Perl 6 - the future is here, just unevenly distributed

IRC log for #openframeworks, 2014-03-20

| Channels | #openframeworks index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:16 pizthewiz joined #openframeworks
00:41 vade joined #openframeworks
01:00 dantheman joined #openframeworks
01:17 AceSlowman joined #openframeworks
01:17 boltR joined #openframeworks
01:22 dantheman joined #openframeworks
01:57 boltR joined #openframeworks
02:01 dantheman joined #openframeworks
02:06 admsyn joined #openframeworks
02:16 mendo joined #openframeworks
02:16 mendo hello!
02:31 dantheman joined #openframeworks
02:51 AceSlowman joined #openframeworks
03:24 Joel_re joined #openframeworks
03:40 Joel_re joined #openframeworks
03:56 AceSlowman joined #openframeworks
04:02 pizthewiz joined #openframeworks
04:21 dantheman joined #openframeworks
04:29 vade joined #openframeworks
05:00 _mj joined #openframeworks
05:01 lmccart joined #openframeworks
05:01 lmccart_ joined #openframeworks
05:11 pizthewiz joined #openframeworks
05:25 _mj joined #openframeworks
05:43 lmccart joined #openframeworks
05:43 lmccart_ joined #openframeworks
07:14 ncls joined #openframeworks
07:44 boltR joined #openframeworks
08:42 ofarturo joined #openframeworks
09:12 tgfrerer_ joined #openframeworks
09:21 Joel_re joined #openframeworks
09:29 arthurc joined #openframeworks
09:46 air_cr4ft joined #openframeworks
10:03 kflux joined #openframeworks
10:05 neue joined #openframeworks
10:58 air_cr4f_ joined #openframeworks
12:06 jvcleave joined #openframeworks
12:20 Joel_re_ joined #openframeworks
12:31 lmccart_ joined #openframeworks
12:31 lmccart joined #openframeworks
12:33 arthurc joined #openframeworks
12:43 Joel_re_ http://google.github.io/liquidfun/
12:43 Joel_re_ is neat
12:44 vade joined #openframeworks
12:49 jvcleave_ joined #openframeworks
12:54 robot joined #openframeworks
13:20 _mj joined #openframeworks
13:40 lmccart_ joined #openframeworks
13:40 lmccart joined #openframeworks
13:42 jvcleave_ joined #openframeworks
13:44 chrisallans joined #openframeworks
13:49 dantheman joined #openframeworks
14:37 lmccart joined #openframeworks
14:37 lmccart_ joined #openframeworks
14:59 admsyn joined #openframeworks
15:13 jvcleave joined #openframeworks
15:27 vade joined #openframeworks
15:36 dantheman joined #openframeworks
15:46 jvcleave_ joined #openframeworks
15:48 fossterer joined #openframeworks
15:50 fossterer Hi, I'm getting this error for every project -- 'AndroidManifest.xml not found.. But I see every project has one in fact!
16:10 nicedice joined #openframeworks
16:18 admsyn joined #openframeworks
16:20 kflux joined #openframeworks
16:23 robot joined #openframeworks
16:33 air_cr4ft joined #openframeworks
16:49 boltR joined #openframeworks
17:19 dantheman joined #openframeworks
17:23 kflux joined #openframeworks
17:30 pizthewiz joined #openframeworks
17:46 * fossterer Got it solved :) :) :)
17:46 air_cr4ft left #openframeworks
17:46 zanilo joined #openframeworks
17:46 zanilo I'm having some issues using oxfCv in my project
17:46 zanilo I've set up my header search paths in the my config
17:47 zanilo but can't figure out how to add the folders to my addons folder
17:47 zanilo (which folder to add)
17:47 admsyn zanilo: using xcode?
17:49 zanilo YES
17:49 zanilo herm.. I meant "yes"
17:49 admsyn :)
17:49 zanilo I just can't figure out if I whould include the folders listed in kyle readme
17:49 admsyn make a "group" in your addons folder (i.e. right click, "New Group")
17:49 zanilo ok
17:50 admsyn err wait a sec, what are you trying to do here?
17:50 zanilo use ofxCv
17:50 zanilo in another project
17:50 admsyn yeah I just mean this particular step
17:50 admsyn you added the header search paths, so it can find the headers fine?
