| *** NIN102 has quit IRC | 00:11 | |
| *** ZogG_laptop has joined #harmattan | 00:21 | |
| *** Venemo has quit IRC | 00:25 | |
| *** risca has quit IRC | 00:41 | |
| *** xTs has quit IRC | 00:46 | |
| DocScrutinizer | GeneralAntilles: most likely a tuning of the TX LC will yield better results tan soldering random wire to the antenna | 00:51 |
|---|---|---|
| DocScrutinizer | GeneralAntilles: I'd check the RX side antenna (position, orientation, distance to metal objects etc) | 00:53 |
| *** Hamidx9 has joined #harmattan | 01:12 | |
| *** Hamidx9 has quit IRC | 01:12 | |
| *** Hamidx9 has joined #harmattan | 01:20 | |
| *** wirwe has quit IRC | 01:38 | |
| *** HAMIDx9_ has joined #harmattan | 01:40 | |
| GeneralAntilles | DocScrutinizer, already soldered the random wire. *g* It seems to have helped. | 01:40 |
| GeneralAntilles | Don't feel like disassembling the vacuum itself to hunt for the RX board. | 01:41 |
| *** Hamidx9 has quit IRC | 01:41 | |
| *** gri is now known as zz_gri | 01:47 | |
| HAMIDx9_ | i installed develsh (deb) manually but when i execute it i haven't root access | 01:50 |
| HAMIDx9_ | i installed it because of my internet problem and i cant enable dev-mode from setting | 01:51 |
| HAMIDx9_ | any suggestion ? | 01:51 |
| mgedmin | develsh doesn't give you root | 01:51 |
| mgedmin | devel-su gives you root | 01:51 |
| mgedmin | (after you've enabled developer mode, at least) | 01:51 |
| mgedmin | it *sucks* that one can't enable dev mode while offline | 01:52 |
| HAMIDx9_ | so it's impossible to enable dev-mode offline ? | 01:52 |
| mgedmin | I dunno; I assume so from what I hear people saying here | 01:53 |
| itsnotabigtruck | more or less impossible, yeah | 01:53 |
| itsnotabigtruck | since you only have to do it once, i doubt see why it's such a huge point of contention | 01:54 |
| itsnotabigtruck | a phone that's offline isn't exactly good for much anyway | 01:54 |
| itsnotabigtruck | also HAMIDx9_: you'll find yourself using this command a lot :p | 01:54 |
| itsnotabigtruck | devel-su -c develsh | 01:55 |
| *** niqt has quit IRC | 01:55 | |
| HAMIDx9_ | yeah , but my problem is about that nokia doesn't support our country | 01:55 |
| HAMIDx9_ | manually i cant install devel-su | 01:55 |
| itsnotabigtruck | HAMIDx9_: what do you mean? i'm pretty sure you can install packages from anywhere on the planet | 01:56 |
| DocScrutinizer | GeneralAntilles: I had good results with following procedure: open up RC-TX, increase distance to RX until fail, then some meters more, ude PLASTIC screwdriver to adjust the small variable capacitor in TX while sending, until RX reacts | 01:56 |
| itsnotabigtruck | if your home internet is dead, you could go find an internet hotspot or something | 01:56 |
| DocScrutinizer | s/ude/use/ | 01:56 |
| infobot | DocScrutinizer meant: GeneralAntilles: I had good results with following procedure: open up RC-TX, increase distance to RX until fail, then some meters more, use PLASTIC screwdriver to adjust the small variable capacitor in TX while sending, until RX reacts | 01:56 |
| HAMIDx9_ | but it seems its impossible and here we have problem | 01:57 |
| itsnotabigtruck | also HAMIDx9_: you should uninstall devel-sh also | 01:57 |
| HAMIDx9_ | i have to use VPN or Socks or HTTP proxy for enable dev-mode | 01:57 |
| itsnotabigtruck | if you install develsh directly from the .deb, it won't work | 01:57 |
| itsnotabigtruck | do develsh -c 'accli -I' | 01:58 |
| itsnotabigtruck | it should have about 20 entries | 01:58 |
| DocScrutinizer | GeneralAntilles: find the two positions of C where RX stops to react, adjust C to middle between those 2 pos | 01:58 |
| itsnotabigtruck | if it's more like 5 entries, develsh is misinstalled | 01:58 |
| *** leinir has quit IRC | 01:58 | |
| DocScrutinizer | GeneralAntilles: redo with some meters more distance | 01:58 |
| DocScrutinizer | GeneralAntilles: until RX never reacts, then go back to prev pos and setting | 01:59 |
| itsnotabigtruck | HAMIDx9_: are you saying your country blocked the nokia site? | 02:00 |
| DocScrutinizer | s/pos/distance/ | 02:00 |
| infobot | DocScrutinizer meant: GeneralAntilles: until RX never reacts, then go back to prev distance and setting | 02:00 |
| itsnotabigtruck | because i highly doubt nokia put regional restrictions on the repository (i'm not talking about the ovi store) | 02:00 |
| HAMIDx9_ | no it seems Nokia block my country | 02:00 |
| itsnotabigtruck | HAMIDx9_: apparently you already obtained a terminal? | 02:00 |
| itsnotabigtruck | what happens if you do apt-get update | 02:00 |
| itsnotabigtruck | also, you can configure apt to use an http proxy | 02:01 |
| * DocScrutinizer idly wonders where's the shell login of this homematic CCU (ARM based linux controller) | 02:01 | |
| HAMIDx9_ | i installed terminal manually | 02:01 |
| *** leinir has joined #harmattan | 02:01 | |
| itsnotabigtruck | you should do the following | 02:01 |
| itsnotabigtruck | dpkg -r develsh | 02:01 |
| itsnotabigtruck | apt-get install aegis-dev-mode developer-mode | 02:01 |
| HAMIDx9_ | thanks let me try | 02:01 |
| itsnotabigtruck | oh crap | 02:01 |
| itsnotabigtruck | you need to be root first, i forgot | 02:02 |
| mgedmin | heh | 02:02 |
| HAMIDx9_ | :) | 02:02 |
| mgedmin | pkgmgr lets you install stuff as non-root | 02:02 |
| DocScrutinizer | haha | 02:02 |
| mgedmin | dunno if it would let you install developer-mode | 02:02 |
| itsnotabigtruck | oh yeah, that's right | 02:02 |
| itsnotabigtruck | pkgmgr can invoke apt | 02:02 |
| *** rm_work has quit IRC | 02:02 | |
| HAMIDx9_ | how to use it ? | 02:02 |
| DocScrutinizer | actually I guess sudo & aegis make that work | 02:03 |
| itsnotabigtruck | HAMIDx9_: pkgmgr remove develsh | 02:03 |
| itsnotabigtruck | HTTP_PROXY='http://yourproxy:port' pkgmgr install developer-mode | 02:03 |
| HAMIDx9_ | thanks :) | 02:03 |
| itsnotabigtruck | give that a try | 02:03 |
| *** Natunen has quit IRC | 02:04 | |
| itsnotabigtruck | HAMIDx9_: did it work? | 02:05 |
| itsnotabigtruck | if it doesn't work (e.g. if your N9 hasn't been online recently), try doing this before the install | 02:06 |
| itsnotabigtruck | HTTP_PROXY='blahblah' pkgmgr refresh-cache | 02:07 |
| * itsnotabigtruck is anxious to find out if this worked | 02:09 | |
| *** beford has quit IRC | 02:24 | |
| djszapi | heh, my apps was rejected in Ovi again | 02:26 |
| HAMIDx9_ | itsnotabigtruck : so sorry for my lateness | 02:26 |
| HAMIDx9_ | when i try this command "HTTP_PROXY='http://yourproxy:port' pkgmgr install developer-mode" | 02:27 |
| HAMIDx9_ | it shows the pkgmgr USAGE | 02:27 |
| rhkfin | djszapi: tough luck - what's the reason? | 02:28 |
| djszapi | rhkfin: http://paste.kde.org/423218/ | 02:30 |
| djszapi | 1) -> I cannot reproduce on any device with any firmware | 02:31 |
| djszapi | 2) -> Weird why they did not scream last time because I had the same file name. | 02:31 |
| djszapi | I did change nothing about the debian file generation. | 02:31 |
| *** djszapiN9 has quit IRC | 02:32 | |
| djszapi | Could someone please try to reproduce the first issue ? | 02:33 |
| djszapi | I am seriously not able to do that ... | 02:33 |
| djszapi | tried on 3 different devices with 3 different firmware versions ... | 02:33 |
| rhkfin | djszapi: where to download .deb? | 02:34 |
| djszapi | second point -> I do not know what they are talking about ... | 02:35 |
| rhkfin | it makes no sense to me either.. | 02:35 |
| djszapi | kanagram_0.1-1.debian.tar.gz -> this is the file name. | 02:39 |
| djszapi | (just like as the the last time when they did not complain about it) | 02:39 |
| djszapi | sorry, this: kanagram_0.1-1_armel.deb | 02:43 |
| *** adlan has quit IRC | 02:51 | |
| djszapi | interesting, I cannot log in to my Ovi Publisher account anymore. | 02:56 |
| djszapi | Heh, they silently changed my user account for my request | 03:04 |
| djszapi | without notifying me LOL | 03:04 |
| djszapi | Err...what the heck ? | 03:04 |
| djszapi | rhkfin: which version (firmware, that is) do you have ? | 03:05 |
| rhkfin | 1.2beta | 03:07 |
| rhkfin | file name isn't enough for me to find it, could you give URL..? | 03:07 |
| *** HAMIDx9_ has quit IRC | 03:07 | |
| * djszapi wonders why Ovi decided what user name they chose for me ... | 03:07 | |
| djszapi | rhkfin: no sorry | 03:08 |
| djszapi | I do not have it uploaded anywhere. | 03:08 |
| rhkfin | oh ok | 03:08 |
| djszapi | the filename obviously was not meant for url | 03:08 |
| djszapi | it was meant for showing the file name, not more since they had a QA issue with the filename | 03:08 |
| djszapi | according to the point two for insane reasons. | 03:08 |
| rhkfin | OK, I thought you needed people to try if it runs | 03:08 |
| djszapi | what is wrong with them ? | 03:08 |
| djszapi | that filename makes complete sense to me ... | 03:09 |
| *** Hamidx9 has joined #harmattan | 03:09 | |
| djszapi | I seriously wonder why they decided to give me a username instead of taking the selected one by me. | 03:18 |
| djszapi | if every service provider was doing it, no people could remember for every individual stuff... | 03:19 |
| *** pinheiro has quit IRC | 03:21 | |
| itsnotabigtruck | djszapi: you're probably more than tired of my packaging tips but consider changing the source format to 3.0 (native) | 03:26 |
| itsnotabigtruck | native package = intended for debian only = pretty much anything built specifically for harmattan | 03:26 |
| djszapi | that does not make too much sense | 03:26 |
| itsnotabigtruck | though actually since kanagram is a port it should be a non-native package... | 03:27 |
| djszapi | the whole KDE Harmattan packaging is quilt in the community repository. | 03:27 |
| djszapi | and my packages too in the platform. | 03:27 |
| djszapi | "native" package is the only thing should not be used actually. | 03:27 |
| Hamidx9 | <itsnotabigtruck> : hi , the command didnt work ! | 03:27 |
| itsnotabigtruck | well, nm, keep using quilt, but use native for harmattan-only stuff | 03:27 |
| itsnotabigtruck | since you don't actually want quilt on harmattan-only packages, since all patches should be part of the main source | 03:28 |
| djszapi | /home/user/MyDocs # /opt/kanagram/bin/kanagram.sh | 03:28 |
| djszapi | invoker: Invoking execution: '/opt/kanagram/bin/kanagram-harmattan' | 03:28 |
| djszapi | /opt/kanagram/bin/kanagram-harmattan: error while loading shared libraries: libXcursor.so.1: cannot open shared object file: No such file or directory | 03:28 |
| djszapi | itsnotabigtruck: except that we do not use harmattan only packages in the community | 03:29 |
| djszapi | that is the main point actually. | 03:29 |
| djszapi | we port debian stuff. | 03:29 |
| djszapi | so 3.0 (native) makes zero sense | 03:29 |
| djszapi | and this is not even the issue for Ovi either ... | 03:30 |
| djszapi | I do really wonder what is wrong about that file name. | 03:30 |
| *** Anssi138 has quit IRC | 03:32 | |
| djszapi | rhkfin: I modified my source since that, but here you go: http://minus.com/mvYkdWYGi#1 | 03:34 |
| djszapi | it is now from master HEAD without any testing for the last 10 commits | 03:34 |
| djszapi | it does not load here for some obscure reasons. | 03:35 |
| djszapi | I could launch the application after installing libxcursor, but it is weird why I would need to do that. | 03:39 |
| djszapi | rhkfin: can you reproduce what Ovi said after installing libxcursor1 from the community repository ? I fail to do so | 03:49 |
| *** adlan has joined #harmattan | 03:52 | |
| djszapi | http://minus.com/mbXiCCJOs#1 -> Could someone please try to reproduce this blank screen issue ? | 03:56 |
| djszapi | It is killing me I cannot, and the testers apparently do not have a button for sending the accumulated (console log, syslog and so forth) information to the developer automatically for rejections. | 03:57 |
| *** adlan has quit IRC | 04:05 | |
| *** adlan has joined #harmattan | 04:05 | |
| *** xnt14 has joined #harmattan | 04:08 | |
| itsnotabigtruck | btw is there any way to change the package auto/non-auto state on harmattan | 04:09 |
| itsnotabigtruck | apt-mark is the normal utility for that, but it's broken | 04:09 |
| itsnotabigtruck | because it relies on python-apt, which for some reason wasn't shipped with the N9 even though it's in the sdk repo | 04:10 |
| djszapi | so why not install ? | 04:12 |
| itsnotabigtruck | djszapi: because that's an extra step | 04:13 |
| itsnotabigtruck | i was hoping there'd be a "proper" way | 04:13 |
| itsnotabigtruck | maybe something in pkgmgr or something | 04:14 |
| itsnotabigtruck | since pkgmgr does an apt-get autoremove after every operation, i'd *hope* there's some way to control that state | 04:14 |
| djszapi | well | 04:14 |
| djszapi | that is what hte SDK repo is for. | 04:15 |
| djszapi | and it sounds like an uncommon use case anyway | 04:15 |
| itsnotabigtruck | well no, the sdk repo is for loading up dev tools on scratchbox | 04:15 |
| djszapi | you can be happy it is in the public SDK at all. | 04:15 |
| djszapi | they would normally reject the suppor for it at all. | 04:15 |
| djszapi | no | 04:16 |
| djszapi | the image should not be bloated with custom use cases for good | 04:16 |
| djszapi | you can be happy it is at all inside the SDK repository, as I mentioned. | 04:16 |
| itsnotabigtruck | well, it would be in the main repo | 04:16 |
| djszapi | they have missing pieces for common use cases | 04:16 |
| itsnotabigtruck | like all the other python stuff | 04:16 |
| itsnotabigtruck | like pyside | 04:16 |
| itsnotabigtruck | it makes sense not to include it in the image, after all python itself isn't | 04:17 |
| djszapi | there are much more liked packages in the sdk repository. | 04:17 |
| djszapi | I would personally not pray for this one as far as I did not hear anybody requesting it yet | 04:18 |
| djszapi | the last two years at least. | 04:18 |
| djszapi | I personally agree with that it should not even be inside the sdk repository, but in community or so. | 04:19 |
| djszapi | since it is not even needed for developers. | 04:20 |
| itsnotabigtruck | erm, that's the diametric opposite of agreeing | 04:20 |
| itsnotabigtruck | and it's bad form to have utilities broken out of the box | 04:20 |
| itsnotabigtruck | that said, core apt utilities really shouldn't be written in python | 04:20 |
| itsnotabigtruck | using non-essential packages | 04:20 |
| itsnotabigtruck | apt-mark isn't anything terribly complicated, there isn't much advantage to doing it in python | 04:21 |
