototype, but based on the reddit comment, it's somehow significative seeing how everybody involved apologies and promises to fix the mistake
20151009 09:39:32 <xranby_f22> there is also situation when the person dissapear and leave the discussion
20151009 09:39:55 <xranby_f22> and the voice of those people are not part of the list
20151009 09:44:17 <xranby_f22> there is also the issue with asking the victim if the victim feels ok, because then you but the burden of defence on the victim
20151009 09:48:49 <elect> hard to say, link is broken and archive as well
20151009 09:49:51 <xranby_f22> what link are you talking about?
20151009 10:01:54 * netlike (~netlike@anon) Quit (Quit: Leaving)
20151009 10:12:26 * bigpet (uid25664@anon) has joined #jogamp
20151009 10:24:21 * elect (~GBarbieri@anon) Quit (Ping timeout: 252 seconds)
20151009 10:34:14 * elect (~GBarbieri@anon) has joined #jogamp
20151009 10:34:27 <elect> the mainlist thread of that Linus intervention
20151009 10:50:51 <xranby> elect: https://web.archive.org/web/20130715181124/https://lkml.org/lkml/2013/7/13/132
20151009 10:51:39 <xranby> someone decided to remove the page in 2015
20151009 10:51:49 <xranby> from the lkml.org site
20151009 11:01:36 <zubzub> 11:14 < xranby_f22> do you think there is more proof needed than reading the LKML and seeing that maintainers there uses cursing as a way to lecture?
20151009 11:01:47 <zubzub> you have no idea how much I curse at my friend (and vice versa)
20151009 11:01:52 <zubzub> *friends
20151009 11:02:23 <zubzub> the real shit it when we're not cursing :)
20151009 11:02:27 <zubzub> then we know it's serious
20151009 11:06:40 <elect> :D
20151009 11:07:08 <xranby> zubzub: the hard line to see is when the way to address your peers puts you inside the realm of verbal abuse
20151009 11:08:40 <xranby> from human social studies verbal abuse is to be avoided because it creates physical unhealth
20151009 11:09:01 <xranby> i feel physical unhealth
20151009 11:09:09 <xranby> therefore i care to have this discussion
20151009 11:09:19 <xranby> because i do not want to feel physical unhealth
20151009 11:12:54 <sgothel> guys ..
20151009 11:13:34 <sgothel> that depends what you consider abuse :)
20151009 11:14:09 <sgothel> story: in basketball team, we usually kicked each other ass .. like 'come on you lazy ass .. get up'
20151009 11:14:21 <sgothel> showing that you do care!
20151009 11:14:40 <sgothel> then later, while studying .. new team says, 'sven, you cannot be so rude' :)
20151009 11:14:49 <sgothel> me: I do care!
20151009 11:15:03 <sgothel> similar discussion w/ parenting ..
20151009 11:15:25 <sgothel> always depends on your upbringing .. IMHO
20151009 11:15:51 <sgothel> reading those emails - IMHO only a moron would think this is abuse .. well, my opinion ofc
20151009 11:15:58 <sgothel> but we all have ours ..
20151009 11:16:07 <zubzub> me, I hate double faced, indirect, beat-around-the-bush conversations. Especially in engineering
20151009 11:16:09 <sgothel> and those people where being ensured - all is good
20151009 11:16:17 <sgothel> right ..
20151009 11:16:17 <zubzub> why?
20151009 11:16:20 <sgothel> so blablablba
20151009 11:16:25 <sgothel> Xerxes: how is 2.3.2 ?
20151009 11:16:36 <sgothel> I would like to let it be 2.3.2
20151009 11:16:37 <xranby> sgothel: 2.3.2 is superb compared to non 2.3.2
20151009 11:16:45 <sgothel> anybody else ?
