RST38h | no? | 00:00 |
---|---|---|
RST38h | what version of qemu should I install and from where? | 00:00 |
RST38h | (none of this stuff appears to be documented btw) | 00:00 |
kimju | ii qemu-user-static 0.14.50-2011.03-1-0ubuntu2 QEMU user mode emulation binaries (static version) | 00:01 |
*** willer_ has quit IRC | 00:01 | |
kimju | I had to install that manually (on ubuntu 11.04) | 00:01 |
RST38h | scratchbox-devkit-qemu is already the newest version. | 00:01 |
RST38h | no such package | 00:02 |
*** baraujo has quit IRC | 00:02 | |
*** hardaker has quit IRC | 00:03 | |
*** fiferboy has quit IRC | 00:03 | |
RST38h | shit , this proves to be way more complicated than I thought =( | 00:03 |
kimju | on the sdk site there is links to some tarballs too. I preferred to use the packaged version | 00:04 |
RST38h | okit was not obvious but I did it | 00:06 |
RST38h | omething wicked happened resolving 'harmattan-dev.nokia.com:80' | 00:07 |
RST38h | and the same thingfor each harmattan repo | 00:07 |
RST38h | naaah, it does not start | 00:15 |
*** piggz has quit IRC | 00:15 | |
RST38h | has anyone managed to run this stuff successfully? | 00:17 |
rm_work | i ran the SB installer on my ubuntu natty machine | 00:20 |
rm_work | and it installed everything fine, | 00:20 |
rm_work | including qemu / xephyr / etc | 00:20 |
rm_work | was able to run a session and launch an app | 00:20 |
rm_work | after which i immediately shut it off and just used my device anyway | 00:20 |
rm_work | i've been coding on-device most of today and last night | 00:20 |
RST38h | and compiling too? | 00:21 |
RST38h | I managed to boot the desktop in X86 target now, but qmlviewer complains about missing dbus and quits | 00:21 |
mgedmin | if you're developing a qml app, I recommend qt creator | 00:22 |
mgedmin | it's android-level easy | 00:22 |
mgedmin | hit a "run" button in the ide, it builds a .deb, deploys it over ssh over wifi, and runs it on your n950 | 00:23 |
mgedmin | I believe it also supports integrated debugging, although I haven't tried | 00:23 |
mgedmin | bliss | 00:23 |
mgedmin | of course if you've got a preexisting .deb package, I've no clue how to make qt creator play nice with it, or if that's even possible | 00:23 |
RST38h | I am acquainted with Android dev now, it is not "easy" | 00:23 |
RST38h | "Messy" would be a better word. | 00:23 |
mgedmin | well, I'm not acquainted with android | 00:24 |
mgedmin | I've played with it for 5 minutes, thought it was a bit more polished than scratchbox | 00:24 |
RST38h | In a nutshell, it is Symbian rewritten in Java | 00:24 |
mgedmin | I'm not acquainted with symbian either :) | 00:24 |
RST38h | I guess, Harmattan desktop sources are not public? :) | 00:26 |
*** piggz has joined #harmattan | 00:27 | |
mgedmin | who knows? | 00:31 |
mgedmin | what are 'desktop sources' -- meegotouch home screens? | 00:31 |
*** apol_ has quit IRC | 00:34 | |
*** stroughtonsmith has quit IRC | 00:35 | |
*** piggz has quit IRC | 00:37 | |
*** gri_ has quit IRC | 00:38 | |
*** hardaker has joined #harmattan | 00:38 | |
mgedmin | brain not working... | 00:39 |
mgedmin | what's the screen size of a n950? | 00:40 |
mgedmin | 854? | 00:40 |
mgedmin | xwininfo -root tells me it's 854 | 00:44 |
macmaN | 848x480 | 00:47 |
*** CaCO3 has quit IRC | 00:47 | |
macmaN | it takes less time to google it than it takes to write the question here | 00:48 |
mgedmin | but the answer is wrong :) | 00:48 |
*** epage has joined #harmattan | 00:49 | |
MohammadAG | 854*480 me thinks | 00:50 |
mgedmin | yes | 00:50 |
*** hardaker has quit IRC | 00:54 | |
*** antman8969 has joined #harmattan | 01:00 | |
mgedmin | \o/ ccache \o/ | 01:01 |
mgedmin | crud, couldn't fix the window size flicker on startup | 01:06 |
mgedmin | and couldn't add a "fullscreen" button to the toolbar.xml either | 01:10 |
lcuk | serious question, could symbian be made to work inside a virtual machine on harmattan? | 01:21 |
lcuk | ie virtual s60 | 01:21 |
lcuk | or whichever is the required version | 01:21 |
SpeedEvil | It would be interesting! | 01:21 |
DocScrutinizer | well, a special qemu shouldn't expose any showstoppers | 01:23 |
* DocScrutinizer idly wonders what became of vmware for arm / N900 | 01:24 | |
* vandenoever is running a plain Qt application and getting a gray bar below the application, any way to remove that? | 01:24 | |
DocScrutinizer | I guess that's the space for status bar supposed to show up on top | 01:25 |
mgedmin | vandenoever++, I want to know too | 01:25 |
macmaN | DocScrutinizer: at least there's a wine build now. that took a while. | 01:25 |
antman8969 | that gray bar is the toolbar menu | 01:25 |
antman8969 | use qml! | 01:25 |
DocScrutinizer | hah, sure | 01:25 |
DocScrutinizer | toolbar without any tools | 01:26 |
*** Reffyy has joined #harmattan | 01:26 | |
antman8969 | I didn't say it was a USEFUL toolbar | 01:26 |
antman8969 | the kind that just takes up space | 01:26 |
vandenoever | my widget uses QWidget, i dont want to port it, that would be a huge task | 01:26 |
* DocScrutinizer neither | 01:26 | |
antman8969 | you can use QWidget in qml | 01:26 |
antman8969 | theres a tutorial somewhere | 01:27 |
ieatlint | yeah, you can, but it's considered a bad idea | 01:27 |
*** jcharpak has joined #harmattan | 01:28 | |
antman8969 | lol well so is not using qml.... ok I'm not being helpful I'm sorry | 01:28 |
* vandenoever is running warptree on his n950, performance is great, but it's a qwidget | 01:28 | |
* macmaN *still* doesnt know the difference between all these | 01:29 | |
* mgedmin doesn't even know how to tell what is what -- fbreader is using qt4, that's all I know | 01:29 | |
antman8969 | lol it's all qt4 | 01:29 |
rm_work | lol well | 01:30 |
SpeedEvil | lcuk: I wonder if there are any binary compatible chipsets, so wine, not qemu would work. | 01:30 |
vandenoever | so this toolbar, how can it be turned off? | 01:30 |
rm_work | I have the same problem with pyRadio, big grey bar, since it uses PySide Qt4 libs | 01:30 |
antman8969 | i'm guessing using the meegotouch libraries | 01:30 |
antman8969 | but those are deprecated | 01:30 |
*** smoku1 has quit IRC | 01:30 | |
antman8969 | afaik | 01:31 |
ieatlint | in short: the whole dev environment is a clusterfuck | 01:31 |
mgedmin | s/dev environment/platform/, I suspect, but then I'm a dour pessimist | 01:32 |
ieatlint | nah, it's more the conflicting dev environments and similar | 01:32 |
antman8969 | pretty much | 01:32 |
vandenoever | hiding some of the QApplication::topLevelWidgets() does not help | 01:34 |
antman8969 | yea i'm pretty sure the answer doesn't lie in standard Qt | 01:34 |
*** rm_work has quit IRC | 01:34 | |
ieatlint | it's truly a: 1. don't use qwidgets, they won't work. 2. don't use mtf, it's deprecated (but you can use it, it'll work). 3. use qml and the non-portable harmattan qml components. they'll probably work, but you'll never be able to port your app | 01:34 |
antman8969 | yes... but 3 is exaggerated | 01:35 |
ieatlint | (and by qwidgets not working, i mean they don't provide any style sheets, and certain widgets are so horribly not touch friendly that they are completely unusable) | 01:35 |
antman8969 | if you stick to com.meego in the QML the app isn't hard to port at all | 01:36 |
antman8969 | from another QML | 01:36 |
mgedmin | so, if I've got a desktop app that uses qt 4, and I don't know qt myself, and I don't know qml, where do I start? | 01:37 |
mgedmin | hello world qml tutorial? | 01:37 |
mgedmin | wrap the custom qt classes of that app as some sort of components and use them in my hello world qml app? | 01:37 |
mgedmin | other things I don't know at all include meegotouch and qt widgets | 01:38 |
mgedmin | or was that qwidgets | 01:38 |
ieatlint | theoretically yes, but you're biting off a HUGE task | 01:38 |
mgedmin | my ignorance is just staggering | 01:38 |
antman8969 | imo... learning QML is very easy and very fast, and making a Ui with it is even faster | 01:38 |
vandenoever | mgedmin: which desktop app? | 01:38 |
antman8969 | you can use the same application logic | 01:38 |
ieatlint | yes, you can pick up qml in a week or two if you don't hit any walls | 01:39 |
mgedmin | vandenoever, fbreader | 01:39 |
ieatlint | but when you get into more complicated things, like embedding qwidgets into qml, you'll have a lot more trouble | 01:40 |
mgedmin | yeah, my other goal is to port a pygtk app to a n950 ;) | 01:40 |
lcuk | mgedmin, is it written in qt4 or is qt4 just a shim around the library code? | 01:40 |
mgedmin | although I plan to take a deep look at gpodder, since it managed to do that somehow | 01:40 |
mgedmin | lcuk, there's an abstraction layer on top of qt4 | 01:40 |
mgedmin | with side-by-side alternative implementations for gtk, hildon, etc... that I intend to mostly ignore | 01:41 |
ieatlint | yeah, so typically here you'd write a new implementation in qml and then have a non-gui qt4 backend to interface with the library | 01:41 |
mgedmin | non-gui, mwahahahaha | 01:43 |
lcuk | mgedmin, shame you are not in #meego, we could have discussed a version suitable for any of the meego uxes | 01:43 |
mgedmin | I used to be in #meego, but didn't have the time to read what people were writing | 01:43 |
mgedmin | I'm still lurking in meego-dev@ | 01:43 |
mgedmin | turns out without a physical device I can't make myself interested in a platform | 01:44 |
mgedmin | if it picks technologies I didn't like (rpm, qt) | 01:44 |
ieatlint | harmattan isn't rpm :P | 01:44 |
SpeedEvil | "" | 01:44 |
SpeedEvil | :/ | 01:44 |
mgedmin | harmattan is not meego :) | 01:45 |
SpeedEvil | Lurking here, and attempting to keep sorta-current is the limit of what I can bring myself to do ATM | 01:45 |
ieatlint | that gets far more complicated when you try to define what meego is :P | 01:45 |
jcharpak | fbreader has a gtk+ based ui, a qt4 ui and a qt3 ui of all things | 01:45 |
mgedmin | lots more, I think | 01:45 |
mgedmin | it supports bazillion PDAs | 01:46 |
SpeedEvil | fbreader would be awesome. | 01:46 |
SpeedEvil | I use it a lot. | 01:46 |
mgedmin | the desktop version mostly works | 01:46 |
SpeedEvil | I hope you're not going down the route of scrolly bling ? | 01:46 |
mgedmin | it crashes on task switch -- a segfault in someclassIforgot->device() | 01:46 |
jcharpak | btw I'm fairly sure the maemo ui is gtk based,partially hildonized | 01:48 |
ieatlint | maemo5 is, maemo6/harmattan is not | 01:49 |
ieatlint | there aren't any gtk or hildon libs installed | 01:49 |
jcharpak | well yeah maemo6 version of fbreader doesn't exist yet, right? | 01:49 |
ieatlint | oh, i'm not thinking clearly :) | 01:50 |
ieatlint | i misunderstood what you meant | 01:50 |
jcharpak | sorry lost context the maemo ui of *fbreader* is gtk based | 01:50 |
ieatlint | yeah, i realized that after your comment | 01:51 |
lcuk | mgedmin, where is latest fbreader code? | 01:51 |
*** seif has quit IRC | 01:52 | |
jcharpak | my bad, typing using the vkb on my n800 | 01:52 |
mgedmin | lcuk, that's an ambiguous version | 01:52 |
mgedmin | s/version/question/ | 01:52 |
infobot | mgedmin meant: lcuk, that's an ambiguous question | 01:52 |
mgedmin | upstream's latest is on github | 01:52 |
mgedmin | my latest .debs are at http://mg.pov.lt/770/dists/harmattan/experimental/ | 01:52 |
mgedmin | they've the same level of functionality as fiferoboy's, currently | 01:52 |
lcuk | mgedmin, i have liqbook | 01:53 |
lcuk | pondering since fbreader has been ported to many uis | 01:53 |
lcuk | if it would be feasible to do a liq* ui | 01:53 |
mgedmin | but I have a source .deb you can dpkg-buildpackage | 01:53 |
lcuk | then it would work on all maemo/meego etc | 01:53 |
mgedmin | yeah, I haven't looked deeply into fbreader's code | 01:53 |
mgedmin | I'm sure it's possible, given the number of platforms fbreader's already been ported to | 01:53 |
lcuk | yeah | 01:54 |
lcuk | where on github? | 01:54 |
jcharpak | upstream has mostly given up on linux and seems to be mostly working on his port to android | 01:54 |
*** hardaker has joined #harmattan | 01:54 | |
mgedmin | that's where the most users are, I suspect | 01:54 |
ieatlint | don't say that, you're ruining our blindness to the outside world | 01:55 |
