=== chriadam|away is now known as chriadam === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [06:31] good morning [08:36] Good morning all; happy Monday and happy Let It Go Day! :-D [08:40] JackYu: [08:40] oops wrong [08:41] JamesTait: the cold never bothered me anyway [08:42] justCarakas, I'm glad to know that, having seen that "celebration" pop up in my browser, I'm not the only person who'll have that song stuck in my head all day. ;) [08:42] :D [08:42] its an awesome song :D [08:45] It's not bad the first couple of times, but it gets a bit much when your kids play it on repeat for hours on end. :-P [08:50] hehe :D [08:50] I've worked a couple of hours with that song on repeat [09:09] justCarakas, https://www.youtube.com/watch?v=2bVAoVlFYf0 [09:10] Google Translate is always fun. [09:10] listening :p [09:12] hahha :p === chriadam is now known as chriadam|away === MacSlow is now known as MacSlow|lunch [11:31] oSoMoN: do you want the list of bookmarks to be returned sorted by time now, instead of ur ? [11:31] url ? [11:32] nerochiaro, that probably makes sense === ara is now known as Guest7746 [11:51] oSoMoN: how do I actually bookmark something from the webbrowser UI right now ? [11:52] oSoMoN: oh, from the activity [11:52] nerochiaro, in the activity view, each open tab has a star icon next to the title of the page === MacSlow|lunch is now known as MacSlow [12:13] oSoMoN: besides the lack of tests, do you see anything missing here ? https://code.launchpad.net/~phablet-team/webbrowser-app/webbrowser-app-bookmark-timestamp/+merge/224117 [12:15] nerochiaro, sqlite has a DATETIME type, why did you use INTEGER? [12:15] Ubuntu SDK: Template HTML5 app shows warning | http://askubuntu.com/q/487121 [12:16] oSoMoN: it doesn't, at least not according to the documentation. it's a timestamp anyway, the number of msec since epoch [12:17] oSoMoN: http://www.sqlite.org/datatype3.html [12:17] oSoMoN: "SQLite does not have a storage class set aside for storing dates and/or times. Instead, the built-in Date And Time Functions of SQLite are capable of storing dates and times as TEXT, REAL, or INTEGER values:" [12:18] nerochiaro, yes, but the datetime keyword is valid (and implicitly converted to numeric), and more explicit than just integer [12:18] nerochiaro, and we’re using it already in the history model, so for consistency I would recommend using it in the bookmarks model as well [12:19] oSoMoN: ok [12:19] oSoMoN: anything else ? [12:19] nerochiaro, and entry.created should store a QDateTime, not a number of msecs since epoch [12:20] nerochiaro, and as you pointed out, unit tests :) [12:27] oSoMoN: i don't see why, but as you prefer [12:28] nerochiaro, because manipulating a QDateTime is much more developer-friendly than a number of milliseconds since epoch, that’s why [12:29] oSoMoN: i beg to differ, but ok === renato_ is now known as Guest20000 === _salem is now known as salem_ === salem_ is now known as _salem [12:57] hi fginther, would you have some time today to join the core apps review call to discuss the addition of the Evernote API key from Jenkins? [12:58] dpm, I have a meeting convlict. Is this something we can discuss first thing, then I'll jump to my other meeting? [12:59] fginther, if that works for you, that'd be perfect. Otherwise we can schedule a quick call that doesn't conflict with your meeting [13:00] hi rpadovani, popey, mzanetti, Reminders app meeting? [13:00] dpm, if we could meet a few minutes early, that would work for me, but it's not a critical confilect [13:01] * fginther just can type 'conflict' right today [13:02] fginther, sure. Or if you want, we've got a Reminders call going on now. You're welcome to join now if that works better [13:02] dpm, sure I can jump in now [13:02] fginther, excellent: https://plus.google.com/hangouts/_/calendar/ZGF2aWQucGxhbmVsbGFAY2Fub25pY2FsLmNvbQ.4gsate3vsqlj03tqqnsqgbl8i8? [13:22] Hi, I have a question about publishing programs in the Ubuntu Store. Is this the right channel? [13:24] barmi, go ahead and ask the question [13:26] SO I have an app published in the store, and I've submitted an update (well, 3 actually) === _salem is now known as salem_ [13:27] The app is still listed as "Further changes awating review" [13:27] Been like