| *** rZr is now known as rzr | 00:01 | |
| *** rzrmob has joined #harmattan | 00:01 | |
| *** rzrmob has quit IRC | 00:04 | |
| *** rzr is now known as rZr | 00:04 | |
| *** ajalkane has quit IRC | 00:12 | |
| *** oberling_ is now known as oberling | 00:15 | |
| Stecchino | djszapi: no, but I didn't investigate further after seeing with strace that there might be a plugin search pad problem | 00:28 |
|---|---|---|
| Stecchino | didn't have time since then | 00:28 |
| *** b0unc3 has quit IRC | 00:29 | |
| *** b0unc3 has joined #harmattan | 00:31 | |
| *** NIN101 has joined #harmattan | 00:44 | |
| *** b0unc3 has quit IRC | 00:56 | |
| *** b0unc3 has joined #harmattan | 00:58 | |
| *** DrGrov has joined #harmattan | 01:06 | |
| *** NIN101 has quit IRC | 01:14 | |
| *** rZr is now known as rzr | 01:29 | |
| *** rzrmob has joined #harmattan | 01:29 | |
| *** Aranel has quit IRC | 01:37 | |
| *** DrGrov has left #harmattan | 01:39 | |
| *** rzrmob has quit IRC | 01:52 | |
| *** rzr is now known as rZr | 01:52 | |
| *** Natunen has quit IRC | 02:03 | |
| *** acidjunkie has quit IRC | 02:03 | |
| *** juhaj has quit IRC | 02:03 | |
| *** [XeN] has quit IRC | 02:03 | |
| *** juhaj has joined #harmattan | 02:03 | |
| *** [XeN] has joined #harmattan | 02:03 | |
| *** acidjunkie has joined #harmattan | 02:04 | |
| *** jaywink has quit IRC | 02:15 | |
| *** [XeN] has quit IRC | 02:15 | |
| *** M4rtinK has quit IRC | 02:35 | |
| *** rcg1 has quit IRC | 02:41 | |
| *** [XeN] has joined #harmattan | 02:43 | |
| djszapi | Stecchino: oh ok | 02:53 |
| *** magog has joined #harmattan | 02:59 | |
| ZogG_laptop | tgalal: lol | 03:32 |
| ZogG_laptop | why didn't you tell me it was #gentoo | 03:32 |
| tgalal | I eventually did :P | 03:32 |
| tgalal | ZogG_laptop: aparantly the whole awardspace.com is down | 03:34 |
| ZogG_laptop | tgalal: i dunno those sites =\ | 03:35 |
| tgalal | ZogG_laptop: awardspace is my host | 03:35 |
| ZogG_laptop | tgalal: why don'y you use github pages instead? | 03:35 |
| tgalal | ZogG_laptop: I might do so | 03:35 |
| ZogG_laptop | tgalal: few questions. | 03:35 |
| tgalal | sure | 03:36 |
| ZogG_laptop | 1) is there option to get notifications regardless to app status? | 03:36 |
| *** b0unc3 has quit IRC | 03:38 | |
| tgalal | ZogG_laptop: app status, as in running/closed ? | 03:39 |
| ZogG_laptop | yes | 03:39 |
| ZogG_laptop | as sometimes it crashes and we don't know i it's running | 03:39 |
| tgalal | ZogG_laptop: Ah no, that's not possible, unless I'm using Notification API, which I'm not sure Whatsapp supports it for Nokia's Notification API, or if we could use Android's | 03:45 |
| *** b0unc3 has joined #harmattan | 03:45 | |
| tgalal | ZogG_laptop: server is back :)\ | 03:46 |
| ZogG_laptop | tgalal: i got it from github | 03:46 |
| ZogG_laptop | tgalal: but thanks. btw i don't find all emoji, those with people (princess and faces) | 03:47 |
| ZogG_laptop | and in preview of textbox they are displayed different — i like it mroe than original | 03:48 |
| tgalal | ZogG_laptop: from within over 400 emoji, this is the one you're looking for :P | 03:48 |
| tgalal | ZogG_laptop: you mean you like the current way of display in textbox? | 03:48 |
| ZogG_laptop | tgalal: i would like it as default one | 03:49 |
| ZogG_laptop | tgalal: let's say those more usefull than numbers =) | 03:49 |
| tgalal | ZogG_laptop: how many hours did you say it took you to fix blocks on gentoo ? :P | 03:51 |
| ZogG_laptop | tgalal: not really a lot, it was more waiting till it builds as i did build few packages separetly, before i could run emerge -uND world | 03:55 |
| ZogG_laptop | common problems | 03:55 |
| ZogG_laptop | tgalal: need help? | 03:56 |
| tgalal | ZogG_laptop: it's getting stable now | 03:56 |
| tgalal | actually I decided couple of days ago to leave everything as it is (broken) untill I have more time | 03:56 |
| tgalal | but then the laptop battery accidentally ran out of charge and it turned off | 03:57 |
| ZogG_laptop | tgalal: lol | 03:57 |
| ZogG_laptop | tgalal: if need help - tell me | 03:57 |
| ZogG_laptop | i'm pretty good with gentoo | 03:58 |
| tgalal | and then I realised how much the system is broken after switching it on again lol | 03:58 |
| ZogG_laptop | i can be gentooul with you =P | 03:58 |
| ZogG_laptop | tgalal: what didn't run? | 03:58 |
| tgalal | Well udev wasn't starting | 03:58 |
| ZogG_laptop | tgalal: you needed to recompile kernel right? | 03:59 |
| tgalal | it turned later it needed an kernel upgrade | 03:59 |
| ZogG_laptop | i got it today (as last time i just restarted X) | 03:59 |
| ZogG_laptop | tgalal: btw can you check something, where does it mount media? | 03:59 |
| ZogG_laptop | i found out that after restart of udev /dev/sde1 on /run/media/zogg/New | 04:00 |
| tgalal | oh I I'll check | 04:00 |