20151009 11:16:47 <xranby> sgothel: i filed one bugreport
20151009 11:16:49 <zubzub> because if are friendly and don't get your point across, the bad solution will be used just not to offend a person
20151009 11:17:03 <zubzub> I've seen it happen
20151009 11:17:07 <zubzub> again and again and again
20151009 11:17:15 <sgothel> right
20151009 11:17:17 <xranby> sgothel: my fedora f22 system still have one test that still trigger an visibility issue
20151009 11:17:39 <xranby> however, its superior compared to yesterday
20151009 11:17:46 <xranby> hence i am very happy
20151009 11:17:47 <xranby> joy
20151009 11:17:48 <sgothel> w/ my poor english skills in 2005 - I didn't undertstand what 'a polite no' means :)
20151009 11:18:31 <sgothel> then I learned the fine line of 'asking' and 'maybe not' 'better not' :)
20151009 11:18:46 <sgothel> or 'I am concerned' :) hahaha
20151009 11:19:01 <sgothel> I thought like, 'yeah, so you are concerned .. whatever ..'
20151009 11:19:07 <sgothel> have to check this bug!
20151009 11:19:16 <xranby> https://jogamp.org/bugzilla/show_bug.cgi?id=1250 <- added VERBOSE_ON log
20151009 11:19:18 <sgothel> 1250 that is
20151009 11:19:37 <sgothel> b/c I cannot release 2.3.2 w/ such an issue
20151009 11:20:14 <xranby> this is the only test of the 1170 test in jogamp that fail with a visibility issue
20151009 11:20:17 <xranby> tests
20151009 11:20:28 <xranby> makes me wonder why
20151009 11:20:33 <xranby> all parenting tests for example pass
20151009 11:20:39 <sgothel> that is great then .. hmm - yup, will retest here
20151009 11:21:01 <sgothel> and could you manually test w/ '-loopvt 100' or more on those 2 tests ?
20151009 11:21:08 <sgothel> I did w/ 100 a few times ..
20151009 11:21:44 * xranby at work atm.. will run tests using ubuntu + compiz unity to verify tonights fixes on this platform as well
20151009 11:22:00 <xranby> i will be back on my fedora 22 system during the evening
20151009 11:22:30 <sgothel> thank you
20151009 11:22:38 <elect> what do you need, sgothel ?
20151009 11:24:03 <xranby> elect: try pick a machine.. do cd jogl/make ant junit.run when it say everything is ok then unzip the build/jogl-2.3-bmanual-20151007-linux-amd64-test-results-local.7z file
20151009 11:24:11 <elect> ok
20151009 11:24:35 <xranby> and run some grep command on that directly check for Exception , Asserts and the like
20151009 11:24:48 <elect> ok
20151009 11:24:50 <xranby> there is probably some tool available that can sumamrize it as well
20151009 11:25:01 <xranby> the directory is filled with junit xml files
20151009 11:25:06 <elect> on the 2.3.2?
20151009 11:25:07 <xranby> with the test result for each test
20151009 11:25:45 <elect> or I just pull from sven jogl?
20151009 11:25:52 <xranby> pull from sven's jogl
20151009 11:26:10 <xranby> i do think the build bot have produced a build that you may use
20151009 11:26:10 <elect> ok, I'll do this evening if I make it
20151009 11:29:30 <xranby> elect: thank yoU!
20151009 11:30:56 <elect> wait to see if I'm successfull first ^^
20151009 11:32:22 <xranby> i applause your intention
20151009 11:43:15 <sgothel> @Xerxes: last event is LeaveNotify
20151009 11:43:24 <xranby> my computer here at work have managed to get a fun opengl + opengl es setup, gl by nvidias driver and es by mesa3d
20151009 11:43:30 <sgothel> looks like we have to check visibility state here as well
20151009 11:44:01 <xranby> both work however
20151009 11:44:04 <sgothel> (a WM bug though .. should update that state at ConfigureNotify)
20151009 11:44:32 <sgothel> the log helps a lot
20151009 11:46:01 <sgothel> so Leave|Enter Notify .. as well
20151009 11:46:15 <sgothel> was it reproducible?