mgedmin | wouldn't it be nice if harmattan could run android apps? | 01:55 |
ieatlint | i seem to recall a lot of effort to get the n900 to run some android apps | 01:56 |
jcharpak | nitdroid apps work, mostly, calls don't :) | 01:57 |
ieatlint | alien dalvik | 01:57 |
jcharpak | vaporware | 01:57 |
ieatlint | i suspect we'll see something do the job, but it'll be so painfully slow that it's not practical | 01:59 |
* lcuk goes to bed to read more niven | 02:03 | |
lcuk | nn \o | 02:03 |
mgedmin | what are you reading, lcuk? | 02:04 |
*** Rizzer has joined #harmattan | 02:04 | |
lcuk | mgedmin, prequals to ringworld | 02:04 |
lcuk | then for the first time in years, ringworld :) | 02:05 |
* lcuk is looking forward to it | 02:05 | |
* mgedmin has read ringworld, but no sequels or prequels | 02:06 | |
*** seif has joined #harmattan | 02:08 | |
SpeedEvil | I recommend 'protector' | 02:08 |
* mgedmin liked the mote in god's eye best from all niven's output, IIRC | 02:09 | |
SpeedEvil | That was niven/pournelle | 02:09 |
mgedmin | author collaborations break my mental model of books, damnit | 02:10 |
SpeedEvil | http://www.baen.com/library/ | 02:10 |
SpeedEvil | Baen free library. Assorted SF. | 02:10 |
mgedmin | also http://baencd.thefifthimperium.com/, more freely redistributable stuff | 02:10 |
mgedmin | baen has sane DRM policies, i.e. no DRM at all | 02:11 |
SpeedEvil | Indeed. | 02:11 |
SpeedEvil | I also use #ebooks over on undernet to get copies of books I own in hardcopy as ebooks. | 02:11 |
SpeedEvil | Also no DRM. | 02:11 |
SpeedEvil | Though only books I actually own. | 02:12 |
SpeedEvil | (including those that I've bought that I can't rea because they have DRM) | 02:12 |
DocScrutinizer | :-D | 02:13 |
* SpeedEvil sighs and wishes he could find a nice CO2 sensor. (to measure expired CO2 for metabolic rate computation. | 02:14 | |
SpeedEvil | err - mischa | 02:14 |
SpeedEvil | n | 02:14 |
DocScrutinizer | errr | 02:14 |
mgedmin | why doesn't the n950 come with a CO2 sensor, really? :D | 02:14 |
DocScrutinizer | mgedmin: because it's not designed by me :-D | 02:15 |
vandenoever | SpeedEvil: you have one in your body | 02:15 |
SpeedEvil | vandenoever: I know. Unfortunately, I have no easy way of reading it. I can maintain my weight well, when I'm not active. But I have real difficulty working out how many calories I'm burning when doing random stuff like picking rasperries or ... | 02:16 |
*** hardaker has quit IRC | 02:17 | |
DocScrutinizer | SpeedEvil: I'm *sure* you can get fairly exact air CO2estimations when you know about air's original CO2 on breath intake (should be ~0%) and blood saturation that's easily obtainable with those IR clips aiui | 02:18 |
*** andre__ has quit IRC | 02:19 | |
SpeedEvil | You can't. Blood oxygenation (carbonation is harder to measure) does not closely track. Bloodflow - for one - compounds it. | 02:19 |
SpeedEvil | Heartrate is a not-so-good proxy that I should probably do instead. | 02:20 |
*** Rizzer has quit IRC | 02:20 | |
*** jcharpak has left #harmattan | 02:20 | |
mgedmin | brainnotworky, sleeptime zzzz | 02:24 |
SpeedEvil | Night. | 02:24 |
DocScrutinizer | SpeedEvil: recal that face color detection? | 02:26 |
SpeedEvil | You mean tracking heartrate and breathing rate through differential colour on exposed skin? | 02:26 |
DocScrutinizer | yo, as a first shot | 02:27 |
SpeedEvil | In principle possible, in practice not really in this app. | 02:27 |
SpeedEvil | (at least for me) | 02:27 |
DocScrutinizer | just wild brainstorming - felt like it :-) | 02:28 |
*** deimos has quit IRC | 02:29 | |
DocScrutinizer | given the premise you already looked for CO2 sensors, so I won't uselessly duplicate your work | 02:29 |
DocScrutinizer | usually you find the better chips than me ;-) | 02:29 |
SpeedEvil | It woul be an interesting add-on to the camera - informed by the face-tracking, also pop up a little heartrate balloon next to the person. | 02:30 |
DocScrutinizer | hehe | 02:30 |
* DocScrutinizer glares to left: N950. To right: Torx-TX6 and other tools | 02:37 | |
* DocScrutinizer ponders | 02:37 | |
Tronic | Now, did the agreement say anything about physically disassembling the device? | 02:38 |
DocScrutinizer | nope | 02:38 |
DocScrutinizer | :-P | 02:38 |
Tronic | Looks like you could get to the battery by opening those four T6 screws. | 02:38 |
DocScrutinizer | those are TX-4 | 02:39 |
Tronic | Ok, couldn't tell by eye. | 02:39 |
DocScrutinizer | been there :-D | 02:39 |
DocScrutinizer | followed achipa's unboxing instructions faithfully :-D | 02:39 |
DocScrutinizer | not really, the screws are still available | 02:40 |
*** wicket64 has quit IRC | 02:49 | |
realsportcars | MohammadAG? | 02:51 |
* vandenoever ported warptree to n950 http://www.vandenoever.info/software/warptree/ would be interesting as a touchfriendly file selector | 02:54 | |
vandenoever | http://oeba.nl/warptree_0_0_1_armel.deb | 02:54 |
vandenoever | must be run as '/opt/dirview/bin/dirview' atm | 02:54 |
realsportcars | what is this deb? | 02:54 |
vandenoever | basically this: http://www.vandenoever.info/software/warptree/images/rect.png | 02:55 |
vandenoever | with drag you move to different folders, if an image is selected, it shows up in background | 02:55 |
*** rcg has quit IRC | 02:58 | |
realsportcars | nice! | 03:01 |
*** apol_ has joined #harmattan | 03:01 | |
vandenoever | if you pass '/' as argument, you can see all your files | 03:02 |
realsportcars | Does an official "repo" exists? | 03:02 |
* vandenoever is very suprised by the performance | 03:02 | |
*** qgil has quit IRC | 03:05 | |
*** Termana has quit IRC | 03:11 | |
antman8969 | i've compiled libouath and liboauth-dev for harmattan (without curl) if anyone is interested | 03:11 |
antman8969 | http://umcs.maine.edu/~naddeoa/packages/ | 03:12 |
*** Termana has joined #harmattan | 03:15 | |
*** Termana is now known as Guest14180 | 03:15 | |
*** epage has quit IRC | 03:40 | |
*** realsportcars has quit IRC | 03:45 | |
*** NIN101 has quit IRC | 03:59 | |
*** Rizzer has joined #harmattan | 04:00 | |
antman8969 | wish qtcreator would actually use my rules file like it promises it will... | 04:09 |
*** M4rtinK2 has quit IRC | 04:19 | |
*** leinir has quit IRC | 04:34 | |
*** Reffyy has quit IRC | 04:39 | |
*** jcharpak has joined #harmattan | 05:09 | |
*** antman8969 has quit IRC | 05:18 | |
*** antman8969 has joined #harmattan | 05:20 | |
*** Guest14180 is now known as Termana | 05:25 | |
Termana | morning | 05:25 |
*** epage has joined #harmattan | 05:33 | |
*** apol_ has quit IRC | 05:42 | |
*** meegoexperts has quit IRC | 06:11 | |
*** Reffy has joined #harmattan | 06:34 | |
*** epage has quit IRC | 06:47 | |
*** jcharpak has left #harmattan | 06:51 | |
*** hardaker has joined #harmattan | 06:58 | |
*** Reffy has quit IRC | 07:03 | |
*** hardaker has quit IRC | 07:25 | |
*** DocScrutinizer has quit IRC | 07:34 | |
*** DocScrutinizer has joined #harmattan | 07:37 | |
rm_you | http://maemo.org/packages/package_instance/view/fremantle_extras-testing_free_armel/pyradio/3.7/ | 07:44 |
rm_you | please vote for this package... | 07:44 |
rm_you | >_> | 07:44 |
rm_you | been sitting there for weeks now | 07:44 |
rm_you | needs to move to extras... | 07:44 |
* Stskeeps downvotes :o | 07:44 | |
rm_you | >_< | 07:44 |
*** achipa has quit IRC | 07:44 | |
* rm_you kries at Stskeeps | 07:44 | |
rm_you | this 10 votes thing is so painful now | 07:46 |
rm_you | with so few people paying attention to extras-testing... | 07:46 |
DocScrutinizer | RM680-22-6_PR_RM680:~# lsof|grep bme | 07:59 |
DocScrutinizer | bme_RX-71 2929 bme cwd unknown /proc/2929/cwd (readlink: Permission denied) | 08:00 |
DocScrutinizer | MEH!!! | 08:00 |
GAN900 | Ooh, I like it when an icon concept can be salvages. | 08:00 |
* DocScrutinizer looks for good sledgehammer selling websites | 08:01 | |
GAN900 | http://thousandsparrows.com/meego/GeoSnap-earth.png | 08:01 |
GAN900 | fiferboy suggested dropping the aperture down to f/22 or so. | 08:01 |
DocScrutinizer | RM680-22-6_PR_RM680:~# ps |grep bme_RX | 08:04 |
DocScrutinizer | 2929 bme /usr/sbin/bme_RX-71 -u -l syslog -v 5 -c /usr/lib/hwi/hw/rx71.so | 08:04 |
DocScrutinizer | RM680-22-6_PR_RM680:~# strace -p 2929 | 08:04 |
DocScrutinizer | attach: ptrace(PTRACE_ATTACH, ...): Operation not permitted | 08:04 |
rm_you | lol | 08:04 |
rm_you | develsh / devel-su combinations not help? | 08:05 |
rm_you | GAN900: let me know tomorrow what the plan is on that | 08:05 |
rm_you | GAN900: since i was going to help, and all :P | 08:05 |
rm_you | bedtime now tho | 08:05 |
* rm_you sleeps | 08:05 | |
GAN900 | rm_you, night. | 08:05 |
DocScrutinizer | RM680-22-6_PR_RM680:~# i2cdetect -y 2 | 08:05 |
DocScrutinizer | Error: Can't use SMBus Quick Write command on this bus (ISA bus?) | 08:05 |
DocScrutinizer | RM680-22-6_PR_RM680:~# i2cget -y 2 0x55 0x02 w | 08:06 |
DocScrutinizer | Error: Read failed | 08:06 |
DocScrutinizer | a HUGE sledgehammer | 08:06 |
DocScrutinizer | rm_you: nope | 08:07 |
DocScrutinizer | at least I'm out of ideas what to combine in which sequence, to further escalate permissions | 08:08 |
DocScrutinizer | I can run all binaries, but obviously not access all domains, neither reading nor writing | 08:09 |
DocScrutinizer | dead end | 08:10 |
DocScrutinizer | I'm tempted to switch off the device and stow away in attik. Fetch it when there's a way to get ***root*** rights on the crap | 08:12 |
Stskeeps | or send it back in developer programme circulation | 08:12 |
Stskeeps | so others can have use of it | 08:12 |
rm_you | DocScrutinizer: just crack aegis | 08:13 |
rm_you | done. | 08:13 |
rm_you | you win. | 08:13 |
rm_you | if the right people put their minds to it, i imagine it shouldn't be TOO hard | 08:13 |
rm_you | i mean, pretty much every scheme involving any sort of master auth key has been cracked AFAIK | 08:14 |
rm_you | anywho, seriously off to bed now | 08:14 |
npm | there are some issues beyond the documentation that may be affecting some aegis usage | 08:23 |
*** divan_ has joined #harmattan | 08:28 | |
*** divan_ has quit IRC | 08:31 | |
*** aleksander_m has joined #harmattan | 08:55 | |
*** smoku1 has joined #harmattan | 09:19 | |
*** achipa has joined #harmattan | 09:19 | |
*** achipa has joined #harmattan | 09:19 | |
*** antman8969 has quit IRC | 09:28 | |
*** Wirta has joined #harmattan | 09:29 | |
*** CaCO3 has joined #harmattan | 09:48 | |
*** spenap has joined #harmattan | 09:52 | |
*** harbaum has joined #harmattan | 09:52 | |
npm | http://nielsmayer.com/meego/qml/qtzibit_0_0_1_armel.deb .... enjoy... | 10:24 |
*** rcg has joined #harmattan | 10:31 | |
*** CaCO3 has quit IRC | 10:43 | |
*** smoku1 has left #harmattan | 10:43 | |
*** leinir has joined #harmattan | 10:47 | |
*** slaine has joined #harmattan | 11:11 | |
*** veli has quit IRC | 11:20 | |
*** seif has quit IRC | 11:22 | |
MohammadAG | DocScrutinizer, develsh -c /bin/ash? | 11:30 |
*** veli has joined #harmattan | 11:31 | |
*** andre__ has joined #harmattan | 11:32 | |
*** andre__ has joined #harmattan | 11:32 | |
*** aleksander_m has quit IRC | 11:32 | |
*** Rizzer has quit IRC | 11:34 | |
*** aleksander_m has joined #harmattan | 11:35 | |
*** seif has joined #harmattan | 11:45 | |
seif | hey guys | 11:46 |
seif | is there twitter support coming anytime soon to the n950 | 11:47 |
hiemanshu | yes, as soon as a new image is launched | 11:47 |
hiemanshu | atleast thats what I heard | 11:48 |
seif | hiemanshu, leaves us with the question when i the next one coming out | 11:48 |
hiemanshu | that one no one knows :P | 11:49 |
dcarr_home | s/knows/can disclose/g | 11:50 |
hiemanshu | its the same to us :P | 11:51 |
vandenoever | if someone feels like testing a nice puzzle game for n950: http://oeba.nl/cubetest_0_9_5_armel.deb | 11:51 |
hiemanshu | but there was a hint at one in the next two weeks | 11:51 |
harbaum | vandenoever: Please start a wiki page or similar with links like this | 11:52 |
vandenoever | harbaum: on http://wiki.meego.com/ ? | 11:52 |
harbaum | vandenoever: Also i trie dot implement the "up one dir by left swipe". This is not possible due to qml limitations :-( | 11:53 |