that for about 4 months :( [13:27] Do I need to do something to prompt the review, so the update can go live? [13:43] barmi, poke popey with a stick [13:45] who's popey, and how can I poke him remotely? === Guest20000 is now known as renato__ [13:54] dpm, wait, why the meeting was today? I'm sorry, but I was not informed about that [13:56] rpadovani, ah, weird, I chose to send an e-mail to all attendees. Would you have time for a call in about 30 minutes? [13:56] I can update you then on what we discussed [13:57] dpm, cool, thanks [13:57] whenever you want, I have a free afternoon [13:58] dpm, , mhh, I have the event on my calendar, but I have no email about it, so I didn't notice it, sorry [13:59] rpadovani, no worries, we can talk in about 30 mins, I can ping you then if you're free [13:59] ok, thanks! === salem_ is now known as _salem [14:13] barmi: if it's a desktop app, poke davmor2 === _salem is now known as salem_ [14:20] popey, where can I find app logs on device? [14:20] rpadovani: /home/phablet/.cache/upstart/ iirc [14:23] davmor2, yeah, thanks. The name of the file is application-click-com.ubuntu.calculator_calculator_1.3.283, for your information :-) [14:30] rpadovani: trust me when I say I'm not typing that out for each different dev ;) [14:30] ahahaha [14:31] barmi: what's the app? [14:32] barmi: if it is in the desktop queue then the updates are just starting to be processed under the new rules so you should hopefully get a reply by the end of the week unless all hell breaks loose in the meantime. [14:41] popey: Pixeluvo === salem_ is now known as _salem [14:52] popey: nik90: https://bugs.launchpad.net/ubuntu-clock-app/+bug/1333299 [14:52] Ubuntu bug 1333299 in Ubuntu Clock App "Recurring alarms only work once" [Undecided,New] [14:53] popey: yes it's a desktop app. What exactly is the best way of poking people - just writing on here? Or is there a more direct way to contact particular people? [14:53] barmi: davmor2 is indeed the guy to poke, and you've done that now ☻ [14:54] thanks mhall119 [14:54] davmor2: Cheers, how do I know if it's in the queue? I submitted the update over 4 months back [14:55] barmi: they are all in the queue, I'm about 10 in on 160ish + so asap but it takes a while. [14:55] ok, thanks. [14:55] so if I haven't heard anything by this time next week, I'll pop back here and re-poke. [15:01] there is someone who has unicorn on desktop and has 5 minutes to help me? :-) [15:02] rpadovani: help with what? [15:02] kalikiana, to see if a bug is reproducible on desktop too... Because there is a bug in calc app on phone, but not on trusty [15:03] kalikiana, could you take last version of lp:ubuntu-calculator-app and try to do 1/3, please? [15:03] barmi: yes should be fine [15:04] oSoMoN: wasn't the standup moved to now ? [15:06] nerochiaro, that was only on Friday [15:06] rpadovani: just getting the latest trunk [15:06] oSoMoN: oh, so i totally missed today's ? [15:06] yup [15:06] intersting no one called me [15:09] rpadovani: so, 1/3 is 0,33333333 [15:09] what bug were you expecting? [15:09] kalikiana, thanks, so the error is only on device, the result is 0.3 [15:10] my ubuntu phone ran out of juice so I can't compare it [15:10] ah [15:11] The error seems to be in an external javascript we use, BigNumber.js, seems I have some debug to do === _salem is now known as salem_ === salem_ is now known as _salem === _salem is now known as salem_ [15:37] rpadovani, i noticed the test passed in CI today. i guess you didn't push a fix yet? [15:38] brendand, no, I'm investigating on it, but I have no idea on what's wrong [15:38] rpadovani, although i can still reproduce it on the device [15:39] brendand, do you have any idea on what I can investigate on? I'm pretty sure that error is created in a file, but it's very big and I'm doing some random check, because I have no idea [15:40] rpadovani, how much do you know at the moment? where is the code that does the calculation? [15:40] rpadovani, if you're stuck somewhere i can have a look, but i don't want to start from the beginning [15:42] brendand, well, the formula is managed in formula.js, that pass the input to engine.js, that parse the formula and pass the value to calc to bignumber.js [15:42] Line 572 of engine.js calls