| ZogG_laptop | when it was /media/New before | 04:00 |
| tgalal | linuxioctl.h:22:44: fatal error: linux/ext2_fs.h: No such file or director | 04:00 |
| tgalal | crap | 04:00 |
| ZogG_laptop | tgalal: where did you get it? | 04:01 |
| ZogG_laptop | and do you hve ext on kernel? | 04:01 |
| tgalal | compiling syslinux | 04:01 |
| tgalal | oh | 04:01 |
| ZogG_laptop | how did you build new? from scratch or oldconfig? | 04:01 |
| tgalal | 1 sec | 04:01 |
| tgalal | old config | 04:01 |
| ZogG_laptop | tgalal: what DE you use? | 04:02 |
| ZogG_laptop | WM? | 04:02 |
| tgalal | currently fluxbox | 04:02 |
| tgalal | gnome was one of casualties lol | 04:03 |
| tgalal | ZogG_laptop: do I need to config from scratch? | 04:07 |
| ZogG_laptop | tgalal: config what? | 04:07 |
| ZogG_laptop | DE? | 04:07 |
| ZogG_laptop | i dont think so | 04:08 |
| tgalal | kernel | 04:08 |
| ZogG_laptop | i use awesome on latop and xfce4 on desktop | 04:08 |
| ZogG_laptop | tgalal: no you don't just thought that error was related to kernel | 04:08 |
| *** beford has quit IRC | 04:13 | |
| *** arcean has quit IRC | 04:14 | |
| *** beford has joined #harmattan | 04:26 | |
| tgalal | ZogG_laptop: masked it | 04:29 |
| *** elldekaa has joined #harmattan | 04:34 | |
| *** npm has quit IRC | 04:47 | |
| *** [XeN] has quit IRC | 04:50 | |
| *** npm has joined #harmattan | 04:51 | |
| *** [XeN] has joined #harmattan | 04:51 | |
| *** adlan has quit IRC | 05:10 | |
| *** oberling_ has joined #harmattan | 05:15 | |
| *** oberling has quit IRC | 05:19 | |
| *** [XeN] has quit IRC | 05:21 | |
| *** natunen has joined #harmattan | 05:48 | |
| *** adlan has joined #harmattan | 06:26 | |
| *** tgalal_ has joined #harmattan | 06:29 | |
| *** tgalal has quit IRC | 06:31 | |
| *** djszapi has left #harmattan | 06:35 | |
| *** hardaker has joined #harmattan | 07:12 | |
| *** hardaker has quit IRC | 07:38 | |
| *** hardaker has joined #harmattan | 07:55 | |
| *** b0unc3 has quit IRC | 08:12 | |
| *** b0unc3 has joined #harmattan | 08:14 | |
| *** beford has quit IRC | 08:17 | |
| *** natunen has quit IRC | 08:33 | |
| *** beford has joined #harmattan | 08:41 | |
| *** GonzoTheGreat_ has joined #harmattan | 08:52 | |
| *** GonzoTheGreat_ has quit IRC | 08:56 | |
| *** GonzoTheGreat_ has joined #harmattan | 08:57 | |
| *** GonzoTheGreat_ has quit IRC | 09:05 | |
| *** GonzoTheGreat_ has joined #harmattan | 09:05 | |
| *** hardaker has quit IRC | 09:07 | |
| GonzoTheGreat_ | Does anybody know how to make a QML LandmarkCategoryFilter work? I have custom landmarks added in c++ which I'd like to display in a MapObjectView. | 09:09 |
| *** Natunen has joined #harmattan | 09:12 | |
| *** DrGrov has joined #harmattan | 09:15 | |
| *** tomma has joined #harmattan | 09:16 | |
| GonzoTheGreat_ | In c++ it works, but I can't get it to work in qml | 09:25 |
| *** acidjunkie has quit IRC | 09:46 | |
| *** acidjunkie has joined #harmattan | 09:47 | |
| *** niqt has joined #harmattan | 09:59 | |
| *** adlan has quit IRC | 10:03 | |
| *** niqt has quit IRC | 10:16 | |
| *** adlan has joined #harmattan | 10:17 | |
| *** NIN101 has joined #harmattan | 10:33 | |
| *** piggz has quit IRC | 10:40 | |
| *** aparaatti has joined #harmattan | 10:43 | |
| *** piggz has joined #harmattan | 10:53 | |
| *** rcg1 has joined #harmattan | 11:06 | |
| *** M4rtinK has joined #harmattan | 11:17 | |
| *** piggz has quit IRC | 11:18 | |
| *** wirwe_ has quit IRC | 11:44 | |
| *** wirwe has joined #harmattan | 11:44 | |
| *** harukomoto has joined #harmattan | 11:55 | |
| *** elldekaa has quit IRC | 11:58 | |
| *** arcean has joined #harmattan | 12:02 | |
| *** elldekaa has joined #harmattan | 12:11 | |
| *** jaywink has joined #harmattan | 12:21 | |
| *** psycho_oreos has joined #harmattan | 12:25 | |
| *** elldekaa has quit IRC | 12:26 | |
| *** elldekaa has joined #harmattan | 12:28 | |
| *** NIN102 has joined #harmattan | 12:34 | |
| *** NIN101 has quit IRC | 12:34 | |
| *** djszapi has joined #harmattan | 12:35 | |
| djszapi | beford: ping | 12:35 |
| *** faenil has joined #harmattan | 12:40 | |
| * djszapi is thinking of publishing an "Ovi daily fun reviews" app to Ovi... | 12:42 | |
| *** rZr is now known as rzr | 12:42 | |
| *** rzrmob has joined #harmattan | 12:42 | |
| *** rzrmob has quit IRC | 12:59 | |
| *** harukomoto has left #harmattan | 12:59 | |
| *** elldekaa has quit IRC | 13:01 | |
| *** psycho_oreos has quit IRC | 13:03 | |
| *** rzr is now known as rZr | 13:12 | |
| *** mschlens has quit IRC | 13:30 | |
| *** mschlens has joined #harmattan | 13:31 | |
| *** beford has quit IRC | 13:44 | |
| *** bef0rd has joined #harmattan | 13:44 | |
| *** djszapi has left #harmattan | 13:55 | |
| *** bef0rd has quit IRC | 14:05 | |
| *** faenil has quit IRC | 14:06 | |
| *** arcean_ has joined #harmattan | 14:10 | |
| *** arcean has quit IRC | 14:10 | |