harbaum | Yes, why not? I'll add my stuff as well | 11:53 |
vandenoever | harbaum: that's a shame about the swipe problem, do events not bubble up, is that the limitation? | 11:54 |
harbaum | I can make events be forwarded, but if i forward the pressed events, then the release events aren't delivered anymore | 11:54 |
harbaum | Something about "mouse focus" i've been told. They want me to implement a special mouse area in c++ for this. But that doesn't mke much sense if i want to write a universal file selector | 11:55 |
vandenoever | harbaum: yeah, that's not pretty | 11:56 |
harbaum | It's nice, they said "you can do everything in qml since you can extend it with c++" | 11:56 |
harbaum | Like in "you can go as fast as you want with your car since you can always go by plane" | 11:56 |
vandenoever | harbaum: there's always the webview ... | 11:57 |
harbaum | Yeah, offloading also saves memory, cpu, battery etc .... | 11:57 |
*** M4rtinK2 has joined #harmattan | 12:03 | |
*** jussi has quit IRC | 12:05 | |
*** jussi has joined #harmattan | 12:06 | |
harbaum | Khertan: http://www.youtube.com/watch?v=es504w-ZKqw | 12:10 |
thp | harbaum: nice file selector :) did you do that? | 12:23 |
harbaum | yepp | 12:25 |
harbaum | do you want it? | 12:26 |
harbaum | it's partly derived of the "SelecitonDialog" and the File Browser of the qmlviewer | 12:28 |
vandenoever | harbaum: does it use model/view? | 12:30 |
harbaum | sure | 12:31 |
*** smoku has joined #harmattan | 12:31 | |
harbaum | FileListModel and ListView | 12:31 |
harbaum | s/File/Folder/ | 12:32 |
infobot | harbaum meant: FolderListModel and ListView | 12:32 |
thp | harbaum: you should publish it as standalone lib / code drop somewhere | 12:34 |
thp | don't need it right now, but might come in handy in the future :) | 12:34 |
harbaum | is there a place for qml code snippets like this? | 12:34 |
*** smoku has left #harmattan | 12:35 | |
*** khertan1 has joined #harmattan | 12:36 | |
khertan1 | Morning | 12:36 |
thp | hmm i'm not sure. just create a small git/hg/bzr repo and push it on github or something | 12:37 |
harbaum | nah, noone would find that | 12:37 |
khertan1 | someone have port osc to harmattan ? | 12:37 |
khertan1 | someone have packaged pyside-assistant for harmattan ? | 12:38 |
Stskeeps | i wonder if aegis restricts chroot | 12:38 |
khertan1 | someone say it does | 12:39 |
khertan1 | but didn't remember who : | 12:39 |
khertan1 | :) | 12:39 |
khertan1 | qole i think | 12:39 |
vandenoever | harbaum: wiki for snippets perhaps | 12:41 |
*** djszapi has joined #harmattan | 12:45 | |
vandenoever | has anyone tried 3d qml on n950 yet? | 12:45 |
harbaum | http://wiki.meego.com/QML/FileSelector | 12:49 |
khertan1 | maybe i should post mine | 12:50 |
harbaum | Sure | 12:50 |
harbaum | Perhaps this will lead to one | 12:50 |
khertan1 | ouch ... your qml is complex | 12:50 |
khertan1 | :) | 12:50 |
khertan1 | did you keep the last opened directory ? | 12:51 |
khertan1 | :) | 12:51 |
harbaum | The animations .... | 12:51 |
khertan1 | grrr gitorious doesn't seems to be up | 12:51 |
djszapi | gitorious is like that quite often. | 12:51 |
khertan1 | a bit too much currently | 12:52 |
khertan1 | and github is closed source ... i prefer promoting and using opensource solution | 12:52 |
djszapi | there are gazillion other ways | 12:53 |
khertan1 | djszapi: indeed ... :) | 12:53 |
vandenoever | khertan1: +1 on foss git repo | 12:55 |
vandenoever | i added a link to fileselector i wrote, it remembers the subdirs that you visited | 12:56 |
vandenoever | so if you go up, you are at the prev position in the list | 12:56 |
vandenoever | eh, i mean down/into a subdir you were previously | 12:57 |
harbaum | mine does that as well. nothing to remember explicitely, jut select the entry in the upper dir that has the same path as the folder you jst left | 12:57 |
*** djszapi has left #harmattan | 12:59 | |
khertan1 | vandenoever: i mean when you instantiate a new file selector it go by default at the last location you were previously, not just, going to parent folder | 13:01 |
vandenoever | the one i wrote does not remember positions between instantiations | 13:01 |
khertan1 | vandenoever: http://www.youtube.com/watch?v=ysLYqoLqcyw | 13:02 |
khertan1 | :) | 13:02 |
khertan1 | this video didn't show the new things implemented | 13:02 |
khertan1 | was made some days ago | 13:02 |
khertan1 | last version is available in gitorious/khteditor branch qml | 13:02 |
khertan1 | :) | 13:02 |
* vandenoever is watching http://www.youtube.com/v/OXcxFZbKUNI, wondering if there's harmattan package | 13:03 | |
khertan1 | noone build pygame for harmattan ? | 13:06 |
*** NIN101 has joined #harmattan | 13:09 | |
MohammadAG | khertan1, is it me, or did you (at some point), sync navigation in KhtEditor with the music in the background? :P | 13:10 |
khertan1 | MohammadAG: lol | 13:11 |
khertan1 | MohammadAG: music was made by my nephew | 13:12 |
khertan1 | and no, there is no link ... i was recording the video with headset | 13:13 |
khertan1 | so didn't notice the sound arround before posting it :) | 13:13 |
*** trx has quit IRC | 13:13 | |
*** trx has joined #harmattan | 13:14 | |
*** trx has quit IRC | 13:14 | |
*** trx has joined #harmattan | 13:14 | |
khertan1 | MohammadAG: maybe i should add some sound to the fileselector | 13:16 |
khertan1 | :) | 13:16 |
Tronic | How are scratchbox bind mounts supposed to be setup? The installer creates them but the login script doesn't, so I need to reinstall scratchbox after each reboot. | 13:30 |
Tronic | Perhaps there is some other script for setting up the mounts? | 13:31 |
Tronic | Additionally, and possibly related, is it possible to put scratchbox somewhere else then /scratchbox? For now I've accomplished this via a symlink and I wonder if this would be causing some trouble. | 13:32 |
*** povbot has joined #harmattan | 14:37 | |
divan | hmm, it's possible to add hotkey on layout switch for hardware keyboard: setxkbmap -model nokiarm680 -layout us,ru -variant ,cyrillic -option "grp:ctrl_shift_toggle" | 14:48 |
*** jussi01_ is now known as jussi | 14:57 | |
Termana | WOOOO | 14:59 |
Termana | WOOO | 14:59 |
Termana | WOOOO | 14:59 |
Stskeeps | got n950? | 14:59 |
Stskeeps | :P | 14:59 |
Termana | No, but they have finally sent it :p | 14:59 |
SpeedEvil | Congrats. | 14:59 |
Termana | Still waiting on a tracking number | 14:59 |
Termana | SpeedEvil, thanks :) | 14:59 |
*** djszapi has joined #harmattan | 15:00 | |
*** djszapi has left #harmattan | 15:00 | |
*** djszapi has joined #harmattan | 15:02 | |
RST38h | meanwhile: Koreans produce $3m glow-in-the-dark dog | 15:03 |
*** skrrr has quit IRC | 15:03 | |
divan | Does anyone familiar with libaccounts/signon implementation on Harmattan? I see only one good implmentation - facebook stuff, but it's closed source(why, btw?). Seems like you'll need to do a lot of guesswork in order to implement own accounts plugin. :( | 15:08 |
djszapi | I always forget the name of the browser on N950, can someone remind me, please ? :p | 15:09 |
*** Stskeeps has quit IRC | 15:10 | |
lcuk | morning djszapi \o | 15:10 |
djszapi | ahh grob | 15:10 |
*** zarlino has joined #harmattan | 15:11 | |
*** Stskeeps has joined #harmattan | 15:12 | |
djszapi | heyhey lcuk :) | 15:12 |
khertan1 | zarlino: every property are signal on qml | 15:20 |
khertan1 | zarlino: example text property is onTextChanged | 15:20 |
khertan1 | font property is onFontChanged | 15:21 |
khertan1 | :) | 15:21 |
leinir | divan: yeah, i've got that same problem... been pondering on doing some status.net stuff, but without docs it's... not exactly straight forward | 15:21 |
*** leinir has quit IRC | 15:22 | |
zarlino | khertan1: oh didn't know that | 15:22 |
zarlino | so why some components do have signals in the docs? | 15:22 |
khertan1 | because no property related to the signal :) | 15:22 |
zarlino | i see | 15:23 |
khertan1 | and text isn't the most feature complete component :) | 15:23 |
zarlino | but still, I see in the QtCreator code completion many "on..." signals | 15:23 |
divan | leinir, I wrote the mail to some developers in Nokia, asked whether we can get the sources of facebook plugin at least to use it as startup point and write tutorials on how to develop social plugins for Harmattan. At least, we've signed NDA already. | 15:24 |
zarlino | many of which could not be found the way you say | 15:24 |
khertan1 | zarlino: http://doc.qt.nokia.com/4.7-snapshot/qdeclarativeintroduction.html#property-change-notifications | 15:24 |
khertan1 | zarlino: because heritance :) | 15:24 |
djszapi | divan: leinir is out. | 15:25 |
*** leinir has joined #harmattan | 15:25 | |
frals | divan: doesnt accounts gitorious have examples? | 15:25 |
khertan1 | zarlino: like Component.onCompleted ? | 15:25 |
zarlino | thank you khertan1 you shed some light | 15:25 |
khertan1 | your welcome ... reading the documentation help a lot :) | 15:26 |
frals | divan: for NDA stuff i wouldnt discuss anything here ;) | 15:26 |
khertan1 | i ve started learning qml by example ... and documentation explain many dark things :) | 15:26 |
djszapi | frals: :) | 15:26 |
zarlino | khertan1: I must have missed the docs explaining this naming convention... | 15:27 |
khertan1 | héhé ... the doc explain things so simply than when you know a bit some part of the language jumping some things that look like simple result in missing important information lost in well know things | 15:28 |
khertan1 | : | 15:28 |
khertan1 | :) | 15:28 |
zarlino | a problem with this approach you don't actually have info on the signal, for example when it will fire | 15:28 |
divan | frals, yes, it has a couple of very primitive examples, mainly used as tests. So, it gives nothing for understanding the frameworks. | 15:28 |
zarlino | khertal1: I found by trial and error the "onTextChanged" fires when you close the virtual keyboard | 15:29 |
zarlino | khertal1: but I initially thought it would fire on each keypress | 15:29 |
djszapi | Elleo: any progress with a shared repository ? :p | 15:30 |
djszapi | btw, What HTML5 video formats does Grob support? Is there some way to play Ogg Theora? | 15:30 |
zarlino | khertal1: so I found the docs somewhat lacking | 15:30 |
djszapi | I was about to package libtheora.. | 15:30 |
zarlino | ops i wrote khertal1 but I meant khertan1 | 15:32 |
djszapi | :) | 15:32 |
khertan1 | zarlino: :) | 15:33 |
khertan1 | zarlino: the doc is far to be complete | 15:33 |
* djszapi is making a nice splitted vim package :) | 15:34 | |
divan | djszapi, with latest vim version? | 15:34 |
djszapi | define latest. | 15:35 |
djszapi | I am not sure it has the bleeding edge patches. | 15:35 |
djszapi | 7.3.154 | 15:35 |
*** CaCO3 has quit IRC | 15:36 | |
*** Wirta has quit IRC | 15:36 | |
divan | djszapi, latest stable. 7.3 is ok :) Thanks. | 15:37 |
khertan1 | a working preview of khteditor qml edition is also available on my home project repository for harmattan | 15:38 |
khertan1 | :) | 15:38 |
djszapi | I am wondering whether others also have issues with test cases in case of qemu while packaging something containing tests ? | 15:39 |
*** hardaker has joined #harmattan | 15:40 | |
djszapi | lcuk: do you have libqbase packages ? :) | 15:43 |
lcuk | libliqbase you mean | 15:43 |
lcuk | :) I only have the fremantle packages | 15:43 |
lcuk | and there are some xvideo integration and bugs to sort out | 15:44 |
lcuk | sgx is frustrating | 15:44 |
djszapi | ahh, I see =) | 15:44 |
lcuk | fixing them first on meego n900-ce before seeing if the patch will apply on harmattan | 15:44 |
lcuk | bug 13084 in meego | 15:44 |
MeeGoBot | Bug https://bugs.meego.com/show_bug.cgi?id=13084 maj, Medium, ---, carsten, ASSI, [n900] Horizontal tearing with xvimagesink | 15:44 |
Stskeeps | lcuk: since you're working on it, could you assign it to yourself? | 15:45 |
lcuk | since harmattan is capacitive and closed source think better to fix on meego open version | 15:45 |
lcuk | Stskeeps, i don't have a clue how to fix it | 15:45 |
lcuk | and keep hoping somebody who knows will take it | 15:45 |
lcuk | but sure, assign it to me since you aren't able to advance it | 15:46 |
