=== m_conley is now known as m_conley_away [03:33] Good morning [03:50] pitti: gigi [03:50] also: hihi [03:52] Morning, pitti. Good $TIME_OF_DAY, desrt :) [03:53] hey desrt, how are you? [03:53] hey RAOF, good day [03:54] pitti: good. just getting ready for an early train ride tomorrow [05:36] robert_ancell: hey Robert, how are you? feeling comfy in your old "desktop team" chair? :-) [05:55] good morning [06:06] Salut didrocks ! [06:07] salut jibel :) [06:07] Comment ça va après ces quelques jours ? [06:07] ça va bien, on se sent encore fatigué, mais le week-end sera vite là :) [06:07] tout s'est super bien passé, c'est ce qui compte ;) [06:09] et toi, ça va? pas eu de surprises particulières ces derniers jours? [06:15] didrocks, rien de spécial, la routine. phone testing, encore et encore, comme je le disais à Martin, c'est un peu comme dans 'un jour sans fin' :) [06:16] tous les jours une nouvelle images avec des trucs un peu différents à chaque fois [06:17] ouai, je trouve l'analogie très juste ;) j'espère que ça ne deviendra pas trop routinier à la longue et qu'on pourra avoir une automation digne de ce nom bientôt [06:30] pitti: mmmh, I think libgphoto transition will require source full uploads :/ [06:31] the name of the -dev pkg has changed too [06:34] bigon: right [06:35] bigon: but only these three packages need actual new upstream versions [06:36] bigon: I wish the -dev wouldn't contain the soname :/ [06:37] I'll open the transition bug later today [07:00] pitti, not quite, back on monday [07:12] Good morning! [07:12] didrocks: hello! How are you? How was the celebration? ;) [07:12] hey sil2100! [07:12] g'day! [07:13] sil2100: I'm still a little bit tired TBH, but everything was fine, we were really lucky in term of weather in the end :) [07:13] so celebration and everything ran perfectly! [07:13] hey mlankhorst [07:13] didrocks: I guess it was really lucky, the weekend and the beginning of this week were really warm and nice! [07:13] morning [07:14] Glad to hear it was the same ;) [07:15] well, we had small rains at the start of the ceremony [07:15] but then, sunny weather [07:15] so it was "just in time" :) [07:16] but yeah for the warm part, as part of the dinner was outside… ;) [07:22] Oh, hehe, sounds nice! [08:00] seb128, how do you build ubuntu-system-settings and run it locally? Having fun trying to decode qmake... [08:00] robert_ancell, bzr bd [08:01] seb128, I was trying qmake PREFIX=`pwd`/install etc, that would be faster but everyone just packages and installs it? [08:03] robert_ancell, I install a package and then when I hack on a panel I tend to "sudo cp fileIhackon /usr/...; system-settings " on a command line [08:03] robert_ancell, and "up-enter" [08:03] seb128, classy :) [08:03] yeah... [08:03] running from tree doesn't find the plugins [08:03] we should fix that [08:03] but until then... [08:03] yeah, and qmake doesn't seem to apply the prefix to everything (e.g. translations) [08:04] robert_ancell, https://code.launchpad.net/~jpakkane/ubuntu-system-settings/cmake/+merge/184123 [08:04] robert_ancell, if you like cmake better [08:04] morning [08:04] happy friday [08:04] robert_ancell, we plan to merge that soon (there is a qtcreator small integration issue to resolve first though) [08:05] Laney, hey, happy thursday to you [08:05] Laney, are you on vac tomorrow? [08:05] seb128, I don't like cmake either but one less build system is better :) [08:05] yep [08:05] damn, you didn't fall into my trap [08:06] hehe [08:06] robert_ancell, just looking around or do you plan to work on something there? [08:06] seb128, was looking if any bugs to fix [08:06] and just generally getting up to speed [08:08] robert_ancell, ok, let me know if you want me to recommend you stuff to do on settings [08:09] seb128, the bug list didn't seem to capture all the work to do - is there another list of things to do somewhere? [08:09] robert_ancell, one thing we need is tests [08:09] seb128, autopilot? [08:09] there's a branch for some initial ones [08:09] we should probably review that [08:09] Laney, yeah, I saw that yesterday [08:10] robert_ancell, well, any test would be good :p but yeah, autopilot for the CI would be useful [08:10] Laney, lp:~vrruiz/ubuntu-system-settings/autopilot? [08:10] sounds right [08:10] I wasn't set up to run them so didn't do any review [08:10] Laney, I