| itsnotabigtruck | it's just calling out to the apt c library anyway | 04:21 |
| djszapi | no, it is not opposite of agreeing | 04:22 |
| djszapi | unless you change your mind every single second | 04:22 |
| djszapi | 04:15 < itsnotabigtruck> well no, the sdk repo is for loading up dev tools on scratchbox -> this is exactly the reason why it should be taken out of the SDK repository. | 04:23 |
| djszapi | actually I will ask the SDK team today for that ... | 04:23 |
| itsnotabigtruck | the sdk repo is for having a proper development environment on scratchbox | 04:23 |
| itsnotabigtruck | things desperately need to be migrated from the sdk repo to the main repo | 04:23 |
| itsnotabigtruck | i want things made more available | 04:24 |
| itsnotabigtruck | you want things made less available | 04:24 |
| itsnotabigtruck | and that's why we're both always at each other's throats | 04:24 |
| djszapi | this is not needed | 04:24 |
| djszapi | nobody requested the last two years | 04:24 |
| djszapi | so why bloat things with uncommon case ? That is what the community repository is about. | 04:25 |
| djszapi | you can put any corner case in there. | 04:25 |
| itsnotabigtruck | just because nobody requested doesn't mean nobody needs it, and moreover, most of the time 'requests' don't even make it to the people with the decision power anyway | 04:25 |
| itsnotabigtruck | it's like how microsoft says "none of our customers wanted this major issue fixed so we left it" | 04:25 |
| itsnotabigtruck | by "customers" they mean fortune 500 customers with premier support contracts | 04:25 |
| djszapi | I am really happy qt and Harmattan are not supposed to work this way | 04:26 |
| djszapi | to bloat the things with everything possible. | 04:26 |
| djszapi | we do not really wanna violate the KISS principles. | 04:26 |
| itsnotabigtruck | you seem to have a strange idea of bloat | 04:26 |
| itsnotabigtruck | rolling anything and everything into the default image = bloat | 04:27 |
| itsnotabigtruck | having an optional package in the repo != bloat | 04:27 |
| itsnotabigtruck | something that's opt-in is by definition not bloat | 04:27 |
| djszapi | of course it is | 04:27 |
| djszapi | because it is way much more maintenance we obviously do not have enough man power | 04:28 |
| djszapi | and that is what the damn community repository is for anyway | 04:28 |
| djszapi | Let us see after talking to the SDK team anyway. | 04:28 |
| djszapi | on top of maintenance hell, we would not really like the common case customers' life really hard to find one relevant package out of the zillion. | 04:29 |
| djszapi | you do not have too much experience if you say that | 04:30 |
| djszapi | afaict that is also one consideration behind the project "Mer": learn some mistakes of MeeGo to provide the full stack with zillion packages | 04:30 |
| djszapi | their principle is heavily to provide only the core, what is /really/ needed. | 04:30 |
| *** risca has joined #harmattan | 04:31 | |
| djszapi | Q: How can I remove created test accounts inside my main publisher account ? | 04:34 |
| *** Guest25259 is now known as Termana | 05:03 | |
| *** Guest44609 has joined #harmattan | 05:18 | |
| *** Guest44609 has left #harmattan | 05:18 | |
| *** oberling_ has quit IRC | 05:18 | |
| *** Hamidx9 has quit IRC | 05:20 | |
| *** oberling has joined #harmattan | 05:20 | |
| *** Guest44609 has joined #harmattan | 05:22 | |
| Guest44609 | hello:),can anyone tell me if memscanner works in n9 like it does in n900 | 05:23 |
| Tronic | I don't think you can scan the memory on N9. | 05:43 |
| Guest44609 | ok | 05:44 |
| *** Guest44609 has left #harmattan | 05:44 | |
| *** Natunen has joined #harmattan | 05:48 | |
| itsnotabigtruck | fwiw it looks like /proc/pid/mem indeed doesn't work on the N9 | 05:52 |
| itsnotabigtruck | hmm, do any of these bots have !tell | 06:25 |
| itsnotabigtruck | !tell itsnotabigtruck 'sup | 06:25 |
| *** faenil has joined #harmattan | 06:25 | |
| itsnotabigtruck | ~tell itsnotabigtruck 'sup | 06:25 |
| faenil | still up? | 06:26 |
| faenil | it's 5:26am here XD | 06:26 |
| itsnotabigtruck | it's only 10pm here | 06:26 |
| faenil | oh I see... | 06:26 |
| faenil | well, time for me to go to bed | 06:26 |
| faenil | been coding all night | 06:26 |
| itsnotabigtruck | btw any idea if any of these bots have a tell command | 06:26 |
| faenil | dunno | 06:26 |
| itsnotabigtruck | infobot does but it's for spamming people with trite crap | 06:26 |
| itsnotabigtruck | instead of storing and forwarding a message | 06:26 |
| faenil | :D | 06:27 |
| *** faenil has quit IRC | 06:30 | |
| DocScrutinizer | ~tell itsnotabigtruck about query | 06:36 |
| itsnotabigtruck | DocScrutinizer: right, things like that | 06:38 |
| itsnotabigtruck | ಠ_ಠ| 06:40 |
| * itsnotabigtruck 's terminal font can't display the look of disapproval :( | 06:41 | |
| djszapi | itsnotabigtruck: have you understood the second issue Ovi gave to me ? | 06:43 |
| itsnotabigtruck | djszapi: the filename? | 06:44 |
| itsnotabigtruck | think i missed that part, but it might have to do with your version numbering | 06:45 |
| itsnotabigtruck | while you're right that your package is better off not native, i think ovi expects everything to use simple version numbering that is supposed to be reserved only for native packages | 06:45 |
| djszapi | so I cannot push the same package name after a rejection ? | 06:45 |
| itsnotabigtruck | they probably want kanagram_1.0_armel.deb | 06:45 |
| itsnotabigtruck | and the control file to match that | 06:46 |
| djszapi | is that written somewhere in the requirements ? | 06:46 |
| itsnotabigtruck | i was going over them a while back, i think it said that | 06:46 |
| itsnotabigtruck | hold on | 06:46 |
| djszapi | http://labs.qt.nokia.com/2012/02/15/introducing-qbs/ | 06:46 |
| *** risca has quit IRC | 06:46 | |
| itsnotabigtruck | http://support.publish.ovi.com/wp-content/uploads/2011/08/Ovi-Store-Content-Guidelines-version-1.2-August-22-2011.pdf | 06:48 |
| itsnotabigtruck | djszapi: i was probably thinking of 2.2.8.1, but rereading that it seems to permit the proper version scheme | 06:48 |
| itsnotabigtruck | *however* you do need to give the app a 3 component version | 06:48 |
| itsnotabigtruck | so, kanagram_1.0.0-1_armel.deb is ok | 06:49 |
| itsnotabigtruck | kanagram_1.0-1_armel.deb isn't | 06:49 |
| itsnotabigtruck | but neither is kanagram_1.0_armel.deb | 06:49 |
| djszapi | heh | 07:03 |
| djszapi | Their issue description could be clearer then, if that is the case. | 07:03 |
| *** benares_98 has quit IRC | 07:10 | |
| djszapi | itsnotabigtruck: I think it would be nice if I can make a community page about ovi store rejection pitfalls. | 07:22 |
| djszapi | so that others could go through them with real examples. | 07:22 |
| djszapi | not a boring and very long ovi guide, but just the points. | 07:22 |
| djszapi | Also, I do not understand why they accepted the filename previously. | 07:23 |
| djszapi | actually, I would even say, such a "mistake" should be caught automatically while uploading. | 07:23 |
| djszapi | Q: Can I hold back my OVI QA submission I made few hours ago so that I do not need to wait yet another 3-4 days for getting any responses the file name is wrong ? I would like to hold back the one submitted with arguably "wrong" filename, and push the "correct" one. | 07:31 |
| djszapi | A: Ahh yes, there is a way of doing that. | 07:31 |
| djszapi | itsnotabigtruck: wonder whether you have a bit of time and sake to test my app again: http://minus.com/mbhZOOJGaj#1 | 07:35 |
| djszapi | You still have PR1.1 ? | 07:35 |
| *** luke-jr_ has joined #harmattan | 07:42 | |
| *** luke-jr has quit IRC | 07:42 | |
| *** hardaker has quit IRC | 07:45 | |
| *** xarcass has joined #harmattan | 08:14 | |
| *** xarcass has joined #harmattan | 08:15 | |
| itsnotabigtruck | djszapi: yeah, i do | 08:16 |
| djszapi | itsnotabigtruck: does it work for you ? | 08:20 |
| itsnotabigtruck | djszapi: crap, i should probably get some sleep | 08:24 |
| itsnotabigtruck | i'll see about loading it tomorrow | 08:24 |
| djszapi | do not bother really | 08:25 |
| djszapi | it should be accepted by that time | 08:26 |
| djszapi | so if you cannot do it right, nvm. | 08:26 |
| djszapi | now* | 08:26 |
| djszapi | someone else might help here... | 08:26 |
| *** Khaled has joined #harmattan | 08:27 | |
| *** Khaled has quit IRC | 08:39 | |
| *** mece has joined #harmattan | 08:55 | |
| *** longthen1 has joined #harmattan | 08:57 | |
| *** longthen has quit IRC | 08:58 | |
| *** zk8 has joined #harmattan | 09:06 | |
| *** e-yes has quit IRC | 09:17 | |
| *** hhartz has joined #harmattan | 09:18 | |
| *** e-yes has joined #harmattan | 09:19 | |
| djszapi | X-Fade ping | 09:43 |
| *** sony123 has joined #harmattan | 09:55 | |
| sony123 | hi djszapi | 09:55 |
| sony123 | I saw yesterday you mentioned there's a way to get root access from aegis? | 09:56 |
| sony123 | Well what I want to do is to get a GUI to edit the allow_adhoc flag under /sys/..... folder | 09:56 |
| sony123 | So I only need to gain root for every QProcess launched within the app | 09:58 |
| *** gabriel9 has joined #harmattan | 09:58 | |
| matrixx | sony123: I would have needed a similar scenario for a reboot from a GUI app, by starting another process for that | 09:59 |
| matrixx | but didn't get it working :/ | 10:00 |
| *** jreznik has joined #harmattan | 10:00 | |
| sony123 | hi matrixx, do you know any alternative solution? | 10:00 |
| matrixx | nope, I just didn't do it :( | 10:01 |
| *** ZogG_laptop has quit IRC | 10:01 | |
| matrixx | let the "reboot part" of my app to the user's responsibility :D | 10:01 |
| sony123 | ah I see | 10:01 |
| sony123 | Wait, I think I read somewhere in harmattan security guide that you can use script during installation for the reboot? | 10:02 |
| *** jreznik has quit IRC | 10:08 | |
| *** ab has joined #harmattan | 10:10 | |
| *** ab has joined #harmattan | 10:10 | |
| *** virtuald has quit IRC | 10:15 | |
| jonni | sony123: add it to aegis manifest, you can request root for your binaries like http://harmattan-dev.nokia.com/docs/library//html/guide/html/Developer_Library_Developing_for_Harmattan_Harmattan_security_Security_guide_Using_Harmattan_security_features.html#Requesting_specific_user_accounts_and_groups | 10:15 |
| sony123 | I totally missed this section, thanks for pointing out, I will give it a try. | 10:19 |
| sony123 | Yesterday I briefly tried setuid/setgid but somehow it didn't work.... | 10:20 |
| sony123 | I had problems getting QProcess to execute mutiple commands..... | 10:20 |
| *** zz_gri is now known as gri | 10:25 | |
| jonni | sony123: yes setuid/setgit only works if you have setup permissions in aegis manifest, like if in manifest you have both credentials for user and root, then setuid between root and user is allowed, without manifest it just fails | 10:26 |
| jonni | I just assumed that you have done manifest that has both credentials | 10:27 |
| *** TheBootroo has joined #harmattan | 10:30 | |
| *** jreznik has joined #harmattan | 10:34 | |
| sony123 | Got it, thanks a lot jonni | 10:34 |
| TheBootroo | hello | 10:35 |
| TheBootroo | i've bought a N9 black 64GB | 10:35 |
| TheBootroo | but i think it is switzerland one, so it don't show me the PR1.1 and 1.1.1 update | 10:35 |
| TheBootroo | i want to flash it but i'm under linux, so i don't have navifirm (i have flasher) | 10:36 |
| TheBootroo | how can i dl a PR1.1.1 for N9 64GB Black west europe (france) ? | 10:36 |
| jonni | pr 1.1.1 is only for middle asian margets, there is no european version for 1.1.1 | 10:37 |
| jonni | pr1.1 is the latest for europe | 10:37 |
| TheBootroo | ok | 10:37 |
| TheBootroo | so where can i download the image directely so i don't have to use Navifirm which is unavaible on my platform | 10:38 |
| jonni | no official download links are available, but if you google search you can find direct links | 10:38 |
| *** hhartz has quit IRC | 10:44 | |
| TheBootroo | i've found naviFirmEx, which is coded in pure C++ (no .net) and runs perfectly with Wine emulator | 10:49 |
| TheBootroo | so i'm dowloading the 20.2011.40.4 | 10:50 |
| TheBootroo | then i'll flash with the native linux flasher bin | 10:50 |
| TheBootroo | i'll to be patient, only 11.6% download completed | 10:53 |
| *** sony123 has quit IRC | 10:56 | |
| gri | djszapi: Now I tried debugging with gdb commandline on the phone: aegis credentials missing, like when using QtCreator ... | 11:00 |
| djszapi | gri: so what ? | 11:10 |
| djszapi | you do not know how the credentials work ? | 11:10 |
| gri | djszapi: Tell me a way how to debug without losing them | 11:10 |
| djszapi | it is really pretty simple. | 11:11 |
| gri | attaching to a running process? | 11:11 |
| djszapi | gri: please read the security guide because I am pretty busy now, sorry. | 11:13 |
| djszapi | it is properly documented. | 11:13 |
| TheBootroo | it is important to keep the same local variant code ? mine is switzerland but i don't it, i want french one | 11:14 |
| *** ab has quit IRC | 11:15 | |
| TheBootroo | if i choose the same model and capacity no problem normally ? | 11:15 |
| djszapi | I guess no for Swiss and French | 11:15 |
| TheBootroo | djszapi: ok | 11:16 |
| TheBootroo | becasue swiss sux, no update for them | 11:16 |
| TheBootroo | hope we'll get 1.2 soon | 11:16 |
| TheBootroo | (OTA) | 11:16 |
| djszapi | I am sorry but I cannot comment on that when we push it out.. | 11:17 |
| djszapi | gri: isn't it trivial you need to pass the credentials to gdb if you use them ? | 11:19 |
| gri | djszapi: No that isn't trivial - how do I do this? | 11:20 |
| djszapi | why isn't it not trivial ? | 11:20 |
| djszapi | please read the manifest manual how to request credentials. | 11:20 |
| djszapi | so on harmattan security: if you would like to use a credential for an app, you need to make sure it is available | 11:21 |
| djszapi | you said it is not: so you need to request it. | 11:21 |
| gri | djszapi: Well my app has the credentials when started without gdb. I was looking for a word about debugging on harmattan-dev but there is none | 11:21 |
| gri | not even a hint in the "debugging tools" section | 11:22 |
| djszapi | gdb != your app. | 11:22 |
| gri | so I have to add the credentials for gdb and remove them after debugging?! | 11:23 |
| djszapi | no | 11:23 |
| djszapi | you need to request a credential for it. | 11:23 |
| djszapi | it is that simple. | 11:24 |