| *** b0unc3 has quit IRC | 14:28 | |
| *** b0unc3 has joined #harmattan | 14:30 | |
| *** GonzoTheGreat_ has quit IRC | 14:46 | |
| *** faenil has joined #harmattan | 14:50 | |
| *** DocScrutinizer has quit IRC | 14:51 | |
| *** DocScrutinizer has joined #harmattan | 14:52 | |
| *** suy has quit IRC | 14:53 | |
| *** Enforcer has quit IRC | 14:53 | |
| *** decibyte has quit IRC | 14:53 | |
| *** AndrewX192 has quit IRC | 14:53 | |
| *** tomyri has quit IRC | 14:53 | |
| *** bambee has quit IRC | 14:53 | |
| *** fignew has quit IRC | 14:53 | |
| *** alterego has quit IRC | 14:53 | |
| *** AndrewX192 has joined #harmattan | 14:53 | |
| *** tomyri has joined #harmattan | 14:53 | |
| *** suy has joined #harmattan | 14:53 | |
| *** AndrewX192 has quit IRC | 14:53 | |
| *** AndrewX192 has joined #harmattan | 14:53 | |
| *** pa has quit IRC | 14:55 | |
| *** Facefox has quit IRC | 14:55 | |
| *** Facefox has joined #harmattan | 14:59 | |
| *** Enforcer has joined #harmattan | 15:16 | |
| *** decibyte has joined #harmattan | 15:16 | |
| *** bambee has joined #harmattan | 15:16 | |
| *** fignew has joined #harmattan | 15:16 | |
| *** alterego has joined #harmattan | 15:16 | |
| *** piggz has joined #harmattan | 15:23 | |
| *** piggz has quit IRC | 15:34 | |
| *** piggz has joined #harmattan | 15:37 | |
| *** GonzoTheGreat_ has joined #harmattan | 15:40 | |
| GonzoTheGreat_ | /msg NickServ identify ircsucks | 15:40 |
| *** GonzoTheGreat_ has quit IRC | 15:42 | |
| *** jaywink has quit IRC | 15:43 | |
| *** elldekaa has joined #harmattan | 15:47 | |
| *** oberling_ is now known as oberling | 15:47 | |
| ZogG_laptop | lol | 15:52 |
| ZogG_laptop | nice password he used there | 15:52 |
| Tronic | I have always been afraid of making that very specific mistake. | 15:52 |
| SpeedEvil | BTDT | 15:52 |
| Tronic | Either somehow blurbing that on a channel or doing it on another network where some wiseguy is using the nick "nickserv". | 15:52 |
| SpeedEvil | - though with my most secret rot password | 15:53 |
| SpeedEvil | Get up - unplug DSL modem, swear a bit. | 15:53 |
| dm8tbr | Tronic: most networks that don't have services have those nicks banned/reserved. btdtmto | 15:57 |
| ZogG_laptop | Tronic: that's why you do it in status tab | 15:57 |
| dm8tbr | also a classic is: your screen is off, you think it's locked itself, you blindly hack in your password and press enter. whoops it wasn't locked and your irc shell had the focus... btdt | 15:58 |
| Tronic | ZogG_laptop: Actually I've configured Irssi to do it automatically for me. | 16:00 |
| Tronic | Works most of the time at least, maybe not quite 100 % but almost. | 16:00 |
| Tronic | dm8tbr: I never press Enter before the screen turns on. | 16:01 |
| Tronic | Still bad enough if you have someone standing next to you, looking at the screen. | 16:01 |
| Tronic | Another common one: you ssh somewhere and start typing your password a bit too early, so that it appears on screen instead of being hidden on the password prompt. | 16:02 |
| dm8tbr | nowadays my password is too long for the keyboard buffer, so I have to wait... | 16:02 |
| Tronic | Why would a screen locker app take more than a millisecond to react is beyond me, though. | 16:03 |
| Tronic | (well, make that more than 17 ms, the time to draw a single frame, to be fair) | 16:03 |
| *** gabriel9 has joined #harmattan | 16:13 | |
| *** e-yes has joined #harmattan | 16:15 | |
| *** piggz has quit IRC | 16:21 | |
| *** rcg1 has quit IRC | 16:22 | |
| *** hardaker has joined #harmattan | 16:22 | |
| *** mikelima has joined #harmattan | 16:27 | |
| *** tgalal_ has quit IRC | 16:34 | |
| *** elldekaa has quit IRC | 16:35 | |
| *** b0unc3 has quit IRC | 16:47 | |
| aparaatti | btdt? | 16:59 |
| aparaatti | Been There Done That | 17:00 |
| *** gabriel9 has quit IRC | 17:03 | |
| *** swerige has quit IRC | 17:11 | |
| *** e-yes has quit IRC | 17:22 | |
| *** jaywink has joined #harmattan | 17:24 | |
| *** elldekaa has joined #harmattan | 17:27 | |
| *** e-yes has joined #harmattan | 17:31 | |
| *** b0unc3 has joined #harmattan | 17:37 | |
| *** adlan has quit IRC | 17:47 | |
| *** piggz has joined #harmattan | 18:39 | |
| *** aquarius has joined #harmattan | 18:42 | |
| ZogG_laptop | Tronic: me too here with autologin in irssi and autologin in xchat on desktop, but in case you do it, do it on nickserv or status tab =) | 18:43 |
| aquarius | when running a harmattan QML app in scratchbox with Xephyr, I can include QtMultimediaKit which gives me the QML audio element, but... I don't actually get any actual sound when I use it, which makes it hard to tell if it's working :) Can I get sound from a scratchbox environment? | 18:43 |
| *** piggz has quit IRC | 18:47 | |
| *** piggz has joined #harmattan | 18:48 | |
| *** cvaldemar has joined #harmattan | 18:53 | |