didn't see it because it was not on https://code.launchpad.net/~system-settings-touch/ubuntu-system-settings/trunk/+activereviews [08:10] feel free if you want [08:11] was when I looked, that's how I found it [08:11] wait what [08:11] it's listed on https://code.launchpad.net/ubuntu-system-settings though [08:11] my URL is different https://code.launchpad.net/ubuntu-system-settings/+activereviews [08:11] oh [08:11] and it's not MPed for u-s-s trunk and doesn't have us as reviewer [08:11] Laney, "lp:~canonical-platform-qa/ubuntu-system-settings/autopilot" is the target vcs [08:11] I guess that's why :P [08:11] yeah [08:12] shouldn't those tests be merged in trunk? [08:12] robert_ancell, otherwise we had workitems in https://blueprints.launchpad.net/ubuntu/+spec/client-s-system-settings-panels ... though we didn't update it recently ... /me does that [08:13] seb128, is there time past final freeze for this, or is this t work? [08:13] robert_ancell, it's mostly "rolling work" from my understanding [08:13] that was what I thought :) [08:13] robert_ancell, we plan to roll regular updates to the phone and upgrades all phones to t automatically after some weeks [08:14] so just get work in the review pipes, that's going to land regularly [08:14] seb128, via the software center / click packages? [08:14] robert_ancell, no, system-images [08:14] that's part of the base image [08:14] seb128, so, does the phone switch to t for building images? [08:14] and we upload new versions to t? [08:15] yes [08:15] aha [08:15] well, I guess we are going to go SRU mode for some week [08:15] then going on T [08:20] robert_ancell, btw, export USS_SHOW_ALL_UI=1 [08:20] robert_ancell, if you want the complete settings UI [08:20] ta [08:20] robert_ancell, we do hide the non working/non stable controls for the release [08:21] seb128: i think dev-channel phone will switch to T straight away, and stable channel will switch to T some time after S release (e.g. sometime 1-3 months into T-cycle) but yeah by christmas all phone update channels will be on T [08:21] seb128, is it one developer per control or everyone working on everything? [08:21] robert_ancell, we sort of have people owning panels [08:21] well "owning" [08:22] but most panels have a main contributor that did most of the work/knows the codebase a bit better/is looking at its issues [08:24] robert_ancell, the one that are mostly free to pick are: phone (though I'm not sure we have enough backend/ofono feature to support that yet), security&privacy, reset, and probably timedate (Laney did most of the work but I think he said he doesn't really wants to own it, he just helped to get things moving) and updates (didrocks did the initial work but he's too busy to work on it since) [08:24] ok, will look at those on monday [08:24] great [08:24] Updates could do with some work [08:25] yes, especially after that release [08:25] it has a new design [08:25] but that came to late to change for saucy [08:25] to merge click and system updates [08:25] I wanted to refactor some of the work into the cpp backend [08:26] didrocks, did you want to keep that one for you or do you think things are too crazy and it makes sense to hand it over? [08:26] at the moment it gets a signal and then munges the arguments in js [08:26] would be nicer to handle that in cpp [08:26] robert_ancell, https://wiki.ubuntu.com/SystemSettings has the designs [08:26] seb128: well, I can continue on that one after this cycle, (I hope) [08:26] didrocks, ok, great ;-) [08:28] hey didrocks, wb and congratulations ;-) [08:29] Laney, does http://paste.ubuntu.com/6217169/ make sense to you? [08:29] Laney: thanks a lot! [08:29] seb128: what's the idea? [08:29] Laney, in the current version both images change the user's background [08:30] Laney, having the homescreen control act on the homescreen :p [08:30] arg, if (x === true)! [08:30] just do if (x) [08:30] both images? [08:30] I guess (I like the verbose way, easier to read) [08:31] Laney, USS_SHOW_ALL_UI=1 system-settings [08:31] Laney, click on the "homescreen" thumbnail [08:31] (well that one is already wrong for me, it displays the fallback image, always) [08:31] can you change the home screen? [08:31] yes [08:31] you wrote that code [08:31] through as [08:31] I thought it didn't work though [08:31] but the code is never called [08:31] no