20151009 11:46:26 <sgothel> or 'just one in a million' ?
20151009 11:46:53 <xranby> it is reproducible
20151009 11:47:24 <sgothel> lunch now .. errands - will look after it later on ..
20151009 11:47:40 <sgothel> maybe there is an event .. more determined .. hmm
20151009 11:47:52 <sgothel> but again, it should be ConfigureNotify
20151009 11:49:34 <xranby> i can reproduce it both from comand line and from inside eclipse every time on my f22 ystem
20151009 12:01:36 <eclesia> new demo coming up : 4layers frame : decoration > 3d scene > widget panel > cursor(not visible here http://postimg.org/image/4wcsvsq4b/
20151009 12:08:56 <xranby> eclesia: :)
20151009 12:09:13 <xranby> [exec] Everything is Ok <- the most misleading thing in jogl hehe
20151009 12:09:13 <xranby> BUILD SUCCESSFUL
20151009 12:09:13 <xranby> Total time: 40 minutes 37 seconds
20151009 12:09:29 <xranby> thus is the output from ant junit.run
20151009 12:10:18 <xranby> at least it show that your system has not locked up
20151009 12:10:46 <xranby> lets investigate
20151009 12:10:49 <xranby> 7z x build/jogl-2.3-bmanual-20151009-linux-amd64-test-results-local.7z
20151009 12:14:32 <xranby> ./TEST-com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.xml: <failure message="expected:<false> but was:<true>" type="junit.framework.AssertionFailedError">junit.framework.AssertionFailedError: expected:<false> but was:<true>
20151009 12:14:32 <xranby> ./TEST-com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.xml- at com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.testDisplayCreate02Impl(TestDisplayLifecycle02NEWT.java:299)
20151009 12:14:32 <xranby> ./TEST-com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.xml- at com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.testDisplayCreate02(TestDisplayLifecycle02NEWT.java:357)
20151009 12:15:12 <xranby> ok this is https://jogamp.org/bugzilla/show_bug.cgi?id=1250 triggering on ubuntu + compiz unity
20151009 12:22:12 * xranby enable the VERBOSE_ON at src/newt/native/X11Common.h and retests
20151009 12:28:43 <xranby> eclesia: is this a top level window undecorated newt window?
20151009 12:29:00 <eclesia> yes
20151009 12:29:22 <xranby> eclesia: :)
20151009 12:30:09 <xranby> do draging the window feel natural?
20151009 12:30:26 <xranby> you are then doing the work of the window manager
20151009 12:30:28 <xranby> ?
20151009 12:31:27 <eclesia> dragging if fine if the mouse don't moves out of the frame border (that's a limitation on my side, not jgol)
20151009 12:31:41 <sgothel> @Eclesia: good stuff - I assume its alpha in the corners
20151009 12:31:43 <eclesia> but resizing is a total mess on xfce
20151009 12:31:59 <sgothel> next step: one desktop sized window .. w/ virtual windows inside :)
20151009 12:32:12 <sgothel> oh .. resize .. hmm
20151009 12:32:50 <eclesia> the resize call moves the frame anywhre on the screen :/
20151009 12:32:59 <eclesia> + crashes from time to time
20151009 12:33:01 <sgothel> that is a bug!
20151009 12:33:17 <sgothel> you probably have autopos on ..
20151009 12:33:22 <sgothel> i.e. no position set yet
20151009 12:33:30 <sgothel> hmm
20151009 12:33:41 <eclesia> don't know, works fine on windows
20151009 12:34:13 <sgothel> so programmatic resize of undecorated window on X11 moves window?
20151009 12:34:42 <eclesia> decorated or not it messes up with xfce
20151009 12:35:18 <sgothel> so: programmatic resize of window on X11 moves window?
20151009 12:35:30 <sgothel> using xfce WM
20151009 12:35:32 <sgothel> ?
20151009 12:35:38 <sgothel> is that the issue?