| *** elldekaa has quit IRC | 19:00 | |
| Elleo | aquarius: I've had sound out of the scratchbox by using phonon directly | 19:02 |
| Elleo | aquarius: that was in an X86 target, not sure if it'd work in an armel target | 19:02 |
| aquarius | Elleo, ah, phonon. I'm doing all this from QML, rather than from C++... | 19:02 |
| aquarius | (in an x86 target) | 19:02 |
| Elleo | I haven't tried doing any sound from QML, but sound from scratchbox in general seems possible at any rate | 19:04 |
| aquarius | that's encouraging, at least in theory :) | 19:04 |
| Elleo | heh | 19:04 |
| aquarius | sound works fine *on* the n9, which is also nice | 19:04 |
| *** thp_ has quit IRC | 19:04 | |
| *** jaywink has quit IRC | 19:05 | |
| *** thp has joined #harmattan | 19:05 | |
| aquarius | grr, head on the n9 doesn't have -c | 19:05 |
| *** e-yes has quit IRC | 19:08 | |
| *** freemangordon has joined #harmattan | 19:09 | |
| aquarius | ok, now comes the fun test... if I alter an mp3 on the filesystem *while it's playing*, does Qt notice that and work? :) | 19:09 |
| Elleo | heh | 19:09 |
| aquarius | crap. No it doesn't. | 19:11 |
| Elleo | what're you trying to achieve? | 19:11 |
| *** elldekaa has joined #harmattan | 19:11 | |
| aquarius | stream music from the U1 servers, and cache it on the filesystem *and* play it at the same time. Without writing any C++, or ideally any Python :) | 19:12 |
| aquarius | I thought it might be a bit hopeful. So, I either (a) can't use QML Audio for this, or (b) you can only *play* a song once it's downloaded, rather than while it's download*ing*. | 19:12 |
| Elleo | I suppose qml's magic network transparency doesn't extend to streaming music? (with a similar cache option to the one for images) | 19:12 |
| aquarius | well. I can't get a QML Audio element to play something direct from a URL *at all* | 19:13 |
| Elleo | ah, that sucks | 19:13 |
| *** jaywink has joined #harmattan | 19:13 | |
| aquarius | even if I could, there appears to be no way to cache it. | 19:13 |
| aquarius | which is, y'know, infuriating | 19:13 |
| Elleo | yeah | 19:13 |
| aquarius | I really, really, really do not want to have to write a playback thingy in python when QML Audio exists. | 19:13 |
| Elleo | python and gstreamer ;) | 19:14 |
| aquarius | I mean, I *can*, it's just bally hard. :) | 19:14 |
| Elleo | tee the source off to a playing pipeline and a filesink | 19:14 |
| aquarius | by comparison with saying: QML, here is an Audio element, play it. | 19:14 |
| *** piggz has quit IRC | 19:14 | |
| *** freemangordon has left #harmattan | 19:14 | |
| aquarius | ah well. So, now to debate whether to do it the proper but hard way, or just take the chicken's way out. :) | 19:15 |
| Elleo | yourpipeline = gst.parse_launch("urisrc location='%s' ! tee name=t t. ! queue ! filesink location='%s' t. ! queue ! decodebin ! audioconvert ! autoaudiosink" % (mp3url, cachedestination); yourpipeline.set_state(gst.STATE_PLAYING) | 19:16 |
| Elleo | or something to that effect | 19:16 |
| Elleo | (completely untested) | 19:16 |
| *** rcg1 has joined #harmattan | 19:16 | |
| *** e-yes has joined #harmattan | 19:17 | |
| aquarius | ya; that's obviously the Right Thing To DO | 19:18 |
| aquarius | although... hm, doing that means that if I stop playback it'll stop downloading and caching it too, won't it? | 19:18 |
| aquarius | I wonder if *gstreamer* will correctly adapt if I give it a file and then overwrite that file with a longer one while it's playng? | 19:19 |
| Elleo | aquarius: I expect it probably reads everything it can as quickly as possible, so probably won't be reading the file anymore by that point | 19:19 |
| Elleo | also s/urisrc/souphttpsrc/ in my previous pipeline | 19:20 |
| Elleo | and it appears to work (minus your problem of stopping it) | 19:20 |
| *** Shaan7 has quit IRC | 19:20 | |
| *** Shaan7 has joined #harmattan | 19:23 | |
| *** beford has joined #harmattan | 19:27 | |
| *** suy has quit IRC | 19:33 | |
| aquarius | coolio | 19:36 |
| aquarius | and you are correct -- gstreamer does not correctly adapt to a file which changes while it's playing. | 19:40 |
| aquarius | darn. | 19:40 |
| aquarius | obviously the master plan was to just download to a file, and then point gstreamer *at* that file while it was still downloading, which would be ideal if it worked, which it doesn't :) | 19:41 |
| *** mikelima has quit IRC | 19:41 | |
| aquarius | and I think I need to do the download and play steps separately anyway, because you might do this: add two songs to the queue, neither of which are downloaded, and start playing. Song 1 starts downloading and starts playing, but it downloads faster than realtime... when song 1 finishes downloading, song 2 should start downloading, while song 1 is still playing | 19:42 |