lcuk | since it effects playing default video in n900-ce | 15:47 |
*** Rizzer has joined #harmattan | 15:47 | |
*** djszapi has left #harmattan | 15:47 | |
*** djszapi has joined #harmattan | 15:53 | |
*** fiferboy has joined #harmattan | 16:04 | |
*** epage has quit IRC | 16:05 | |
*** Wirta has joined #harmattan | 16:05 | |
fiferboy | Anyone interested in testing some apps could add my repo: http://repo.pub.meego.com/home:/fiferboy/Harmattan/ | 16:14 |
fiferboy | Lexicon is a word look-up and definition app and is pretty well finished | 16:14 |
fiferboy | GeoSnap tacks a GPS track for use in geotagging photos (or whatever else you want) | 16:14 |
fiferboy | It needs some UI changes to export projects easily, but is pretty fully function | 16:15 |
*** baraujo has joined #harmattan | 16:15 | |
djszapi | vgrade: interesting, I am getting again rpc timeout even for the smallest packages. | 16:16 |
*** aleksander_m has quit IRC | 16:26 | |
*** aleksander_m has joined #harmattan | 16:27 | |
*** meegoexperts has joined #harmattan | 16:31 | |
*** aleksander_m has quit IRC | 16:32 | |
*** Arkenoi has joined #harmattan | 16:33 | |
*** zarlino has quit IRC | 16:34 | |
Arkenoi | will e7 keyboard fit to n950? they look similar but i doubt about buttons size | 16:34 |
divan | oh, nokia engineers responded on my request regarding closed facebook plugins. They told it uses Nokia only authentication and it's proprietary. But proposed to share (under NDA) some code snippets for other similar plugins. | 16:35 |
*** aleksander_m has joined #harmattan | 16:35 | |
divan | BTW, I guess we need some frontend app for packrat :) | 16:36 |
*** w00t has joined #harmattan | 16:36 | |
Termana | That seems like a pretty lame excuse. Could we just packet sniff it to see what interaction they are having with facebook? | 16:36 |
Termana | Couldn't we* | 16:36 |
divan | Termana, I'm less interested in what namely they're doing with Facebook, and more for the well explained structure of sharing/chatting/news plugins to integrate it easily into Harmattan. | 16:38 |
Termana | yeah, I was just commenting on what they said | 16:38 |
divan | In that context, I would be pretty happy with Picase plugin sources for example. | 16:38 |
divan | frals, I do not share! | 16:38 |
divan | frals, I just describe the politeness of Nokia engineers and their desire to help :) | 16:39 |
Arkenoi | is usb host mode feasible on n950? | 16:42 |
harbaum | at least the connector says "no" | 16:43 |
*** Rizzer has quit IRC | 16:44 | |
Arkenoi | maybe there is some ugly hack possible like on n900 :-) | 16:44 |
harbaum | maybe, but that instantly decreases the number of possible users | 16:45 |
harbaum | with 250 n950 devices out there ther userbase for such a hack is probably 2-3 | 16:45 |
vandenoever | +future N9 users | 16:45 |
fiferboy | harbaum: That is 250 community device program users, there are lots distributed through different channels | 16:46 |
Arkenoi | are there really just 250 of them? | 16:46 |
harbaum | still not the best setup for niche hardware solutions | 16:46 |
achipa | the real question is how different the actual hardware is in that regard between N9 and N950 | 16:47 |
fiferboy | harbaum: Yeah, I'm not saying that necessarily increased the userbase significantly :) | 16:47 |
achipa | might be the same, or might just as well not... | 16:47 |
*** zarlino has joined #harmattan | 16:47 | |
SpeedEvil | The connector doesn't really say no. | 16:48 |
SpeedEvil | At the moment, aegis is annoying | 16:48 |
SpeedEvil | As you need to - likely replace the kernel. | 16:48 |
SpeedEvil | Or - at least - get 'real' root to screw with it. | 16:48 |
Arkenoi | what is the name of xmpp telepathy plugin in harmattan repo? | 16:50 |
harbaum | SpeedEvil: The connector is a usb b recepticle which is target mode only. OTG would be a AB-recepticle | 16:51 |
SpeedEvil | It's quite irrelevant. | 16:51 |
SpeedEvil | You can use the same scheme that the n900 or n810 does | 16:51 |
SpeedEvil | a 'reverse' cable. | 16:51 |
harbaum | No, that's what the physical connector sais about the capabilities of the device. A hacked kernel may know better ... | 16:52 |
harbaum | the n810 has a real ab connector | 16:52 |
SpeedEvil | Do I mean 800? | 16:52 |
* SpeedEvil forgets. | 16:52 | |
divan | Arkenoi, telepathy-gabble | 16:53 |
harbaum | the n800 imho had a mini usb, not a micro usb | 16:53 |
Arkenoi | divan, thanx | 16:53 |
divan | Arkenoi, actually it's not a plugin, it's a provider. The conversations plugin for pure Jabber seems not available yet. | 16:54 |
*** Wirta has quit IRC | 16:54 | |
SpeedEvil | Key aspects are - does the battery charger support inherently boost mode, or does it require external components. Similar question about the USB phy | 16:55 |
*** zarlino has quit IRC | 16:56 | |
*** zarlino has joined #harmattan | 16:56 | |
*** rm_work has joined #harmattan | 17:08 | |
*** rm_work has quit IRC | 17:08 | |
*** rm_work has joined #harmattan | 17:08 | |
*** crevetor has joined #harmattan | 17:11 | |
rm_work | vandenoever / fiferboy: answering both of your questions from like 20 hours ago: ~packrat | 17:13 |
*** achipa has quit IRC | 17:13 | |
rm_work | ~packrat | 17:13 |
infobot | well, packrat is a cute little rodent that keeps everything it finds. http://www.ageofikon.com/prh/ | 17:13 |
rm_work | fiferboy: has a de-facto section list in one of the dropdowns | 17:13 |
rm_work | vandenoever: basically a catalog of harmattan apps | 17:14 |
fiferboy | rm_work: Ah, thanks! | 17:14 |
rm_work | :P | 17:14 |
*** harbaum has quit IRC | 17:14 | |
rm_work | no promises on whether those are technically valid | 17:14 |
*** razvanpetru has joined #harmattan | 17:14 | |
rm_work | but each of those are currently used by at least one package | 17:14 |
fiferboy | rm_work: Both my packages are discoverble through there :) | 17:14 |
rm_work | :P | 17:15 |
rm_work | just got through backlogs from YESTERDAY T_T | 17:16 |
rm_work | since i was away from my computer basically until i left work | 17:16 |
rm_work | also, pyRadio *actually* works now | 17:17 |
rm_work | i was lieing before when i said it worked apparently | 17:17 |
rm_work | but 3.6.2 actually works | 17:17 |
rm_work | i need to update to 3.9 upstream tonight though T_T | 17:17 |
razvanpetru | where do you guys save writeable data for your apps? | 17:18 |
rm_work | i think ~/.appname/ is appropriate | 17:19 |
razvanpetru | is that equal to /home/user/MyDocs/.app ? | 17:21 |
rm_work | err | 17:22 |
rm_work | ^^ that one | 17:22 |
rm_work | forgot about MyDocs | 17:22 |
razvanpetru | I think I tried something like that, but QDir::mkpath failed to create a path with . in it | 17:22 |
* lcuk puts mind back to bug 13084 | 17:22 | |
MeeGoBot | Bug https://bugs.meego.com/show_bug.cgi?id=13084 maj, Medium, ---, liquid, ASSI, [n900] Horizontal tearing with xvimagesink | 17:22 |
lcuk | crap e-chan | 17:22 |
razvanpetru | and... without . all my cached images pop up in the gallery app :D | 17:22 |
rm_work | yeah... | 17:23 |
rm_work | there might be a way to block them from being picked up by tracker in the tracker config | 17:23 |
rm_work | but | 17:23 |
rm_work | :/ | 17:23 |
rm_work | it shouldn't fail to create . folders | 17:23 |
rm_work | actually | 17:25 |
rm_work | you might just try /home/user/.app | 17:25 |
rm_work | it's a 2Gb EXT4 partition | 17:25 |
razvanpetru | well my db is < 1MB so no problem there | 17:25 |
rm_work | yeah | 17:25 |
razvanpetru | but my app didn't have write access to anything except MyDocs | 17:25 |
rm_work | it is 3% used on my system <_< | 17:25 |
rm_work | hrm | 17:25 |
rm_work | should have access to /home/user? | 17:26 |
rm_work | not sure why it wouldn't | 17:26 |
razvanpetru | developer or user acct? | 17:27 |
razvanpetru | :) | 17:27 |
rm_work | yeah tons of .app folders there | 17:27 |
rm_work | AH erm | 17:27 |
rm_work | you'll want to use something like ~/ | 17:27 |
rm_work | since when you run it qtcreator it'll be developer | 17:27 |
razvanpetru | ~/.app :) | 17:27 |
rm_work | yes | 17:27 |
razvanpetru | I'll try | 17:27 |
spenap | razvanpetru, if you're using QML, check http://www.developer.nokia.com/Community/Wiki/How-to_create_a_persistent_settings_database_in_Qt_Quick_(QML) and http://doc.qt.nokia.com/4.7/qdeclarativeglobalobject.html | 17:28 |
razvanpetru | thanks ;) | 17:28 |
rm_work | dunno if it will resolve ~/ | 17:28 |
rm_work | you might have to figure it out | 17:28 |
razvanpetru | spenap: using QML, but the backend is all C++ | 17:28 |
spenap | I see: it could be an interesting read for you, anyway | 17:29 |
spenap | :) | 17:29 |
razvanpetru | hehe, I could just call QDeclarativeEngine::offlineStoragePath() and use that :P | 17:30 |
rm_work | err | 17:30 |
rm_work | spenap: page appears to be empty? | 17:30 |
rm_work | >_> | 17:30 |
razvanpetru | although the Path object is much lower level than qdeclengine... | 17:31 |
spenap | both pages, rm_work ? | 17:31 |
spenap | ah, check the end | 17:31 |
razvanpetru | works for me in opera | 17:31 |
spenap | this (QML) | 17:31 |
spenap | looks like xchat breaks the URL | 17:31 |
spenap | you'll have to copy the URL by hand | 17:31 |
Arkenoi | damn, where do i configure xmpp account on the harmattan? telepathy-gabble is installed but it is not on the list to "add account" | 17:32 |
rm_work | ah | 17:34 |
rm_work | got it | 17:34 |
*** qgil has joined #harmattan | 17:36 | |
razvanpetru | what's the developer/user pass? :) | 17:44 |
divan | Arkenoi, I told you before, telepathy-gabble is just a provider for XMPP, it's not a harmattan Accounts implementation. I didn't see the implementation yet, and (guess)working on it. | 17:45 |
divan | razvanpetru, developer pass is generated via SDK Connecrivity Tool | 17:45 |
razvanpetru | yeah, figured... | 17:45 |
razvanpetru | but user? | 17:45 |
divan | razvanpetru, I wonder too | 17:45 |
divan | razvanpetru, in /etc/passwd the password field for user is emoty | 17:46 |
divan | s/emoty/empty/ | 17:46 |
infobot | divan meant: razvanpetru, in /etc/passwd the password field for user is empty | 17:46 |
razvanpetru | entering twilight zone | 17:47 |
divan | but I'm not sure it's safe to change user password.. | 17:47 |
razvanpetru | so I did as rm_work suggested, used ~/.app as path... app works brilliantly | 17:47 |
razvanpetru | now I log in as developer and there is no ~/.app folder... | 17:47 |
rm_work | lol | 17:47 |
razvanpetru | ls -a shows no .app inside ~/ | 17:47 |
rm_work | hrm | 17:48 |
rm_work | divan: yeah, wouldn't recommend changing password for user | 17:48 |
rm_work | it doesn't have a password, and isn't log-in-able | 17:48 |
rm_work | you can su to it tho | 17:48 |
divan | rm_work, any idea how to het user password? | 17:48 |
divan | rm_work, no, I can't | 17:48 |
rm_work | you don't want to | 17:48 |
rm_work | and yes you can :P | 17:48 |
rm_work | i do it all the time | 17:48 |
divan | rm_work, do you use exec relaxed mode? | 17:49 |
rm_work | ? | 17:49 |
divan | rm_work, for aegis | 17:49 |
rm_work | I ssh in as root and use "su user" | 17:49 |
*** achipa has joined #harmattan | 17:49 | |
*** achipa has quit IRC | 17:49 | |
*** achipa has joined #harmattan | 17:49 | |
rm_work | i don't even know what that is (relaxed mode, not aegis) | 17:49 |
divan | ahh.. that way.. | 17:49 |
divan | rm_work, I ssh as developer and then get root via devel-su | 17:50 |
divan | rm_work, http://wiki.meego.com/Migrating_from_N900_to_N950#Security_framework | 17:50 |
*** zarlino has quit IRC | 17:50 | |
divan | rm_work, yep ssh via root@ allows to do su - user | 17:50 |
qgil | hi, I'm trying to 'connect' my Qt Creator project with gitorious. Is there somewhere a step-by-step guide or any docs? | 17:51 |
rm_work | yeah never considered logging in as developer | 17:52 |
*** mgedmin has joined #harmattan | 17:52 | |
rm_work | sorry qgil, don't use git personally :( | 17:52 |
rm_work | qgil: you having fun making your app? :) | 17:53 |
divan | qgil, I used github, but seems that gitourious has much better integration with qtcreator. Should be intuitive. | 17:53 |
vandenoever | qgil: do you have a local git already? | 17:53 |
divan | rm_work, I guess it's chess program :) | 17:53 |