17:51 zanilo I've set up my headers
17:51 admsyn ok, so now you just need to add ofxCv itself?
17:51 zanilo in my xconfig file
17:51 admsyn cool
17:51 admsyn oh
17:51 zanilo (not using the IDE)
17:51 admsyn yeah that'll work
17:51 zanilo yes
17:52 admsyn alright, so the new group you just added, call it "ofxCv"
17:52 zanilo ok
17:52 admsyn then right click it, "Add files to ofApp"
17:52 admsyn then pick the libs/ and src/ folders in ofxCv
17:52 admsyn should be it
17:52 zanilo I've done it
17:52 zanilo doesn't work
17:52 admsyn how so
17:52 zanilo I'm redoing it again
17:52 zanilo also
17:52 zanilo how about opencv?
17:53 zanilo won't ofxcv need to add its folder as well
17:53 admsyn sort of, but not in the same way you added ofxCv
17:53 admsyn you just need the header search paths, and the opencv.a static lib
17:53 admsyn so, if you try and compile now, what do you get?
17:54 zanilo one moment
17:54 zanilo so
17:54 zanilo I shouldn't include opencv group?
17:55 admsyn you don't have to, no
17:55 admsyn you can if you want to have the opencv source in the file browser though
17:55 zanilo ok my folder looks like this now
17:55 zanilo http://i.imgur.com/0bVJDlE.png
17:56 admsyn ok, and if you try and build what do you get?
17:56 zanilo I still get issues on #include statements
17:57 admsyn ok, which files
17:57 admsyn ?
17:57 zanilo file not found
17:57 zanilo #include "ofxCv/Distance.h"
17:57 admsyn ok, in that case you'll need to add a header search path, yeah
17:57 * admsyn looks
17:58 zanilo at the moment they are HEADER_SEARCH_PATHS = $(OF_CORE_HEADERS) "../../../addons/ofxOpenCv/libs/opencv/include/" "../../../addons/ofxCv/libs/ofxCv/include/"
17:58 admsyn oh, "../../../addons/ofxCv/libs/ofxCv/include/" should cover it
17:58 admsyn I'd guess the setting isn't actually taking hold
17:59 admsyn do you know how to look at the build settings for your project in xcode? it'd be worth taking a look to see if it actually shows up there
17:59 zanilo yes I know how to check the header search paths in the IDE
17:59 zanilo (...and they don't show up correctly)
18:01 admsyn alright, well one way to do it would be to just put the setting there in the settings browser
18:01 admsyn the other would be to figure out why the .xcconfig isn't doing anything
18:01 admsyn is the setting in bold? (i.e. overridden)
18:01 zanilo hm I'll add them manually in the ide
18:04 zanilo nope
18:04 pizthewiz OFPlugin!
18:05 admsyn haha
18:05 fossterer Q) Including "ofxOpenCv.h" is throwing errors --
18:05 admsyn zanilo: did you add the setting to the project or to the target? if you added it to the project it's possible the target is overriding it with nothingness
18:05 pizthewiz That or we really need a general guide on Xcode project settings that we can refer people to
18:06 fossterer <Installtion-Path>/android-ndk-r9b/sources/cxx​-stl/gnu-libstdc++/4.8/include/complex:69:30: error: expected unqualified-id before 'const'    template<typename _Tp> _Tp abs(const complex<_Tp>&);
18:07 zanilo ok i'll double check it
18:08 zanilo I've added it to the target
18:08 admsyn can you screenshot it or something? all signs are pointing to "should be working"
18:09 zanilo sure
18:09 admsyn alternatively, try moving it over to user_header_search_paths but that's basically grasping at straws
18:10 zanilo http://i.imgur.com/Qqvcrh9.png
18:11 admsyn are you building in debug?
18:12 zanilo yes
18:12 admsyn yeah I don't know, it looks like you've set it up fine
18:15 zanilo I can now find ofxCv
18:15 zanilo but can't find opencv2/opencv.hpp
18:15 admsyn alright, what was the issue with ofxCv?
18:15 admsyn opencv will be the same deal, just find the parent folder of opencv2 and add the path for it
18:15 zanilo it was that I added the folders suggested in the readme
18:16 ThatCantBe joined #openframeworks
18:16 zanilo instead of libs src
18:16 zanilo ok
18:16 zanilo to add opencv
18:17 zanilo I'm creating a new group called...ofxOpenCv
18:17 zanilo ?