| djszapi | it is exactly the same as applauncher. | 11:24 |
| djszapi | or valgrind and the like. | 11:24 |
| djszapi | those are all separate applications, not your binary. | 11:25 |
| djszapi | and at least applauncher is properly documented, even with an example imo. | 11:25 |
| gri | do you have an exact line for me? <for path="/usr/bin/gdb" id="" /> does not seem to work | 11:25 |
| djszapi | gdb comes from your package ? | 11:25 |
| djszapi | if not, why would you pass that way ? | 11:25 |
| djszapi | sorry, but /please/ read the documentation | 11:26 |
| gri | applauncher is also done this way: <for path="applauncherd-launcher::/usr/bin/applauncherd.bin" id="" /> | 11:26 |
| djszapi | do you seriously think those two lines are the same ? | 11:26 |
| gri | no but I have to find out the group for gdb first | 11:26 |
| * djszapi wonders whether gri sees the scope there. | 11:26 | |
| djszapi | that is not a group | 11:26 |
| djszapi | that is a package scope... | 11:27 |
| gri | I just asked you first before looking up since you seem to know the solution but won't tell me :) | 11:27 |
| djszapi | I can copy/paste the documentation to you. | 11:27 |
| djszapi | do the job for you, but will you learn something out of it ? | 11:27 |
| djszapi | or will you come next time with valgrind or strace ? | 11:28 |
| djszapi | this way, you understand the thing next time. | 11:28 |
| djszapi | (once you already came to me with applauncher, same issue) | 11:28 |
| *** DocScrutinizer has quit IRC | 11:28 | |
| *** DocScrutinizer has joined #harmattan | 11:28 | |
| gri | well the applauncher thingy was somehow documented for gdb I have no clue | 11:29 |
| djszapi | we should document every single binary ? | 11:29 |
| djszapi | or we should document how to request creds for apps not coming from the same package ? | 11:30 |
| gri | well, essential ones like "how to use the debugger since it does not work out of the box", yes | 11:30 |
| djszapi | no | 11:30 |
| djszapi | we do not duplicate and clutter the documentation | 11:30 |
| djszapi | the security documentation has already been huge | 11:30 |
| gri | well, there is a FAQ where one sentence would help :) | 11:30 |
| djszapi | one sentence for everything ? | 11:31 |
| gri | atleast people who just want to develop a small app without reading through hundeds of pages, this would be helpful | 11:32 |
| * djszapi thinks people do not need documentation sections just FAQ since they skip the documentatio part | 11:32 | |
| djszapi | gri: it is inside the "how to write a manifest file" section | 11:33 |
| djszapi | and no, we do not wanna have another hundreds of FAQs | 11:34 |
| djszapi | Q: any debian experts here ? https://build.pub.meego.com/package/live_build_log?arch=armv7el&package=grantlee&project=home%3Arzr%3Aharmattan&repository=MeeGo_1.2_Harmattan_Maemo.org_MeeGo_1.2_Harmattan_standard Asked already this morning, but I do not know what makeshlibs complains about. :/ | 11:34 |
| TheBootroo | will there be the Qt4.8 version in final PR1.2 for N9 ? | 11:35 |
| djszapi | doubt that | 11:36 |
| djszapi | 4.8 is broken | 11:36 |
| djszapi | I mean...it broke many apps already in Ovi | 11:36 |
| djszapi | we do not need such regressions on Harmattan for no real gain. | 11:36 |
| djszapi | maybe 4.8.1. when they revert the breakages, but I am not sure it is worth dealing with it | 11:37 |
| djszapi | too risky. | 11:37 |
| djszapi | no real benefits that much | 11:37 |
| djszapi | QtQuick comes with 4.7.4 anyway | 11:37 |
| djszapi | 1.1, that is | 11:37 |
| TheBootroo | 4.8 broke apps ? not for me, just huge perf gain for QML and network access ! | 11:39 |
| djszapi | check out the development mailing list | 11:39 |
| TheBootroo | which breakage N? | 11:39 |
| djszapi | it is really well written on the mailing list | 11:39 |
| TheBootroo | dunno where it is | 11:39 |
| TheBootroo | (i don't use mailing list, too much unreadable) | 11:40 |
| TheBootroo | if only we could subscribe via RSS | 11:40 |
| djszapi | I wll not copy paste it out for you, sorry :) | 11:40 |
| TheBootroo | would be could | 11:40 |
| *** Khaled has joined #harmattan | 11:40 | |
| gri | djszapi: "The Aegis manifest file can also request credentials for an executable in another package. For example, this is the case with an application invoker/launcher, when the application launcher daemon needs to have all the credentials for launching an application. In such cases, the path must be prefixed with the target package name." and as example below I see "packagename::/path/to/binary" - and gdb::/usr/bin/gdb is not accepted while installing, again | 11:40 |
| djszapi | you do not need to subscribe for reading a mailing list. | 11:40 |
| TheBootroo | s/could/cool | 11:40 |
| TheBootroo | djszapi: i know | 11:40 |
| TheBootroo | but much unreadable in the web browser | 11:41 |
| djszapi | lol | 11:41 |
| TheBootroo | have to next next next next to find somethings | 11:41 |
| djszapi | not really | 11:41 |
| TheBootroo | and there are too many quote of previous msg making it hard to find the intersting parts | 11:41 |
| djszapi | there are one page threading, too. | 11:41 |
| djszapi | is* | 11:41 |
| TheBootroo | djszapi: maybe they have enhanced the mailing list system but last time is used it i found it not very helpfull | 11:42 |
| djszapi | but you do not wanna read the information you need because it is on the web ? :) | 11:43 |
| gri | djszapi: gdb does not have an "Object: /path/to/somebinary" entry in restok.conf and if I read the docs correctly I can only request credentials for binaries listed there? | 11:44 |
| TheBootroo | djszapi: i only ddont want to spend hours in searching something i'll take age to find | 11:44 |
| TheBootroo | (yeah i know, i'm lazy) | 11:44 |
| djszapi | gri: huh ? | 11:47 |
| djszapi | gri: you do realize you should use develsh right ? | 11:48 |
| djszapi | TheBootroo: am I supposed to do the job for you ? ;) | 11:48 |
| gri | as far as I remember this did not help me. Also I have to run as "user" | 11:48 |
| TheBootroo | trying to flash the n9 | 11:49 |
| djszapi | no | 11:49 |
| djszapi | never | 11:49 |
| TheBootroo | with flasher | 11:49 |
| djszapi | you need to run it as developer | 11:49 |
| TheBootroo | when i plug the N9 to the pc, the flasher says Error claiming USB interface: Device or resource busy | 11:49 |
| gri | well then I have a problem :) | 11:49 |
| djszapi | TheBootroo: you do realize it is not much information right ? :) | 11:50 |
| djszapi | can be many problem and I would need to ask zillion information | 11:50 |
| TheBootroo | yeah that's why i can't understand | 11:50 |
| TheBootroo | i tried to plug the phone being on or off, no change | 11:50 |
| djszapi | no | 11:50 |
| gri | since the accounts are stored per user and running as developer results in two nokia accounts fighting for the push notification service | 11:50 |
| djszapi | I mean I would need the OS, whether you tried cold flash, what sort of flashing way, which images, what image is now on the device, whether you followed the official step, etc | 11:51 |
| djszapi | gri: user is not supposed to debug | 11:52 |
| *** Khaled has quit IRC | 11:52 | |
| djszapi | actually it could, but only with user credentials. | 11:52 |
| djszapi | gri: and please please try to provide more information, which token we are talking about etc | 11:52 |
| TheBootroo | djszapi: on linux | 11:52 |
| gri | djszapi: Only the TrackerReadAccess token | 11:52 |
| TheBootroo | with flasher form nokia dev | 11:53 |
| TheBootroo | 3.12 | 11:53 |
| *** ZogG_laptop has joined #harmattan | 11:53 | |
| djszapi | gri: damn, I should not have spent time with you then ... | 11:53 |
| TheBootroo | tryed to flash N9 64 GB Black to pR1.1 | 11:53 |
| djszapi | gri: have you checked out what credentials gdb has ? | 11:53 |
| djszapi | accli -I `pidof gdb` | 11:53 |
| djszapi | TheBootroo: have you read how to flash a device ? | 11:54 |
| djszapi | please provide the whole output /from/ prompt (yes, the command involved) | 11:54 |
| gri | djszapi: http://pastebin.com/9WfhRxpA | 11:55 |
| gri | debugging works fine except for the missing token | 11:56 |
| djszapi | so the debugging does not work then ? | 11:56 |
| gri | it works but my application does not display anything since it can't load the data | 11:56 |
| djszapi | so it does not work | 11:57 |
| djszapi | well, actually It was me reporting a bug about gdb wrt this very issue | 11:57 |
| djszapi | but it was long ago | 11:57 |
| djszapi | so it should have been released | 11:57 |
| *** ab has joined #harmattan | 11:57 | |
| *** ab has quit IRC | 11:57 | |
| *** ab has joined #harmattan | 11:57 | |
| djszapi | please check out the gdb manifest | 11:57 |
| djszapi | (it should not have manifest) | 11:58 |
| djszapi | apt-get source gdb | 11:58 |
| gri | E: You must put some 'source' URIs in your sources.list | 11:58 |
| djszapi | aye, I remember | 11:59 |
| djszapi | we decided to not stuff it | 11:59 |
| djszapi | gri: obviously in your madde stuff | 11:59 |
| djszapi | so yeah, as I said, you need to use develsh | 11:59 |
| gri | http://harmattan-dev.nokia.com/pool/harmattan-beta3/free/g/gdb/ this does not help? | 12:00 |
| gri | djszapi: As I remember I had the same problem with "developer" but I have to check that again | 12:01 |
| djszapi | it is probably because the request policy is not add by default | 12:01 |
| TheBootroo | djszapi: after blacklist some phonet modules i can start the process of flasher | 12:01 |
| djszapi | and the guys were claiming that is a good idea | 12:02 |
| djszapi | sdk* | 12:02 |
| TheBootroo | but i get and error | 12:02 |
| TheBootroo | here is the paste | 12:02 |
| djszapi | gri: well, I will not the job for you :) | 12:02 |
| djszapi | extract the stuff and see | 12:02 |
| djszapi | either the binary (_aegis) or the source (gdb.aegis) | 12:02 |
| TheBootroo | djszapi: http://pastebin.com/LEBcQrnd | 12:02 |
| djszapi | TheBootroo: phonet is also mentioned | 12:02 |
| djszapi | but ohye on the internet you dislike ;) | 12:03 |
| TheBootroo | i search on internet | 12:03 |
| TheBootroo | i like internet | 12:03 |
| TheBootroo | i don't like mailing lsit layout, that's all | 12:03 |
| TheBootroo | so i blacklisted phonet | 12:03 |
| djszapi | I mostly find the stuff on mailing lists | 12:03 |
| TheBootroo | so what is the problem nox | 12:04 |
| TheBootroo | *now | 12:04 |
| djszapi | since the qt/kde/linux_kernel/webkit are developed there. | 12:04 |
| TheBootroo | what is this bb5_rdc_cert_read failed sutff ? | 12:04 |
| djszapi | sorry, I do not know | 12:04 |
| TheBootroo | i have image file : DFL61_HARMATTAN_20.2011.40-4_PR_LEGACY_001-OEM1-958_ARM.bin | 12:05 |
| TheBootroo | and RM696_059J1V6_20.2011.40.4_001.dcp | 12:05 |
| TheBootroo | and that RM696_059J1V6_20.2011.40.4_001.vpl | 12:05 |
| TheBootroo | and taht RM696_059J1V6_20.2011.40.4_001_signature.bin | 12:05 |
| djszapi | paste.kde.org ... | 12:05 |
| djszapi | have you tried the cold flash ? | 12:05 |
| TheBootroo | what is that ? i tries to flash with -F -f | 12:06 |
| TheBootroo | and with phone off | 12:06 |
| TheBootroo | then laucnh command, plug phone and waited | 12:06 |
| *** japplo has joined #harmattan | 12:07 | |
| djszapi | flasher --help | 12:08 |
| djszapi | -F is for the image | 12:08 |
| djszapi | -f is for flashing | 12:08 |
| * djszapi is back to do something | 12:09 | |
| gri | djszapi: the gdb .deb does not have a _aegis file | 12:09 |
| djszapi | as I said :D | 12:09 |
| japplo | hi guys, is it possible to check via command line if developer mode is enabled? | 12:10 |
| djszapi | also develsh != developer user | 12:10 |
| djszapi | whoami returns user to me ... | 12:10 |
| djszapi | japplo: yes | 12:10 |
| japplo | djszapi: ok how? | 12:11 |
| djszapi | well, there are many ways of doing that ... | 12:11 |
| TheBootroo | omfg i think my n9 is bricked | 12:12 |
| djszapi | checking credentials available in developer mode only, checking log files, binaries etc | 12:12 |
| djszapi | TheBootroo: blame yourself for using NaviFirm. | 12:12 |
| djszapi | or direct imaging. | 12:12 |
| djszapi | Nokia does not recommend that. | 12:13 |
| japplo | djszapi: and what is the common choice? | 12:13 |
| djszapi | japplo: no clue, I have never heard this request from anybody yet. | 12:13 |
| *** coffetime has joined #harmattan | 12:13 | |
| TheBootroo | djszapi: what can we do if they don't provied update ? the only tip i swa every where was NaviFirm | 12:13 |
| djszapi | TheBootroo: huh ? they do provide update. | 12:14 |
| djszapi | I have just updated my N9 yesterday actually that I got at the Qt Developer Days. | 12:14 |
| TheBootroo | djszapi: not for the switzerland variant | 12:14 |
| TheBootroo | i didn't i bought a swiss one | 12:15 |
| TheBootroo | (i bought it on a french website) | 12:15 |
| djszapi | bad luck, but please remember Nokia warned you. | 12:15 |
| TheBootroo | warnes nothing | 12:15 |
| TheBootroo | i saw the warning AFTER rebooting from missed flash | 12:15 |
| *** ab has quit IRC | 12:16 | |
| TheBootroo | now flasher don't want to reflash it | 12:16 |
| djszapi | no | 12:17 |
| djszapi | the flasher gives a warning for you. | 12:17 |
| *** ab has joined #harmattan | 12:17 | |
| japplo | djszapi: which binaries are not available? | 12:19 |
| TheBootroo | djszapi: i try to reflash warm | 12:19 |
| TheBootroo | it seems it starts to flash rootfs but fails | 12:19 |
| djszapi | japplo: syslog | 12:19 |
| TheBootroo | others images worked | 12:19 |
| djszapi | japplo: but there might be a more precise way of doing it. I do not personally know this that well, sorry. | 12:20 |
| TheBootroo | means i should try another iamge ? | 12:20 |
| djszapi | gri: heh the plugin icon thingie is still not fixed | 12:21 |
| *** ab has quit IRC | 12:21 | |
| djszapi | TheBootroo: it is perhaps that you try to flash an older kernel | 12:23 |
| djszapi | try to reflash the same. | 12:23 |
| djszapi | or newer. | 12:23 |
| TheBootroo | suck it, i need to redownload the image | 12:23 |
| TheBootroo | djszapi: i tried to install FR PR1.1 on a device with SW PR1.0 | 12:23 |
| djszapi | you can downgrade the firmware. | 12:23 |
| TheBootroo | i'll try to flash SW PR1.1 | 12:23 |
| djszapi | it is very specific version (232 and so on) | 12:23 |
| djszapi | so it might be that PR1.0 is newer than PR1.1 | 12:24 |
| TheBootroo | but i hope i'll not need to flash again for PR1.2 (was for that reason i wanted to flash FR variant instead of swiss one) | 12:24 |