20151009 12:35:42 <eclesia> sgothel: don't you remember, you made me run some particular tests once, asking if the frame was 'jumping' everywhere
20151009 12:36:00 <eclesia> sgothel: es it's related to xfce wm
20151009 12:36:02 <eclesia> yes*
20151009 12:36:03 <sgothel> be formal now - answer question
20151009 12:36:06 <sgothel> good
20151009 12:36:18 <sgothel> so I have to check this w/ gnome + KDE
20151009 12:36:31 <sgothel> if reproducible .. I will fix, otherwise .. hmm
20151009 12:36:41 <sgothel> but IMHO .. xfce is a good WM
20151009 12:37:02 <sgothel> simple and correct .. I should check
20151009 12:37:16 <eclesia> thanks :)
20151009 12:37:39 <sgothel> good .. so both earmarked for today - before 2.3.2 .. officially .. hmm, have to remove binaries
20151009 12:37:39 <eclesia> back to work for me
20151009 13:14:18 * monsieur_max uses xfce too
20151009 13:52:17 <eclesia> the perfect desktop, small,fast and nice
20151009 13:53:54 * monsieur_max could not agree more
20151009 14:04:48 * guillaum1 uses dwm
20151009 14:05:34 <guillaum1> 20 kB binary
20151009 14:08:51 * guillaum1 uses also xfce when not trying esoteric WMs
20151009 14:23:08 <eclesia> guillaum1: it uses a lot of memory ?
20151009 14:34:39 <guillaum1> eclesia: ps command says 168 kB on its own (RSS) and 28 Mb virtual size (caused by libX11 linking)
20151009 15:04:17 * xranby_f22 (~familjen@anon) Quit (Ping timeout: 250 seconds)
20151009 15:04:49 <elect> xfce makes sense only on very weak computers
20151009 15:04:51 <elect> cmon
20151009 15:07:17 * xranby_f22 (~familjen@anon) has joined #jogamp
20151009 15:07:58 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20151009 15:09:56 <eclesia> elect: no so weak, have a dual CPU i7 quadcore here for work, with xfce :D
20151009 15:10:56 <elect> sorry, I cant hear you from my gnome3 highness
20151009 15:21:46 <xranby_f22> elect: we are discovering bugs in gnome here daily :)
20151009 15:21:59 <xranby_f22> elect: do you want to know the best way to crash gnome3 ?
20151009 15:22:07 <xranby_f22> i know the secret
20151009 15:22:07 <elect> ^^
20151009 15:22:45 <xranby_f22> grab a window border and while doing that let your application open a new window
20151009 15:23:07 <xranby_f22> and tada instant gnokme 3 crash
20151009 15:23:18 <xranby_f22> when you do it three times in a row, computer reboots
20151009 15:25:13 <elect> you need to be very mean to do that :p
20151009 15:25:46 <eclesia> mouhahahaha
20151009 15:25:47 <xranby_f22> i make gnome apology
20151009 15:26:34 <xranby_f22> and it actually do, it say it is sorry
20151009 15:30:17 * xranby_f22 (~familjen@anon) Quit (Quit: Leaving.)
20151009 15:59:26 <elect> btw, today I finished testing all the textures gli had, sgothel
20151009 16:00:03 <elect> everything here is supported except where differently specified
20151009 16:00:21 <elect> I ported the 0.7.0.0, the bloody went obviously further in the meanwhile ^^
20151009 16:00:39 <elect> I'll track him to be synchronized and updated :p
20151009 16:00:43 <elect> https://github.com/elect86/jgli/blob/master/Jgli/src/test/Main.java#L107-L279
20151009 16:01:21 * monsieur_max (~maxime@anon) has joined #jogamp
20151009 16:09:43 * eclesia (~husky@anon) has left #jogamp
20151009 16:16:38 * elect (~GBarbieri@anon) Quit (Ping timeout: 268 seconds)
20151009 16:23:06 * xranby_f22 (~familjen@anon) has joined #jogamp
20151009 17:15:34 * xranby_f22 (~familjen@anon) Quit (Ping timeout: 244 seconds)
20151009 17:17:13 * xranby_f22 (~familjen@anon) has joined #jogamp
20151009 17:30:05 * Eclesia (~eclesia@anon) has joined #jogamp
20151009 20:21:38 <sgothel> any incoming test results?