| aquarius | then when song 1 finishes playing, song 2 should start playing, even if it's only partially downloaded. | 19:43 |
| aquarius | grr. | 19:43 |
| aquarius | how the hell do our android and iOS clients handle this nightmare, I wonder? | 19:43 |
| *** aparaatti has quit IRC | 19:46 | |
| *** arcean_ has quit IRC | 19:54 | |
| *** arcean_ has joined #harmattan | 19:54 | |
| *** beford has quit IRC | 19:55 | |
| aquarius | heh, QDeclarativeNetworkAccessManagerFactory. I wonder if Audio uses that. | 19:57 |
| aquarius | ha! | 20:02 |
| aquarius | for extra comedy, QML Audio elements on my N9 don't support playing from an SSL url. | 20:02 |
| aquarius | so that's that idea gone for a Burton, then. I shall *have* to download it myself :) | 20:02 |
| Elleo | heh | 20:03 |
| aquarius | is gstreamer the way you're *supposed* to do media stuff on harmattan? | 20:05 |
| aquarius | on the grounds that there are no python bindings for it :) | 20:06 |
| mankeli | is python a supposed way to do anything? :) | 20:07 |
| aquarius | or am I meant to use QSound ? | 20:07 |
| mankeli | gstreamer is afaik the lower level api | 20:07 |
| mankeli | and qt uses that | 20:07 |
| aquarius | mankeli, well, if someone wants to say "writing N9 apps in Python is a hopeless task, give up and use C++ for everything", then I will happily throw my phone away and buy something else ;-) | 20:07 |
| aquarius | but I am currently fighting a lonely battle against the C++ war machine :P | 20:07 |
| mankeli | afaik qt-wrapped stuff is the only "supposed" way of doing anything on harmattan | 20:08 |
| mankeli | kind of official way | 20:08 |
| aquarius | mankeli, right, yeah. I am reading about QSound now | 20:09 |
| aquarius | since right at the moment I don't think doing more low-level direct-to-gst work will actually buy me anything :) | 20:10 |
| mankeli | well especially if there is a python api for qsound, it's probably better to use | 20:10 |
| aquarius | there seems to be | 20:11 |
| aquarius | although QSound is the most barebones API ever. Play a sound... or stop it. That's it. Duration? onFinished event? Metadata? not for you :-) | 20:12 |
| mankeli | yuck. | 20:12 |
| * aquarius rtf-pyside/phonon-m | 20:12 | |
| mankeli | would qtmultimedia help you? | 20:14 |
| aquarius | hm, gpodder must do this simultaneous download-and-playback stuff. | 20:14 |
| mankeli | and there's also that phonon module | 20:14 |
| aquarius | mankeli, yeah, that's what I was looking at before; QML's Audio object was where I started, but it doesn't do https, usefully. Maybe I can use the QtMobility stuff from python, though | 20:14 |
| mankeli | maybe | 20:15 |
| aquarius | QMediaPlayer looks like the answer, here. Now I get to read loads more docs. :) | 20:16 |
| mankeli | yey :) | 20:16 |
| *** swerige has joined #harmattan | 20:25 | |
| *** tom_ has joined #harmattan | 20:25 | |
| Elleo | aquarius: if you do want to use gstreamer from python I can send you a precompiled version of the bindings that you can build into your package | 20:31 |
| aquarius | Elleo, thanks, but no. Doing something like that is about four miles beyond the "this is way too much effort, just use the website" horizon ;) | 20:31 |
| Elleo | heh | 20:31 |
| aquarius | I think this QMediaPlayer stuff might help, but I want to talk to our android/ios people and work out how they're handling this caching thing. | 20:32 |
| *** arcean_ has quit IRC | 20:36 | |
| *** diorahman has joined #harmattan | 20:42 | |
| diorahman | my n950 doesn't give any response to flasher -i, is there a way to re-flash the bootloader (NOLO?) via other ports except USB? | 20:42 |
| *** hardaker has quit IRC | 21:01 | |
| *** arekinath has quit IRC | 21:05 | |
| *** arekinath has joined #harmattan | 21:06 | |
| *** arekinath has joined #harmattan | 21:06 | |
| *** e-yes has quit IRC | 21:06 | |
| aquarius | thp, ping? | 21:07 |
| *** swerige has quit IRC | 21:08 | |
| thp | aquarius: pong | 21:09 |
| aquarius | thp, in gpodder, if someone starts playing a podcast before it's completely downloaded, how do you handle that? When the audio stuff says "ok, that's finished playing", do you look at the file and think "hang on, it only got halfway through the full duration; I'll load that file and play it again, but seek to their last position"? | 21:12 |
| thp | aquarius: when you play it before it has been (completely) downloaded, the playing is streaming | 21:13 |
| *** b0unc3 has quit IRC | 21:13 | |
| thp | i think right now, i don't switch to the file when the download is finished | 21:13 |
| aquarius | hm. How do you stream it *and* save the file to disk at the same time? :) | 21:13 |
| thp | well yeah, that's kind of wasteful.. but easier | 21:13 |
| thp | streaming happens via QML's Audio element | 21:14 |
| thp | downloading via the gpodder core | 21:14 |