the div function [15:43] result = lhs.value.div(rhs.value); <--- Here the result is wrong [15:44] brendand, unfortunately I don't know well how bignumber.js works, is an extern library [15:44] https://github.com/MikeMcl/bignumber.js [15:44] brendand, maybe is something related to split() function? I read there were some problems, calculator has used to work, but maybe the fix brokes something? [15:46] rpadovani, what provides BigNumber.js? [15:46] rpadovani, which package that is [15:46] brendand, we included it in the calculator code, it's in the root folder [15:49] rpadovani, have any changes gone in recently to that file? [15:50] brendand, nope, there are no change in calculator since a while === chihchun is now known as chihchun_afk === salem_ is now known as _salem [16:00] rpadovani, i can't trace where lhs.value.div is [16:00] rpadovani, js is hard to read for me :) === gatox is now known as gatox_lunch === _salem is now known as salem_ === salem_ is now known as _salem === _salem is now known as salem_ [16:39] =48 [16:39] oops [16:56] mhall119: which image are you running to see the bug 1333299 [16:56] bug 1333299 in Ubuntu Clock App "Recurring alarms only work once" [Undecided,New] https://launchpad.net/bugs/1333299 [16:58] mhall119: I created a weekly recurring alarm which rang an hour ago, and I still see it in my indicator. [16:58] mhall119: I would still need to wait until tomorrow to check if it actually goes of again though [16:59] jdstrand: ping === gatox_lunch is now known as gatox [17:10] Can anyone help me verify that my source tree is well setup before i resubmit my application? The reviewer said that many of my files have the /usr/bin path still (especially in my postinst file) but i commented out those areas and now i have removed them completely. I've also looked over everything else and it looks fine to me, can anyone do me a huge favor and look at it as well? [17:11] the setup is pretty basic i don't think you will get confused about it, should be easy but i am new to this hehe [17:17] nik90: latest devel image, r87 [17:17] bluezone: is this a desktop .deb or a phone .click package? [17:17] mhall119: I was running 89. Let me wait until tomorrow to check if the alarm goes off to confirm the bug. [17:18] nik90: ok [17:18] mhall119, it is a desktop application .deb, but i can give you the source package if you want to build it yourself :) [17:19] bluezone: there's a tool called arb-lint that was developed to make checking desktop app packages easier [17:19] ok [17:20] https://launchpad.net/arb-lint is the project page, I don't know if it's in the archives or not [17:20] but it'll check your package for common errors like pointing to /usr/bin/ [17:25] nik90: hey, what's up? [17:25] jdstrand: hey, I had a confined app question. let me phrase it [17:26] jdstrand: The clock app needs to be able to read AND write into the indicator-datetime schema which will be used to store a user setting. [17:26] jdstrand: to provide a bit of context, the clock app will read and write the alarm volume into this schema which indicator datetime will use to ring the alarm in the volume set by the user. [17:26] jdstrand: is that possible for a confined clock app to do? [17:28] charles: you might be interested in following this conversation ^^ [17:31] mhall119, had to make some changes myself to get it to install, do you recall how to use this arb-lint? which script should i be using in arb-lint/lint [17:33] Hi, how do we take a screenshot these days? my script doesn't want to run against #94 [17:35] nik90: no, that is not possible with app confinement if I understand your question [17:35] bluezone: the executable itself is called arb-lint, you just point it at your .deb package I think [17:36] nik90: why is the alarm app setting the volume? shouldn't this be up to the user? [17:36] hmm [17:36] nik90: jdstrand: since only the clock app woul dbe using this, could we use a non-standard security policy to allow it, rather than running the whole app unconfined? [17:37] what is 'it' in this context? a file? [17:37] jdstrand: well we are thinking of exposing a setting where the user can set the alarm volume through the clock app. [17:38] the first thought was the setting would be in indicator-datetime's