#jogamp @ irc.freenode.net - 20131124 05:05:54 (UTC)
20131124 05:05:54 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20131123050554.html
20131124 05:05:54 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20131124050554.html
20131124 06:33:07 * xranby (~xranby@anon) Quit (Ping timeout: 246 seconds)
20131124 06:33:47 * xranby (~xranby@anon) has joined #jogamp
20131124 08:05:28 * xranby (~xranby@anon) Quit (Read error: Operation timed out)
20131124 08:21:05 * xranby (~xranby@anon) has joined #jogamp
20131124 09:17:16 * monsieur_max (~maxime@anon) has joined #jogamp
20131124 11:51:31 * hija (~hija@anon) has joined #jogamp
20131124 13:06:48 * xranby (~xranby@anon) Quit (Ping timeout: 245 seconds)
20131124 13:08:12 * xranby (~xranby@anon) has joined #jogamp
20131124 13:21:34 * xranby (~xranby@anon) Quit (Ping timeout: 246 seconds)
20131124 13:33:55 * hija (~hija@anon) Quit (Quit: hija)
20131124 13:34:58 * hija (~hija@anon) has joined #jogamp
20131124 13:36:19 * xranby (~xranby@anon) has joined #jogamp
20131124 14:02:42 * Eclesia (~eclesia@anon) has joined #jogamp
20131124 14:02:46 <Eclesia> hi
20131124 14:16:11 * Eclesia found another awesome projet in public domain : https://github.com/wilkie/djehuty
20131124 14:16:47 <Eclesia> ithas regex, mp2, mp3, readers, time/date api and many more :D
20131124 14:45:35 * hija (~hija@anon) Quit (Ping timeout: 272 seconds)
20131124 15:42:58 * Eclesia (~eclesia@anon) Quit (Quit: Leaving.)
20131124 16:19:40 * hija (~hija@anon) has joined #jogamp
20131124 18:36:23 <hija> sgothel: can a test be osx specific?
20131124 18:52:14 <sgothel> sure
20131124 18:58:19 <hija> how?
20131124 19:05:36 <hija> Assume.assumeTrue(NativeWindowFactory.TYPE_MACOSX == NativeWindowFactory.getNativeWindowType(false));
20131124 19:05:47 <hija> in a @Before method?
20131124 19:11:55 <sgothel> one sec .. we have a way ..
20131124 19:13:30 <sgothel> look for UITestCase.setTestSupported(boolean)
20131124 19:13:53 <sgothel> @BeforeClass
20131124 19:13:53 <sgothel> public static void initClass() {
20131124 19:13:53 <sgothel> if( !satisfied ) {
20131124 19:13:53 <sgothel> setTestSupported(false);
20131124 19:13:53 <sgothel> }
20131124 19:13:53 <sgothel> }
20131124 19:14:41 <sgothel> 'false' will trigger: Assume.assumeTrue(testSupported); // abort
20131124 19:15:16 <sgothel> so yes: you found the right stuff :)
20131124 19:16:06 <sgothel> Platform's OSType etc .. can also be used to identify a system .. windowing toolkit agnostic - hence the 'almost' duplication
20131124 19:25:27 * hija (~hija@anon) Quit (Quit: hija)
20131124 19:46:35 * hija (~hija@anon) has joined #jogamp
20131124 19:48:40 <hija> so just call setTestSupported in a method (I guess that a @Before one) in my own test and give it my boolean
20131124 19:49:15 <sgothel> yup .. it does the same as you figured .. UnitTest simply checks it @Before
20131124 19:49:34 <sgothel> set @BeforeClass simply disables the whole thing
20131124 19:49:42 <hija> oh
20131124 19:50:02 <hija> then what does @Before do?
20131124 19:50:07 <hija> each test?
20131124 19:50:10 <sgothel> before each test - yup
20131124 19:50:22 <sgothel> IDE mouse-over should give you the help
20131124 19:50:34 <sgothel> (at least w/ Eclipse)
20131124 19:50:51 <hija> yup
20131124 19:51:07 <hija> ok tests are almost ready
20131124 19:57:21 <sgothel> great !
20131124 19:57:24 <sgothel> dinner .. laters
20131124 21:01:28 * hija (~hija@anon) Quit (Ping timeout: 246 seconds)
20131124 21:05:11 * hija (~hija@anon) has joined #jogamp
20131124 22:37:45 <hija> a complete pull request finally sent
20131124 22:48:15 <hija> meh, previous commit was wrong...
20131124 22:48:23 <hija> fixed now
20131124 23:04:51 * monsieur_max (~maxime@anon) has left #jogamp
20131124 23:42:08 * xranby (~xranby@anon) Quit (Ping timeout: 246 seconds)
20131124 23:43:50 <sgothel> Great Hija .. reading ..