| thp | whenever playback of an episode starts, it's checked whether to play it from MyDocs or stream it from the download url | 21:14 |
| dm8tbr | diorahman: what did you do to the device that you thing this is what would help you? | 21:14 |
| *** b0unc3 has joined #harmattan | 21:14 | |
| dm8tbr | s/thing/think/ | 21:15 |
| infobot | dm8tbr meant: diorahman: what did you do to the device that you think this is what would help you? | 21:15 |
| aquarius | thp, what I'm not quite following is, er, if I say "play this file" and it's not downloaded, you just stream it from the download URL and it is never downloaded and saved into MyDocs at all? Or do you download it to the device *twice*, once direct into the QML Audio element and then again by gpodder core onto the filesystem? | 21:16 |
| *** swerige has joined #harmattan | 21:17 | |
| thp | aquarius: yes, if you just stream it, the stream is not saved | 21:18 |
| aquarius | thp, oh. Darn. I was hoping you had some super-clever solution that I hadn't thought of ;-) | 21:18 |
| thp | ;) | 21:18 |
| thp | i can't really feed "incomplete" or streaming data to QML Audio AFAIK | 21:18 |
| thp | otherwise that would be cool | 21:18 |
| aquarius | I can't stream the music i want to play, because QML Audio doesn't support https on N9, uselessly. :( | 21:18 |
| *** e-yes has joined #harmattan | 21:18 | |
| thp | aquarius: you could create a fifo file and have your program write to it (reading from https) and let qml audio "play back" that fifo? | 21:19 |
| aquarius | ooh. a fifo. | 21:19 |
| thp | what's your use case btw? | 21:19 |
| aquarius | cor. | 21:19 |
| aquarius | that's mildly complex. :) | 21:19 |
| aquarius | use case: playing streaming music from Ubuntu One | 21:19 |
| thp | and i guess you want to avoid doing "two downloads" of the same data? | 21:21 |
| aquarius | actually, the fifo trick won't help me save and play back at the same time, 'cos the thing which saves to the filesystem will read all the bytes out of the fifo and then the audio player won't have anything left to play :) | 21:21 |
| aquarius | thp, well. I'd do two downloads if I could -- it's inefficient and I'd certainly like to avoid it, but I'm quite lazy. However, I can't -- QML Audio doesn't do https, and Ubuntu One is all https. So I *have* to do something other than just point QML's Audio at the download URL :) | 21:22 |
| thp | aquarius: couldn't that app that reads from https just save both to disk (as a file) and write the same data to the fifo? | 21:22 |
| aquarius | it could, but we might be downloading a song well in advance of playing it, which means we'll run out of buffer in the fifo | 21:23 |
| aquarius | this would all be perfect if any audio system at all anywhere coped with you overwriting a file while it's playing. But they do not ;) | 21:24 |
| thp | aquarius: i haven't checked how qml audio deals with reading from a download-in-progress file | 21:24 |
| aquarius | I have. :) It plays up to the point where the file ended when it started playing | 21:25 |
| aquarius | annoyingly enough :) | 21:25 |
| aquarius | I think maybe I'll do the idea of checking when a playback hits EOF whether it's the same EOF as it was when we started | 21:25 |
| thp | of course! ;) otherwise that would have been too easy :p | 21:25 |
| aquarius | and if it isn't, then play the file *again* but seek to the current point. | 21:25 |
| thp | i mean, you could probably use gstreamer's low-level api, but that's probably overkill | 21:25 |
| aquarius | having to do this is a pain in the hole by comparison with how gstreamer should do it for me, but there it is :) | 21:25 |
| thp | yeah, and qml audio is just so simple and easy to use (and also takes care of stuff like pausing when the headphones are unplugged) | 21:26 |
| aquarius | raw gstreamer does the same thing as QML, here; if you start playing a file and then overwrite the file while still playing, it plays up to where the file ended when it started playing | 21:26 |
| aquarius | exactly. I really don't want to write a whole complex gst thing when QML's just there for me :) | 21:26 |
| aquarius | so, that sounds like the way. Darn it. I was hoping it'd be less fiddly than this. Never mind :) | 21:27 |
| *** diorahman has quit IRC | 21:28 | |
| thp | but i think the fifo solution would be kind of cool | 21:28 |
| thp | you would have to have some coordination between the gui and the downloader, though | 21:28 |
| thp | and check if you can write non-blocking to the fifo | 21:28 |
| thp | or! | 21:29 |
| thp | write a http proxy that sits locally and exposes the audio via http, let the QML audio connect to it and forward the requests to u1's https (and save the stream while passing data through) | 21:29 |
| * aquarius gives thp a long-suffering look | 21:30 | |
| aquarius | that'd work, I suppose | 21:30 |
| aquarius | overkill, much? Stupid QML. :-) | 21:30 |