gsettings schema [17:38] how are you making the adjustment? you mentioned scema, do you mean gsettings? [17:38] jdstrand, yes [17:38] schema* [17:38] mhall119's suggestion is not possible at this time [17:39] the proper way to do this is to have a service that allows adjusting the volume. that service does the gsettings stuff, not the clock. the clock then talks to this service [17:40] this service may or may not already exist [17:40] this service should be a trusted helper [17:41] then we could add this access to the service to the 'calendar' policy group [17:41] is this something coming from design? it is the first I've heard of it [17:42] jdstrand, right, it's a new change from design [17:42] jdstrand: https://docs.google.com/presentation/d/1JvDyhsW17d1-Mz8OY1YMBKwfRI2z9qgyRjbujEsxEMk/edit#slide=id.g18895458d_024 [17:42] jdstrand: we need 2 settings options. 1 for alarm volume and the other for setting the silence time [17:43] jdstrand: i.e how long the alarm should ring before being silenced automatically [17:43] someone needs to be assigned to design the implementation then [17:44] basically, there is no gsettings support for apps [17:44] yeah I was told that gsettings cannot be used by confined apps [17:44] jdstrand, indicator-datetime-service is always running. it could expose these settings through its public state s.t. clock-app doesn't go through gsettings [17:45] sure [17:45] we would need to add policy to calendar policy group to allow talking to indicator-datetime-service [17:45] jdstrand, can you take care of the policy part of this? [17:46] note that the medium term (ie, not rtm) plan is to have the calendar abstraction not be a reserved policy group, which means that indicator-datetime-service will need to become a trusted helper [17:46] but for rtm, it will remain reserved, so the clock app can continue to use it and nothing has changed from a policy POV [17:47] ok [17:47] charles: I can adjust apparmor-easyprof-ubuntu, yes. please file a bug for it once the dbus interface is defined for it [17:47] jdstrand, ok [17:48] charles: how would I access it from the clock app when the dbus interface is defined for it? [17:48] charles: I suggest filing a bug against indicator-datetime-service to implement the change there and adding an apparmor-easyprof-ubuntu task [17:51] nik90, it would be a D-Bus property, ie, org.freedesktop.DBus.Properties [17:53] charles: can I access it in qml? [17:54] charles: fyi, apparmor can only mediate to the method, not message contents. if you implement this as a property then any app with access to the calendar policy group would be able to edit any property since we could only mediate down to Set() [17:54] jdstrand, ok. maybe better to add explicit getters & setters then [17:54] this doesn't affect the short term [17:55] but it would affect when we want calendar to move out of being a reserved policy group [17:55] nik90, I believe so, though I haven't done enough qml to have done this firsthand -- [17:55] charles: I think that would be best. Then we can just add to the calendar policy group: dbus bus=session interface=... method=GetAlarmLength, (or whatever) [17:56] charles: ack. I will take a look into it to see if it can be done [17:56] * nik90 goes to prep dinner [17:56] nik90, dednick would be one person to ask, though he's EODed [17:57] implementing the getters and setters now means no API change down the line [17:57] nik90, I'll hold on this until you ping me back [17:57] jdstrand, agreed [17:59] mhall119, okay it seems to be complaining about "Could not find .changes file. Can't do binary package tests." [17:59] haven't seen this file mentioned in my tut, is it okay if i leave it out? [18:01] otherwise every else looks good, doesn't complain about files in /usr/bin at all [18:17] bluezone: have you created a source package? [18:17] with debuild -S [18:18] that should create the .changes file [18:25] ok [18:26] davmor2, really sorry about msging you here i know you are a busy person :). Question is regarding the sshtunnel application you reviewed, you say that my postinst script is writing to /usr/bin but those areas have been commented out (now i removed them completely). I've also updated DEBIAN/sshtunnel.dirs to correspond to the