| TheBootroo | djszapi: yes its _232 version | 12:24 |
| TheBootroo | but how can a PR1.0 be newer than PR1.1 ? | 12:25 |
| djszapi | trust me | 12:25 |
| djszapi | I do not have time to explain it now | 12:25 |
| TheBootroo | i didn't get the "can't downgrade error" | 12:25 |
| djszapi | and I am probably not entitled either. | 12:25 |
| djszapi | Q: Also I would like to get help too with this: https://build.pub.meego.com/package/live_build_log?arch=armv7el&package=grantlee&project=home%3Arzr%3Aharmattan&repository=MeeGo_1.2_Harmattan_Maemo.org_MeeGo_1.2_Harmattan_standard | 12:25 |
| TheBootroo | djszapi: hope reflashing with swiss variant will solve my bricked device problem | 12:27 |
| djszapi | japplo: I got it from bb5 in dpkg wrapper | 12:27 |
| djszapi | when I worked on it | 12:27 |
| djszapi | but I bet you cannot do the same. | 12:27 |
| *** pinheiro has joined #harmattan | 12:31 | |
| japplo | djszapi: what do you mean with bb5? | 12:32 |
| djszapi | library | 12:33 |
| japplo | djszapi: it is possible to check this via command line? | 12:34 |
| djszapi | check dpkg wrapper | 12:34 |
| djszapi | whether it has such an option. | 12:34 |
| djszapi | japplo: why are you interested in this btw ? | 12:34 |
| *** ab has joined #harmattan | 12:35 | |
| *** ab has joined #harmattan | 12:35 | |
| japplo | djszapi: do you know ipadd-widget ? | 12:37 |
| japplo | djszapi: I like to add this info | 12:38 |
| djszapi | no, sorry. | 12:38 |
| djszapi | what is that ? | 12:38 |
| djszapi | X-Fade you did not get the bug report | 12:40 |
| japplo | djszapi :http://thp.io/2011/ipaddr-widget/ | 12:40 |
| X-Fade | djszapi: ? | 12:40 |
| djszapi | X-Fade if you check out the modification of my rules file, it was added later | 12:40 |
| djszapi | X-Fade: how could I get it work without that hack ? | 12:40 |
| djszapi | OBS returns Coverage | 12:41 |
| djszapi | official environment returns "" (notype) | 12:41 |
| djszapi | adding coverage to the rules file is a very nasty hackaround | 12:41 |
| djszapi | it is indeed an OBS issue | 12:41 |
| X-Fade | djszapi: Nothing in OBS sets that. | 12:41 |
| djszapi | X-Fade even if I set it to release OBS hard keeps it to Coverage | 12:41 |
| djszapi | I spend almost my whole day yesterday with this issue | 12:41 |
| djszapi | OBS thinks Coverage is cool, and that is not cool | 12:41 |
| X-Fade | fgrep doesn't return anything. | 12:41 |
| djszapi | and we cannot even override | 12:42 |
| X-Fade | So it is not OBS. | 12:42 |
| djszapi | it is an obs issue | 12:42 |
| djszapi | obs does not behave like the target is supposed to behave. | 12:42 |
| djszapi | you will get notype | 12:43 |
| djszapi | compile this package in scratchbox, madde or the phone | 12:43 |
| X-Fade | It must me some other thing. Like picking up some env var or thinking it is running on some system. | 12:43 |
| djszapi | obs returns Coverage | 12:43 |
| djszapi | anyway, OBS needs to be fixed | 12:43 |
| X-Fade | obs has no settings for cmake | 12:43 |
| djszapi | whatever the root cause is. | 12:43 |
| X-Fade | obs doesn't use cmake | 12:43 |
| djszapi | since it only does not work on obs | 12:43 |
| djszapi | not just this package, but anything | 12:43 |
| djszapi | it is a generic issue | 12:43 |
| X-Fade | And I checked, it is the latest cmake from beta3. | 12:43 |
| X-Fade | Anyway, nothing I can do about it. | 12:44 |
| djszapi | that is a sad consequence | 12:44 |
| djszapi | because you have only the access to c-obs | 12:44 |
| djszapi | for checking deep details. | 12:44 |
| djszapi | I cannot know what it messes up | 12:44 |
| X-Fade | djszapi: Does it do that in a local build? | 12:44 |
| djszapi | it does not work as the official environments. | 12:44 |
| djszapi | X-Fade yes | 12:45 |
| djszapi | if you meant to write osc | 12:45 |
| X-Fade | Then you can debug it too. | 12:45 |
| djszapi | how ? | 12:45 |
| djszapi | what ? | 12:45 |
| X-Fade | Just go into the chroot and check? | 12:45 |
| djszapi | check what ? | 12:45 |
| X-Fade | Well, whatever ;) | 12:46 |
| *** virtuald has joined #harmattan | 12:46 | |
| djszapi | meh blah D | 12:46 |
| djszapi | if I knew what to check I have already done it. | 12:46 |
| djszapi | I tried to check the whole c-obs project config | 12:46 |
| X-Fade | Just run the what you see in the build log? | 12:46 |
| djszapi | X-Fade: done that many times. | 12:46 |
| djszapi | same cmake options passed | 12:46 |
| djszapi | since I did not set anything | 12:46 |
| X-Fade | In the chroot? | 12:46 |
| djszapi | so cmake is not a culsprit | 12:46 |
| djszapi | chroot ? | 12:46 |
| djszapi | /var/tmp/build/.. | 12:47 |
| X-Fade | osc build | 12:47 |
| X-Fade | Yeah.. | 12:47 |
| X-Fade | Chroot into it. | 12:47 |
| djszapi | and ? | 12:47 |
| X-Fade | run the commands | 12:47 |
| djszapi | what commands ? | 12:48 |
| X-Fade | Whatever you need to do to check. | 12:48 |
| djszapi | to check what ? :D | 12:48 |
| X-Fade | Run BUILD for instance. | 12:48 |
| djszapi | I have seriously no clue | 12:48 |
| djszapi | cmake is not my project, nor is c-obs | 12:49 |
| djszapi | what I can do is to write to the cmake mailing list including you in the loop | 12:49 |
| djszapi | so that if they say something you can tell from c-obs pov | 12:49 |
| X-Fade | run override_dh_auto_configure: $(overridden_command) -- -DCMAKE_BUILD_TYPE=Coverage | 12:50 |
| X-Fade | ehm.. | 12:50 |
| X-Fade | run .build/build | 12:50 |
| X-Fade | And see what it does. | 12:50 |
| djszapi | what do we expect ? | 12:50 |
| X-Fade | Well, let it run until you see the cmake line. | 12:50 |
| djszapi | X-Fade why not check the build log ? | 12:51 |
| djszapi | I mean why do we need to do it again ? | 12:51 |
| X-Fade | Because locally you can use strace or whatever. | 12:51 |
| X-Fade | Or read the build script. | 12:51 |
| * djszapi has no clue about strace | 12:51 | |
| X-Fade | Anyway, you have an identical enviroment there. If you want you can debug it. | 12:52 |
| X-Fade | I don't have time to dive into it, but it is not that you are limited because you don't have access. | 12:52 |
| djszapi | I do not have access to project config I am afraid | 12:52 |
| X-Fade | Sure you do. | 12:52 |
| djszapi | or well, honestly: I do not know how OBS works. | 12:53 |
| *** adlan has quit IRC | 12:53 | |
| djszapi | so I do not know what to check | 12:53 |
| X-Fade | Read the build script | 12:53 |
| djszapi | that is shipped by obs ? | 12:53 |
| X-Fade | yep | 12:53 |
| *** faenil has joined #harmattan | 12:53 | |
| X-Fade | .build/build | 12:53 |
| djszapi | ./build folder is empty in chroot | 12:54 |
| X-Fade | run osc build and ctrl-c it. | 12:54 |
| djszapi | well I need to ask cmake developers what to check | 12:54 |
| djszapi | since I have seriously no clue | 12:54 |
| djszapi | and I would not like to read the cmake code as well | 12:54 |
| djszapi | X-Fade got a clue for this ? https://build.pub.meego.com/package/live_build_log?arch=armv7el&package=grantlee&project=home%3Arzr%3Aharmattan&repository=MeeGo_1.2_Harmattan_Maemo.org_MeeGo_1.2_Harmattan_standard | 12:56 |
| X-Fade | djszapi: Yeah, I triggered a rebuild. | 12:56 |
| djszapi | X-Fade is the build script generated or static ? | 12:57 |
| X-Fade | static | 12:57 |
| djszapi | isn't it available somewhere then publicly ? | 12:57 |
| djszapi | instead of ctrl-c and chroot | 12:57 |
| djszapi | also, does c-obs have diff to upstream obs ? | 12:58 |
| X-Fade | Sure, it is github in the OBS repo. | 12:58 |
| X-Fade | no | 12:58 |
| djszapi | right | 12:58 |
| djszapi | so c-obs is just a different configuration of obs ? | 12:58 |
| X-Fade | It is just an install, yes. | 12:58 |
| djszapi | than for instance, arch obs or debian obs ? | 12:58 |
| djszapi | X-Fade: so it is just like setting a bugzilla or trac ? | 12:59 |
| X-Fade | djszapi: Yes, you can download a vm appliance too. Then you have your own. | 12:59 |
| djszapi | is there a way of searching for a file on github ? | 13:01 |
| djszapi | in a desired repository for instance ? | 13:01 |
| djszapi | X-Fade why did you re-trigger the build without modification ? | 13:02 |
| djszapi | Did you change something on c-obs side ? | 13:02 |
| djszapi | X-Fade: Also, I had many time issue yesterday | 13:03 |
| X-Fade | djszapi: No, I wanted to see the live log. | 13:03 |
| X-Fade | djszapi: I fixed those, I hope. | 13:03 |
| djszapi | X-Fade c-obs does not build packages having rules file touched in the future. | 13:03 |
| djszapi | like in 1-2 hours. | 13:03 |
| djszapi | because here we have utc + 2 hours | 13:04 |
| djszapi | and it works with utc | 13:04 |
| X-Fade | Clock is running utc. | 13:04 |
| X-Fade | https://github.com/openSUSE/obs-build/tree/db16c8b79524b5be3990a7ddfda77672ca18cb15 | 13:04 |
| X-Fade | check the file: build | 13:04 |
| X-Fade | That is all, just a shell script. | 13:05 |
| djszapi | X-Fade any hope you get it fixed ? | 13:05 |
| djszapi | I would not like to set my time back 2 hours | 13:05 |
| djszapi | nor waiting 2 hours. | 13:05 |
| djszapi | "just" a shell script :D 2128 LOC :P | 13:06 |
| X-Fade | djszapi: Run you clock in UTC? | 13:06 |
| X-Fade | That is the only real time anyway :) | 13:06 |
| djszapi | X-Fade will my time still aligned to my timezone ? | 13:06 |
| djszapi | be* | 13:06 |
| X-Fade | djszapi: display can be changed with TZ env var. | 13:07 |
| djszapi | true, that is what I do with my friend's server, but still | 13:07 |
| *** pinheiro has quit IRC | 13:07 | |
| djszapi | is it a dpkg limitation, or obs ? | 13:07 |
| *** jluisn has joined #harmattan | 13:07 | |
| X-Fade | It is the same as that make is complaining. | 13:08 |
| X-Fade | I guess it is build into dpkg | 13:08 |
| djszapi | I wonder why I had no such issues previously. | 13:08 |
| djszapi | but this might be the reason as well for the late publishing ? | 13:08 |
| X-Fade | But some of the clocks of the workers work off. | 13:08 |
| X-Fade | I fixed those. | 13:09 |
| X-Fade | So it might have been just that. | 13:09 |
| djszapi | so it is now published properly what we discussed precedingly ? | 13:09 |
| X-Fade | Did not check that yet. | 13:09 |
| djszapi | well we can check after grantlee built. | 13:09 |
| djszapi | but did not build, same issue. | 13:10 |
| djszapi | X-Fade: this is what ubuntu/debian uses for this package: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/precise/grantlee/precise/view/head:/debian/rules | 13:11 |
| djszapi | prolly related to this: DEB_DH_MAKESHLIBS_ARGS_ALL = -V -u-c0 | 13:11 |
| djszapi | but not sure what that is for and how to interpret into dh | 13:11 |
| djszapi | just copy/paste ? | 13:11 |
| X-Fade | Can't hurt :) | 13:11 |
| djszapi | I have no idea what that is supposed to mean | 13:12 |
| X-Fade | Anyone here familiar with libcontentaction and desktop files? | 13:12 |
| *** TheBootroo has quit IRC | 13:12 | |
| X-Fade | Apps client doesn't come to the front when you put it in the background and click on the icon again. | 13:13 |
| X-Fade | Eventhough we use invoker --single-instance. | 13:13 |
| *** miroslav has joined #harmattan | 13:16 | |
| RST38h | Probably not handling some event right | 13:16 |
| X-Fade | We register a dbus service in the same desktop file | 13:17 |
| X-Fade | Might be that it gets confused that it is already running then. | 13:17 |
| X-Fade | https://meego.gitorious.org/meego-applications/meego-apps-client/blobs/master/appsformeego/xdg-harmattan/org.formeego.apps.desktop | 13:18 |
| RST38h | Ah | 13:18 |
| RST38h | I would suggest removing dbus stuff completely, Harmattan appears to deprecate it | 13:19 |
| X-Fade | No, this is dbus for an urischeme | 13:19 |
| RST38h | if you do have dbug, carefully check what kind of messages you are getting from it | 13:19 |
| X-Fade | Not to start the app. | 13:19 |
| RST38h | It is quite possible that the OS tries "waking up" your app by sending some specific dbus message | 13:19 |
| RST38h | Some Fremantle remnant nobody cared to clean up | 13:20 |
| X-Fade | Ah, right.. It might be smart ;) | 13:20 |
| X-Fade | Well, tries to be :) | 13:20 |
| djszapi | X-Fade: does it make sense to merge the rpm and debian based kde packages into one repository ? | 13:23 |
| djszapi | PA and Harmattan specially. | 13:23 |
| X-Fade | djszapi: It might help you share resources (human) | 13:23 |
| djszapi | X-Fade: they do not know deb packaging and I do not know rpm either. I was thinking of consistency pov | 13:26 |
| djszapi | and the fact if there is a package update, I can see what line they added even if rpm | 13:26 |
| X-Fade | djszapi: That does not matter as you can keep packaging separate. | 13:26 |
| djszapi | so that I can reflect to that in debian simpler. | 13:26 |
| djszapi | and this is true for sure vica versa. | 13:26 |
| X-Fade | you can have a debian.tar.gz, dsc and a spec file. | 13:27 |
| djszapi | what human resource sharing do you mean ? | 13:27 |
| djszapi | yes | 13:27 |
| X-Fade | Then you can target both in one project. | 13:27 |
| djszapi | that is the idea | 13:27 |
| X-Fade | Sure, that can work. | 13:27 |
| djszapi | the orig.tar.{bz2/gz} would be shared. | 13:27 |
| X-Fade | yep | 13:27 |
| djszapi | so we can share the suckyness | 13:27 |
| djszapi | if there is some serious issue with updating | 13:28 |
| djszapi | we can share the update suckyness | 13:28 |
| djszapi | and interpret it into debian or rpm | 13:28 |
| *** pinheiro has joined #harmattan | 13:28 | |
| djszapi | instead of going through the same theoritical issue | 13:28 |
| djszapi | that came to my mind previously. | 13:28 |
| djszapi | whereas it might take a little bit of time in the beginning to get in somewhat sync. | 13:29 |
| djszapi | X-Fade: the future stuff again: https://build.pub.meego.com/package/live_build_log?arch=armv7el&package=grantlee&project=home%3Arzr%3Aharmattan&repository=MeeGo_1.2_Harmattan_Maemo.org_MeeGo_1.2_Harmattan_standard | 13:29 |
| X-Fade | Yeah, that is make | 13:29 |
| X-Fade | Will happen on any system locally too. | 13:30 |
| djszapi | X-Fade can I keep UTC in sb, but UTC+2 outside ? | 13:37 |
| djszapi | for the clock, that is. | 13:37 |
| X-Fade | djszapi: yes | 13:37 |
| djszapi | X-Fade hwclock -w after setting the time in sb ? | 13:39 |
| djszapi | to UTC, that is. | 13:39 |
| djszapi | X-Fade also, are you familiar with this spectacle thing ? | 13:48 |
| *** jluisn has quit IRC | 13:58 | |
| *** faenil has quit IRC | 13:58 | |
| djszapi | X-Fade fakeroot date 021612062012 | 14:06 |