20151009 20:21:51 <sgothel> cont working on the issue
20151009 20:23:24 <xranby_f22> sgothel: added logs for ubuntu compiz unity, the bug is reproducible using both gnome and compiz
20151009 20:23:38 <xranby_f22> and the bug can occasionally pass, thus i added one log when it fail and one when it pass
20151009 20:26:17 <sgothel> look at your log FAIL ubuntu ..
20151009 20:26:32 <sgothel> X11: event . ConfigureNotify call 0x4400002 (parent 0x4400002, above (nil)) 75/62 640x480 0, child-event: 0
20151009 20:26:32 <sgothel> **************** X11: WM_STATE of 0x4400002: 0 props -> 0x0
20151009 20:26:32 <sgothel> X11: event . ConfigureNotify call 0x4400002 - isMapped 1, visibleChanged 1, map_state 2
20151009 20:26:46 <sgothel> should be not visible
20151009 20:27:02 <xranby_f22> compare that with the one that pass and you see
20151009 20:27:03 <xranby_f22> X11: event . ConfigureNotify call 0x4400002 (parent 0x4400002, above (nil)) 75/62 640x480 0, child-event: 0
20151009 20:27:03 <xranby_f22> ...... [0] -> [8/25]: _NET_WM_STATE_HIDDEN
20151009 20:27:03 <xranby_f22> **************** X11: WM_STATE of 0x4400002: 1 props -> 0x100
20151009 20:27:03 <xranby_f22> X11: event . ConfigureNotify call 0x4400002 - isMapped 1, visibleChanged 0, map_state 2
20151009 20:27:05 <sgothel> isMapped means that the hidden state shall be used ..
20151009 20:27:19 <sgothel> ah right
20151009 20:27:30 <sgothel> and sadly no other event :-/
20151009 20:27:40 <sgothel> i.e. where we could linger on ..
20151009 20:27:55 <sgothel> i.e. not the Leave/Enter pair
20151009 20:28:09 <sgothel> VisibilityNotify <- trying this one .. hmm
20151009 20:28:25 <sgothel> http://tronche.com/gui/x/xlib/events/structures.html <- the pool we have :)
20151009 20:30:37 <sgothel> w. the debug && freeze .. property .. did it become invisible?
20151009 20:31:18 <xranby_f22> the extra command line argument you mentioned?
20151009 20:31:37 <sgothel> -Dnewt.debug.Window.visibility.failure.freeze
20151009 20:31:38 <xranby_f22> i have not tested that since i was able to reproduce the issue without it
20151009 20:32:17 <sgothel> just helps analyzing the window state
20151009 20:32:31 <sgothel> visually or via cmd-line .. one sec
20151009 20:32:45 <sgothel> xwininfo
20151009 20:35:21 <xranby_f22> you suggest running the test in a debugger and with the info use that tool to peek at it
20151009 20:35:26 <xranby_f22> one second
20151009 20:41:29 <sgothel> VisibilityNotify always comes before FocusIn
20151009 20:42:29 <sgothel> and it is not coming w/ GL 'attached' :)
20151009 20:43:42 <xranby_f22> (22:30:37) sgothel: w. the debug && freeze .. property .. did it become invisible? <- the test passes when i use that property.. thus yes it becomes invisible
20151009 20:44:29 <sgothel> ok .. I hook visibility to expose/repaint as well .. and thats it .. hmm
20151009 20:44:46 <sgothel> what we could do .. is manual query before failing
20151009 20:45:01 <sgothel> yeah .. i.e. if TO is reached .. we query visibility
20151009 20:45:26 <sgothel> those rare cases will be a bit stuttering, but at least we pass - iff window is visible
20151009 20:45:59 <xranby_f22> w. the debug && freeze now it did freeze
20151009 20:46:32 <sgothel> and .. its still visible visually ?