correct installs that looks like this now: http://bpaste.net/show/mettt3LWepvBIKZ4NN2j/ . You also mention you see many fil [18:26] es that are still using the /usr/bin path but i don't see any such files i'm beginning to think i gave you the wrong upstream source or something :O. Anyway i will double check everything again and reupload, let me know if you can shed some light on this, sorry for pestering you with noob questions. Thanks :) [18:27] ouph that's a large amount of text [18:29] bluezone: I think there was the postinst and check the desktop file and I think there was another but I can't remember off hand I think there were at least 3 files where /usr/bin was in use iirc [18:34] hmm well i am using /usr/share/applications for the desktop file but it points to the correct location to launch the program 'python3 /opt/sshtunnel/SSHTunnel.py'. I'm also using /usr/share/icons for icons but i don't see anything using /usr/bin. I really think i'm stupid and gave you the wrong tar source haha [18:35] postinst also the areas where it was writing to /usr/bin were comment out with """ """ and now i have removed them completely [18:40] charles: I will check with dednick about the dbus usage in QML. Was there anything else you wanted to tell m? [18:40] nik90, I think we're good for tonight [18:40] nik90, I'm working on handling custom settings in the snap decisions / alarm sounds now [18:41] not the expose parts, obviously [18:41] but the private gsettings schema + testing volume levels etc [18:41] charles: ok, just let me know when you have the alarm sounds ready, and I can test it with my branch [18:41] nik90, ok [18:41] charles: I was referring to the different alarm sounds through the ALarms API === sasasasa is now known as bluezone_ [18:55] connection dropped -.- === bluezone_ is now known as bluezone [19:00] the postinst is making use of #!/usr/bin/python3 on line 1 [19:00] that's allowed, no? :O === salem_ is now known as _salem [19:08] well i triple-checked everything now and changed what was wrong or confusing should be fine now i hope [20:37] So, I'm trying to build cdfs-src (mostly 'cause no prebuilt packages exist) and I'm running into some difficulties [20:38] like, trying make prepare on /usr/src/linux and I get [20:38] No rule to make target `/usr/src/linux-headers-3.11.0-22-generic/arch/x86/syscalls/syscall_32.tbl', needed by `arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h' [20:38] I was trying that 'cause m-a a-i cdfs was failing on linux/autoconf.h missing [20:38] I'm kinda wondering what the next move is [20:39] this is my first time trying this sorta thing in ubuntu [20:39] install autoconf / autotools-dev [20:39] What are you trying to build? [20:40] cdfs [20:40] ki7mt: wanna try mounting this CD w/ it [20:40] is mostly a shot in the dark. the CD isn't displaying any files under linux, works fine on mac and windows [20:40] and cdfs-src is the only package I can find for ubuntu 13.10 [20:40] nemo, wow, there is like a stupid number of deps to build cdfs from source. [20:40] hm :-/ [20:41] well, there's a decent chance I might have them [20:41] this machine has a ton of dev packages already [20:41] I would downlod the source package, then sudo apt-get build-dep first [20:41] ki7mt: oh. I was trying to do things a bit more "managed" [20:41] so my dpkg -i of cdfs-src was a bad idea? [20:41] nemo, That is managed, as it will install only the deps needed to build cdfs [20:41] ok. so I got that far at least 😃 [20:42] I'm pretty sure I have autoconf/autotools already [20:42] but let's see [20:42] also. the error in linux doesn't *seem* related [20:42] From the error you sent in #ubuntu, your a header file somewhere it seems. [20:42] yeah. I have them already [20:42] .. missing .. [20:42] ki7mt: I just installed linux-headers, switched to /usr/src/linux and tried make oldconfig make prepare [20:43] Yeah, then I've do: sudo apt-get build-dep linux-image-$(uname -r) [20:43] O_o [20:43] ok.... [20:43] * nemo tries [20:43] that should clear any of the missing kernel deps, then the same for cdfs [20:43] oh well. this machine has a ton of space [20:43] weird set of deps. tex and such. like for document generation [20:44] Dev boxes eat disk space, it's just the way it is. I try to