18:17 admsyn sure, yeah, same deal as ofxCv
18:18 zanilo ok it still can't find opencv2/opencv.hpp
18:18 zanilo I'm also adding the opencv include folder in the header paths
18:18 admsyn so, for my curiosity, let's back up a sec :3
18:19 admsyn what'd you do from before to change ofxCv from not-working to working
18:19 zanilo I've added just the folder src and libs inside its folder
18:19 zanilo (group)
18:20 admsyn and that got rid of the #include error?
18:20 zanilo yes
18:20 admsyn alright.
18:22 admsyn well anyway it should be the same deal with ofxOpenCv
18:23 admsyn you're getting issues with it not being able to #include "opencv2/opencv.hpp", it's because Xcode doesn't have a search path for ../../../addons/ofxOpenCv/libs/opencv/include
18:24 admsyn I'm not sure how your project is set up, but if you work backwards from there it should be figure-out-able
18:24 zanilo ok
18:24 zanilo I may be including a mistyped header search path
18:25 admsyn that'd be possible
18:25 admsyn also try with and without the trailing slash
18:25 admsyn I forget if it matters but I've been burned by subtle shit like that in the past
18:25 admsyn w/early Xcode 4
18:25 admsyn don't know if it's still the case
18:29 zanilo "Build Succeeded"
18:29 admsyn nice
18:30 zanilo \m/
18:30 zanilo (it WAS a typo)
18:30 admsyn ah
18:30 zanilo but man, you gotta love Xcode
18:30 zanilo not
18:30 zanilo btw
18:31 zanilo I'm still working on the Xcode instructions for noobies
18:31 zanilo https://github.com/tezzutezzu/open​Frameworks/blob/master/docs/osx.md
18:31 zanilo (from a noob)
18:31 admsyn nice, looking good
18:32 zanilo ta!
18:36 jedahan joined #openframeworks
18:38 jedahan trying to fix the tr1/memory missing error with makefiles - https://github.com/openframeworks/openFrameworks​/commit/4b669b86c635257af6a764772d0d46df37677244 fixed it for xcode, but I am not sure where to tell clang to link using libstdc++ in the makefiles...
18:39 pizthewiz joined #openframeworks
18:41 fossterer q) Using this method -- void ofImage_::setFromPixels(const PixelType *pixels, int w, int h, ofImageType type, bool bOrderIsRGB=true)
18:41 fossterer in Android, it crashes the app
18:42 fossterer Help!!
18:43 dantheman fossterer: don't do it
18:43 dantheman lol
18:43 jedahan woohoo! PLATFORM_CFLAGS += -stdlib=libstdc++ in libs/openFrameworksCompiled/proj​ect/osx/config.osx.default.mk:78 did it!
18:44 dantheman I don't think you can get the pixels from the screen on android
18:44 fossterer I'm not getting the pixels from screen..
18:44 fossterer I set pixels from an external array
18:44 fossterer and trying to construct an image from them
18:45 fossterer dantheman:
18:45 dantheman yeah I don't think you can grab the pixels from screen that way
18:45 dantheman on Android
18:47 fossterer I'm not grabbing... I've a lot "unsigned char va" values... which I want to look into.. rather than see them individually.. I'm constructing an ofImage
18:47 dantheman huh?
18:47 fossterer ?
18:47 dantheman are you trying to grab the screen?
18:47 fossterer no.. no.. no..
18:47 dantheman or just set an image from pixels
18:47 fossterer yes
18:48 dantheman where is the image?
18:48 dantheman is it local or generated?
18:48 fossterer would you like to see the code?