| aquarius | if I have to write a mini caching http proxy just to get QML to play things then I'm going to find someone who works for Nokia and pour a cup of tea into their lap | 21:31 |
| *** gabriel9 has joined #harmattan | 21:32 | |
| aquarius | actually...if QML does stuff with range requests then I wouldn't even need to make it a *proxy*, per se -- I'd just open an http server which serves the filesystem, but if you request a file which is in the process of downloading, I'll just lie about how long it is. :) | 21:33 |
| thp | yep | 21:35 |
| thp | and write it in a generic way, so that i can steal it and use it for gpodder :p | 21:35 |
| aquarius | :) | 21:35 |
| ZogG_laptop | thp: do you code only in pyhton? | 21:35 |
| thp | ZogG_laptop: no, not only but still mainly | 21:36 |
| thp | aquarius: https://github.com/gpodder/gpodder/blob/master/src/gpodder/qmlui/helper.py#L75 <- the MediaButtonsHandler class might be helpful if you want to support bluetooth headset buttons (that's not handled by qml audio directly) | 21:40 |
| aquarius | ooh, that sounds nice | 21:40 |
| aquarius | although I don't have a BT headset, so testing will be hard:) | 21:40 |
| thp | if you plan to get one, I'd recommend something like the BH-214 - it has a 3.5mm slot, so you can use it with all kinds of line in stuff, not just headphones | 21:43 |
| *** rcg1 has quit IRC | 22:08 | |
| *** arcean has joined #harmattan | 22:14 | |
| *** arcean has left #harmattan | 22:18 | |
| *** arcean has joined #harmattan | 22:19 | |
| aquarius | bah, I wish QML Audio had an onDurationIsNowAvailable event. | 22:20 |
| *** gabriel9 has quit IRC | 22:28 | |
| *** jaywink has quit IRC | 22:38 | |
| thp | aquarius: it does | 22:40 |
| aquarius | ! | 22:40 |
| thp | "somehow" | 22:41 |
| thp | let me find the link | 22:41 |
| aquarius | not according to the documentation it doesn't ;) | 22:41 |
| thp | aquarius: onDurationChanged and then check for duration > 0 | 22:41 |
| aquarius | I am eager to see this! | 22:41 |
| aquarius | hmph. Entirely *undocumented* signal. Thanks, QML. | 22:41 |
| thp | actually that's an implicit signal | 22:42 |
| thp | for every "property", you have a "onPropertyChanged" signal whenever it changes | 22:42 |
| aquarius | omg! | 22:42 |
| aquarius | honestly??? | 22:42 |
| thp | yep | 22:42 |
| aquarius | wow, no-one tells me anything | 22:42 |
| aquarius | OK, that's the most useful piece of information I've heard ever in the whole of today. | 22:43 |
| thp | http://doc.qt.nokia.com/4.7-snapshot/qdeclarativeintroduction.html#property-change-notifications | 22:43 |
| * aquarius adds a tally mark to the "beers I owe thp" list. | 22:43 | |
| thp | ;) | 22:45 |
| *** arcean has quit IRC | 22:49 | |
| *** arcean has joined #harmattan | 22:50 | |
| *** DocScrutinizer2 has joined #harmattan | 22:51 | |
| *** DocScrutinizer2 has quit IRC | 22:53 | |
| *** DocScrutinizer2 has joined #harmattan | 22:53 | |
| *** DocScrutinizer2 has joined #harmattan | 22:53 | |
| *** Stecchino has quit IRC | 22:54 | |
| *** Stecchino has joined #harmattan | 22:57 | |
| *** Stecchino has joined #harmattan | 22:57 | |
| *** Sniper_swe has joined #harmattan | 22:58 | |
| *** elldekaa_ has joined #harmattan | 23:02 | |
| *** elldekaa has quit IRC | 23:05 | |
| *** faenil has quit IRC | 23:08 | |
| aquarius | grr! | 23:09 |
| aquarius | "this" isn't bound. | 23:09 |
| aquarius | how can I call a QML widget's method from inside another method without using the widget's ID? | 23:09 |
| *** niqt has joined #harmattan | 23:09 | |
| Jaffa | aquarius: By "method" you mean function? If so, you just do foo() - rather than this.foo() | 23:10 |
| aquarius | (that is: I could just do this... Thingy { id: thingy1; onSomething: { thingy1.somethingElse() } } ) | 23:10 |
| aquarius | Jaffa, oh, really? | 23:10 |
| aquarius | I just don't qualify it at all? | 23:10 |
| aquarius | heh. | 23:10 |
| Jaffa | aquarius: Exactly. | 23:10 |
| aquarius | Excellent, that works | 23:11 |
| aquarius | I didn't think of that | 23:11 |
| aquarius | because it is mad. :-) | 23:11 |
| aquarius | thanks, Jaffa :) | 23:11 |
| Jaffa | aquarius: It's standard JavaScript and, from your "this" I assume a Java background, Java too | 23:12 |
| *** piggz has joined #harmattan | 23:13 | |
| aquarius | Jaffa, nah, I'm a JS guy. It's standard JS if the functions are closed over, right enough, but then they're not available from outside :) | 23:13 |
| Jaffa | aquarius: Indeed. QML is file-level scope, I think (rather than item-level) | 23:13 |
| aquarius | var Thingy = function() { var inner = function() {}; this.outer = function() { inner(); } }; var thingy = new Thingy(); :-) | 23:14 |
| aquarius | so I instinctively think of things declared as visible from outside to require "this." on the front :) | 23:14 |