20131124 23:44:17 * xranby (~xranby@anon) has joined #jogamp
20131124 23:44:32 <sgothel> your git author tag contains your real name ? I would like that .. if possible
20131124 23:45:53 <hija> How do I do that? :P
20131124 23:46:10 <hija> in eclipse?
20131124 23:46:21 <sgothel> Author: Sven Gothel <sgothel@jausoft.com> <- mine ..
20131124 23:46:36 <sgothel> yours - fine !
20131124 23:46:41 <sgothel> thx
20131124 23:47:07 <sgothel> Bug672NewtCanvasSWTSashFormComposite why only OSX ?
20131124 23:47:41 <hija> because everything appears in osx
20131124 23:47:51 <sgothel> *scratch head* ..
20131124 23:48:05 <hija> mm
20131124 23:48:19 <hija> the bug only appears in osx
20131124 23:48:20 <sgothel> well, the test must pass all platforms usually to prove no regression happened .. :)
20131124 23:48:43 <sgothel> sure, if it is special to OSX .. but you remember your 1st fix broken Windows, Linux ..
20131124 23:49:04 <sgothel> so I assume - you tests always worked well on Windows/Linux ?
20131124 23:49:20 <hija> yeap they did
20131124 23:49:25 <hija> no problems there
20131124 23:49:34 <hija> it only appeared in osx
20131124 23:49:36 <sgothel> is there any reason why they now should not work on those ?
20131124 23:49:43 <sgothel> right ..
20131124 23:49:57 <sgothel> so the unit tests shall [still] work on such other platforms ..
20131124 23:50:01 <hija> hmm
20131124 23:50:02 <sgothel> so let them run there :)
20131124 23:50:04 <hija> hard to explain
20131124 23:50:07 <sgothel> be generous
20131124 23:50:19 <hija> the thing is
20131124 23:50:29 <sgothel> go ahead
20131124 23:50:43 <hija> NewtCanvasSWT uses this "getLocationOnScreen" hack
20131124 23:50:52 <sgothel> ok ..
20131124 23:50:56 <sgothel> and ?
20131124 23:51:07 <sgothel> (you are more into the details now)
20131124 23:51:22 <hija> and it returns a point
20131124 23:51:26 <hija> if
20131124 23:51:34 <hija> you are in osx
20131124 23:51:36 <hija> 0 if not
20131124 23:51:37 <sgothel> ah .. n/a on other platforms ..
20131124 23:51:50 <hija> the problem seems to lie on a lower level actually
20131124 23:51:56 <sgothel> so .. this is the assertion right ?
20131124 23:52:02 <sgothel> i.e. the actual test
20131124 23:52:05 * xranby (~xranby@anon) Quit (Ping timeout: 248 seconds)
20131124 23:52:21 <sgothel> great ..
20131124 23:52:24 <hija> yes, the assertion checks that method
20131124 23:52:41 <sgothel> so we would only need to support getLOS for other platforms .. maybe
20131124 23:52:54 <sgothel> besides - visual confirmation would work right ?
20131124 23:53:08 <sgothel> i.e. see no regression and the fix for OSX
20131124 23:53:11 <hija> when in other platforms the method is not really used...
20131124 23:53:17 <sgothel> right
20131124 23:53:32 <hija> on other platforms the offset is placed automagically
20131124 23:53:43 <sgothel> is there a visual confirmation w/ this test ?
20131124 23:53:49 * xranby (~xranby@anon) has joined #jogamp
20131124 23:53:51 <sgothel> (manual and visual)
20131124 23:53:59 <sgothel> (can I SEE it ?)
20131124 23:54:04 <hija> I asked you about that method actually
20131124 23:54:07 <hija> LOS
20131124 23:54:14 <sgothel> yup - I remember
20131124 23:54:16 <sgothel> all good
20131124 23:54:18 <hija> let me find that
20131124 23:54:57 <sgothel> simple Q: Can I see that the test is working ?
20131124 23:55:00 <hija> 20131117 20:03:30 <sgothel> ah .. it's from the inner SWTNativeWindow ..
20131124 23:55:01 <hija> 20131117 20:03:33 <sgothel> memories ..