20151009 20:46:34 <xranby_f22> was it supposed to freeze every time?
20151009 20:46:49 <sgothel> only if expected result is not coming within TO
20151009 20:47:09 <xranby_f22> yes i have a visible window that i can use the xwininfo on
20151009 20:47:22 <sgothel> damn .. it is supposed to be invisible?
20151009 20:47:23 <xranby_f22> xwininfo: Window id: 0x2c00004 "Newt Window"
20151009 20:47:23 <xranby_f22> Absolute upper-left X: 267
20151009 20:47:23 <xranby_f22> Absolute upper-left Y: 161
20151009 20:47:23 <xranby_f22> Relative upper-left X: 10
20151009 20:47:23 <xranby_f22> Relative upper-left Y: 45
20151009 20:47:24 <xranby_f22> Width: 640
20151009 20:47:24 <xranby_f22> Height: 480
20151009 20:47:24 <xranby_f22> Depth: 24
20151009 20:47:25 <xranby_f22> Visual: 0x27
20151009 20:47:26 <xranby_f22> Visual Class: TrueColor
20151009 20:47:26 <xranby_f22> Border width: 0
20151009 20:47:26 <xranby_f22> Class: InputOutput
20151009 20:47:54 <sgothel> kicked you out?
20151009 20:48:11 <xranby_f22> still here
20151009 20:48:26 <xranby_f22> yes it was supposed to go invisible
20151009 20:49:04 <sgothel> so action even didn't work - so that is a 'good' thing .. sort of :)
20151009 20:49:42 <sgothel> test case ?
20151009 20:49:58 <sgothel> which setVisible(true) line?
20151009 20:50:04 <xranby_f22> com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT
20151009 20:50:14 <xranby_f22> at com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.testDisplayCreate02Impl(TestDisplayLifecycle02NEWT.java:299)
20151009 20:50:29 <sgothel> wow ..
20151009 20:50:58 <xranby_f22> java.lang.AssertionError: expected:<false> but was:<true>
20151009 20:51:05 <sgothel> nothing special here .. hmm
20151009 20:51:39 <sgothel> well, we cannot help this issue .. I am afraid
20151009 20:52:20 <xranby_f22> it occasionally pass
20151009 20:52:27 <xranby_f22> so i say.. ship it
20151009 20:52:41 <sgothel> is it pass-1 or pass-2 ?
20151009 20:52:54 <sgothel> i.e. next line of stack ?
20151009 20:52:56 <xranby_f22> OK (2 tests)
20151009 20:53:03 <xranby_f22> oh
20151009 20:53:04 <xranby_f22> w8
20151009 20:53:09 <sgothel> 355 or 257 ?
20151009 20:53:14 <sgothel> 357
20151009 20:54:00 <xranby_f22> at com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT.testDisplayCreate02(TestDisplayLifecycle02NEWT.java:355)
20151009 20:54:10 <sgothel> so a normal case
20151009 20:54:40 <sgothel> dinner .. will think about it .. thank you
20151009 20:54:46 <xranby_f22> youre welcome
20151009 22:03:06 <xranby_f22> its 00.00 and every dns server on the planet goes down... is it caused by excessive dns lookup during webserver log parsing?
20151009 22:03:34 <xranby_f22> probably not but it at least feels that way
20151009 22:14:32 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20151009 22:30:12 <xranby_f22> https://www.eff.org/deeplinks/2015/10/final-leaked-tpp-text-all-we-feared
20151009 22:31:47 <sgothel> mine are still running :)
20151009 22:32:41 * xranby_f22 is experimenting how to best run all junit test on embedded devices..