limit it also,but its easier, to just install them. [20:44] yeah, Im not sure what uses Tex, I saw that allot also. [20:45] And AsciiDoc A2x for manpages in other, all sorts of stuff. [20:47] This is a descent roll-your-own kernel guide too: https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [20:49] I'm actually mildly puzzled why a mount module requires kernel anything... [20:49] I kinda figured that reading the filesystem and presenting it would be more userland [20:49] ki7mt: I don't actually want to build my own kernel. just want to get enough of the config in place to satisfy the cdfs build [20:50] Im not sure why it wants that either, Im sure there is a reason though. [20:50] Maybe it's just the headers it needs or something [20:53] ki7mt: I'm sure that's pretty much it, yeah [21:11] ki7mt: hm. I was reallly hoping I could just do "make prepare" ☹ [21:11] ki7mt: even after installing deps still can't get that to run [21:12] ki7mt: I don't suppose you have any suggestions to get past that error? [21:12] nemo, whats the error getting now? [21:12] same [21:12] make[1]: *** No rule to make target `/usr/src/linux-headers-3.11.0-22-generic/arch/x86/syscalls/syscall_32.tbl', needed by `arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h'. Stop. [21:13] I wonder if blowing away .config might help... [21:13] not sure why _32 anything is being use [21:13] is a 64 bit system [21:14] That's good question, did you run make on the kernal before configure and make on cdfs? [21:16] ummm [21:17] for CDFS I'm just trying m-a a-i cdfs [21:17] since I know fairly little about the ubuntu build process [21:17] for the kernel, I was just trying make oldconfig, make prepare [21:17] since I just want the headers [21:17] I don't see why I need anything else [21:17] Just curious, why are your trying to build it from source? [21:18] ki7mt: because I can't find anything else ☹ [21:18] https://launchpad.net/ubuntu/saucy/amd64/cdfs-src/2.6.23-4 [21:18] oh [21:18] lol [21:18] deleted [21:18] * nemo facepalms [21:19] huh. strange. the referenced bug in the deletion notice is junk. maybe is really an upstream kernel bug? [21:19] ah. debian bug [21:19] well. damn [21:20] Yeah, something not right about this, I can't find a new version of the ewrc package anywhere. [21:20] 'course. I have noooo idea if that's actually the issue. my shot-in-the-dark at cdfs was due to that's what windows was identifying the filesystem as [21:20] Ok, maybe time to back up a bit, what's the original issue that brought you too build cdfs [21:26] aw hell [21:26] this one CD reads ok [21:26] ki7mt: ok. I think the CDs are just crap [21:27] ki7mt: I had a DVD that was showing up as blank, but looked fine on coworker's windows 7 machine, so I thought the problem was possibly a filesystem linux didn't recognise [21:27] ki7mt: windows 7 claimed it was "cdfs" which was not a CD filesystem I was familiar with [21:28] so I hunted around for something like that for linux [21:28] ki7mt: but... I've now tested a few of these dvds on a couple of machines here, and another windows machine wasn't seeing anything either [21:28] and... the mac that was reading one of them just locked up [21:28] nemo, Ok, I see, to be honest, Im not up to speed on that FS either, i've never used it to my knowledge [21:28] and now linux is reading one of them fine [21:29] so... I'm going to go with "these stupid DVDs which were a dumb idea anyway weren't burned properly, or the media are bad" [21:29] I see read errors in dmesg which is suggestive of that [21:31] I looked a fare bit for cdfs in later version of UB, either it's included in the kernel somewhere, or it's obsolete or something, I'd have to research that more. [21:31] I dunno. I'm going with "crap dvds" at the moment [21:31] since, yeah, I got this other one to read [21:31] maybe my drive needs cleaning too [21:32] isn't like I use it much anymore [21:32] find a can of air and see if there's dust on the laser or something [21:32] Yeah, but on *Nix and Mac ... both having issues, seems fishy. [21:32] ki7mt: and windows! [21:32] Use them as beer coasters :-) [21:48] yeah, woudln't that be nice [21:49] popey: hi [21:58] is there anyone here who is working on the files webapp [21:58] correction mobile app