20131124 23:55:03 <hija> 20131117 20:04:13 <sgothel> inner getLocationOnScreen() is used w/ NEWT's OSX impl. which fetches the parent LOS
20131124 23:55:06 <sgothel> hehe
20131124 23:55:07 <hija> 20131117 20:04:38 <sgothel> since OSX requires us to use the LOS .. (ugly ugly)
20131124 23:55:07 <hija> 20131117 20:04:58 <sgothel> so it should be OK'ish .. hmm maybe it helps
20131124 23:55:07 <hija> 20131117 20:05:29 <sgothel> -> NEWT's OSX WindowDriver impl .. calls getLocationOnScreen() of this inner NativeWindow impl.
20131124 23:55:10 <sgothel> jaja
20131124 23:55:47 <hija> without the assertion yes
20131124 23:55:52 <sgothel> I guess we do have getLOS impl. for all platforms - so I / you can add this at a later time .. no problem
20131124 23:55:56 <hija> visual confirmation is enough
20131124 23:56:03 <sgothel> so we can simply disable the assertion on OSX
20131124 23:56:10 <sgothel> *oops* vice versa
20131124 23:56:21 <sgothel> enable on OSX, disable on others
20131124 23:56:26 <sgothel> ok w/ you ?
20131124 23:57:05 <hija> yeap
20131124 23:57:29 <hija> will you do it? or should I pull that back and fix it?
20131124 23:57:29 <sgothel> getLOS for other platforms:
20131124 23:57:36 <sgothel> no .. I merge
20131124 23:57:45 <sgothel> if you like to do it .. I pull again
20131124 23:58:07 <sgothel> and I check the getLOS for Windows/Linux ..
20131124 23:58:10 <sgothel> protected Point getLocationOnScreenImpl(int x, int y) {
20131124 23:58:10 <sgothel> return GDIUtil.GetRelativeLocation( getWindowHandle(), 0 /*root win*/, x, y);
20131124 23:58:10 <sgothel> }
20131124 23:58:35 <hija> so just wrap an if arround the assertion and commit again?
20131124 23:58:49 <sgothel> and do another commit .. yes please
20131124 23:59:02 <sgothel> many commits == good :)
20131124 23:59:05 <hija> :D
20131124 23:59:34 <sgothel> ok, I guess I have to add the getLOS thingy for X11Util .. will do .. will do
20131125 00:00:18 <sgothel> return X11Lib.GetRelativeLocation(getDisplayHandle(), getScreenIndex(), getWindowHandle(), 0 /*root win*/, x, y);
20131125 00:00:21 <sgothel> we have it :)
20131125 00:00:26 <sgothel> sweet
20131125 00:00:36 <sgothel> so I will test on X11 and Windows .. nice nice
20131125 00:02:21 <hija> ok then, i'm off to sleep!
20131125 00:02:24 <sgothel> thank you for the fix - and for the great git commit !
20131125 00:02:30 <sgothel> good night - good job!
20131125 00:02:46 <sgothel> .. git commit message as well!
20131125 00:02:58 <hija> :D
20131125 00:06:44 <sgothel> merged - I will refine a bit - i.e. unit tests must start w/ 'Test' to be run automatically .. etc - will also test the getLOS() on X11/Windows
20131125 00:06:53 <sgothel> so you can focus on the other many SWT things ..
20131125 00:07:05 <sgothel> wiki now ..
20131125 00:13:57 * xranby (~xranby@anon) Quit (Ping timeout: 248 seconds)
20131125 00:15:38 <sgothel> https://jogamp.org/wiki/index.php/Maintainer_and_Contacts
20131125 00:15:57 <sgothel> welcome Petros Koutsolampros .. will make an email to forum
20131125 00:19:28 * xranby (~xranby@anon) has joined #jogamp
20131125 00:20:25 <sgothel> http://forum.jogamp.org/New-Maintainer-JOGL-SWT-Petros-Koutsolampros-td4030687.html
20131125 00:22:21 <sgothel> @Hija: wiki account created - pls always use https !
20131125 00:22:29 <sgothel> you need to change your password
20131125 00:23:30 <sgothel> @Xerxes: Good evening - did you have a chance to test IcedTea-Web 1.5'ish ?
20131125 00:23:59 <sgothel> I hope my bug description can help a bit to fix it the issue
20131125 00:32:26 <sgothel> @Hija: Looks like you are a professional - w/ sense for art and design (as an architect .. duh!) - so I see already how we can do some UI things .. always great to have one pro regarding such things.
20131125 00:49:51 * void256 (~void@anon) has joined #jogamp
20131125 01:03:23 * void256 (~void@anon) Quit (Remote host closed the connection)
20131125 02:27:45 * hija (~hija@anon) Quit (Quit: hija)
20131125 02:56:47 * hija (~hija@anon) has joined #jogamp
20131125 03:21:57 * hija (~hija@anon) Quit (Quit: hija)
20131125 05:05:54 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20131125050554.html