20151009 22:33:08 <xranby_f22> running the ant junit.run do work... but i do not have controll to skip some tests
20151009 22:33:27 <sgothel> yeah, I have gotten one fast 32bit arm .. to be run w/ mesa soft .. will considering it after 2.3.2
20151009 22:33:48 <sgothel> we have diff targets in build-test.xml ..
20151009 22:33:53 <sgothel> try them ..
20151009 22:34:29 <sgothel> btw ... checking manual test run, there is a scripts/check-junit.sh .. :)
20151009 22:35:12 <xranby_f22> it is possible to run single tests using java -cp jogamp-fat-test.jar:jogamp-fat.jar org.junit.runner.JUnitCore com.jogamp.opengl.test.junit.newt.TestDisplayLifecycle02NEWT
20151009 22:35:16 <sgothel> the one-click (or command line) unit test run ..
20151009 22:35:33 <sgothel> yup
20151009 22:35:34 <xranby_f22> thanks to that we have junit core inside the fat jar
20151009 22:36:43 <xranby_f22> scripts/check-junit.sh <- ah!
20151009 22:36:50 <xranby_f22> sgothel: thank you
20151009 22:41:57 <xranby_f22> https://gist.github.com/xranby/3983f866a77d320cfc5d <-- this is what my fedora f22 machine looks like only one failed test and that is the setVisible(false) thing
20151009 22:42:38 <xranby_f22> the tree tests that errored is flagged unsupported.. however we should maybe use the Assume so that they get skipped
20151009 22:42:56 * Eclesia (~eclesia@anon) Quit (Quit: Leaving.)
20151009 22:43:51 <xranby_f22> ah the AWTAWT is also flagged unsupported
20151009 22:43:56 <xranby_f22> SWTAWT test
20151009 22:46:47 <xranby_f22> all newt.parenting tests that do a lot of visibility tests passes, which is great
20151009 22:47:27 <sgothel> having an update in a bit .. < 30 min
20151009 22:47:31 <xranby_f22> oh
20151009 22:47:50 <xranby_f22> is it the release announcement or a new commit?
20151009 22:47:58 <sgothel> new commit, attempt for a fix ..
20151009 22:48:10 <sgothel> which we then can discuss and make a decision
20151009 23:00:36 <sgothel> TestDisplayLifecycle02NEWT.java:299 <- on debian 8 w/ gnome as well, good
20151009 23:01:06 <xranby_f22> sgothel: while you are running gnomeshell
20151009 23:01:15 <sgothel> ay
20151009 23:01:19 <xranby_f22> try grab and hold a terminal window border while running the junit tests
20151009 23:06:47 <xranby_f22> sgothel: https://twitter.com/xranby/status/652621063742967808 - on my machine... this happens
20151009 23:13:10 <sgothel> hmm
20151009 23:13:20 <sgothel> will test after push
20151009 23:13:24 <sgothel> nice icon though :)
20151009 23:13:58 <xranby_f22> the triangle? i made it
20151009 23:14:36 <xranby_f22> the icon pointer stays as a grabbing hand when gome turns into this nice image
20151009 23:14:41 <xranby_f22> gnome
20151009 23:14:48 <xranby_f22> the logou button do work
20151009 23:32:15 <xranby_f22> good night
20151009 23:47:19 <sgothel> ah
20151009 23:47:25 <sgothel> pushed - good night
20151009 23:48:50 <sgothel> yes -- reproduced the crash .. omg
20151009 23:49:49 <sgothel> "main-Display-.x11_:0.0-1-EDT-2" #17 daemon prio=5 os_prio=0 tid=0x00007f3c706f8000 nid=0x5f8b runnable [0x00007f3c190ed000]
20151009 23:49:49 <sgothel> java.lang.Thread.State: RUNNABLE
20151009 23:49:49 <sgothel> at jogamp.newt.driver.x11.WindowDriver.CreateWindow0(Native Method)
20151009 23:52:08 <sgothel> clearly a gnome multithreading issue
20151009 23:53:38 <sgothel> ofc no crash w/ KDE
20151010 01:30:05 <sgothel> building 2.3.2 .. probably final
20151010 01:37:32 <sgothel> https://jogamp.org/bugzilla/show_bug.cgi?id=1251 <- xfce is nice, like ole fvwm days :) - could not reproduce issue though ..