| djszapi | date: cannot set date: Operation not permitted | 14:06 |
| djszapi | Thu Feb 16 12:06:00 UTC 2012 | 14:06 |
| *** coffetime has quit IRC | 14:08 | |
| djszapi | ok timezone helped. | 14:24 |
| *** xmlich02 has quit IRC | 14:25 | |
| *** PaulePanter has quit IRC | 14:27 | |
| *** phako has quit IRC | 14:27 | |
| *** PaulePanter has joined #harmattan | 14:28 | |
| *** phako has joined #harmattan | 14:28 | |
| *** DocScrutinizer51 has quit IRC | 14:28 | |
| *** DocScrutinizer51 has joined #harmattan | 14:29 | |
| *** befr0d has joined #harmattan | 14:36 | |
| *** etrunko has joined #harmattan | 14:36 | |
| *** NsaneCoffe has joined #harmattan | 14:48 | |
| NsaneCoffe | 92 mAh idle =) | 14:51 |
| RST38h | too much | 14:51 |
| RST38h | prolly running an RSS pull or something | 14:52 |
| djszapi | X-Fade: why does the apps4meego client handle only one external repository for dependencies ? Where does this limitation come from ? Design or implementation ? | 15:01 |
| X-Fade | djszapi: It does not do any dependencies, I told you that before. | 15:04 |
| X-Fade | djszapi: It is only listing things from ocs. Installing is all done by the OS. | 15:05 |
| *** faenil has joined #harmattan | 15:07 | |
| NsaneCoffe | RST38h, no its messageserver | 15:11 |
| NsaneCoffe | used top command in terminal | 15:12 |
| NsaneCoffe | on top of the list is messageserver with 95% cpu usage | 15:12 |
| NsaneCoffe | Any way i can force it to close | 15:12 |
| NsaneCoffe | kill command wont let me | 15:13 |
| RST38h | yeah, another good candidate | 15:14 |
| RST38h | it won't let you because aegis prevents you from killing useful processes | 15:14 |
| RST38h | Have you tried shutting down messaging accounts from the UI? | 15:14 |
| djszapi | X-Fade why did we discuss aggregated repository then ? | 15:14 |
| X-Fade | djszapi: That is just to have one repo. | 15:15 |
| * RST38h yawns and loudly wonders if X-Fade & Co are planning a repository-neutral app manager UI | 15:15 | |
| RST38h | That would let you add repos a la HAM | 15:15 |
| X-Fade | djszapi: So you can guarantee that everything in there actually installs. | 15:15 |
| X-Fade | RST38h: No plans | 15:16 |
| RST38h | Can the current Apps4Meego client be hacked to allow other repos? | 15:17 |
| djszapi | X-Fade: I am not getting why we mentioned shared (aggregated) repository. | 15:17 |
| djszapi | RST38h: ofc | 15:17 |
| X-Fade | RST38h: It doesn't use repos. | 15:17 |
| djszapi | RST38h: without any hack | 15:17 |
| RST38h | You said it is using ob repos | 15:17 |
| djszapi | read above what X-Fade said :) | 15:17 |
| djszapi | RST38h: no | 15:18 |
| X-Fade | Apps for MeeGo website imports from OBS. | 15:18 |
| djszapi | you can put for instance the Harmattan SDK repository | 15:18 |
| djszapi | which is not obs | 15:18 |
| *** jluisn has joined #harmattan | 15:18 | |
| djszapi | if we are speaking about dependencies | 15:18 |
| X-Fade | The client talks to Apps website over OCS api. | 15:18 |
| RST38h | Oh | 15:18 |
| RST38h | custom web api then | 15:18 |
| djszapi | X-Fade in my understand, please correct me, but we do not need to have aggregated repository. | 15:18 |
| X-Fade | And uses the native backends to do the installing etc. | 15:19 |
| djszapi | X-Fade just proper sources.list on users' stuff. | 15:19 |
| RST38h | djszapi: I think the reasoning he is to force users to use the OBS repo and the web site | 15:19 |
| X-Fade | djszapi: The problem with that is that the repos can conflict with each other. | 15:19 |
| RST38h | s/he/here | 15:19 |
| RST38h | No carrot if pesky users start installing stuff from elsewhere | 15:19 |
| djszapi | X-Fade: yes they can | 15:20 |
| RST38h | + of course it is for users' safety | 15:20 |
| djszapi | RST38h: I tend to disagree | 15:20 |
| RST38h | just like aegis, you know, right? | 15:20 |
| djszapi | especially because that is not why we established our standard. | 15:20 |
| djszapi | this OCS | 15:20 |
| djszapi | it exactly has the opposite purpose | 15:20 |
| djszapi | to talk to any servers | 15:20 |
| djszapi | implementing the same API | 15:20 |
| * RST38h knows of only one standard - debian repository | 15:20 | |
| RST38h | no idea why anyone needs another standard | 15:21 |
| X-Fade | This is for metadata, not the actual install data. | 15:21 |
| *** jluisn has quit IRC | 15:21 | |
| RST38h | you can keep metadata in the same debian repository | 15:21 |
| RST38h | With the apps index holding small metadata (as fields in package sections) | 15:21 |
| X-Fade | Yeah, those 30MB Packages files work so well :) | 15:21 |
| djszapi | RST38h: Nemo does not use meego | 15:22 |
| djszapi | RST38h: nor does Mer. | 15:22 |
| djszapi | *debian | 15:22 |
| RST38h | you mean does not use debian? | 15:22 |
| RST38h | there are rpm repositories | 15:22 |
| RST38h | another platform, another standard, indeed | 15:22 |
| RST38h | X-Fade: Well, remove icon data, refer to it by URL | 15:22 |
| RST38h | X-Fade: And you suddenly truncate your 30MB to something like 5-7MB | 15:23 |
| X-Fade | RST38h: Yeah, the thing is that that is still way too much. | 15:23 |
| RST38h | Do not keep outdated package metadata. Store metadata gzipped, CPUs are now fast enough to gunzip on the fly | 15:23 |
| X-Fade | A browsing app should not have to go over all the dependency stuff etc. | 15:24 |
| RST38h | X-Fade: Ok. How much is too much? (want to make an estimate)' | 15:24 |
| *** jluisn has joined #harmattan | 15:24 | |
| RST38h | Your apt-get goes over all the dependency stuff anyway | 15:24 |
| X-Fade | So it would need to be slimmed down more, at which point you end up with something we are now doing with ocs. | 15:24 |
| RST38h | dpkg too | 15:24 |
| X-Fade | Sure, but the store client doesn't need that info. | 15:24 |
| X-Fade | But as it is one big text file, it needs to go through it anyway. | 15:25 |
| RST38h | But it is there, on the device | 15:25 |
| RST38h | You can index text files | 15:25 |
| RST38h | Should I show you how? | 15:25 |
| X-Fade | Sure, that takes you 2 minutes :) | 15:25 |
| RST38h | When you download a new file, yes, if you do not go into advanced stuff | 15:25 |
| RST38h | If you do, it does not take you 2 mins | 15:26 |
| X-Fade | Servers should do processing, not the client. | 15:26 |
| RST38h | And if servers are down?... | 15:26 |
| X-Fade | then you are out of luck anyway. But you just use the cached data. | 15:26 |
| RST38h | (it is a very practical case BTW, remember Nokia's AGPS server for Diablo disappearing?) | 15:26 |
| X-Fade | RST38h: Yeah, but then you can't install the app anyway. | 15:27 |
| X-Fade | I understand where you are getting at, but wrong example ;) | 15:27 |
| RST38h | well the app is in the repo | 15:27 |
| RST38h | not on the apps4meego server, right? | 15:27 |
| X-Fade | Yeah and comes from repo. | 15:27 |
| RST38h | so if I handle all metadata locally (as I already have it thanks to apt) I should not need apps4meego | 15:27 |
| RST38h | And *here* is what I am coming to, indeed | 15:28 |
| X-Fade | Anyway, RESTfull api should do the same. | 15:28 |
| RST38h | The main reason to do all processing on apps4meego would be to...well...do all processing on apps4meego. | 15:28 |
| *** miroslav has quit IRC | 15:29 | |
| RST38h | Because once you get rid of icons inside metadata, you get something like <=1kB of metadata per package, so 30MB of matadata = 30000 packages | 15:29 |
| *** snowpong has joined #harmattan | 15:30 | |
| djszapi | X-Fade: sorry for the distraction, but we are about to add JSON support to the standard fyi | 15:30 |
| X-Fade | djszapi: For V2, I guess? | 15:30 |
| djszapi | nop | 15:30 |
| djszapi | for both. | 15:30 |
| X-Fade | I'd rather see V2 become stable at some point :) | 15:31 |
| *** japplo has left #harmattan | 15:31 | |
| djszapi | X-Fade: well...we need a maintainer | 15:31 |
| djszapi | which might well be me if Henri cannot take it. | 15:31 |
| djszapi | Frank is about to step down. | 15:31 |
| X-Fade | Hmm ok | 15:32 |
| djszapi | I would personally love Henri taking it | 15:32 |
| djszapi | and I only care about the client qt bits. | 15:32 |
| *** etrunko has quit IRC | 15:34 | |
| *** jluisn has quit IRC | 15:34 | |
| *** etrunko has joined #harmattan | 15:34 | |
| *** jluisn has joined #harmattan | 15:34 | |
| *** zk8 has quit IRC | 15:35 | |
| *** black_tass has left #harmattan | 15:36 | |
| *** lizardo has joined #harmattan | 15:36 | |
| djszapi | X-Fade do you have an example for merged deb/rpm repository ? | 15:37 |
| *** psycho_oreos has quit IRC | 15:37 | |
| *** xarcass has quit IRC | 15:38 | |
| *** jluisn has quit IRC | 15:39 | |
| djszapi | RST38h: if you would do what you say you would lose the advantage of the API | 15:41 |
| djszapi | did* | 15:41 |
| djszapi | and you would need to ahck differends together to get it done as it nicely work on its own. | 15:42 |
| djszapi | hack different standards* | 15:42 |
| *** jluisn has joined #harmattan | 15:44 | |
| X-Fade | djszapi: Setting up the project is easy, just add multiple repositories. | 15:45 |
| X-Fade | djszapi: Then in the package you add .dsc and debian.tar.gz and spec file. | 15:46 |
| djszapi | X-Fade: I would probably merge the Harmattan packages into the PA repository. | 15:47 |
| X-Fade | djszapi: I'd suggest you test it in your home project or so. | 15:47 |
| X-Fade | djszapi: But sure, you can do it there too ;) | 15:47 |
| *** jluisn has quit IRC | 15:49 | |
| djszapi | X-Fade: is it possible to just rename a stuff, like s/CE:MW:PlasmaActive/CE:MW:KDE/ ? | 15:49 |
| djszapi | X-Fade: Although, speaking of which, we would kinda depend on each other with orig.tar.{bz2/gz} updated, which might not be a cool thing. | 15:50 |
| X-Fade | I don't think renaming is possible. | 15:51 |
| djszapi | so it might introduce a testing repository concept. | 15:51 |
| X-Fade | Yeah and a maintainer.. | 15:51 |
| *** Venemo_N950 has joined #harmattan | 15:51 | |
| djszapi | maintainer of what ? | 15:51 |
| Venemo_N950 | hey | 15:51 |
| X-Fade | Not that just everybody updates whenever they want :) | 15:51 |
| Venemo_N950 | Jaffa, ping | 15:52 |
| djszapi | X-Fade: mmm I dislike this bureaucracy personally. | 15:52 |
| djszapi | that is what made Surround die. | 15:52 |
| djszapi | and that is why our community repository got a little bit of a success | 15:52 |
| X-Fade | djszapi: You noticed yourself what happened when someone decided to update openssl ;) | 15:53 |
| X-Fade | djszapi: It is a pain, that needs communication. | 15:53 |
| djszapi | definitely. | 15:53 |
| djszapi | but as far as it is not a common case, just in every blue moon, I think it is more overhead to communicate everything in this special case. | 15:54 |
| X-Fade | So that is why project like Mer etc, do automated checks and then need a review before a SR is accepted. | 15:54 |
| X-Fade | You can have multiple reviewers/maintainers though | 15:54 |
| djszapi | they did not make 500 packages that fast as we did regarding the available human resource. | 15:55 |
| djszapi | yes, but that is the thing | 15:55 |
| djszapi | nobody likes packaging; we are all developers; it is the neccesary evil for us ;P | 15:55 |
| X-Fade | But maintaining is more than just dump some packages in a repo. | 15:55 |
| *** jluisn has joined #harmattan | 15:55 | |
| X-Fade | And that is why maintainer is a whole different job :) | 15:55 |
| Jaffa | Venemo_N950: Pong | 15:56 |
| djszapi | X-Fade: the problem is that mobile stuff are changing packaging too fast | 15:57 |
| djszapi | I spent 2 years with debian packaging | 15:57 |
| djszapi | and I would need to relearn everything for rpm | 15:57 |
| RST38h | djszapi: right. because I might not need the api in the first place. | 15:58 |
| *** jreznik has quit IRC | 15:58 | |
| X-Fade | djszapi: if you have a proper yaml, the rpms are easy. | 15:59 |
| *** ab has quit IRC | 15:59 | |
| djszapi | X-Fade you mean spectacle ? | 16:01 |
| X-Fade | djszapi: An example: https://build.pub.meego.com/package/view_file?file=apps-client.yaml&package=apps-client&project=CE%3AApps&srcmd5=d510aa31a84b0f6bf73c9361fad35577 | 16:01 |
| X-Fade | djszapi: Yeah. | 16:01 |
| djszapi | it is just damn that | 16:01 |
| djszapi | I spent a decent in my life with debian stuff | 16:02 |
| djszapi | and I need to re-learn everything | 16:02 |
| djszapi | I am not happy about it | 16:02 |
| * djszapi has never understood the deb->rpm or rpm->deb like changes | 16:02 | |
| djszapi | it is just way too big change. | 16:02 |
| X-Fade | djszapi: Well, it is basically the same. | 16:02 |
| X-Fade | The app doesn't change, so you only need a few lines of packaging changes. It isn't that complicated. | 16:03 |
| djszapi | I need to drop all my work | 16:03 |
| djszapi | for zillion packages, like 300 | 16:03 |
| X-Fade | Check the example, you can copy/paste and adapt that easily. | 16:03 |
| djszapi | rm -rf */debian | 16:03 |
| djszapi | and set up 300 rpm stuff | 16:03 |
| djszapi | even if I know nothing about rpm | 16:03 |
| djszapi | it is a very huge mentality, area and everything change. | 16:04 |
| X-Fade | Or write a tool to create yaml from /debian dir and you are a hero :) | 16:04 |
| djszapi | especially because rpm sucks ;) | 16:04 |
| djszapi | not really | 16:04 |
| djszapi | that is not gonna work since they are just ... different | 16:04 |
| djszapi | I do not think it can be automated. | 16:04 |
| X-Fade | It can be automated to to most of the work. | 16:04 |
| X-Fade | But then you need to tune some things. | 16:04 |
| djszapi | it is like JSON or XML if you make a simple s tandard output for them as we do in OCS | 16:05 |
| djszapi | they are just not interchangable on purpose. | 16:05 |
| X-Fade | But a lot of data you can just copy over, field names are just different. | 16:05 |
| djszapi | X-Fade: so you are saying the rules file has a similar concept in rpm ? | 16:05 |
| djszapi | say, "dh" way ? | 16:05 |
| djszapi | or source for mat ... | 16:06 |
| djszapi | or the control entries. | 16:06 |
| djszapi | it is a whole bunch of distinct stuff imho | 16:06 |
| X-Fade | You don't need a lot of that stuff in rpm. | 16:06 |
| X-Fade | Check my yaml | 16:06 |
| X-Fade | That is all that is needed for our Qt/Qml app to be built and packaged. | 16:07 |
| *** hardaker has joined #harmattan | 16:07 | |
| djszapi | I do not see any binary compatibility consideration it | 16:07 |
| djszapi | in it* | 16:07 |
| djszapi | and that is prolly the most powerful side of the debian packaging system | 16:07 |