backend [08:31] or nothing reading the setting even [08:31] mterry pinged me yesterday asking why we hide the control [08:32] the greeter supports it apparently [08:32] so I wanted to test [08:32] voilà ! c'est un vieil homme marié ! [08:32] oh [08:32] someone told us it didn't work [08:32] * pitti donne une accolade à didrocks [08:32] and I ran into "the as code is never cold" [08:32] Laney, yeah, that's what I told him ... he said they fixed it since [08:32] so I stopped tracking it when the content hub stuff arrived [08:32] Laney, so communication screwups [08:32] well anyway [08:32] it should work [08:32] so I'm trying to test it [08:33] the way the current code seems to work (not sure I understand it all) is that both selector trigger the same callback when the transfert is done [08:33] and that callback only sets the gsettings key [08:33] see in http://paste.ubuntu.com/6217169/ [08:33] l40 [08:33] * Laney looks [08:34] that's why I'm adding the logic to handle the different cases [08:34] and I added the variable, that I set when you click, to know what image got clicked/what to do [08:37] seb128: Hi Seb! Can you please take a look at bug 1231529 [08:37] Launchpad bug 1231529 in language-selector (Ubuntu) "ubuntu-standard is uninstallable with only security repo enabled : language-selector-common depends unavailable version of accountsservice" [Undecided,Confirmed] https://launchpad.net/bugs/1231529 [08:37] GunnarHj, hey, I just subscribed the security team to it, we need their input [08:40] seb128: Ok, I see. Was it a mistake to include everything up to l-s 0.79.4 in the security fix, or does this type of conflicts happen from time to time? [08:41] GunnarHj, it's on purpose, security needs to be newer in version than -updates, otherwise users with updates wouldn't get the security fix version [08:41] seb128: Hmm.. Sounds logical. ;-) [08:43] seb128: I think it would be best if the onClicked handlers were factored out into a js function which takes a closure that is the action to take when the image is returned [08:44] Laney, want to work on that? ;-) [08:44] can do [08:44] after pping though [08:44] * Laney gets back to that [08:44] Laney, thanks [08:44] Laney, yeah, wait a bit anyway, sil2100 has work in that panel ongoing, it would be nice to not conflict too much with that [08:44] Does it actually work once you set the as property even though? [08:45] well, sil2100 could do that if he wants ;-) [08:46] Laney, not sure if it works, when I stopped yesterday for some reason the as was updating when the gsettings key was changing so I couldn't test them independently (that was on my desktop) [08:46] I'm going to do another round of playing with that [08:46] What's up? ;) [08:47] sil2100, hey [08:47] * sil2100 wants to do everything! [08:47] yeah they're synced on desktop [08:47] Let me backlog once the hangout is over [08:47] we might want to undo that or not have the feature there [08:47] sil2100, read 15 minutes of backlog, we are discussing the background panel [08:48] Laney, they are synced by nautilus right? [08:48] sil2100: the onClicked handlers in the background panel are duplicated and could be factored out; additionally nothing updates the as value (backgroundPanel.backgroundFile) [08:48] not sure [08:48] (e.g if I close that one the syncing should stop) [08:48] Laney, IIRC we made g-c-c set both and we made nautilus react on changes by write the new background to as [08:48] Laney: I actually made a function out of that in one of my merges so that the onClick handlers do not duplicate code [08:49] cool [08:49] well, to make the separate greeter background work it (IMHO) needs to take a closure which tells it what the 'successful' action is [08:49] Laney: but seb128 wasn't sure if it's the place for that, as I did it 'by the way' of fixing another issue, as I didn't want code to be duplicated 3 times already [08:49] s/it/the Connections component/ [08:49] sil2100, that needs a parameter telling what image got clicked so we know what to call [08:50] Laney: but I actually checked unity8 and didn't see any code related to having support for different backgrounds - where should I look for that? [08:50] seb128: ^ [08:50] onClicked: transferContent(function() { background.pictureUri = imageUrl }) for example [08:50] seb128, Laney: since I wanted to see how those 2 different