razvanpetru | :D | 17:53 |
rm_work | yeah i saw the screenshots on G+ :) | 17:53 |
razvanpetru | just convince chessbase to port fritz and you're set | 17:53 |
vandenoever | if so you need to "git remote add $gitoriousurl" | 17:53 |
qgil | I'm about to create the local git vandenoever but I wanted to check first if there was something I had to consider since the goal is to have that cvode in gitorious - I have never used git before :) | 17:54 |
qgil | razvanpetru: we should also convince them to do it open source? | 17:54 |
rm_work | oh vandenoever you're here now | 17:55 |
rm_work | you asked about a catalog of harmattan apps yesterday | 17:55 |
razvanpetru | qgil: I would be happy with closed source also | 17:55 |
rm_work | did you get my note this morning? | 17:55 |
vandenoever | qgil: in your code dir do "git init;git add -A;git commit -a -m 'Initial commit'" | 17:55 |
vandenoever | then connect to gitorious with above command | 17:55 |
razvanpetru | considering they didn't port to any platform, it could be a differentiator | 17:55 |
qgil | razvanpetru: considering they didn't port to any platform, I would be wasting my time with them ;) | 17:55 |
vandenoever | rm_work: not sure, what about? | 17:56 |
qgil | thanks vandenoever will try now | 17:56 |
rm_work | you asked about a catalog of harmattan apps yesterday | 17:56 |
vandenoever | rm_work: ah, the pack rat | 17:56 |
vandenoever | it's opened in a tab, not looked to well, yet | 17:56 |
*** zarlino has joined #harmattan | 17:56 | |
qgil | vandenoever: stupid question: the default directory offered for the git repo is the same dir as my app project - is that fine or should I better put it else where? | 17:57 |
rm_work | if the repo is listed on the wiki | 17:57 |
rm_work | it will be indexed there | 17:57 |
rm_work | or I can add some manually | 17:57 |
vandenoever | rm_work: nice, so what's easiest way to install (deb -i ?) | 17:58 |
vandenoever | qgil: is this on your local file system? i'm not sure i understand the question | 17:58 |
qgil | vandenoever: send the package attached in an email and open the attachment from the email client :) | 17:58 |
vandenoever | qgil: :-) | 17:58 |
rm_work | vandenoever: just click the install button in browser | 17:58 |
vandenoever | qgil: i dont have mail configured on the n950, i'm still testing the water for the extent of things i'll use it for | 17:59 |
qgil | yes, my local git repo. Qt Creator offers to create in the same directory as my Qt Quick project, but I'm not sure if this is ok or will mess my project | 17:59 |
vandenoever | qgil: i'll use mass storage | 17:59 |
vandenoever | qgil: i have let qtcreator create the repo too and i think it just does the three commands i said (init,add,commit) | 17:59 |
qgil | vandenoever: open browser, find the deb listed somewhere, click the deb link, proceed | 18:00 |
vandenoever | but without the connection to gitorious | 18:00 |
vandenoever | qgil: ok, and in the right order when there are dependencies, no doubt | 18:00 |
razvanpetru | rm_work, I think mkpath created the path as /home/developer/~/.app | 18:00 |
vandenoever | or add things to sources.list, i guess | 18:00 |
qgil | I *believe* proper packages can define repos to be installed, but I'm not sure vandenoever | 18:01 |
*** zarlino has quit IRC | 18:01 | |
razvanpetru | this is what QDir expands the absolute path to... Qt, I am disappoint | 18:01 |
vandenoever | is "Beginning Nokia Apps Development: Qt and HTML5 for Symbian and MeeGo" good reading? | 18:01 |
vandenoever | since it's from 1.5 years ago, i guess it's outdated ... | 18:02 |
razvanpetru | I bet it uses QWidget :P | 18:02 |
vandenoever | it does talk about meego | 18:02 |
* mgedmin suspects books like these are outdated before they get published | 18:02 | |
vandenoever | yeah, i'm looking for a book of 15 euro to get free shipping of a usb cable :-) | 18:03 |
razvanpetru | try advanced qt | 18:03 |
razvanpetru | by mark summerfield | 18:03 |
vandenoever | well, on the Qt front, i know tons already | 18:04 |
DocScrutinizer | MohammadAG: develsh allows strace and lsof, but then again /bin/ash: i2cdetect: Operation not permitted | 18:04 |
razvanpetru | how the hell can QDir create a path like /home/developer/~/.app/filecache :-/ | 18:05 |
razvanpetru | vandenoever - then try something about WWII | 18:05 |
vandenoever | razvanpetru: :-) | 18:05 |
qgil | vandenoever: http://library.developer.nokia.com/topic/MeeGo_1.2_Harmattan_API/html/guide/html/Developer_Library_Welcome_page_a381.html | 18:06 |
vandenoever | qgil: thank, i bookmarked it | 18:06 |
*** zarlino has joined #harmattan | 18:07 | |
rm_work | razvanpetru: makes sense actually | 18:14 |
rm_work | vandenoever: lol, from amazon? | 18:14 |
razvanpetru | rm_work: this did the trick - QDir(QDir::homePath()).filePath(".app"); | 18:17 |
mgedmin | razvanpetru, have you considered using XDG paths, i.e. ~/.cache, ~/.local/share, ~/.config? | 18:19 |
mgedmin | I wonder if Qt has abstractions for them | 18:20 |
mgedmin | (since the default locations can be overriden with environment variables) | 18:20 |
vandenoever | rm_work: yes, i settled for a book about modelling ontologies in rdfs | 18:20 |
*** gri has joined #harmattan | 18:20 | |
razvanpetru | mgedmin - no, my other idea was to user tmp for the cache :) | 18:22 |
rm_work | vandenoever: lol, get amazon prime | 18:23 |
rm_work | vandenoever: soooo worth it... free 2-day shipping on EVERYTHING... | 18:23 |
*** CaCO3 has joined #harmattan | 18:23 | |
razvanpetru | ...in US | 18:23 |
rm_work | ah :( | 18:23 |
rm_work | sad | 18:23 |
rm_work | if you have access to an email with a .edu extension, it's free :P | 18:23 |
*** zarlino has quit IRC | 18:24 | |
rm_work | i love things that check TLD for student verification | 18:24 |
razvanpetru | I don't think amazon will ship to me from Germany for free :) | 18:24 |
razvanpetru | now I pay 10 EUR/book | 18:24 |
razvanpetru | Safari Book Online is good... | 18:25 |
razvanpetru | mgedmin: AFAIK Qt doesn't know about those, I think they may be OS-specific | 18:25 |
mgedmin | being OS specific is what makes them good abstraction targets, IMHO | 18:26 |
mgedmin | "where should my app keep its cache files on this OS?" | 18:26 |
*** djszapi has left #harmattan | 18:27 | |
razvanpetru | QDesktopServices::StandardLocation has something | 18:27 |
razvanpetru | QDesktopServices::DataLocation | 18:27 |
w00t | QDesktopServices::storageLocation is indeed what you're probably after2 | 18:27 |
razvanpetru | yes, mgedmin was right :), there's even a QDesktopServices::CacheLocation | 18:28 |
mgedmin | w00t! | 18:28 |
razvanpetru | :D | 18:29 |
*** zarlino has joined #harmattan | 18:30 | |
rm_work | :P | 18:31 |
npm | try my package, and give me a clue on replacing COPY_DIR with something that doesn't copy all my .svn dirs http://nielsmayer.com/meego/qml/qtzibit_0_0_1_armel.deb | 18:35 |
*** Arkenoi has quit IRC | 18:35 | |
npm | makes the package larger than expected by duplicating a lot of files | 18:35 |
razvanpetru | that's why I hate svn/cvs | 18:36 |
npm | blame me... i helped hire the guy that wrote it :-) | 18:36 |
npm | he rescued my cat as part of the job interview :-) | 18:36 |
rm_work | lol | 18:37 |
* rm_work still uses svn... | 18:37 | |
npm | he was sick of it meowling... i guess that's why businesses normally use offices | 18:37 |
razvanpetru | and I using SVN myself for work... | 18:38 |
razvanpetru | *I am using... | 18:38 |
npm | yeah a lot of people do. i think it needs better suport in qtcreator | 18:38 |
mgedmin | svn 1.veryrecentversion has only one .svn subdir at the top of the tree, like everyone else | 18:39 |
razvanpetru | really? :) | 18:39 |
*** hardaker has quit IRC | 18:39 | |
razvanpetru | I have <quite recent version> myself... | 18:39 |
mgedmin | speaking of those, how do I tell dh_make to skip the .hg when it's creating an .orig.tar.gz? | 18:39 |
mgedmin | my vim .orig.tar.gz is over 50 megs because I ran dh_make in a hg checkout | 18:39 |
mgedmin | upstream doesn't package point versions, and I did not want to untar 7.3.0 and apply 260 patches by hand | 18:40 |
rm_work | mgedmin: i packaged vim a few days ago... | 18:40 |
mgedmin | rm_work, where? | 18:40 |
npm | what's the syntax for replacing COPY_DIR found in a Makefile inside one's project.pro file | 18:40 |
rm_work | it's in packrat | 18:41 |
npm | just COPY_DIR=... | 18:41 |
rm_work | ~packrat | 18:41 |
infobot | i heard packrat is a cute little rodent that keeps everything it finds. http://www.ageofikon.com/prh/ | 18:41 |
npm | or something more | 18:41 |
mgedmin | I searched in c-obs, couldn't find anything, gave it a try | 18:41 |
mgedmin | and djszapi(sp?) also packaged vim late last night | 18:41 |
rm_work | this is why packrat should be in the topic | 18:41 |
razvanpetru | ~qt | 18:41 |
infobot | methinks qt is TrollTech's QT library, available at http://www.trolltech.com | 18:41 |
rm_work | i did it like... last friday | 18:41 |
rm_work | nah, sunday | 18:41 |
* mgedmin didn't know about packrat | 18:41 | |
mgedmin | oh, it's version 7.2 -- I've got 7.3.260 :) | 18:42 |
rm_work | :P | 18:42 |
rm_work | then trump it :P | 18:42 |
mgedmin | at http://mg.pov.lt/770/dists/harmattan/user/binary-armel/ | 18:42 |
rm_work | fine by me | 18:42 |
rm_work | i should get your repo in packrat | 18:42 |
rm_work | especially if you're going to have anything else in there | 18:42 |
rm_work | is that the final path for it? | 18:42 |
mgedmin | I don't plan to remove anything, no | 18:43 |
rm_work | k | 18:44 |
mgedmin | but I hope we'll get something like Maemo Extras, and all these little repos are going to become mostly obsolete | 18:44 |
rm_work | aww yours doesn't have an icon :( | 18:44 |
mgedmin | I've some fbreader debs in /experimental/binary-armel | 18:44 |
mgedmin | yeah, no icon, blame me | 18:44 |
rm_work | it shows up in packrat now BTW | 18:44 |
razvanpetru | can Harmattan's web browser still install debs like in Maemo? | 18:45 |
leinir | razvanpetru: it can :) | 18:46 |
razvanpetru | leinir: thanks, that makes updating rather easy I think... | 18:46 |
mgedmin | fbreader consists of maybe 7 debs, I'd rather not click on them one by one in the right order | 18:46 |
razvanpetru | although there's probably a better solution... | 18:46 |
leinir | Or, well, it can download them, and then you can launch them from the download list (once it's completed, tap on the item and it'll ask if you're sure and all that jazz) | 18:46 |
rm_work | unfortunately it doesn't add repos via .install like maemo5 did, but it will install debs straight (no dep handling :() | 18:46 |
lcuk | lardman|gone, ping when you get back | 18:48 |
razvanpetru | fails for me over web: downloads, says installing and then says installation package not found | 18:52 |
rm_work | erm | 18:53 |
rm_work | something must be wrong with the package | 18:53 |
rm_work | i've had it work on 10+ debs | 18:53 |
razvanpetru | normally yes, but I just installed it with qtc :( | 18:53 |
rm_work | if they have unresolved deps... | 18:53 |
*** zarlino has quit IRC | 18:54 | |
*** spenap has quit IRC | 19:01 | |
*** qgil has quit IRC | 19:22 | |
*** cpscotti has joined #harmattan | 19:36 | |
*** razvanpetru has left #harmattan | 19:39 | |
*** CaCO3 has quit IRC | 19:41 | |
*** aleksander_m has quit IRC | 19:43 | |
*** deimos has joined #harmattan | 19:47 | |
*** meegoexperts has quit IRC | 20:00 | |
*** hardaker has joined #harmattan | 20:02 | |
*** djszapi_ has joined #harmattan | 20:06 | |
*** cpscotti has quit IRC | 20:09 | |
djszapi_ | Is there any package wishlist somewhere ? | 20:11 |
mgedmin | I've a private one in my mind | 20:11 |
mgedmin | e.g. pwsafe would be nice | 20:11 |
djszapi_ | are you fine with 0.2.0 or do you need newer version ? | 20:14 |
*** qgil has joined #harmattan | 20:15 | |
mgedmin | I'm using 0.2.0 on my laptop | 20:17 |
djszapi_ | obs seems to be down this time | 20:17 |
mgedmin | and while it'd be nice to have a GUI, porting PasswordSafe from maemo5 to harmattan won't be trivial, I expect | 20:17 |
mgedmin | uh oh, I hope pwsafe's password integration will work with the web browser | 20:18 |
mgedmin | since meego-terminal has no clipboard support | 20:18 |