| X-Fade | How do you mean? | 16:07 |
| X-Fade | >= version X etc ? | 16:08 |
| djszapi | I am referring to the symbols file. | 16:08 |
| djszapi | and the control file management wrt this | 16:08 |
| *** jluisn has quit IRC | 16:09 | |
| X-Fade | Have not needed that yet, that you need to ask others :) | 16:09 |
| *** jluisn has joined #harmattan | 16:09 | |
| *** xmlich02 has joined #harmattan | 16:09 | |
| djszapi | anybody having an idea why "dpkg-buildpackage -rfakeroot" build the package, but "dpkg-buildpackage -rfakeroot | tee buildlog" fails at the quilt stage ? | 16:10 |
| djszapi | http://paste.kde.org/423482/ | 16:10 |
| *** adlan has joined #harmattan | 16:13 | |
| djszapi | XFade ^ | 16:16 |
| djszapi | meh, X-Fade | 16:16 |
| *** Anssi138 has joined #harmattan | 16:18 | |
| *** b00^wk has joined #harmattan | 16:19 | |
| b00^wk | hello | 16:19 |
| *** jluisn has quit IRC | 16:19 | |
| b00^wk | anyone knows here how I can if at all configure the Wifi Hotspot start WITHOUT it requesting me to connect to a WLAN ? | 16:19 |
| b00^wk | it does not run without a running 3,3g connection | 16:20 |
| *** jluisn has joined #harmattan | 16:20 | |
| djszapi | Could you please rephrase the question ? I am not getting it. | 16:20 |
| b00^wk | Ok, here is this app in my Nokia, called Wifi Hotspot | 16:21 |
| djszapi | similar to joikuspot ? | 16:21 |
| djszapi | or is that your app ? | 16:21 |
| b00^wk | i want to use it just for that - a wifi ap, so clients can connect and exchange data | 16:21 |
| b00^wk | without the bloody wlan / 3g running | 16:21 |
| b00^wk | and all of them downloading stuff off the net | 16:21 |
| djszapi | joikuspot for instance does not run with 3g, if you do not want it to. | 16:22 |
| b00^wk | but, this app does not start unless there is an established WLAN, which is the 3g in my case | 16:22 |
| djszapi | but please clarify: are you the author of joikuspot or do you have a similar app ? | 16:22 |
| b00^wk | djszapi, i see | 16:22 |
| *** Hamidx9 has joined #harmattan | 16:22 | |
| b00^wk | djszapi, as you asking me ? | 16:22 |
| b00^wk | no , i've never seen that app | 16:22 |
| djszapi | well, the hotspot needs to have internet connection somehow.... | 16:23 |
| djszapi | let it either 3g or wifi.... | 16:23 |
| b00^wk | for what ? | 16:23 |
| djszapi | for accessing the internet from clients. | 16:23 |
| b00^wk | why can't i just setup a software ap ? | 16:23 |
| b00^wk | no, i don't want any internet | 16:23 |
| b00^wk | just an access point | 16:23 |
| djszapi | well it is a phone, not a router after all :) | 16:23 |
| b00^wk | whats the difference when my clients can still talk to each other ? | 16:24 |
| b00^wk | its an unneccessary restriction | 16:24 |
| b00^wk | that I have to have a wlan running to do this | 16:24 |
| djszapi | I do not think that app was meant for a router replacement. | 16:24 |
| b00^wk | but does not the app rely on the kernel driver to actually do the software access point ? | 16:25 |
| b00^wk | i think the app's work here is minimal in that regard | 16:25 |
| Jaffa | djszapi: He's asking for an app which willprovideDHCP on an ad-hoc WLAN | 16:25 |
| djszapi | b00^wk it does | 16:25 |
| djszapi | and that is why people cannot customize it. | 16:26 |
| djszapi | because the kernel is frozen to externals. | 16:26 |
| djszapi | but I am not in an intimidate relation with the kernel driver honestly :) | 16:26 |
| Jaffa | b00^wk: If you configure each client with an ad-hoc WLAN with the same details, it should Just Work. You do not need a router-type device. | 16:26 |
| b00^wk | i still dont' understand where is that in kernel restriction to START wlan to do this | 16:26 |
| b00^wk | Jaffa, problem is, the Hotspot does not start otherwise :( | 16:26 |
| b00^wk | and, so none of the clients can connect | 16:27 |
| djszapi | b00^wk: have you looked at the kernel code ? | 16:27 |
| Jaffa | b00^wk: I'm saying do not use hots[ot | 16:27 |
| djszapi | the relevant driver, I mean. | 16:27 |
| b00^wk | djszapi, i beleive this is the texas instruments wlan adapter, and, | 16:27 |
| b00^wk | the iwl whatever driver ? | 16:27 |
| Jaffa | b00^wk: Youonlyneed it when wanting to roite the other devices to a3G connection. | 16:27 |
| djszapi | b00^wk: I think so. | 16:27 |
| Jaffa | b00^wk: you want the very definitipn of an ad-hoc WLAN | 16:28 |
| *** Hamidx9 has quit IRC | 16:28 | |
| b00^wk | ok, without any definitions, if you can connect for me two clients to the phone, without that nasty app running | 16:28 |
| b00^wk | then i don't care what definition of yoru wlan is | 16:29 |
| djszapi | =) | 16:29 |
| *** jluisn has quit IRC | 16:29 | |
| *** mece has quit IRC | 16:29 | |
| *** adlan has quit IRC | 16:29 | |
| *** jluisn has joined #harmattan | 16:29 | |
| *** briglia has joined #harmattan | 16:30 | |
| *** Free-MG has joined #harmattan | 16:30 | |
| *** adlan has joined #harmattan | 16:32 | |
| *** divan has joined #harmattan | 16:33 | |
| *** ab has joined #harmattan | 16:35 | |
| *** ab has quit IRC | 16:35 | |
| *** ab has joined #harmattan | 16:35 | |
| *** divan1 has quit IRC | 16:36 | |
| *** jluisn has quit IRC | 16:38 | |
| *** jluisn has joined #harmattan | 16:38 | |
| *** rm_work has joined #harmattan | 16:47 | |
| *** rm_work has joined #harmattan | 16:47 | |
| *** Venemo_N950 has quit IRC | 16:50 | |
| *** aloof_jezzster has joined #harmattan | 16:56 | |
| djszapi | X-Fade ping | 16:59 |
| djszapi | X-Fade have you seen this new QBS build system ? | 16:59 |
| *** Jeffrey04 has joined #harmattan | 17:00 | |
| *** nebulon has joined #harmattan | 17:01 | |
| *** gri is now known as zz_gri | 17:01 | |
| *** aloof_jezzster has quit IRC | 17:06 | |
| *** zz_gri is now known as gri | 17:08 | |
| djszapi | gri: the icon and plugin thingy might be fixed for PR 1.3 :D | 17:16 |
| gri | djszapi: Will this ever be released? :D | 17:16 |
| djszapi | how should I know ? :D | 17:17 |
| gri | Aren't you a fortune teller? | 17:19 |
| Jeffrey04 | lol | 17:19 |
| djszapi | gri: I am, but the crystal is at home :) | 17:20 |
| Jeffrey04 | dev an app for that then :) | 17:20 |
| Jeffrey04 | tot 1.3 will be out anyway? https://twitter.com/#!/melsop74/status/159161735549427712 | 17:20 |
| Jeffrey04 | unless he is not telling the truth lol | 17:20 |
| gri | wow, I just found a cool shortcut in QtCreator: holding ctrl and shift lets you move your current line up/down with the cursor keys :) | 17:23 |
| *** jabis has quit IRC | 17:25 | |
| djszapi | gri: I even knew this and I really rarely use qtcreator... | 17:26 |
| *** jabis has joined #harmattan | 17:26 | |
| gri | since you're a fortune teller that does not count | 17:28 |
| djszapi | I think this option is even in the first video intro really :) | 17:29 |
| djszapi | it is just .. such a basic stuff. | 17:29 |
| gri | yeah, stuff I never looked or searched for | 17:30 |
| gri | I just happened to find it by accident | 17:30 |
| djszapi | gri: you can soon try out the plasma components for Harmattan if you wanna to. | 17:33 |
| djszapi | not that it makes too much sense, but it at least can be experimented with. | 17:33 |
| gri | is there anything special about them? | 17:34 |
| djszapi | well, it is a brand new component set | 17:35 |
| djszapi | it is like asking what the difference between oxygen icons or other icon set, dunno, taste probably. | 17:35 |
| djszapi | this is what used on Plasma Active. | 17:36 |
| *** jluisn has quit IRC | 17:37 | |
| *** jluisn has joined #harmattan | 17:37 | |
| djszapi | gri: I think I will now implement soon the main page for my game with Play/Resume and the Help labels | 17:40 |
| DocScrutinizer | http://share.ovi.com/album/joerg900.unboxingGTA04 | 17:42 |
| *** longthen1 has quit IRC | 17:43 | |
| *** longthen has joined #harmattan | 17:44 | |
| *** gabriel9|work has joined #harmattan | 17:53 | |
| djszapi | X-Fade how could I trigger -j5 or even more for kdelibs build on c-obs ? | 17:54 |
| *** Jeffrey04 has left #harmattan | 17:54 | |
| djszapi | X-Fade parallel option is already in the rules file, but it is still slow. | 17:54 |
| *** gabriel9 has quit IRC | 17:57 | |
| *** risca has joined #harmattan | 17:58 | |
| *** Hamidx9 has joined #harmattan | 17:59 | |
| *** jluisn has quit IRC | 17:59 | |
| *** jluisn has joined #harmattan | 18:00 | |
| gri | djszapi: http://dl.dropbox.com/u/4691970/2012-02-16_16-56-27.png is the icon on the top right self-speaking or should I write "update available" or something like that next to it (left or below)? | 18:01 |
| djszapi | gri: check the updater | 18:03 |
| djszapi | to replicate that if you wanna well integrate it | 18:03 |
| gri | they only show the icon - then I won't assume the user is too stupid to guess what it means | 18:07 |
| djszapi | well they spam the feed | 18:09 |
| djszapi | so it is all okay to do it that way | 18:09 |
| *** faenil has quit IRC | 18:12 | |
| *** benares_98 has joined #harmattan | 18:12 | |
| *** Hamid_ has joined #harmattan | 18:15 | |
| *** Hamid_ is now known as Guest47645 | 18:15 | |
| *** faenil has joined #harmattan | 18:15 | |
| *** Hamidx9 has quit IRC | 18:17 | |
| *** zk8 has joined #harmattan | 18:18 | |
| *** jluisn has quit IRC | 18:20 | |
| *** jluisn has joined #harmattan | 18:20 | |
| *** Guest47645 has quit IRC | 18:22 | |
| *** aloof_jezzster has joined #harmattan | 18:22 | |
| djszapi | gri: what do you think about usability POV for the help option ? Shall I pop up a Dialog or just push a help page with back toolicon ? | 18:24 |
| gri | djszapi: I'd love bubble popups ... I think my N9 showed such on the first start but I don't know for sure | 18:26 |
| gri | the system always does the page thingy with back button elsewhere | 18:27 |
| djszapi | I am not in favor of popup dialogs... | 18:28 |
| djszapi | gri: to be honest I am not sure what certain components are in Extra, like CountBubble or Tumbler | 18:30 |
| *** xTs has joined #harmattan | 18:30 | |
| itsnotabigtruck | i believe a tumbler is a digit selector you can scroll up and down through | 18:31 |
| itsnotabigtruck | mtimer has tumblers | 18:31 |
| djszapi | screenshot ? | 18:31 |
| djszapi | I need a horizontal slider for the alphabet | 18:31 |
| djszapi | wonder whether I could use it for that. | 18:31 |
| djszapi | uh-huh, I found a screenshot, cool | 18:33 |
| djszapi | well, it is useless for my goal :) | 18:33 |
| djszapi | I need a horizontal listview simply after all | 18:33 |
| djszapi | gri: the tour what you are referring to is too much work :D | 18:36 |
| djszapi | I admit, it is fancy, but come on, it is just an anagram game :) | 18:36 |
| itsnotabigtruck | oh yeah, speaking of that, time to load up that deb | 18:37 |
| djszapi | itsnotabigtruck: can you wait few minutes ? | 18:37 |
| djszapi | and you could help a mainpage with help/play options | 18:37 |
| djszapi | so that the game does not start automatically after the launch | 18:38 |
| djszapi | and gri complained it is not intuitive | 18:38 |
| *** ab has quit IRC | 18:38 | |
| djszapi | so I would add a help, too | 18:38 |
| itsnotabigtruck | i still don't understand why kanagram could possibly need a postinst script that restarts pulseaudio | 18:39 |
| *** zk8 has quit IRC | 18:40 | |
| djszapi | itsnotabigtruck: because it is install a pulse config file ? | 18:41 |
| djszapi | installs* | 18:42 |
| *** eean has quit IRC | 18:43 | |
| itsnotabigtruck | djszapi: it does? i'm not finding that anywhere in the package... | 18:45 |
| djszapi | gri: actually help sucks :) | 18:45 |
| djszapi | people do not read help, so I should rather make it intuitive by default, but not sure how | 18:45 |
| djszapi | gri: what would make you guess you can click on the letter ? http://minus.com/mAdl0bZqR#1 | 18:45 |
| itsnotabigtruck | oh there it is, ./opt/kanagram/share/policy/etc/pulse/xpolicy.conf.d/kanagram.conf | 18:46 |
| djszapi | itsnotabigtruck: I pasted the policy framework documentation last time | 18:46 |
| djszapi | please read it for details. | 18:46 |
| djszapi | I could just well restart it with pactl and not full pulseaudio restart | 18:46 |
| *** torarne has joined #harmattan | 18:46 | |
| *** gabriel9 has joined #harmattan | 18:46 | |
| torarne | javispedro: ping | 18:46 |
| djszapi | but I would need to make a hackery to identify the process | 18:47 |
| torarne | oh, not here | 18:47 |
| djszapi | but I did not have time, and I am not sure it is worth it | 18:47 |
| *** azbest_ has joined #harmattan | 18:47 | |
| djszapi | what is wrong about restarting pulse ? | 18:47 |
| *** gabriel9|work has quit IRC | 18:48 | |
| itsnotabigtruck | djszapi: i'm reading up on this...are you sure you're putting the file in the right place | 18:48 |
| itsnotabigtruck | angry birds has its pulseaudio file at /usr/share/policy/etc/pulse/xpolicy.conf.d/ | 18:48 |
| *** gabriel9 has quit IRC | 18:48 | |
| *** aloof_jezzster has quit IRC | 18:49 | |
| djszapi | itsnotabigtruck: no, I am putting it into the wrong place. | 18:49 |
| *** ab has joined #harmattan | 18:51 | |
| *** ab has joined #harmattan | 18:51 | |
| itsnotabigtruck | also it looks like the ./opt/kanagram/include/ dir ought to be dropped from the package | 18:51 |
| *** aloof_jezzster has joined #harmattan | 18:52 | |
| djszapi | disagree | 18:52 |
| djszapi | the plasma components based standalone will pick up the header from there. | 18:52 |
| djszapi | the engine, that is. | 18:52 |
| itsnotabigtruck | hrm? what good would c headers for a game be on the phone itself | 18:53 |
| djszapi | oh you meant include, no lib | 18:54 |
| djszapi | sure. | 18:54 |
| DocScrutinizer | javispedro left #maemo (didn't contribute much) in favour of #harmattan and his struggle to fight aegis introduced limitations. It seems he got fed up with this as well now, judging by the frequency he shows up here. So: weldone Nokia - another brilliant devel distracted from work for fremantle by this abomination, then finally seems to have turned away completely from Nokia-based device/OS "ecosystem" | 18:54 |
| djszapi | even I should drop the oggs | 18:54 |
| djszapi | but I did not care about such unproductive stuff so far | 18:54 |
| itsnotabigtruck | also djszapi: did you ever try chmod +x'ing your launcher script before the build | 18:55 |
| itsnotabigtruck | i'd expect the package builder would keep the permissions and you wouldn't need to mess with that in postinst | 18:55 |
| djszapi | yes we all tried, at least 10-20 people so far | 18:55 |
| djszapi | just on this channel | 18:55 |
| djszapi | hah, no | 18:55 |