| Jaffa | aquarius: Those are anonymous functions. Same as when you do onSomething: { somethingElse() }. But "somethingElse" is just defined with "function somethingElse() { ... }" | 23:14 |
| Jaffa | QML syntax edge cases: you could write a book on 'em ;-) | 23:15 |
| aquarius | true enough, on both counts :) | 23:16 |
| aquarius | the grey area is where QML's idiosyncracies and *my* idiosyncracies collide. :) | 23:16 |
| thp | iirc you can even call functions in "outer" qml objects deeper in the object tree | 23:19 |
| Jaffa | thp: Yes, the scoping of 'id' is something I'm either confused about, worried about or ignoring depending on my mood. | 23:21 |
| aquarius | aaaaaaaaaaaaaaaargh bloody qml! | 23:21 |
| aquarius | overwriting a file while it's playing *sometimes* updates the duration. | 23:21 |
| * aquarius headbutts the keyboard | 23:21 | |
| * Jaffa beds | 23:21 | |
| thp | http://pastebin.com/awRf53tc <- example of that | 23:22 |
| aquarius | that means I can't tell the difference, inside the Audio object, between "we stopped because we'd run out of file, but the file's been updated and there is actually more to play and I ignored it" and "the user pressed stop before the end". | 23:22 |
| thp | Jaffa: yep, you can do "anchors.centerIn: myObject" in a .qml file and the id myObject is defined somewhere else ;) which is kind of weird (and confusing to the reader) | 23:22 |
| thp | aquarius: maybe you can detect it via "status"? http://doc.qt.nokia.com/qtmobility-1.2/qml-audio.html#status-prop | 23:23 |
| thp | EndOfMedia vs InvalidMedia or something maybe? | 23:23 |
| aquarius | well. status seems to be *load* status, not playing status: in particular, if you've played to the end of a file and then start playing it again, status seems to always be 6 for EndOfMedia even while it's playing :( | 23:25 |
| aquarius | mildly pleased with this, though; it seems to work. | 23:25 |
| aquarius | but I can test for that by trapping duration updates which happen while playing, can't I? | 23:26 |
| thp | yep, i guess so | 23:28 |
| aquarius | haha! it works! | 23:32 |
| aquarius | glory be. | 23:32 |
| aquarius | now to decide on a name for the component | 23:32 |
| aquarius | AudioWhichDealsWithQMLsStupidWayOfHandlingChangingFiles | 23:33 |
| aquarius | bit passive-aggressive. | 23:33 |
| aquarius | http://paste.ubuntu.com/1010282/ should be the basics of a reusable component (although I haven't split it out into a separate file yet) -- test it by getting a reasonably short mp3 file and pointing "source" at it, pressing the window to play, and then *while it's playing* overwrite the short mp3 file with a longer mp3 file | 23:38 |
| aquarius | it should play to the point where the short mp3 file ended, and then carry on with the rest of it | 23:39 |
| SpeedEvil | Umm. | 23:40 |
| SpeedEvil | you mean it skips back to the beginning when it hits the end? | 23:40 |
| *** b0unc3 has quit IRC | 23:42 | |
| aquarius | no | 23:43 |
| aquarius | if you get to the "end", but the file has grown in the meantime, then it reloads the file and continues playing from where you got to. | 23:43 |
| aquarius | so it's relatively seamless | 23:43 |
| aquarius | you do get a little click/jump in the music | 23:43 |
| aquarius | but, well, fixing that is absurdly low-level; if someone wanted to fix that they might as well fix gstreamer to handle a playing file changing while playing :P | 23:44 |
| *** b0unc3 has joined #harmattan | 23:45 | |
| SpeedEvil | That seems like insanity. | 23:46 |
| SpeedEvil | It's going to mess up badly on some mp3 changes. | 23:46 |
| ieatlint | you'd need to change how you load the audio... doing it from a fileio isn't going to work | 23:48 |
| ieatlint | you can't buffer something if you change the source | 23:48 |
| SpeedEvil | Depending on how the 'replace' is done - it may not work at all. | 23:48 |
| ieatlint | you'd need to do it on the c++ side, and stream the audio to it | 23:48 |
| SpeedEvil | If something has deleted the file - and replaced it - then you have a deleted file-handle that remains the original file until you close it at which time it's dropped. | 23:49 |
| ieatlint | yeah, i was impressed that it would work... qt is handling the file descriptor change better than i'd expect | 23:49 |
| SpeedEvil | If something overwrites the file - then it's going to depend on caching behaviour. | 23:49 |
| ieatlint | perhaps qt proxies an fd for the qml component, and in the component it detects the "source" changed | 23:50 |
| SpeedEvil | And the decoder not barfing on format changes. | 23:50 |
| ieatlint | would be my guess | 23:50 |
| *** b0unc3 has quit IRC | 23:53 | |
| *** b0unc3 has joined #harmattan | 23:55 | |
| *** ant has joined #harmattan | 23:56 | |
| *** gabriel9 has joined #harmattan | 23:59 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!