=== philipballew is now known as philip === chriadam is now known as chriadam|away [08:42] good morning [08:44] morning all [08:45] dpm: hi! Do you think you can try the same .service experiment on the phone? [08:46] morning mardy, sure. Was the result on the desktop what you were expecting? [08:46] dpm: yes [08:46] cool [08:47] on it, testing it on the phone as soon as unity8 recovers :) === dholbach_ is now known as dholbach [08:53] mardy, good news: it works on the phone too :) [08:53] mardy, quite similar steps as on the desktop: [08:54] 1. Deleted old Evernote account on system settings [08:54] 2. Updated the .service file with the new API key [08:54] 3. Added a new Evernote account on system settings [08:54] 4. Started app, tapped on the account to authorize it [08:55] 5. Got a prompt from system settings to authorize (Evernote's embedded login page) [08:55] 6. The app resumed and communicated to the Evernote service, fetching notes as expected [08:56] dpm: cool! Then I think that this is the way to go! [08:56] dpm: then you should reduce the scopes (permissions) for the key in the .provider file [08:58] mardy, great. Time for a quick hangout? I'd like to understand what exactly we need to change both in the auth plugin and in the app itself (if we have to change anything in there at all) [08:58] dpm: sure [08:58] thanks, let me set that up [09:27] Good morning all; happy Shopping Reminder Day! :-D [09:30] dpm: o/ [09:31] morning mzanetti! [09:31] dpm: good morning [09:31] dpm: do you think this is good enough for a start? https://code.launchpad.net/~mzanetti/reminders-app/html2enml/+merge/196616 [09:31] * dpm looks [09:32] dpm: mzanetti I tried evernote this morning, and signed into sandbox via the accounts plugin (but it didn't show in the list of accounts after I had authorised), and in reminders doesn't show my account in "select evernote account" [09:33] popey: hmm... that sounds like an issue with the accounts plugin [09:33] possibly, yes [09:34] popey: I don't know much about that yet tbh... Maybe mardy or cwayne can help you [09:37] popey: desktop of phone? [09:38] phone [09:38] popey: can you see the account with "account-console list"? [09:38] no [09:39] but I got an email from evernote [09:39] "You have authorized Evernote app on Ubuntu to access your Evernote account with these permissions" [09:39] popey, did you install both the authentication plugin package and the updated signon package from the PPA? [09:40] * popey checks [09:40] ah, not since updating.. drat [09:40] phew [09:45] mzanetti, your MP looks like the way to go to me for now. In the future though, we'll need to display images and tables, and I understand that the TextArea won't help us there. What do you think we should eventually do? Use a WebView? [09:49] nerochiaro, hey, I updated https://code.launchpad.net/~osomon/notes-app/robust-test-server/+merge/196174 according to your suggestion [09:50] oSoMoN: ok, looking into it [09:52] nerochiaro, thanks, after that one is merged, it would be good to request a release of notes-app, hopefully this will improve the test results in the QA dashboard [10:01] dpm: mzanetti added ppa, re-authed and re-ran the app.. http://popey.com/~alan/phablet/device-2013-11-26-100108.png [10:01] popey: yep. this is known [10:01] ah k [10:02] popey: it's on my todo to enable the accounts plugin fetching the account name [10:02] popey: should still work, just that the text is not there [10:02] oh, i didnt realiser there was actually an entry there [10:02] :) [10:02] oSoMoN: approved [10:03] om26er: hello. seen my email ? [10:04] nerochiaro, thanks [10:13] mzanetti: http://popey.com/~alan/phablet/device-2013-11-26-100532.png [10:13] not showing the single note I have - "Getting Started" [10:13] popey: hmm... can you paste the debug output? [10:15] mzanetti: https://pastebin.canonical.com/101046/ [10:15] phablet@ubuntu-phablet:~$ account-console list [10:15] account: id 4, enabled, provider: evernote [10:22] nerochiaro, right, so I think you might need to talk to vruiz on how they are using autopilot to test webapps. [10:28] om26er: it's not a webapp problem. it's a very plain QML problem: how do you access in autopilot a component instantance when this instance is dynamically created by a 3rd party component you don't control [10:29] om26er: if i had autopilot vis working and not crashing I would probably be able to just go through the tree and find it and then figure out a way to do the same from the test code. [10:29] om26er: but vis is currently broken, so i can't do that [11:31] popey, could you review