DocScrutinizer | mgedmin: don't even think about it. AIUI passwords are agis' domain, and you'll probably have a hard time to plumb anything (else) to any authentication event | 20:21 |
djszapi_ | nobody started an IRC client yet ? | 20:23 |
dm8tbr | ssh+screen+irssi for me | 20:23 |
djszapi_ | I do prefer the Qt/KDE solution for it on this gadget. | 20:24 |
djszapi_ | without any server dependency. | 20:24 |
DocScrutinizer | dm8tbr: yeah, the benefits of off-board processing. A very special "web app" | 20:24 |
DocScrutinizer | >>Nokia Developer Newsletter: Ovi Store tops 7 million daily downloads<< Number of maemo downloads per day == number of maemo downloads per month ;-P | 20:28 |
DocScrutinizer | >>Microsoft Mango<< \o/ | 20:29 |
*** djszapi_ has quit IRC | 20:31 | |
*** djszapi_ has joined #harmattan | 20:31 | |
*** crevetor has quit IRC | 20:32 | |
djszapi_ | btw why wasn't the "old" fremantle server used for harmattan purposes in case of c-obs? It has been constructed pretty well. The current Harmattan PC is nowhere close to that. | 20:33 |
DocScrutinizer | >>Champion of the Month: Randy Arnold<< Congrats, texrat! | 20:34 |
djszapi_ | DocScrutinizer: what application do you develop for your N950 ? | 20:36 |
DocScrutinizer | errr, USB hostmode | 20:36 |
DocScrutinizer | though this seems to become more and more an absolute won't_fly | 20:37 |
DocScrutinizer | fist it seemed N9(50), due to more sane hw, had better chances to get USB hostmode than N900, but MSSF is killing the fun in it | 20:38 |
djszapi_ | if it compiles in scratchbox, just ship it xD | 20:38 |
DocScrutinizer | HAHA | 20:38 |
*** antman8969 has joined #harmattan | 20:40 | |
*** slaine has quit IRC | 20:40 | |
DocScrutinizer | djszapi_: this morning I tried to explore the hw a bit, using i2ctools. Alas once more it turns out without *real* root permissions (I.E. with aegis intercepting all things the MSSF team didn't think developers need access to) this doesn't pan out | 20:41 |
antman8969 | docscrutinizer, out of curiosity, what are you trying to accomplish? | 20:42 |
DocScrutinizer | antman8969: ??? USB hostmode | 20:42 |
antman8969 | just logged in | 20:42 |
antman8969 | lol just saw you clndt do something | 20:42 |
DocScrutinizer | aah | 20:42 |
djszapi_ | DocScrutinizer: are you politely offensing us ? :) | 20:42 |
djszapi_ | what is the exact problem ? | 20:42 |
DocScrutinizer | djszapi_: not really, just whining | 20:43 |
DocScrutinizer | ;-D | 20:43 |
antman8969 | I guess a better question would be: you seem to be doing a lot things that require true root access, do you have an ultimate goal in mind? | 20:43 |
djszapi_ | I hope the valgrind will work fine on the device. That is a bit critical question because it needs to have all the credentials for the applications it is used for. | 20:43 |
DocScrutinizer | tbh I somewhat accepted this won't work on harmattan, ever | 20:44 |
djszapi_ | "true root" access hahahaha | 20:44 |
antman8969 | lol an unfortunate distinction on this platform.... | 20:44 |
DocScrutinizer | ~tell antman8969 about h-e-n | 20:44 |
djszapi_ | antman8969: you should really forget thinking of root/superadmin/whatever | 20:45 |
djszapi_ | it is not a unix workstation, it is a user device | 20:45 |
DocScrutinizer | I'm just stating I need *all* permissions on this system, to do anything about USB hostmode | 20:45 |
antman8969 | lol thats why I was asking what it doc needed it for | 20:46 |
antman8969 | but I got his tell | 20:46 |
djszapi_ | DocScrutinizer: well, the "tcb" token is something I am not aware of that much. | 20:46 |
djszapi_ | upstart has it for example, but I think you can get it. | 20:46 |
djszapi_ | * cannot | 20:46 |
DocScrutinizer | djszapi_: again, my device has a FAT PRINTING "DEVELOPER DEVICE" - it's not "user device" | 20:47 |
djszapi_ | Nokia did not care about 250-500 device users. | 20:47 |
djszapi_ | Also, you can flash your own kernel ;-) | 20:48 |
DocScrutinizer | if it was a user device, then we probably all could start development on PC, and happily wait until N9 rollout to test if our apps work on the "user device" | 20:48 |
*** smoku1 has joined #harmattan | 20:48 | |
DocScrutinizer | djszapi_: flashing my own kernel largely defeats the purpose to bring USB hostmode to "user device" - paradox, isn't it ;-D | 20:50 |
DocScrutinizer | I need a user device that'S actually a developer device, to create some low level app that's meant to run on user devices | 20:50 |
DocScrutinizer | developing it for "own kernel" is kinda silly | 20:51 |
DocScrutinizer | trying to develop it on a device that has the final tight user device restrictions is not going to fly | 20:51 |
DocScrutinizer | I'm doomed | 20:52 |
djszapi_ | not really since you could turn off just the security from the kernel that way. Also, I think being able to turn off the security is not much any different. You will develop an application which can behave completely differently without security. Hence I am not getting your point :) | 20:52 |
DocScrutinizer | djszapi_: the point is: Nokia doesn't decument ANY of the low level stuff I need to mess with, so I need to investigate by myself. This isn't feasible on a security restricted system. | 20:53 |
djszapi_ | DocScrutinizer: SDK team did not do a great job :) | 20:54 |
djszapi_ | at leats about the aegis documentation. | 20:54 |
djszapi_ | If you miss something from the SDK, just let me know and I can forward and boost it to the SDK team. | 20:56 |
DocScrutinizer | yeah, you'd think there HAS to be a way to disable aegis, by setting a flag and rebooting. And this should result in a working harmattan that just refuses to run specially protected programs, like angry-birds, or maybe even phone stack | 20:56 |
*** lardman has joined #harmattan | 20:56 | |
infobot | lardman: please check http://mg.pov.lt/harmattan-irclog/%23harmattan.2011-07-22.log.html#t2011-07-22T18:42:19, cheers DocScr | 20:56 |
lardman | evening | 20:56 |
lardman | anyone know how I get infobot to stop telling me about a message? | 20:56 |
lcuk | lol | 20:56 |
lcuk | lardman, by reading the logs :P | 20:57 |
djszapi_ | DocScrutinizer: This device is not designed without security. Hence turning off and writing application how it is not designed to be used is not really a good environment. | 20:57 |
* lardman now has internet at home! Rejoice! ;) | 20:57 | |
DocScrutinizer | lardman: I'm taking care | 20:57 |
lardman | thanks Doc | 20:57 |
djszapi_ | I know it might be worse for you, but will be way cooool for end users. | 20:57 |
lardman | onto other business, the odd packaging info that Qt Creator adds for Harmattan builds, how does one use that with dpkg-buildpackage? | 20:58 |
lardman | s/odd/odd location | 20:58 |
DocScrutinizer | djszapi_: I'm talking about investigating stuff that's definitely not available from Nokia as docs, and probably never will be. I'm *not* talking about developing app in an aegis-free environment. For illustration see my comment about i2ctools above | 20:59 |
SpeedEvil | lardman: Congrats! | 21:00 |
djszapi_ | ok but then gain, le me know what you do not find in the doc, and I can help with that by boosting it at the SDK team. :) | 21:00 |
djszapi_ | * again * let, etc :) | 21:00 |
lardman | SpeedEvil: not sure my wife is pleased mind you | 21:00 |
SpeedEvil | lardman: It can be restful to have it not working. For moments at a time. | 21:01 |
lardman | restful yeah, not so good when you've just moved and need to lookup phone numbers, change addresses, etc | 21:01 |
lardman | glad to have it back anyway | 21:01 |
lardman | bbiab suppertime | 21:01 |
SpeedEvil | :) | 21:01 |
DocScrutinizer | djszapi_: what is missing in the docs: schematics, explanation how to switch bq24153 to boost mode, how to mess with Mentorgraphics musb_hdrc core on a very low level, docs what's the real circuit around USB PHY... | 21:01 |
*** crevetor has joined #harmattan | 21:02 | |
djszapi_ | DocScrutinizer: Could you please send it to me in email ? | 21:02 |
DocScrutinizer | djszapi_: that's probably the best chance I got regarding this issue, since ~2 years. Sure I will | 21:03 |
DocScrutinizer | djszapi_: will take some time though, as I need to make sure my wording is elaborate enough to make clear what's the core problem | 21:04 |
*** Venemo has joined #harmattan | 21:04 | |
DocScrutinizer | djszapi_: and I'm pretty sure your SDK team has NFC about the things I ask for | 21:05 |
djszapi_ | DocScrutinizer: yeah, take your time :) | 21:05 |
DocScrutinizer | djszapi_: probably not even SDK team has access to the documents that hold the info I need. And if they have, these documents are under NDA and SDK team must not help, even if they like to do | 21:06 |
Venemo | hey guys | 21:06 |
djszapi_ | DocScrutinizer: What bugs is that, why c-obs is using so lame hardware :) | 21:07 |
djszapi_ | compared to that what maemo.org used. | 21:07 |
djszapi_ | * bugs me | 21:07 |
djszapi_ | and why they couldn't just re-use that server.. | 21:07 |
DocScrutinizer | djszapi_: it's like asking the manufacturer of your car about the engine controller firmware, as you want to patch it to give higher motor power | 21:08 |
Venemo | how're you guys? | 21:09 |
djszapi_ | DocScrutinizer: well the SDK team leader was a Qt/KDE/Debian guy with well ... lots of skill. | 21:09 |
DocScrutinizer | djszapi_: enabling USB hostmode really means messing with the hw with your arms in the guts to the ellbows | 21:09 |
DocScrutinizer | djszapi_: just to give you an idea: on N900 we had to >> stop bme << to make hostmode work | 21:10 |
SpeedEvil | As another fun example - important docs were found on about 5 versions of the CPU datasheet. | 21:11 |
DocScrutinizer | mention this to your SDK team and they'l leave the room, running and screaming | 21:11 |
SpeedEvil | Different bits of different versions. | 21:11 |
Venemo | hmm, Qt Creator just screwed me. | 21:11 |
Stskeeps | was it good? | 21:12 |
DocScrutinizer | djszapi_: do you still think it's worth sending you a mail about it? | 21:13 |
djszapi_ | DocScrutinizer: I fail to see why not.. | 21:13 |
DocScrutinizer | ok then, I'll do it | 21:13 |
DocScrutinizer | but please take a video of the SDK tema's reaction when you ask them for schematics ;-P | 21:13 |
Venemo | installed the newest Qt Creator nightly, and it says "No tool chain can produce code for this Qt version" about the Harmattan toolchain | 21:14 |
djszapi_ | I am a geek, I do not meet people, just mailing list :p | 21:14 |
Venemo | achipa, maybe you know something about this? | 21:14 |
DocScrutinizer | gnhnhnh | 21:14 |
Venemo | :D:D:D | 21:15 |
Venemo | djszapi_ :D | 21:16 |
tomma | Venemo, it tells you to add toolchain | 21:17 |
Venemo | tomma, all right, but how do I add it? | 21:17 |
Venemo | it's installed by the Qt SDK, I guess it should be added already | 21:17 |
tomma | but did you tell creator where it is? | 21:18 |
tomma | toolchains -> add | 21:19 |
Venemo | nope, the old Qt Creator finds it automatically, but the new one doesn't | 21:19 |
tomma | it should be added automaticly but could be it is broken now | 21:19 |
tomma | ok it adds it after you add qt version | 21:20 |
Venemo | tomma, hm... which option should I select in the Add button? In the older Qt Creator it appears as "Maemo GCC", but there is no such option in the Add button | 21:20 |
tomma | just press apply | 21:20 |
Venemo | hmm | 21:21 |
Venemo | still not there | 21:21 |
RST38h | Ok, Qt SDK does not compile anything either | 21:23 |
Venemo | tomma, Tools/Options/Qt4 has the Harmattan API in the list | 21:23 |
RST38h | Anyone knows how I am supposed to develop for this thing??? | 21:23 |
RST38h | When neither Scratchbox nor QtSDK work? | 21:23 |
Venemo | tomma, but the Tools/Option/Toolchains does not have the Harmattan toolchain and does not allow me to add it | 21:23 |
djszapi_ | RST38h: what is wrong about sb ? | 21:24 |
RST38h | djszapi: fails. | 21:25 |
RST38h | QtSDK sows a dialog saying "Unable to create debian templates: dh_make failed (Unnknown error) | 21:25 |
RST38h | When trying to run an empty project it says: :-1: error: Packaging Error: Command 'c:/qtsdk/madde/bin/mad dpkg-buildpackage -nc -uc -us' failed.Exit code: 1 | 21:25 |