18:48 dantheman would help
18:49 dantheman pastebin it
18:50 fossterer http://pastebin.com/XJickH6a
18:52 admsyn joined #openframeworks
18:52 dantheman you need to allocate your pixels
18:52 dantheman http://openframeworks.cc/documentation​/graphics/ofPixels.html#!show_allocate
18:53 dantheman also use brackets in your For Loops
18:53 dantheman you're getting a bad access
18:53 dantheman and your math is off if you are doing 3 color
18:54 fossterer http://openframeworks.cc/documentation/g​raphics/ofImage.html#show_setFromPixels
18:54 fossterer has a NOTE
18:54 dantheman ofPixels px;
18:54 dantheman ofImage img;
18:54 fossterer saying oyu don't need to allocate
18:54 dantheman do you see that
18:54 dantheman yeah the ofImage
18:55 dantheman what is this doing
18:55 dantheman for(int vo = 0; vo < WRAP_V_SIZE; vo++)
18:55 dantheman for(int uo = 0; uo < WRAP_U_SIZE; uo++)
18:55 dantheman im[k++] = phaseGammaLut[vo][uo][0];
18:55 dantheman where is Im and is I'm allocated?
18:55 dantheman does this run on OSX/Linxu/Windows?
18:55 dantheman you should try it on desktop before you run on mobile
18:56 fossterer It is running too good on desktop and the same fails on mobile!!
18:56 dantheman hmmm
18:56 dantheman huh?
18:56 fossterer Running excellently on Linux Desktop
18:56 dantheman ok
18:56 dantheman not sure what to tell you them
18:56 dantheman then*
18:57 fossterer 'im' is just an array.. allocated on the second line of my paste
18:57 dantheman I see it now
18:57 dantheman oh oh
18:57 dantheman you don't have to case the ofPixles to an unsigned char *
18:57 Joel_re joined #openframeworks
18:58 dantheman you can just give it the ofPixels
18:58 dantheman also might want to try running an img.update()
18:58 dantheman before you save
18:58 dantheman throw something up on the forum or github
18:58 fossterer If I don't cat, -- src/ThreePhaseWrap.cpp:39:86: error: no matching function for call to 'ofImage_<unsigned char>::setFromPixels(ofPixels*, int, int, ofImageType, bool)'              img.setFromPixels(&px, WRAP_V_SIZE, WRAP_U_SIZE, OF_IMAGE_GRAYSCALE, true);
18:58 dantheman this is wonky
18:58 fossterer *If i don't cast
18:58 admsyn jedahan: yeah makes total sense to me
18:59 admsyn just giving it a build, will merge it in a moment
18:59 dantheman transparency.setFromPixels(pix);
18:59 dantheman you can't just do that?
19:00 dantheman img.setFromPixels(pix) and give it the ofPixels?
19:00 dantheman I just did that in the ofImage Example
19:00 dantheman otherwise you are repeating yourself
19:01 dantheman you can just give the img.setFromPixels the char * array
19:02 fossterer Hmm... just as it was failing it that "straight way" I took this 'ofPixel" path!
19:03 fossterer Failed again
19:06 admsyn jedahan: how can I A/B the makefile change?
19:07 admsyn I just ran a current makefile and it was fine, not sure how to set up a scenario where it failed before but now is fine
19:11 dantheman fossterer: not sure
19:11 dantheman what to tell you unless you can get the crash log
19:12 dantheman do you know how to debug JNI on Android?
19:20 admsyn joined #openframeworks
19:27 fossterer How do i debug.. This thing is just showing me 'Fatal Signal 11.."
19:27 fossterer ?
19:28 admsyn before even launching?
19:28 admsyn throw a breakpoint as early as possible
19:28 admsyn that's a segfault i.e. you tried to read from a bad pointer or something similar
19:29 fossterer I can see that just this method -- setFromPixels is causing this segfault
19:29 admsyn sure, so breakpoint that call
19:29 admsyn and then see what the variables are before they go in
19:29 admsyn one's probably BS
19:29 admsyn so trace back to find why it's BS
19:47 lmccart joined #openframeworks
19:48 lmccart_ joined #openframeworks
19:54 robot joined #openframeworks
19:54 jedahan slowly but surely getting oF to play nicer with atom :)
19:54 admsyn nice
20:09 pizthewiz Atom the processor or Atom the (web browser) editor?
20:13 * oneironaut assumes the embedded chips
20:13 pizthewiz Probably the latter, since the processor should more or less work, though the GL story is murky with stuff like Intel's NUC.
20:13 admsyn no probably the editor, since he just submitted a makefile PR :)
20:13 pizthewiz Heh, and here I was thinking it was the editor 😉
20:13 oneironaut yah, try decoding video on windows with an atom :P
20:13 admsyn atom the editor's pretty ok? From what I can tell?