https://code.launchpad.net/~dpm/reminders-app/fix-dependencies/+merge/196696 ? It should fix the issue you had earlier on whereby you had to install the dependencies for authentication manually [11:31] sure [11:32] super, thanks === _salem is now known as salem_ [12:11] om26er: mzanetti: any news on that saucy vesrsion of AP that will allow me to run tests using ubuntuuitoolkit on the desktop ? [12:13] nerochiaro_, isnt that the case already [12:16] om26er: i'm updating now, didn't notice there was a release. i'll let you know in a miniute [12:20] popey, did you have any luck getting the notes displayed? Looking at the log, it seems authentication with the service worked, but not the actual notes fetching [12:21] dpm: nope [12:23] popey, which branch are you using? [12:23] i was using the mzanetti branch with the script in [12:23] om26er: no, it's not working. ubuntu-ui-toolkit-autopilot needs python-autopilot > 1.4, but python-autopilot > 1.4 isn't compatible with libautopilot-qt which is also needed. I can't have both installed. so I can't install ubunut-ui-toolkit-autopilot [12:23] brb -> post office [12:24] om26er: i mentioned this yesterday already, with mzanetti too [12:24] who's working on it ? [12:25] * mzanetti is not === MacSlow_ is now known as MacSlow|lunch [12:29] om26er: are you working on it or planning to ? === MacSlow is now known as MacSlow|lunch [12:41] nerochiaro_, right, I left a bit early yesterday. Will take that to autopilot devs today. Can you file a bug for the packaging issue so that we can track [12:43] dpm: should I have used a differnt branch? [12:46] om26er: ok, against what package in lp ? [12:46] nerochiaro_, report for upstream autopilot-qt [12:46] nerochiaro_, https://bugs.launchpad.net/autopilot-qt/+filebug [12:47] popey, no, that one should have worked already (or at least worked for me when I tried), it's just that in the latest tests I did, I locally merged all pending branches. Let me try again with a clean checkout of the same branch you're using [12:47] dpm: i wiped my phone to #31 so will need to re-setup to test again later [12:50] om26er: https://bugs.launchpad.net/autopilot-qt/+bug/1255077 let me know if you need more info [12:50] Launchpad bug 1255077 in Autopilot Qt Support "Dependency conflict for ubuntu-ui-toolkit-autopilot on Saucy" [Undecided,New] [12:50] What's the "proper" name for the Ubuntu OS on touch-enabled devices, the one we released last month? Is it still Ubuntu Touch, or is it Ubuntu for Phones now? Or is it something else that I missed? [12:51] both [12:51] We still call it Ubuntu Touch all over the wiki and in various other places. [12:52] but the website calls it "Ubuntu for phones" (not "Ubuntu for Phones") [12:54] Hrm, the capitalisation aspect introduces another variable I hadn't taken account of - with a capital P it looks like a formal brand ("Ubuntu for Phones"), but with a lower-case p it looks more like a qualifier ("Ubuntu" for phones). [12:54] I'll stick with Ubuntu Touch for now, I think. [12:57] om26er: also any news on how to interact with other apps in AP integration tests ? interact in the sense of finding a button in the other app's tree and clicking it [12:58] nerochiaro_, sorry missed that. I was working on that on friday but got distracted with other work. I can give you a sample by the end of day [12:58] om26er: thank you,that would help a lot === MacSlow|lunch is now known as MacSlow === ara is now known as Guest1754 [13:32] Kaleo, hey, do you know someone who can help me with animating transition inside of a listview? [13:32] I have a listview item that expands by loading a component inline [13:32] and I can't figure out how to tell the items in the listview to animate their transition [13:32] rickspencer3, nerochiaro_ has some experience with it I believe [13:32] they don't seem to get onYChanged [13:32] rickspencer3, though Saviq is probably the master :) [13:33] and Behvior on y doesn't do anything either :/ [13:33] ah [13:33] hey Saviq thoughts? [13:33] rickspencer3, do you have a code sample? [13:33] rickspencer3, what are you after? [13:34] rickspencer3, we've an Expandable coming into the SDK soon https://code.launchpad.net/~mzanetti/ubuntu-ui-toolkit/expanding-listitem/+merge/195602 [13:34] rickspencer3, which implements the expansion pattern https://docs.google.com/a/canonical.com/document/d/1_c4029C6Jwll_ng8gqp3SRnF4DgXRA3tR_LL3J-zOD8/edit [13:34] mu listview starts at line 223 [13:34] http://bazaar.launchpad.net/~rick-rickspencer3/+junk/DCBusit/view/head:/DCBusit.qml [13:35] Saviq, ok, so the expansion pattern is what I am after [13:35] but when it