20151010 01:37:38 <sgothel> works .. nice
20151010 02:58:09 <sgothel> WWIII - Syria, Russia & Iran - The New Equation https://www.youtube.com/watch?v=W3ZLYUvAZvs
20151010 03:03:43 <sgothel> http://jogamp.org/deployment/archive/master/gluegen_900-joal_626-jogl_1469-jocl_1106/
20151010 03:10:02 <sgothel> jogamp 2.3.2 maven deployed on jogamp
20151010 03:12:04 <sgothel> https://jogamp.org/wiki/index.php/SW_Tracking_Report_Objectives_for_the_release_2.3.2 closed
20151010 03:12:13 <sgothel> version 2.3.2 has been disabled in bugzilla now
20151010 03:12:34 <sgothel> 2.4.0 is next .. and we have 2.4.1, 2.4.2, 2.5.0 available ..
20151010 03:14:12 <sgothel> links are updated -> 2.3.2
20151010 03:14:15 <sgothel> (webpage)
20151010 03:41:14 <xranby_f22> KUDOS! \o/ *POM* *POM *POM*
20151010 03:41:58 <xranby_f22> JOGAMP 2.3.2 *pom* *Pom* *POM*
20151010 03:46:17 <xranby_f22> bruce-: (05:10:01) sgothel: jogamp 2.3.2 maven deployed on jogamp
20151010 03:46:36 <sgothel> @Mark: Pls push 2.3.2 to maven - thank you!
20151010 03:46:58 <sgothel> tagged and sources, changelogs .. provided
20151010 03:47:23 <sgothel> https://jogamp.org/wiki/index.php/Release_2.3.2
20151010 03:56:16 <sgothel> https://jogamp.org/deployment/v2.3.2/archive/API-Changes/
20151010 03:56:46 <sgothel> CompatChanged Field com.jogamp.opengl.[GL3 -> GL3ES3].GL_MAX_GEOMETRY_OUTPUT_COMPONENTS, access[]
20151010 03:57:12 <sgothel> ^^ shows how things got moved upwards during updated headers .. cool :)
20151010 03:57:39 <sgothel> https://jogamp.org/wiki/index.php/Release_2.3.2 - page completed
20151010 04:00:22 <sgothel> http://jogamp.org/ 2.3.2
20151010 04:07:01 <xranby_f22> HURRAY!
20151010 04:09:37 <sgothel> https://jogamp.org/wiki/index.php/Android_Packages .. fixed
20151010 04:18:08 <sgothel> still working on the androids (arm 32bit and 64bit)
20151010 04:23:51 <sgothel> http://forum.jogamp.org/Release-2-3-2-Finally-td4035446.html <- Release Note ..
20151010 04:47:32 <xranby_f22> Today is JogAmp release and Double Ten Day https://en.wikipedia.org/wiki/National_Day_of_the_Republic_of_China
20151010 04:48:18 <sgothel> yes .. double double .. good day, even though w/o any eights :)
20151010 04:48:35 <xranby_f22> 十十
20151010 04:48:53 <sgothel> good night
20151010 04:48:59 <sgothel> taking my 4 hours :)
20151010 04:49:34 <sgothel> https://jogamp.org/git/?p=gluegen.git;a=summary .. post 2.3.2 era :)
20151010 04:49:58 <xranby_f22> sgothel: sleep well and happy celebrations
20151010 04:50:01 <sgothel> breaking the game -> API changes are OK
20151010 05:03:16 <xranby_f22> https://twitter.com/hashtag/%E5%8D%81%E5%8D%81?src=hash
20151010 05:05:16 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20151010050516.html