backgrounds are managed [08:50] seb128: yes it's a patch we have in nautilus [08:50] But couldn't find anything in unity8 [08:51] sil2100, one is in the greeter [08:53] seb128: btw. I updated the ubuntu-settings branch to change to phone_wallpaper instead [08:53] sil2100, great [08:53] seb128: you think it's ok to get that released? [08:53] seb128: https://code.launchpad.net/~sil2100/+junk/ubuntu-settings_background [08:53] sil2100, yes, let me have a look [08:54] seb128: since as I see differences between cropping on Mir and surfaceflinger, I guess my idea of using the tablet wallpaper just died ;) [08:56] Laney: I'll be backlogging more after the meeting, but I'm eager with anything related to background - when working on the reset button, I actually wanted to check how the welcome screen changing works, but I didn't see any logic in the code and thought it's a currently-dead-feature [08:56] the backend code is there [08:56] But since you say it's actually not, then I'll do my best to fix it then in another merge [08:57] backgroundPanel.backgroundFile [08:57] setting that should change the value in accountsservice on the device [08:59] Laney: thanks! Looking at that..! [08:59] sil2100, http://paste.ubuntu.com/6217153/ makes it be called [09:00] sil2100, though Laney recommended a slightly nicer way to do that [09:01] seb128: looks a bit hacky, but making sense indeed ;) But if Laney has a fix ready, I guess he can pick it up - I'm always eager, but don't want to duplicate code/steal karma ;p [09:02] not ready [09:02] I just suggested how I would do it [09:02] sil2100, no he doesn't, he just described how he would do it [09:02] you already have the onClicked handlers factored out and my way should be easy to slot into that [09:02] sil2100, I think we all happy if you want to have a look [09:03] you'll probably need to define another function for the backgroundDuplicate case too, and check that in the handler [09:03] I'm not sure how you can give an argument to the callback of the content-hub call [09:03] or something :-) [09:05] anyways, should be sponsoring, I'll let you play with that [09:05] * Laney looks for important-for-saucy fixes in the queue [09:06] Laney, sil2100: yeah, device supports greeter background just fine [09:06] gdbus call --system -d org.freedesktop.Accounts -o /org/freedesktop/Accounts/User32011 -m org.freedesktop.Accounts.User.SetBackgroundFile '/home/phablet/Pictures/Ubuntu.jpg' [09:06] tried with that in adb [09:06] \o/ [09:06] (the image has to exist, and 32011 is my phablet user uid) [09:08] seb128: you think https://code.launchpad.net/~sil2100/ubuntu-system-settings/background_reset_button/+merge/189924 is anyway fine as it is now, or should I fix it to work also for the greeter image? [09:08] sil2100, let me try it, what did you do with the buttons' alignement? did you get a reply from mpt about that? [09:09] seb128: no reply yesterday, so what I did is I left one pair of buttons centered anyway as I actually thought that the greeter image changing was not implemented [09:10] Just made it work with USS_SHOW_ALL_UI=1 [09:11] * mpt finds the question [09:11] That's why I'm asking if it's still feasible since we want to fix the greeter changing - but I guess it should be in another merge [09:12] mpt: hello! Let me prepare a screenshot of how it looks now [09:12] sil2100, is it different from ? [09:13] . [09:13] oops [09:13] mpt: a bit... but this base 'idea' is the same - it's without the thin dividers [09:13] sil2100, https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1237860 [09:13] Launchpad bug 1237860 in ubuntu-system-settings (Ubuntu) "[background] the greeter selector isn't working" [High,Confirmed] [09:13] Laney, ^ [09:13] merci [09:14] mpt: but there's also a different case with the Welcome screen changer enabled [09:14] mpt, well, on that screenshot, did you have an opinion on the alignements of the button's borders and and on the use of separators? [09:15] mpt, the spacing as well [09:15] seb128, yes, delete all the separators :-) [09:15] and agreed with you on the spacing [09:20] * didrocks donne une accolade en retour à pitti [09:42] mpt: as for the size of the button - should I make it be the same as the preview image, or fit the whole width of the settings? [09:42] mpt: the problem with the first approach that the string 'Use