| djszapi | otherwise we would not obviously do this | 18:55 |
| djszapi | if it had worked oob, why would we have touched this ? | 18:56 |
| *** lmoura has joined #harmattan | 18:56 | |
| * itsnotabigtruck shrugs | 18:56 | |
| torarne | DocScrutinizer: i see :/ | 18:56 |
| itsnotabigtruck | there has to be a way | 18:56 |
| djszapi | you mentioned so far things I would never really like to bother with | 18:57 |
| djszapi | except pulse. | 18:57 |
| djszapi | especially the include stuff | 18:57 |
| djszapi | and the chmod | 18:57 |
| djszapi | chmod is needed, include would cause own install file definition, and manual maintenance of the files. | 18:57 |
| itsnotabigtruck | speaking of aegis - djszapi: maybe add an <aegis><request><for path="x" /></request></aegis> | 18:57 |
| djszapi | I am so happy with the current way, I have no install file. | 18:58 |
| djszapi | why ? | 18:58 |
| djszapi | I do not need any special creds | 18:58 |
| itsnotabigtruck | right, that would drop all creds, precisely because of that | 18:58 |
| itsnotabigtruck | by changing from add behavior to set behavior | 18:58 |
| djszapi | well I implemented that bit in the platform really :) | 18:59 |
| djszapi | I do not need this at all. | 18:59 |
| *** jluisn has quit IRC | 19:02 | |
| *** jluisn has joined #harmattan | 19:02 | |
| torarne | is setcap available in a pacakge on harmattan? | 19:08 |
| djszapi | setcap or setfcap ? | 19:09 |
| torarne | setcap (i think) | 19:10 |
| torarne | actually the issue im trying to solve is sudo not working as user on an open-mode flashed device | 19:10 |
| djszapi | I do not see setcap anywhere. | 19:10 |
| djszapi | where do you see it ? | 19:10 |
| djszapi | I see only setfcap | 19:10 |
| djszapi | and that is available forf ovi apps | 19:10 |
| *** aloof_jezzster has quit IRC | 19:10 | |
| torarne | i get "unable to change to sudoers gid: Operation not permitted", which i assume is because i don't have the cap to become uid 0 | 19:10 |
| djszapi | why would you do that in the first place ? | 19:11 |
| torarne | so i can add 'mount /foo/bar' to sudoers without a password | 19:12 |
| torarne | without opening up every other command do run as root | 19:12 |
| torarne | the alternative is to use something like opensh, which just gives you root without even requiring a password | 19:13 |
| djszapi | I do not really understand where you got this setcap from | 19:13 |
| *** jluisn has quit IRC | 19:13 | |
| djszapi | I only see setfcap | 19:13 |
| djszapi | I opened up the /usr/include/linux/capability.h file | 19:13 |
| djszapi | no mention about setcap | 19:13 |
| *** jluisn has joined #harmattan | 19:13 | |
| *** NIN101 has joined #harmattan | 19:13 | |
| *** arcean has joined #harmattan | 19:13 | |
| torarne | http://linux.die.net/man/8/setcap | 19:14 |
| *** heymaster has quit IRC | 19:14 | |
| djszapi | but you can do anything in open mode anyway | 19:14 |
| djszapi | setcap is not a credential | 19:14 |
| djszapi | it is an app | 19:14 |
| djszapi | so what you actually need to ask is setfcap | 19:14 |
| djszapi | and as I said, that is available. | 19:14 |
| djszapi | for Ovi packages, not for user packages. | 19:14 |
| djszapi | request CAP::setfcap | 19:15 |
| djszapi | though I would not really do it this way | 19:15 |
| torarne | so… not availble using apt-get on device? | 19:15 |
| djszapi | ohh you are interested in a package. | 19:15 |
| torarne | yepp :) | 19:15 |
| itsnotabigtruck | what's the significance to the type subcode (/device/hw-version) | 19:15 |
| torarne | i don't like doing echo 'rootme' | devel-su root -c "foo" in my scripts | 19:15 |
| torarne | nor do i want to use opensh -c 'foo' (which does not ask for a password) | 19:16 |
| *** heymaster has joined #harmattan | 19:16 | |
| itsnotabigtruck | torarne: put your stuff in a package, assert the needed caps | 19:16 |
| djszapi | torarne: if it is not in the community repository, you are out of the luck | 19:16 |
| torarne | itsnotabigtruck: this is not for a ovi store thing, this is just for script to set up a device for development | 19:17 |
| djszapi | you need to port libcap | 19:17 |
| torarne | djszapi: i see, hmm | 19:17 |
| djszapi | but you do not need to bother with this if you have RDC | 19:17 |
| itsnotabigtruck | djszapi: it's called d e v e l o p e r m o d e | 19:17 |
| torarne | djszapi: im trying to make a setup for people without rdc, but with open mode | 19:17 |
| itsnotabigtruck | there's nothing you can do from an ovi app that you can't do from an unsigned app, with devmode on | 19:18 |
| torarne | https://trac.webkit.org/wiki/SettingUpDevelopmentEnvironmentForN9#a2.ConfiguringyourdeviceN9N950 | 19:18 |
| djszapi | itsnotabigtruck: you are not getting the issue | 19:18 |
| djszapi | itsnotabigtruck: there is *NO* available libcap | 19:18 |
| djszapi | so he can *NOT* use setcap | 19:19 |
| djszapi | since it is *NOT* presented. | 19:19 |
| torarne | https://github.com/resworb/scripts/blob/master/setup-device.sh#L101 | 19:19 |
| itsnotabigtruck | well, i did miss the part about this being on an openmoded phone already | 19:19 |
| torarne | see that ugly echo rootme | devel-su | 19:19 |
| *** ab has quit IRC | 19:19 | |
| torarne | i dont' want to do that, but still want the user to be able to mount | 19:19 |
| torarne | which would be solvable with sudo | 19:19 |
| djszapi | torarne: you cannot mount with stock kernel | 19:19 |
| torarne | this is open mode, patched | 19:20 |
| itsnotabigtruck | torarne: afaik doc's "open mode kernel" just nulls out all aegis permission checks | 19:20 |
| itsnotabigtruck | that means that aegis permission grants still work | 19:20 |
| itsnotabigtruck | otherwise loads of stuff would break | 19:20 |
| *** lmoura has quit IRC | 19:20 | |
| djszapi | torarne: you can do everything then if you wish to | 19:20 |
| itsnotabigtruck | so put your thing in a package, then add an _aegis manifest to it | 19:20 |
| itsnotabigtruck | that asserts UID::root | 19:20 |
| itsnotabigtruck | that basically makes your thing setuid root | 19:20 |
| djszapi | torarne: try AEGIS_FIXED_ORIGIN | 19:20 |
| *** benares_98 has quit IRC | 19:21 | |
| itsnotabigtruck | however, are you sure you need root access | 19:21 |
| djszapi | itsnotabigtruck: that is exactly what he wanted to avoid | 19:21 |
| itsnotabigtruck | mounting requires CAP::sys_admin, not uid = 0 | 19:21 |
| torarne | itsnotabigtruck: yes, it nulls out all checks, but look at the last line http://maemo.cloud-7.de/950/openmode_kernel_PR1.1/enable-root-superuser.diff | 19:21 |
| itsnotabigtruck | what happens if you aegis-exec on an openmoded phone | 19:22 |
| torarne | itsnotabigtruck: im trying to get mounting working by having the command in the sudoers file | 19:22 |
| torarne | its doing /* aegis hack: always success for root */ | 19:22 |
| itsnotabigtruck | aegis-exec -a CAP::sys_admin "$0" $@ | 19:22 |
| itsnotabigtruck | try that | 19:22 |
| torarne | cool | 19:22 |
| *** risca has quit IRC | 19:23 | |
| torarne | itsnotabigtruck: fails :/ | 19:24 |
| *** jluisn has quit IRC | 19:25 | |
| *** jluisn has joined #harmattan | 19:25 | |
| torarne | oh, wait | 19:27 |
| torarne | now i get fuse: failed to open /dev/fuse: Permission denied | 19:28 |
| torarne | so i think it's working, somewhat, but i need more permissions | 19:28 |
| djszapi | you can always add tcb | 19:29 |
| torarne | tcb? | 19:29 |
| djszapi | though it sounds a very insecure system to me. | 19:29 |
| torarne | no worse than echo 'rootme' | devel-su :) | 19:29 |
| djszapi | much worse from sec pov. | 19:30 |
| djszapi | but if you would like to avoid the accumulating of the creds, just use tcb | 19:30 |
| djszapi | and you can be truly root. | 19:30 |
| torarne | what is tcb? | 19:30 |
| itsnotabigtruck | torarne: the keys to the castle | 19:33 |
| itsnotabigtruck | it sounds like you need to be uid=0 to access /dev/fuse or something | 19:34 |
| *** jluisn has quit IRC | 19:35 | |
| torarne | chgrp developer /dev/fuse | 19:36 |
| *** jluisn has joined #harmattan | 19:36 | |
| itsnotabigtruck | crw-rw---- 1 root root 10, 229 Dec 31 1969 /dev/fuse | 19:36 |
| torarne | one step closer | 19:36 |
| itsnotabigtruck | lol | 19:36 |
| torarne | nope | 19:37 |
| torarne | [pid 5451] mount("torarne@192.168.2.14:/", ".", "fuse", MS_NOSUID|MS_NODEV, "max_read=65536,fd=5,rootmode=400"...) = -1 EPERM (Operation not permitted) | 19:37 |
| torarne | [pid 5451] write(2, "fusermount: mount failed: Operat"..., 50fusermount: mount failed: Operation not permitted | 19:37 |
| torarne | ) | 19:37 |
| djszapi | torarne: why don't you try tcb ? | 19:37 |
| djszapi | your system is completely insecure anyway | 19:37 |
| torarne | i still don't know what it is :) | 19:37 |
| djszapi | s/CAP::sys_admin/tcb/ | 19:37 |
| RST38h | Ah, Apple finally starts disabling unathorized application development in MacOS | 19:38 |
| torarne | hmm, the previous lines was with strace -f aegis-exec -a CAP::sys_admin -a CAP::sys_admin sshfs torarne@192.168.2.14:/ /mnt/torarne | 19:38 |
| *** azbest_ has quit IRC | 19:38 | |
| RST38h | The end is, indeed, near now. | 19:38 |
| torarne | so i already have that cap :/ | 19:38 |
| itsnotabigtruck | RST38h: link? | 19:39 |
| djszapi | torarne: why don't you request "tcb" ? | 19:40 |
| djszapi | what is wrong about that ? | 19:40 |
| djszapi | as I said, your system has zero security :) | 19:40 |
| djszapi | it is not any hole. | 19:40 |
| itsnotabigtruck | djszapi: because there's no need... | 19:40 |
| djszapi | and probably less trouble workflow-wise. | 19:40 |
| djszapi | itsnotabigtruck: of course there is | 19:41 |
| torarne | djszapi: am i not doing that when runnning aegis-exec -a CAP::sys_admin -a CAP::sys_admin sshfs torarne@192.168.2.14:/ /mnt/torarne ? | 19:41 |
| djszapi | since it does not work for some reason | 19:41 |
| itsnotabigtruck | if you're asserting tcb through the package just to aegis-exec stuff | 19:41 |
| itsnotabigtruck | you might as well assert the capabilities you actually need | 19:41 |
| djszapi | itsnotabigtruck: huh ? | 19:41 |
| itsnotabigtruck | torarne: do aegis-exec -a CAP::sys_admin accli -I | 19:41 |
| djszapi | who said it should be done in the package for testing ? | 19:41 |
| djszapi | sorry, it is too much noise to me, I am off. | 19:41 |
| itsnotabigtruck | see if CAP::sys_admin is actually in the list | 19:42 |
| torarne | itsnotabigtruck: nope, it's not :/ | 19:42 |
| itsnotabigtruck | there's the answer then :/ | 19:42 |
| itsnotabigtruck | well, for now, test with | 19:42 |
| torarne | :) | 19:42 |
| itsnotabigtruck | opensh -c 'aegis-exec -c -a CAP::sys_admin mount ...' | 19:42 |
| torarne | ah, yes, that works :) | 19:43 |
| torarne | both opensh and the hacky echo 'rootme' | devel-su works | 19:43 |
| *** jluisn has quit IRC | 19:43 | |
| djszapi | indeed... | 19:43 |
| torarne | im curious why sudo doesnt, thats all :) | 19:43 |
| djszapi | my point too from the beginning. | 19:43 |
| itsnotabigtruck | i don't understand why the devel-su version works...i guess on open mode devel-su grants all caps | 19:43 |
| *** jluisn has joined #harmattan | 19:44 | |
| djszapi | it is very simple really why it works. | 19:44 |
| torarne | itsnotabigtruck: yes, it's a patched kernel that indeed grants all caps for root | 19:44 |
| itsnotabigtruck | torarne: how is this script being invoked | 19:44 |
| itsnotabigtruck | why don't you just put | 19:44 |
| itsnotabigtruck | if ! /usr/bin/accli -t CAP::sys_admin; then; echo 'sys_admin capability is required'; exit 1; fi | 19:45 |
| itsnotabigtruck | and request that users invoke it as opensh -c 'sh foo.sh' | 19:45 |
| *** gri is now known as zz_gri | 19:45 | |
| itsnotabigtruck | oh wait, this runs from the pc side | 19:46 |
| itsnotabigtruck | weird | 19:46 |
| djszapi | torarne: what is your project about, if I may ask ? | 19:46 |
| torarne | just doing qt5 and webkit development with the n9/n950 as a target | 19:47 |
| torarne | hacking up some scripts to bootstrap that | 19:47 |
| torarne | and the less hacks (regular sudo vs opensh or echo 'rootme') the better | 19:48 |
| djszapi | torarne: maybe you could assign tcb universally to the root user :D | 19:48 |
| itsnotabigtruck | ok, now i see why using a package isn't so convenient | 19:48 |
| djszapi | and you would never really need to worry. :) | 19:48 |
| itsnotabigtruck | torarne: just prompt the user for their root password | 19:49 |
| torarne | yeah, i think opensh is probably even easier | 19:49 |
| torarne | it doesnt care what the pwd is | 19:49 |
| *** jluisn has quit IRC | 19:49 | |
| *** jluisn has joined #harmattan | 19:50 | |
| djszapi | torarne: if I were you I would try to assign tcb to root, and I would not even need opensh and other hacks. | 19:50 |
| djszapi | everything would work as on a linux desktop machine. | 19:50 |
| djszapi | otherwise you always need to hack here and there. | 19:50 |
| *** azbest_ has joined #harmattan | 19:50 | |
| itsnotabigtruck | djszapi: i don't think you can 'assign' privileges to users... | 19:51 |
| itsnotabigtruck | there's various ways you could arrange it to assert tcb when you su to root | 19:51 |
| itsnotabigtruck | that's really the problem with aegis | 19:52 |
| itsnotabigtruck | it has no concept whatsoever of users and roles | 19:52 |
| djszapi | itsnotabigtruck: of course you can | 19:52 |
| itsnotabigtruck | which every real security model is based on | 19:52 |
| djszapi | since we decided what to assign | 19:53 |
| djszapi | but surely, you know better :) | 19:53 |
| itsnotabigtruck | and aegis being entrenched makes it close to impossible to substitute it for an actually functioning security model | 19:53 |
| djszapi | and the beautiness is that with tcb, no need to worry anymore. | 19:53 |
| djszapi | about nothing. | 19:53 |
| djszapi | otherwise you keep repeating the aforementioned dirty hacks. | 19:53 |
| *** azbest_ has quit IRC | 19:54 | |
| djszapi | there is no "su to root" | 19:56 |
| djszapi | and it does have concept of users and roles. | 19:56 |
| djszapi | you can even explicitely request users in the manifest (as zillion apps do btw) even on a normal device. | 19:56 |
| *** torarne has quit IRC | 19:56 | |
| *** faenil has quit IRC | 19:56 | |
| *** jluisn has quit IRC | 19:57 | |
| djszapi | but of course there is no need for more than root and users since it is not a linux workstation | 19:57 |
| djszapi | Harmattan is never meant to be a multiuser unix workstation for good. | 19:57 |