djszapi_ | what fails in sb ? | 21:26 |
lardman | re | 21:27 |
Venemo | RST38h, when I ge thome, you could send me your stuff and I could try compiling it in my Qt SDK | 21:27 |
mgedmin | ok, time to go debug some segfaults | 21:27 |
mgedmin | I hope I can figure out how to build an fbreader deb without stripping debug symbols | 21:28 |
RST38h | Venemo: there is no "my stuff" there: I just created an empty QtQuick project and tried running it. | 21:28 |
RST38h | And it fails. | 21:28 |
RST38h | Epic fail for Qt folks, as far as I am concerned | 21:28 |
RST38h | Managed to turn relatively tolerable SB into...mghm...you know | 21:29 |
Venemo | RST38h, worksforme...... | 21:29 |
lardman | Anyone tried building a project created in Qt Creator in SB? What is the magic command line switch for dpkg-buildpackage to let it know where the "debian" dir is located? | 21:29 |
RST38h | Yes yes, workforyou | 21:29 |
antman8969 | on a usual install of sb, you shouldn't need to specify anything on command line other than -rfakeroot to build | 21:30 |
fiferboy | lardman: ln -s qtc_packaging/harmattan debian ;) | 21:30 |
RST38h | SB does not even see my host file system | 21:30 |
RST38h | SB2 did, but there is no Harmattan rootstrap for SB2 =(( | 21:30 |
djszapi_ | it is not true | 21:31 |
djszapi_ | you can mount it in bind mode andthere you go. | 21:31 |
lardman | fiferboy: is that how Qt Creator does it? | 21:31 |
Venemo | tomma, after removing the MeeGo 1.2 Harmattan target and re-adding it, it seems to work | 21:31 |
Venemo | tomma, thanks for your help :) | 21:31 |
fiferboy | lardman: QtCreator actually seems to copy it to debian rather than link | 21:31 |
Venemo | RST38h, hmm... | 21:31 |
* mgedmin tries DEB_BUILD_OPTIONS=nostrip dpkg-buildpackage -uc -us -rfakeroot -i | 21:32 | |
lardman | fiferboy: interesting, thanks | 21:32 |
Venemo | hm. | 21:32 |
lardman | fiferboy: I'd expected some switch I didn't know about rather than wholesale copying... | 21:32 |
*** djszapi_ has left #harmattan | 21:32 | |
deimos | RST38h, why not compiling for desktop for apps that doesn't need harmattan apis ? The steps are here http://meegoharmattandev.blogspot.com/2011/07/setting-up-qt-sdk-for-meego-harmattan.html | 21:32 |
Venemo | can I use USB networking on the N950 at the same time I'm connecting my laptop to the internet through its modem? | 21:32 |
deimos | RST38h, and Qt Mobility could also be compiled and installed on desktop | 21:33 |
RST38h | Ok, now it has hung in the Package manager | 21:33 |
fiferboy | lardman: Yeah, I always have to delete the debian directory QtCreator leaves lying around | 21:33 |
mgedmin | Venemo, you could do that on the N900; I haven't tried with a N950 yet | 21:33 |
RST38h | EPIC FAIL | 21:33 |
Venemo | mgedmin, ok, I'll try | 21:33 |
mgedmin | fiferboy, lardman, Qt Creator uses a separate build directory by default, I think | 21:34 |
mgedmin | I've ~/src/qmlcompass with crevetor's project, and a ~/src/QMLCompass-build-harmattan with debian/, a Makefile and a bunch of .o files in it | 21:34 |
Venemo | mgedmin, hm, seems that it can't do this | 21:35 |
mgedmin | debian/rules has these nice hardcoded full filesystem paths augh | 21:35 |
lardman | yeah true | 21:35 |
*** harbaum has joined #harmattan | 21:35 | |
Venemo | mgedmin, at least I can't find a way to enable usb network from the GUI without plugging out the cable and reconnecting it in "SDK mode" | 21:35 |
mgedmin | Venemo, Internet tethering on the N900 did not work via USB networking, but rather the N900 pretended to be a USB 3G modem when connected in "PC Suite mode" | 21:36 |
Venemo | mgedmin, yes, N950 does the same on "Sync and Connect" mode | 21:37 |
Venemo | mgedmin, but when in "Sync and Connect" mode, I can't seem to enable the usb network as well | 21:37 |
mgedmin | oh, I misunderstood your question! | 21:38 |
Venemo | mgedmin, I want to develop for the N950 while at the same time I'm tethering it | 21:38 |
*** inean has joined #harmattan | 21:39 | |
inean | Hi everyone | 21:39 |
Venemo | hi | 21:40 |
inean | I need to know UserAgent string for N950 and N9 ... Is there someone here with a N950 device ? :P | 21:41 |
mgedmin | lots | 21:41 |
Venemo | inean, I have an N950, but I can't tell you right now. but if noone does by the time I get home, ping me and I will. | 21:42 |
inean | Venemo: Thanks | 21:42 |
mgedmin | inean, n950 is "Mozilla/5.0 (MeeGo; NokiaN950-00/00) AppleWebKit/534.13 (KHTML, like Gecko) NokiaBrowser/8.5.0 Mobile Safari/534.13" | 21:42 |
Venemo | heh | 21:43 |
Venemo | that's lots | 21:43 |
mgedmin | modern user agent strings are made of insanity | 21:43 |
inean | Ok, I suspect than for N9 will be s/N950/N9 ... | 21:43 |
mgedmin | RST38h, what I do for scratchbox is ln -s /scratchbox/users/mg/home/mg ~/scratchbox | 21:45 |
mgedmin | then I put all my stuff in ~/scratchbox and it's right there in my home after I sb-login | 21:45 |
*** Scifi has joined #harmattan | 21:46 | |
*** inean has quit IRC | 21:47 | |
RST38h | mgedmin: Does it work? | 21:47 |
mgedmin | yes | 21:47 |
RST38h | mgedmin: and what should I link? =) | 21:47 |
mgedmin | you know your username better than I do :) | 21:47 |
* RST38h was already going to bring up the nfs share (!) | 21:47 | |
*** inean has joined #harmattan | 21:47 | |
mgedmin | scratchbox can't see host system, but host system can see scratchbox | 21:47 |
mgedmin | that symlink makes it easier for me to find the scratchbox subtree | 21:47 |
* mgedmin compares the size of the new .debs and decides they have debug symbols after all | 21:49 | |
fiferboy | mgedmin: Yeah, I turned of shadow builds | 21:51 |
mgedmin | what are shadow builds? | 21:51 |
fiferboy | mgedmin: In QtCreator, telling it to build in a separate directory | 21:51 |
mgedmin | ah! | 21:51 |
mgedmin | if I knew how to make QtCreator build fbreader, maybe I could use the integrated debugger... | 21:52 |
* mgedmin drools, then abandons hope | 21:52 | |
* mgedmin gdb -ex run --args fbreader | 21:54 | |
mgedmin | yes! a better traceback! | 21:54 |
Venemo | fiferboy, you turned them off...? it doesn't let me to turn them on at all :P | 21:55 |
mgedmin | hey! | 21:56 |
mgedmin | I can execute only one gdb command, like bt | 21:57 |
mgedmin | but I can't go up a stack frame and look at locals or whatevs -- "The program no longer exists" | 21:57 |
*** Scifi has quit IRC | 22:00 | |
*** hardaker has quit IRC | 22:03 | |
fiferboy | Venemo: Yeah, I don't know what logic lets you turn them on/off | 22:03 |
fiferboy | Venemo: Did you create the project in the current QtCreator version? | 22:04 |
Venemo | fiferboy, nope | 22:04 |
Venemo | nah, I've just arrived, so I'll put the laptop to sleep now | 22:04 |
Venemo | will be back later :) | 22:04 |
*** gri has quit IRC | 22:05 | |
*** hardaker has joined #harmattan | 22:06 | |
*** Venemo has quit IRC | 22:12 | |
*** Scifig has joined #harmattan | 22:15 | |
*** wicket64 has joined #harmattan | 22:18 | |
mgedmin | why do I get a paint event when fbreader becomes invisible? | 22:19 |
mgedmin | the stack trace ends at QApplication::notify(QObject*, QEvent*) in /usr/lib/libQtGui.so.4 | 22:19 |
fiferboy | mgedmin: fbreader controls the paint events itself, right? | 22:24 |
fiferboy | Or does it use widgets to do that? | 22:25 |
mgedmin | I'm not sure | 22:25 |
mgedmin | I see uses of QWidget | 22:25 |
mgedmin | (gdb) p myHolder.myFrame->isVisible() | 22:25 |
mgedmin | Couldn't find method QWidget::isVisible | 22:25 |
mgedmin | but http://doc.qt.nokia.com/latest/qwidget.html#visible-prop | 22:25 |
mgedmin | is this a gdb thing or what? | 22:25 |
mgedmin | my current frame is in ZLQtViewWidget::Widget::paintEvent | 22:26 |
fiferboy | Does ZLQtViewWidget implements its own paint? | 22:26 |
mgedmin | something like that | 22:27 |
*** harbaum has quit IRC | 22:27 | |
fiferboy | Looks like it handles its own painting | 22:27 |
mgedmin | AFAICS it's painting using QPainter into a pixmap, and then blits that pixmap onto the screen, after optional rotation | 22:27 |
mgedmin | I suppose isVisible is an inline function | 22:32 |
fiferboy | mgedmin: isVisible is a access function to the visible property of a QWidget | 22:32 |
fiferboy | http://doc.qt.nokia.com/latest/qwidget.html#visible-prop | 22:32 |
fiferboy | Oh, I see you already got that :) | 22:33 |
npm | the "ready to be checked in version" :-) http://nielsmayer.com/meego/qml/qtzibit_0_0_2_armel.deb | 22:34 |
fiferboy | mgedmin: ZLQtViewWidget inherits ZLViewWidget | 22:35 |
fiferboy | I haven't found that in the git repo yet, do you see it? | 22:35 |
fiferboy | Ah, see it core/src/view | 22:35 |
mgedmin | ZLViewWidget is in zlibrary/core/src/view/ | 22:35 |
mgedmin | ctags rule | 22:35 |
mgedmin | what happens if I put if (!isVisible()) return; at the top of ZLQtViewWidget::Widget::paintEvent()? | 22:36 |
fiferboy | ZLApplication (inherited by ZLQtViewWidget) has a isVisible function... | 22:38 |
qgil | antman8969: ah you're here | 22:38 |
qgil | make[1]: *** No rule to make target `clean'. Stop. make: *** [clean] Error 2 dpkg-buildpackage: failure: debian/rules clean gave error exit status 2 The process "/home/qgil/QtSDK/Madde/bin/mad" exited with code 2. Error while building project testdef (target: Harmattan) When executing build step 'Custom Process Step' | 22:38 |
* lbt follows | 22:38 | |
lbt | just in case... | 22:38 |
* mgedmin wants rsync | 22:40 | |
*** smoku1 has left #harmattan | 22:40 | |
*** antman8969 has quit IRC | 22:40 | |
qgil | antman8969: thank you for all your patience - I think at this point it's better to go to the forum - I can't believe I'm the only one hitting these roadblocks between Qt Creator & Community OBS | 22:41 |
fiferboy | mgedmin: The ZLApplication isVisible function always returns true!? | 22:41 |
mgedmin | heh | 22:41 |
mgedmin | QML has this platformWindow.visible property | 22:41 |
*** antman8969 has joined #harmattan | 22:41 | |
mgedmin | can anybody translate QML to QWidget? | 22:41 |
*** antman8969 has quit IRC | 22:41 | |
* lbt tries to follow qgil to the forum but hits the lack of an email bridge and watches from afar... | 22:42 | |
qgil | lbt: how to setup Qt Creator to provide what the OBS wants: a tar.gz and a .dsc file | 22:43 |
fiferboy | qgil: Hav eyou looked at Jaffa's mud2 script? | 22:43 |
fiferboy | s/Hav eyou/Have you/ | 22:43 |
infobot | fiferboy meant: qgil: Have you looked at Jaffa's mud2 script? | 22:43 |
qgil | fiferboy: nope, URL? | 22:43 |
npm | qgil: pls try my demo of some infrastructure http://nielsmayer.com/meego/qml/qtzibit_0_0_2_armel.deb | 22:43 |
*** antman8969 has joined #harmattan | 22:43 | |
lbt | I don't have Qt Creator running qgil ... otherwise I'd have tried | 22:43 |
npm | will post forum page today | 22:44 |
qgil | lbt: just making sure that you can't blame the lack of email gateway in the forum ;) | 22:44 |
lbt | :P | 22:44 |
fiferboy | qgil: https://gitorious.org/attitude/attitude/blobs/master/mud2 | 22:44 |
fiferboy | qgil: cd to your directory and run mud2 and it will create the files needed for OBS | 22:44 |
lbt | qgil: I need to get some work done ... I will be around if you hit issues with OBS | 22:45 |
* fiferboy notes you should clean up your project directory first | 22:45 | |
lbt | this is a real Qt Creator problem though .... | 22:45 |
qgil | ok, I surrend: in 36h I have started with Qt Creator, Qt Quick, COBS and git... learning how to get an actual executable from a git repo is in my queue | 22:45 |
lbt | good progress - what docs? | 22:46 |
npm | qgil: why not start w/ QtSDK 1.1.2 | 22:46 |
lbt | I really would hate to see the last step fail... | 22:46 |
lbt | you should have a .dsc somewhere | 22:46 |
npm | oh COBS ... | 22:46 |
crevetor | qgil: you're new to developpement ? | 22:46 |
npm | i haven't got there yet | 22:46 |
fiferboy | qgil: click "Raw blob data" | 22:46 |
lbt | npm: if dpkg-buildpackage works then you're ready for C.OBS | 22:46 |
npm | need to bug lbt about getting an account :-) | 22:46 |
qgil | I have no problems with Qt SDK: it gets a package and injects it to my device. The extra hassle is to use the COBS instead of e.g. Ovi directly | 22:46 |