Original Background' would have problems fitting in such a small button [09:42] mpt: so maybe just fit the whole width? [09:43] sil2100, agreed [09:48] sil2100, I merged/upload your ubuntu-settings change btw [09:48] uploaded [09:49] seb128: super awesome time! Thanks ;) [09:49] yw ;-) === tkamppeter_ is now known as tkamppeter [10:46] sil2100, good work on the reset background, I just approved it === MacSlow is now known as MacSlow|lunch [11:14] * desrt watches tmobile throw down the gauntlet [11:18] desrt, good morning [11:18] good morning!! [11:20] * desrt makes the coffee extra-strength this morning [11:20] desrt, do you take the train today? [11:20] yup [11:20] cool [11:20] leaving toronto union at 9:30 [11:20] there is a go train arriving 10 minutes before my train leaves... and i don't want to chance it [11:21] Lars is probably over the ocean at this time [11:21] and because it's during the morning rush hour and i live in the ghetto, all of the express trains skip my station [11:21] so i have to take the train _an hour_ earlier [11:21] baaaah [11:21] stop complaining :p [11:21] Lars had to wake up at 4:30am apparently, to take a plane at 6 [11:21] okay. he wins. [11:21] I would call that loose :p but yeah [11:25] didrocks, there? [11:25] didrocks, what is infoMessage in the system-update panel? [11:26] seb128: around, let me check, I think it's the error/info message transmitted by the daemon [11:26] didrocks, oh, the label to display I guess [11:26] didrocks, you don't implement Info() right? [11:27] seb128: hum, I don't even know about that one, so no ;) [11:27] didrocks, ok, good, going to do that [11:27] I'll have a look then, not sure what this is but it seems really interesting ;) [11:27] Are you doing the version display? [11:27] seb128: [11:28] Laney, I was starting looking at it [11:28] Laney, did you say you would do it? I was unsure [11:28] we need to do it both in about and system-update [11:28] I did but it's OK I didn't really start on it [11:29] Laney, I'm happy to let it to you ;-) [11:29] that's alright, you have fun with it :-) [11:30] hehe [11:30] I'll look at the UriHandler after PPing [11:30] do we know any turkish people? [11:30] or any french people? [11:31] desrt, nein [11:31] or germans... [11:31] pitti: are you around? [11:31] desrt, what's up? ;-) [11:31] on the topic of text matching.... [11:31] like say i have sébastian in my contacts [11:31] and i type 'seb' [11:31] desrt: yes (about to have lunch, though) [11:31] should i have had to have typed 'séb'? [11:32] and what if i have mr. übernickel? can i get away with uber, or would it have to be ueber? [11:32] it would certainly be nice to not have to [11:32] Laney, do you have an opinion on how we should get the build number in "about"? dbus or calling system-image-cli --info? (or reading a file on disk) [11:32] ubernickel and uebernickel should both match ideally [11:32] dbus [11:32] Laney, I though you would say that :p [11:32] "ue" is the widely used/accepted umlaut transliteration for ascii-only [11:33] pitti: do you know a good way of doing this asciiification, ideally in the absense of information about the origin language of the string? [11:33] desrt, for sure "seb" should find/match "séb" [11:33] "ubernickel" is wrong, so matching on that is more like a bonus [11:33] desrt: not off the top of my head; I'm fairly sure that this is a common problem though and that there's some library to transliterate Unicode chars into some ascii counterparts [11:33] (fwiw, uebernickel is actually right, but for the purposes of our example...) [11:34] pitti: "bonus" being a good thing? :) [11:34] seb128: It should fit into the Update class fine [11:34] desrt: yes, as I said Ueber is fine, but Uber isn't; even if the US folks think it is :) [11:34] bonus points if you refactor that class to have proper properties [11:34] pitti: well... that's the other thing of course [11:34] desrt: yes, just leaving out all the accents and stuff is fine for searching [11:34] desrt: so I guess it should really match either [11:34] if this is a search algorithm over my contacts, and my friend is german and i am unaware of the german transliteration rules, ... [11:35] Laney, can we use that from the about panel? (e.g import backends from other panels) [11:35] i would probably try 'uber' rather than 'ueber' [11:35] desrt: but shouldn't the general fuzzy matching take care of this already? [11:35] seb128: should be fine [11:35] Laney, great, let me look at that then ;-) [11:35] pitti: we're adding a new stringmatch algorithm and we don't want fuzziness [11:35] (new glib api) [11:35] desrt: what if I write Uebernikel or Uebernickl or some other typo [11:35] there isn't a public interface but it should be ok to use it internally from u-s-s [11:35] pitti: then you lose [11:35] desrt: aah [11:35] desrt: I though that was for HUD or so [11:35] also: must be a prefix match [11:35] no. [11:35] Laney, yeah, I was just unsure it would be in the import path for other panels, let me try [11:36] https://bugzilla.gnome.org/show_bug.cgi?id=709753 [11:36] Gnome bug 709753 in general "Add helpers for string matches when using GtkSearchBar-like widget" [Normal,Unconfirmed] [11:36] any input is welcome here [11:37] desrt: http://search.cpan.org/~sburke/Text-Unidecode-0.04/lib/Text/Unidecode.pm sounds good, so at least there already is some prior art [11:37] * desrt was browsing for an official table file on unicode.org [11:38] didn't find anything :/ [11:39] some discussion here: http://www.unicode.org/mail-arch/unicode-ml/Archives-Old/UML018/0357.html [11:40] more focused on punctuation characters (like removing fancy quotes) than on language :/ [11:40] * desrt takes a look at the perl module [11:40] The Text::Unidecode motto is: [11:40] It's better than nothing! [11:41] * desrt likes these guys :) [11:41] hm, http://www.cattlegrid.info/blog/2009/10/string-normalization-with-unic.html again refers to the perl module [11:41] ooh! [11:41] desrt: [11:41] libunac1-dev - A C programmer's library that removes accents from a string [11:42] desrt: that might just be what you need [11:42] just what i need is a data file in the same format as the normal unicode tables :p [11:42] yes, I meant "for peeking how they do it" [11:42] there is python-unidecode as well [11:43] does glib already speak iconv? [11:43] yes [11:43] $ echo "l'hôpital été Übernickel" | iconv -f UTF-8 -t ASCII//TRANSLIT [11:43] l'hopital ete UEbernickel [11:44] it speaks the C library function iconv() [11:44] (yes, French folks, I forgot an 's' in hôspital) [11:44] i bet there is a way to compel that behaviour via the API, though [11:44] desrt: I'd hope that /usr/bin/iconv is just a shallow wrapper around iconv() [11:44] me too.. [11:45] lots of fun research for the train ride :D [11:45] desrt: iconv_open(3) describes //TRANSLIT anyway [11:45] pitti: thanks for the infos === MacSlow|lunch is now known as MacSlow [11:45] desrt: so it's built into the library, not the CLI frontend [11:45] desrt: that doesn't give you the "Ubernickel" thing, but at least it's close [11:46] desrt: and for any real user search you need "within this hamming distance" anyway for some fuzzyness [11:48] pitti: i disagree... [11:49] we've all survived with tab complete our entire lives, and generally liked it quite a bit [11:49] i think a solid prefix match is all that is required [11:49] * desrt really doesn't like the search behaviour of the dash and hud [11:50] urg? [11:50] I'm glad we have UI that handle spelling mistake [11:50] sometime I don't know exactly how a city, artist, etc is named [11:51] having "you didn't type the exact correct name, there get no result" is not helpful === psivaa is now known as psivaa-afk [13:18] pitti: i think this is going to work very nicely [13:20] desrt: \o/ [13:20] pitti: i think the decoding is baked-in to the C library.. i can't see any data file that it opens that could possibly contain the information [13:21] but meanwhile, i read that ö -> oe is not the correct thing to do for scandinavians :/ [13:51] bah [13:58] does anyone know the QDBusReply to use for a return type of "isss{ss}"? [13:59] is that a QVariantList? [14:01] Laney, didrocks, Saviq: ^ does any of you know? [14:01] maybe QDBusVariant [14:02] or http://qt-project.org/doc/qt-5.1/qtdbus/qdbusargument.html [14:02] (haven't used either) [14:13] Laney, I tried QDBusVariant but reply.isValid() is invalid [14:13] I wonder if that's the type or if I'm doing something else stupid [14:14] Laney: for the love of god, don't use QDBus [14:14] it's the one dbus binding on earth that actually manages to be worse than dbus-glib [14:15] it's usually working quite fine/easy to use [14:15]