20:14 admsyn I assume they're leaning on community/plugins to make it actually good
20:14 pizthewiz Yeah, I'll bet in a few months it is pretty ok++.
20:14 admsyn did a bit of javascripting in it and it was pretty much the most average thing ever
20:14 pizthewiz So far it feels really sluggish, search isn't great with false positives/negatives.
20:15 pizthewiz My hope like with TextMate is that the plugins do a lot for us.
20:15 admsyn oh pizthewiz you're going to resonate still yeah?
20:16 pizthewiz The weird thing is, the real value seems like it'll come out of the plugins… which are probably externally created :0\
20:16 pizthewiz admsyn: Actually I just started an oF+RPi+Arduino project that is gonna be in an assembly phase right then 😥
20:16 admsyn :C
20:17 pizthewiz I saw that there was a seat available for the CInder workshop too, which is an extra bummer
20:17 admsyn yeah
20:18 pizthewiz We'll cross paths at Eyeo no doubt, which reminds me, I need to sort out travel and what not
20:18 admsyn same
20:18 admsyn I just got my resonate hotel booked like a week ago <_< >_>
20:19 pizthewiz Heh, did you still get the good pricing?
20:19 admsyn yeah, actually
20:19 admsyn not at the 5 star but someone I think I'll manage
20:19 admsyn *somehow
20:21 pizthewiz Maybe some extra excitement in your future… the program looks great though
20:35 ofarturo joined #openframeworks
20:36 austinslominski joined #openframeworks
21:11 kureta joined #openframeworks
21:11 jvcleave ofarturo - I have narrowed down an GL_INVALID_OPERATION at this line when trying to use an Fbo/texture in update https://github.com/openframeworks/openF​rameworks/blob/master/libs/openFramewor​ks/gl/ofGLProgrammableRenderer.cpp#L154
21:21 Alski @admsyn , (this is AceSlowman) I have all of the FBO stuff working now. Video appears perfectly behind my alpha mask and it is running well. I remember you had mentioned using std::map to link id's to different FBOs (since I want a different video beneath different contours). What would be a good start towards treating each shape seperately?
21:21 admsyn yey!
21:21 admsyn well if you're using a blob tracker they'll come with IDs
21:22 admsyn std::map is like a vector, but instead of everything having a numerical address ( vector[3] ) things are addressed with other things ( map[thing] )
21:23 Alski Yep, got the ID and area from each blob. The problem is I don't have them when I build the paths. (because it is pathFromContour, but I would expect I need something like pathFromContour[i] or something)
21:23 Alski which would be my contour id? Would I be able to loop through that to build different maskfbos?
21:23 admsyn yeah
21:24 admsyn yeah to both questions
21:24 admsyn you'd loop through with an iterator
21:25 admsyn this was assuming you were associating a polyline directly with a video, though
21:25 kflux joined #openframeworks
21:25 admsyn if you all you want to do is keep track of which polylines have which IDs, it might be easier to just subclass ofPolyline and add an ID member
21:27 Alski do you know of an oF example that I might be able to see that working? From there I want to build a distinct alpha mask for each shape
21:30 admsyn which, subclassing or std::map?
21:32 Alski if subclassing will suffice. Looking at it, I just need to have an alpha mask with a corresponding video for each shape.
21:33 admsyn unless I'm missing something, you could probably do this way simpler by re-using the same alpha mask and just drawing with it over and over?
21:33 admsyn but subclassing is just class MyPolyline : public ofPolyine { public: int videoId; };
21:34 admsyn the custom events example has subclassing, off the top of my head
21:34 admsyn as do a bunch more I'm sure
21:35 Alski If I used the same alpha mask, I could never use a different video under each shape, or could I?
21:36 admsyn I'd think you could? polyline -> mask -> composite -> screen
21:36 admsyn looped for each polyline
21:36 admsyn the problem you might run into is if you have new masks for every polyline you'll be using tons of video memory
21:40 Alski I'm gonna play with putting where the video is built into my contour loop (because I would assume it will load a video for each time it loops?)
22:13 lmccart joined #openframeworks
22:13 lmccart_ joined #openframeworks
23:26 lmccart joined #openframeworks
23:26 lmccart_ joined #openframeworks

| Channels | #openframeworks index | Today | | Search | Google Search | Plain-Text | summary