lbt | npm: what's your meego.com acct ? | 22:47 |
npm | NielsMayer | 22:47 |
lbt | done | 22:47 |
qgil | crevetor: the page you saw yesterday was my first non-WWW code ever | 22:47 |
npm | (i have a feeling my life just got more complicated) | 22:47 |
npm | thanks | 22:47 |
lbt | npm: ... and "hey" | 22:47 |
npm | lbt | 22:47 |
npm | hi | 22:47 |
mgedmin | from what I've seen (briefly), Qt SDK hackety-hacks the debian build system to the point where it can build a .deb on _your_ machine, but not to the point where it could produce a .dsc | 22:47 |
crevetor | qgil: qtcreator is good for creating packages without having to think but if you try to build packages from outside it's a pita. | 22:47 |
lbt | mgedmin: oh, seriously! | 22:47 |
RST38h | tail: cannot open `debian/changelog' for reading: No such file or directory | 22:48 |
RST38h | dpkg-buildpackage: failure: tail of debian/changelog gave error exit status 1 | 22:48 |
RST38h | Packaging Error: Command 'c:/qtsdk/madde/bin/mad dpkg-buildpackage -nc -uc -us' failed.Exit code: 1 | 22:48 |
RST38h | How do I fix THAT in QtSDK? | 22:48 |
antman8969 | hey now, I thought we got this resolved | 22:48 |
RST38h | not here | 22:48 |
crevetor | qgil: if you want to build a package with cobs from source generated by qtcreator : copy the directory where your project is, cd into it and do qmake | 22:48 |
npm | qgil: i found reading the resulting Makefile and also turning on debugging (commented out in default .pro) helpful | 22:48 |
lbt | RST38h: FDISK | 22:48 |
RST38h | lbt: Sorry? | 22:48 |
crevetor | qgil: then ln -s qtc_packaging/debian_harmattan debian | 22:48 |
lbt | RST38h: any problem starting c:/.... | 22:49 |
crevetor | then dpkg-buildpackage -nc -us -uc -S | 22:49 |
npm | crevetor: thanks | 22:49 |
crevetor | it should generate a tgz and a dsc in the parent directory that you can upload in cobs to build | 22:49 |
mgedmin | thing is, Qt Creator's debian/rules does no building, just a make install | 22:49 |
rm_work | yeah QtCreator builds debs on windows machines O_o iz crazy | 22:49 |
qgil | thank you very much for all your help but I have been following recipes since I can to the office at 7am (6h ago) - now I need lunch and get something done :) | 22:50 |
mgedmin | and it generates a Makefile full of absolute pathnames when you build | 22:50 |
*** antman8969 has quit IRC | 22:50 | |
lbt | qgil: crevetor's suggestion sounds *very* correct | 22:50 |
mgedmin | maybe if I disabled shadow builds it'd be a bit more sane? | 22:50 |
*** antman8969 has joined #harmattan | 22:50 | |
RST38h | lbt: Ok thanks | 22:50 |
qgil | antman8969: crevetor just shared a different method and now I'm not sure what/who to follow | 22:51 |
lbt | RST38h: sorry ... just having fun | 22:51 |
RST38h | Anyone has a better suggestion? | 22:51 |
qgil | and fiferboy shared a third one :) | 22:51 |
antman8969 | lol | 22:51 |
mgedmin | e.g. the build rule for qc_res.cpp is | 22:51 |
mgedmin | /opt/QtSDK/Madde/targets/harmattan-nokia-meego-api/bin/rcc -name res ../qmlcompass/res.qrc -o qrc_res.cpp | 22:51 |
antman8969 | well If it calms you, I'm usiing my method right now qgil | 22:51 |
mgedmin | good luck finding /opt/QtSDK/Madde on the OBS builder | 22:51 |
antman8969 | i'm sure theres overlap | 22:51 |
qgil | :D | 22:51 |
qgil | lunch time for me | 22:51 |
* RST38h googled but just found other people hitting the same problem | 22:52 | |
mgedmin | make install_desktopfile: | 22:52 |
RST38h | No answrs on how to fix it so far =( | 22:52 |
mgedmin | -$(INSTALL_FILE) /home/mg/src/qmlcompass/QMLCompass.desktop $(INSTALL_ROOT)/usr/share/applications/ | 22:52 |
mgedmin | yay /home/mg | 22:52 |
mgedmin | good luck finding _that_ on the OBS builder | 22:52 |
crevetor | mgedmin: that's why I used qmake in the original source dir (not the build dir) otherwise it generates crap | 22:52 |
antman8969 | ^^ that | 22:52 |
RST38h | Am I supposed to CREATE the changelog? If so, where? | 22:52 |
antman8969 | standard debian packaging! a debian driectory goes on the top level of your project | 22:53 |
crevetor | mgedmin: are you adding stuff to qmlcompass ? | 22:53 |
antman8969 | debian/changelog is the changelog | 22:53 |
mgedmin | crevetor, not any more | 22:53 |
RST38h | antman: when I create it, QtSDK still fails | 22:54 |
mgedmin | I jumped on it because it was a perfect way to try out Qt Creator | 22:54 |
npm | http://www.timolassy.com/ makes good programming music for a Finnish phone | 22:54 |
RST38h | Complains of foreign stuff in debian | 22:54 |
npm | :-) | 22:54 |
antman8969 | is this a newly created project? | 22:54 |
mgedmin | more fun than a hello world, anyway | 22:54 |
RST38h | Yes. | 22:54 |
antman8969 | and you have a qtc_packaging directory that it made for you too right | 22:54 |
RST38h | yes | 22:54 |
antman8969 | the new directory is paralelle to qtc_packaging, called debian, correct | 22:55 |
mgedmin | well, if (!isVisible()) return; did not help prevent segfaults | 22:55 |
mgedmin | the widget thinks it's visible | 22:55 |
RST38h | yes | 22:55 |
npm | i just had qmake puke on my changelog... it didn't like multiline changes so i just smashed em into big long lines and it worked | 22:55 |
RST38h | BTW it also says "tail: cannot open `debian/changelog' for reading: No such file or directory" | 22:55 |
antman8969 | did you delete it from the qtc_packaging by any chance? | 22:55 |
RST38h | Wait, no , says this: which: pgp: unknown command | 22:56 |
RST38h | no I did not | 22:56 |
antman8969 | what I'm getting at is, as long as you leave the QtCreator files alone and don't change your config, adding a debian dir shoudln't cause any problems | 22:56 |
antman8969 | as long as you don't add the dir to your .pro file | 22:56 |
antman8969 | at least it's not (and has never) for me | 22:56 |
RST38h | antman:thank you, but I have not touched any files | 22:56 |
RST38h | just created a new project and tried running it | 22:57 |
RST38h | can do it once again | 22:57 |
RST38h | done it twice now | 22:57 |
antman8969 | lol well i'll listen if you up for it, wasn't there the fisrt two times | 22:57 |
antman8969 | got a little bit before the gtm | 22:57 |
antman8969 | gym* | 22:57 |
* RST38h skipped onthe MingW installation (too large), maybe THAT was the reason? | 22:58 | |
npm | antman8969: i agree... there's a leftover 'debian' in http://ytd-meego.googlecode.com/svn/trunk/playground/qmltube/ and it doesn't break the build | 22:58 |
antman8969 | ooh mingw... i didn't know you were on windows | 22:58 |
antman8969 | ^^ | 22:58 |
mgedmin | oh, *barf* | 22:58 |
antman8969 | lol well | 22:58 |
mgedmin | fbreader's sources assume a tabsize of 2! | 22:59 |
antman8969 | in theory it shouldn't matter | 22:59 |
antman8969 | mmm | 22:59 |
RST38h | mgedmin: just globalreplace with two spaces | 22:59 |
antman8969 | changlelog is very picky about formatting | 22:59 |
RST38h | mgedmin: and commit it this way | 22:59 |
* mgedmin always uses dch to edit the changelog | 22:59 | |
RST38h | mgedmin: I am ok with this change, btw | 22:59 |
antman8969 | * uses emacs ;) | 23:00 |
mgedmin | RST38h, are you fbreader's upstream? | 23:00 |
npm | i used emacs' changelog formatting, but because i had text in autowrap it crapped out. (next time i won't autowrap) | 23:00 |
RST38h | mgedmin: no, but you are not committing upstream are you> | 23:00 |
RST38h | ? | 23:00 |
antman8969 | yea, lots of weird issues with that.. | 23:00 |
mgedmin | I've hopes of maybe submitting a patch someday if I can make it work on harmattan | 23:00 |
mgedmin | small steps | 23:00 |
RST38h | mgedmin: naah, unlikely | 23:01 |
mgedmin | gdb is so similar to pdb, but there are treacherous differences | 23:01 |
mgedmin | e.g. 'u' is not shorthand for 'up' | 23:02 |
* mgedmin hates c++ | 23:02 | |
RST38h | When creating a project says "Unable to create debian templates: dh_make failed (unknown error)" | 23:04 |
mgedmin | I love error messages that say "unknown error" | 23:04 |
* RST38h wonders how developers are supposed to use the tool giving such messages when creating an empty project | 23:05 | |
RST38h | And it too ~4GB of my disk space too | 23:06 |
RST38h | took | 23:06 |
Scifig | Is anyone working on qmafw plugins? | 23:06 |
* mgedmin abandons gdb and tries running fbreader under valgrind | 23:07 | |
fiferboy | RST38h: Did you do the maemo5 interface on fbreader? | 23:10 |
RST38h | fiferboy: Maintained it for a while | 23:12 |
RST38h | fiferboy: Fixed power drain, added sensible settings, added scroll-by-drag | 23:12 |
RST38h | Other things too | 23:12 |
fiferboy | RST38h: I thought so | 23:13 |
mgedmin | valgrinded fbreader crash: http://pastie.org/2281152 | 23:13 |
RST38h | mgedmin: well it is Qt not fbreader crashing | 23:14 |
mgedmin | does "Meego graphics system destroyed" mean all QPixmap()s become invalid? | 23:14 |
*** hardaker has quit IRC | 23:14 | |
fiferboy | mgedmin: Is the "MeeGo graphics system destroyed" happening when you swipe out? | 23:16 |
mgedmin | yes | 23:16 |
*** dcarr_home has quit IRC | 23:16 | |
mgedmin | so, QPixmap represents a server-side pixmap | 23:16 |
fiferboy | Hmm, and it is trying to paint after the graphics system is closed... | 23:17 |
*** dcarr_home has joined #harmattan | 23:18 | |
mgedmin | "Note: When using the native X11 graphics system, the pixmap becomes invalid when the QApplication instance is destroyed." | 23:19 |
mgedmin | I don't know if that's relevant | 23:19 |
mgedmin | zlibrary is not destorying the QApplication it creates | 23:20 |
mgedmin | in fact it's leaking it explicitly | 23:20 |
mgedmin | I wish harmattan had -dbgsym packages like Ubuntu | 23:21 |
*** Arkenoi has joined #harmattan | 23:23 | |
*** piggz has joined #harmattan | 23:25 | |
fiferboy | mgedmin: At least you know where to look now, but I'm afraid I won't be much help with that | 23:27 |
mgedmin | I'm way in over my head | 23:28 |
mgedmin | learning ARM assembly now | 23:28 |
mgedmin | http://pastie.org/2281247 | 23:30 |
crevetor | neat | 23:31 |
mgedmin | I'm assuming r0 holds this, r4 is d | 23:31 |
mgedmin | isActive() returns true | 23:32 |
mgedmin | [r4+0x58] is d->engine? | 23:32 |
mgedmin | d_func() is an inline function, since I see no call instructions? | 23:32 |
mgedmin | there was an illegal read at 0x3b51c0f0 | 23:33 |
mgedmin | so d->engine was deleted? | 23:33 |
Arkenoi | do you guys use n950 as the primary phone? if yes, how do you deal with IM functionality being missing? | 23:34 |
mgedmin | and then there was an illegal read at 0x3b51c0f4 because the value read in the previous step was pointing to lala-land | 23:34 |
crevetor | Arkenoi: I do but I don't use IM on my phone | 23:34 |
*** inean has quit IRC | 23:35 | |
fiferboy | mgedmin: Can you tell if it is a Qt arm problem or the way FBReader is using painter? | 23:37 |
mgedmin | hm | 23:38 |
crevetor | mgedmin: you seem to now gdb quite a bit | 23:38 |
mgedmin | do you have any other Qt desktop apps we could test? | 23:38 |
crevetor | mgedmin: how do you see content of memory at address x ? | 23:38 |
mgedmin | if they crash on swipe, that would point towards a Qt bug | 23:38 |
mgedmin | I don't know gdb all that well | 23:38 |
fiferboy | mgedmin: I am pretty sure someone said they had a Qt program that didn't crash on swipe | 23:39 |
mgedmin | p *(sometype *)0x123456? | 23:39 |
*** hardaker has joined #harmattan | 23:44 | |
piggz | how can i take screenshots on the n950? | 23:46 |
fiferboy | piggz: Built-in camera ant two mirrors? | 23:47 |
fiferboy | :) | 23:47 |
fiferboy | piggz: You can open the Boosted Widget Gallery and go to "Debug Tools" "Take a screenshot" | 23:48 |
piggz | fiferboy: cool, cheers | 23:49 |
*** antman8969 has quit IRC | 23:50 | |
*** fiferboy has quit IRC | 23:51 | |
*** crevetor has quit IRC | 23:51 | |
*** crevetor has joined #harmattan | 23:56 | |
mgedmin | also, I've heard that ctrl-shift-p works and puts the screenshot in /home/user/MyDocs/.images | 23:56 |
mgedmin | so, QMLCompass (why is it in /opt/QMLCompass/bin with no symlink in /usr/bin, crevetor???) also prints those "Meego graphics system destroyed" messages, but doesn't segfault | 23:58 |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!