| *** pinheiro has quit IRC | 19:58 | |
| djszapi | 19:51 < itsnotabigtruck> djszapi: i don't think you can 'assign' privileges to users... -> of course not without kernel modification | 19:59 |
| djszapi | but since it is an open kernel, it is quite possible. | 19:59 |
| djszapi | moreover you can request user uid on a stock device and request the credentials for that process which is gonna run as user. | 19:59 |
| djszapi | so I do not see the issue. | 19:59 |
| *** jluisn has joined #harmattan | 20:00 | |
| djszapi | and with open mode kernel, you can do what you would like to do. I would suggest being truely root, if the openness is the purpose, and not bother with credentials here and there. | 20:00 |
| djszapi | 19:52 < itsnotabigtruck> which every real security model is based on -> not really. | 20:02 |
| djszapi | actually the meego and tizen security model, called "Smack" has the same concept wrt this issue in question, like aegis. | 20:02 |
| djszapi | same with Android afaik. | 20:03 |
| *** jluisn has quit IRC | 20:06 | |
| *** lmoura has joined #harmattan | 20:06 | |
| *** jluisn has joined #harmattan | 20:06 | |
| *** vLassi has joined #harmattan | 20:06 | |
| *** vLassi has quit IRC | 20:07 | |
| *** jluisn has quit IRC | 20:12 | |
| *** risca has joined #harmattan | 20:12 | |
| *** jluisn has joined #harmattan | 20:12 | |
| *** arcean has quit IRC | 20:15 | |
| *** jluisn has quit IRC | 20:18 | |
| *** jluisn has joined #harmattan | 20:18 | |
| *** pinheiro has joined #harmattan | 20:18 | |
| *** djszapiN9 has joined #harmattan | 20:21 | |
| DocScrutinizer | er, indeed | 20:28 |
| DocScrutinizer | openmode kernel doesn't care about tcb | 20:30 |
| DocScrutinizer | so opensh basically not needed at all | 20:31 |
| *** arcean has joined #harmattan | 20:32 | |
| DocScrutinizer | opensh was of any use only for "normal mode" if you could succeed to isntall it in a way normalmode doesn't MALF immediately | 20:33 |
| *** jluisn has quit IRC | 20:33 | |
| *** jluisn has joined #harmattan | 20:33 | |
| DocScrutinizer | since any install in openmode seems to cause MALF when you change to "normal mode" later on, opensh in the form we got it now is basically useless | 20:34 |
| *** cclarky85 has joined #harmattan | 20:35 | |
| itsnotabigtruck | DocScrutinizer: it's because installing something = local aegis signatures on various files | 20:36 |
| DocScrutinizer | ?? | 20:36 |
| itsnotabigtruck | if you make an aegis signature in open mode and go back to closed mode, that signature isn't valid anymore | 20:36 |
| itsnotabigtruck | bad signature = malf on boot | 20:36 |
| DocScrutinizer | it's because messing with some of the relevant files will invalidate their hashes | 20:36 |
| DocScrutinizer | yes | 20:37 |
| itsnotabigtruck | but those hashes are authenticated with the 'local signatures' | 20:37 |
| DocScrutinizer | whatever that is | 20:37 |
| *** smokex|away has joined #harmattan | 20:37 | |
| DocScrutinizer | anyway, bbl | 20:37 |
| *** Sazpaimon has quit IRC | 20:41 | |
| *** jluisn has quit IRC | 20:44 | |
| *** jluisn has joined #harmattan | 20:44 | |
| *** miroslav has joined #harmattan | 20:44 | |
| *** oytunx has joined #harmattan | 20:46 | |
| *** benares_98 has joined #harmattan | 20:47 | |
| djszapiN9 | local sign is not the only type of those. | 20:48 |
| djszapiN9 | and yes, opensh makes no sense in open mode kernel | 20:50 |
| *** jluisn has quit IRC | 20:52 | |
| *** jluisn has joined #harmattan | 20:52 | |
| djszapiN9 | and what i would actually for is a kernel with sanely redefined policyset in the first place. | 20:52 |
| djszapiN9 | hope* | 20:52 |
| djszapiN9 | it is not a demand to drop the security completely for doing things. | 20:53 |
| *** benares_98 has quit IRC | 21:07 | |
| *** jesuschrist has joined #harmattan | 21:11 | |
| jesuschrist | hi | 21:11 |
| *** benares_98 has joined #harmattan | 21:12 | |
| jesuschrist | anyone using lpsmagic mod plus zones app ? | 21:13 |
| djszapi | jesuschrist: ask the question directly :) | 21:18 |
| jesuschrist | lpsmagic doesnt work with zones. this happens to anyone else ? <-- question | 21:19 |
| Mek | what is lpsmagic? | 21:20 |
| jesuschrist | Mek : http://talk.maemo.org/showthread.php?t=81923 | 21:21 |
| Mek | well yeah, that can't work since zones replaces the standby screen | 21:22 |
| Mek | (and admittedly doesn't show too much useful information on the standby screen yet) | 21:22 |
| jesuschrist | heh | 21:23 |
| *** Sazpaimon has joined #harmattan | 21:23 | |
| *** jluisn has quit IRC | 21:23 | |
| Sazpaimon | so, does GSM have a phase out year | 21:23 |
| *** jluisn has joined #harmattan | 21:23 | |
| Sazpaimon | out of curiosity | 21:23 |
| djszapi | Mek: good work on the android front with calligra btw :) | 21:23 |
| Mek | I guess I could try to make zones show the operator_logo image, that would then also make this work again... | 21:24 |
| alterego | Sazpaimon: that would very much depend on the country, and no, I doubt anyone is willing to do it for the foreseeable future, as a lot of infrastructure is in place that uses the tech. | 21:24 |
| Sazpaimon | I mean, it has to go away at some point | 21:26 |
| Sazpaimon | its already kind of silly to have GSM, UMTS AND LTE radios on the same device | 21:26 |
| *** vLassi has joined #harmattan | 21:32 | |
| *** gabriel9 has joined #harmattan | 21:32 | |
| *** faenil has joined #harmattan | 21:32 | |
| djszapi | Mek: was it a real hard work to port a qt based project to Android ? | 21:33 |
| Mek | djszapi: not really, most of the hard work was already done by the necessitas project | 21:34 |
| *** jluisn has quit IRC | 21:34 | |
| djszapi | cool | 21:34 |
| *** jluisn has joined #harmattan | 21:34 | |
| faenil | djszapi: my thesis is a Qt android software, if you're interested ;) | 21:34 |
| xmlich02 | faenil, i am interested | 21:34 |
| faenil | I'm here :) | 21:35 |
| faenil | for 10 mins :) | 21:35 |
| xmlich02 | could you upload your thesis? | 21:35 |
| djszapi | faenil: if you do not reinvent the wheel this time ;) | 21:36 |
| faenil | I can't because it's not finished yet, I'm graduating on 24th :) | 21:36 |
| faenil | you'll find it in the Android Market in the next weeks :) | 21:36 |
| faenil | djszapi: it's a 3D model viewer, made for Counsil of National Research here in Pisa :) | 21:37 |
| *** etrunko has quit IRC | 21:37 | |
| djszapi | faenil: you can upload, it is not a problem, if unfinished. | 21:37 |
| faenil | djszapi: it could be :) I don't know if it's only free or opensource too yet :) | 21:38 |
| djszapiN9 | we do not need code, just pdf :) | 21:39 |
| djszapiN9 | but as you wish. | 21:40 |
| faenil | djszapiN9: what do you need the pdf for? XD | 21:40 |
| djszapiN9 | curious | 21:41 |
| faenil | you'll have once it's finished then :D don't want to share it now :D | 21:42 |
| *** tarantism has joined #harmattan | 21:42 | |
| faenil | anyway, it's a 3 CFU thesis...nothing big | 21:43 |
| faenil | still, it took a while | 21:43 |
| faenil | definitely not the same amount of time a 3 CFU exam needs | 21:43 |
| faenil | much more :) | 21:43 |
| *** risca has quit IRC | 21:43 | |
| faenil | xmlich02: you too interested in the pdf? or wanted to ask anything? | 21:44 |
| xmlich02 | just pdf | 21:44 |
| *** risca has joined #harmattan | 21:45 | |
| *** risca has quit IRC | 21:45 | |
| itsnotabigtruck | speaking of zones | 21:45 |
| itsnotabigtruck | is there some reason why ovi store let somebody list a bunch of their apps as being from "nokia" | 21:45 |
| itsnotabigtruck | http://www.appsforn9.com/ < all of those, including zones | 21:46 |
| faenil | xmlich02: oh ok...don't think you'll find anything that interesting in there anyway :) there is almost nothing about implementation details there | 21:46 |
| faenil | djszapi: ^ | 21:46 |
| itsnotabigtruck | apparently a not-insignificant number of people think zones is a nokia app, and some of them were indignant about it costing $5 for something like that | 21:46 |
| djszapi | and some Nokia apps are not mentioned as Nokia apps on the contrary. | 21:47 |
| itsnotabigtruck | also you guys were talking about 'setfcap' capability earlier | 21:48 |
| jesuschrist | so zones isnt from nokia ? uh ? | 21:48 |
| itsnotabigtruck | turns out it's a completely useless capability | 21:48 |
| djszapi | I do not know what it is for | 21:48 |
| djszapi | since it is not documented properly in the header file at least. | 21:48 |
| itsnotabigtruck | it lets you interact with a feature that was off by default until linux 2.6.33 | 21:48 |
| djszapi | and I was not having time to look at the kernel implementation with lxr. | 21:49 |
| itsnotabigtruck | so it's compiled out of harmattan's 2.6.32 kernel | 21:49 |
| djszapi | itsnotabigtruck: note that we have some patches merged into our kernel from later versions. | 21:49 |
| djszapi | but yeah, I do not know what that capability is for without documentation. | 21:49 |
| itsnotabigtruck | it's definitely compiled out though in this case | 21:50 |
| itsnotabigtruck | # CONFIG_SECURITY_FILE_CAPABILITIES is not set | 21:50 |
| itsnotabigtruck | and jesuschrist: nope, it isn't | 21:50 |
| itsnotabigtruck | it's from whoever owns appsforn9.com | 21:50 |
| *** risca has joined #harmattan | 21:50 | |
| djszapi | itsnotabigtruck: that does not make too much sense then | 21:51 |
| *** faenil has quit IRC | 21:51 | |
| *** jluisn has quit IRC | 21:52 | |
| djszapi | because the credential is requestable for ovi apps :D | 21:52 |
| *** jluisn has joined #harmattan | 21:52 | |
| djszapi | I wonder why it was selected that way... | 21:53 |
| Mek | itsnotabigtruck: zones is a nokia app (okay, developed by subcontractors, but all the code is owned by nokia) | 21:53 |
| Mek | and yes, it is silly that it costs $5 | 21:53 |
| itsnotabigtruck | mek: it is? why isn't the webpage on a nokia site then | 21:53 |
| djszapi | itsnotabigtruck: same with hiro | 21:53 |
| itsnotabigtruck | appsforn9.com looks super amateur and doesn't say anything about nokia | 21:53 |
| djszapi | it is developed by Nokia, subcontractors and Nokia owns the code. | 21:54 |
| itsnotabigtruck | it looks like appsforn9.com is registered by a nokia project manager, but with a personal gmail email | 21:55 |
| djszapi | ask them, they probably have a good reason. | 21:56 |
| djszapi | imho it was even more strange to have a big Nokia app development competition on a non-Nokia website :) | 21:57 |
| *** lassi has joined #harmattan | 21:58 | |
| djszapi | Mek: How about the qt-components on Android, if any ? | 21:59 |
| Mek | djszapi: I know there is some work going on, but I haven't tried to use them myself yet | 21:59 |
| djszapi | Mek: do they also work on the qt5 port, or qt4 is still the only focus ? | 22:01 |
| Mek | qt4 only at the moment afaik | 22:01 |
| djszapi | Mek: have you done any packaging ? Is it difficult to get used to ? | 22:02 |
| *** SqRt7744 has joined #harmattan | 22:02 | |
| Mek | it's all trivially easy if you use qmake and the necessitas version of qt creator, otherwise it is a bit tricky to figure out what to do exactly | 22:03 |
| djszapi | I was mostly interested in starting to port kdelibs | 22:03 |
| djszapi | and its dependencies beforehand. | 22:04 |
| *** vLassi has quit IRC | 22:05 | |
| djszapi | Mek: do they have QtQuick 1.1, especially pinch area support ? Qt 4.7.4 | 22:05 |
| Mek | Qt4.8, so I assume yes | 22:06 |
| *** jluisn has quit IRC | 22:07 | |
| *** jluisn has joined #harmattan | 22:07 | |
| *** wirwe has joined #harmattan | 22:10 | |
| *** risca has quit IRC | 22:11 | |
| *** oytunx has quit IRC | 22:11 | |
| *** jluisn has quit IRC | 22:12 | |
| *** jluisn has joined #harmattan | 22:12 | |
| djszapi | Mek, ok thanks | 22:17 |
| djszapi | What is the simplest way in qml to fill in a listview/repeater model with the alphabet so the letters are the elements ? I guess there is some trick instead of enumerating them explicitely ? | 22:18 |
| befr0d | filling the listmodel with a for loop? | 22:19 |
| *** miroslav has quit IRC | 22:19 | |
| djszapi | that is a way, yes for instance accumulating the characters in an array and return that | 22:20 |
| djszapi | though, array cannot be used in the same qml file. | 22:20 |
| Mek | model: 26, and then use some javascript to convert number to letter? | 22:20 |
| *** SqRt7744 has quit IRC | 22:21 | |
| djszapi | the most convenient would be how it is done in regexp :P [a-z,A-Z] :P | 22:21 |
| *** jluisn has quit IRC | 22:22 | |
| djszapi | I am not sure how to mix string and integers according to the ascii table in javascript | 22:23 |
| *** jluisn has joined #harmattan | 22:23 | |
| djszapi | c = 'A'; while( c != ('Z' + 1)) letterList.append(c++); -> something like this would be in c++ | 22:26 |
| *** snowpong has quit IRC | 22:29 | |
| *** jluisn has quit IRC | 22:32 | |
| *** jluisn has joined #harmattan | 22:32 | |
| * GeneralAntilles sighs. | 22:37 | |
| GeneralAntilles | What are we going to do when we run completely out of reasonable mobile operating systems? :( | 22:37 |
| *** jluisn has quit IRC | 22:44 | |
| *** jluisn has joined #harmattan | 22:44 | |
| itsnotabigtruck | GeneralAntilles: use android? | 22:49 |
| itsnotabigtruck | i went N9 for the hipster factor + it not being all tapped out in terms of apps/hacks to implement, but android is a perfectly serviceable os | 22:49 |
| *** JoseLuis has joined #harmattan | 22:54 | |
| *** jluisn has quit IRC | 22:55 | |
| *** arcean_ has joined #harmattan | 22:55 | |
| *** arcean has quit IRC | 22:57 | |
| *** lizardo has quit IRC | 23:03 | |
| *** JoseLuis has quit IRC | 23:08 | |
| *** jluisn has joined #harmattan | 23:08 | |
| *** Myname24 has joined #harmattan | 23:19 | |
| jesuschrist | gn8 | 23:19 |
| *** jesuschrist has quit IRC | 23:19 | |
| Myname24 | Did someone ask the peregrine guys what is the status of the developpement? | 23:20 |
| *** Myname24 has quit IRC | 23:23 | |
| *** gabriel9 has quit IRC | 23:23 | |
| *** gabriel9 has joined #harmattan | 23:25 | |
| *** befr0d has quit IRC | 23:28 | |
| *** jluisn has quit IRC | 23:29 | |
| *** longthen has quit IRC | 23:30 | |
| *** longthen has joined #harmattan | 23:31 | |
| *** niwakame|away has quit IRC | 23:33 | |
| *** niwakame|away has joined #harmattan | 23:36 | |
| *** befr0d has joined #harmattan | 23:41 | |
| *** Hamidx9 has joined #harmattan | 23:42 | |
| *** heymaste_ has joined #harmattan | 23:54 | |
| *** heymaster has quit IRC | 23:56 | |
| *** Hamidx9 has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!