expands, I can't make the items animate [13:35] so it's not smooth at all [13:35] just BAM [13:35] :) [13:35] rickspencer3, you just need a Behavior on height of the item that expands [13:36] rickspencer3: what Saviq said [13:36] * rickspencer3 tries === ara_ is now known as ara === ara is now known as Guest55288 [13:37] Saviq, nerochiaro_ ok, closer [13:37] ! [13:37] the problem is that the other list items don't move at the same rate, and there are some weird overlaps [13:37] should I just tweak the duration of the PropertyAnimation? [13:38] rickspencer3, weird, something else must influence it [13:39] rickspencer3, by default a ListView would just push other items down all in concert [13:39] Saviq, well, I think the items moving out of the way must have some default animation [13:39] like a NumberAnimation [13:39] they are columns though [13:39] rickspencer3, they don't, it's all driven by the height of the expanding one [13:39] oh, I see what's happening [13:39] rickspencer3, ListView just stacks items on top of one another [13:39] Saviq, so the height changes several times [13:40] rickspencer3, so if one of them grows, the rest are pushed down [13:40] like once when it loads the image and then again when it loads the label [13:40] so I just need to figure out how to set the height only once in the component [13:40] rickspencer3, that's possible, yes [13:40] I think it's the missing clipping on the column [13:40] Saviq, well, I slowed it way down and I can see what's happening [13:40] rickspencer3, you can also use SmoothedAnimation on the height, that should deal ~ nicely with multiple height changes [13:41] * rickspencer3 tries [13:41] rickspencer3, and what mzanetti said - you need clip: true on the item [13:41] rickspencer3: wrap the Column into an Item, set clip: true on that item and make it height: column.height [13:41] hey! [13:41] now we're talking [13:41] mzanetti, that shouldn't be necessary, why wouldn't you just apply a Behavior directly on the column? [13:42] mzanetti, and clipping, too? [13:42] rickspencer3, btw, there's a *lot* of indentation going on there, you should think of splitting stuff into smaller chunks [13:42] the clip: true did it, i think [13:42] Saviq, can you explain more? [13:43] oh, in my code, you mean? [13:43] yeah, I'll refactor it into comonents [13:43] components [13:43] rickspencer3, yeah, that [13:43] I've just totally reworked the UI based on some design feedback [13:43] ah right... the wrapping might indeed not be necessary [13:43] so, it's all over the place now [13:43] rickspencer3, about tr.i18n().replace [13:43] mzanetti, it was the clipping ;) [13:44] * rickspencer3 braces [13:44] rickspencer3, Qt's way is "%1 %2".arg('one').arg('two') === Guest55288 is now known as ara_ [13:44] rickspencer3: you would be welcome to use the expanding listitems branch and give some feedback [13:44] mzanetti, well, my code is working, but if you want me to try it out, I will [13:45] rickspencer3, predictionsList.model[index]["RouteID"] should be unnecessary, just "RouteID" should be available, assuming the model exposes that as a role [13:46] rickspencer3, so: [13:46] i18n.tr("{{t}} minutes until bus {{rt}} to {{dir}} in ").replace("{{rt}}",predictionsList.model[index]["RouteID"]).replace("{{dir}}",predictionsList.model[index]["DirectionText"]).replace("{{t}}", predictionsList.model[index]["Minutes"]) [13:46] should be: [13:46] rickspencer3: it also adds some convenience stuff. like make the item scrollable if the expanded height is too big to fit one the screen, scroll it into the visible area on expansion etc. So if you face such issues you know where you find the fixes :) [13:47] mzanetti, maybe I'll try it later [13:47] i18n.tr("%1 minute until bus %2 to %3 in ").arg(Minutes).arg(RouteID).arg(DirectionText) [13:47] after I reconstitute the rest of my app [13:47] Saviq, cool, thanks [13:48] Saviq, is there a replace() in there/ [13:48] ? [13:48] rickspencer3, no need [13:48] coolio [13:48] rickspencer3, .arg() replaces %1 etc. with its argument [13:48] yeah, I get it [13:48] weird, but cool [13:48] :) [13:49] every environment has it's own system for this, not sure that any has really improved on C that much ;) [13:49] rickspencer3, also, you should use the plural variant of i18n [13:49] what is the plural variant? [13:49] rickspencer3, i.e. i18n.tr("%1 minute...", "%1 minutes...", Minutes).arg(Minutes)... [13:50] rickspencer3, which will display "1 minute...", but "2 minutes..." [13:50] wow, that looks nicely complicated lol [13:50] I'll give that a try [13:50] maybe write a little post about it [13:50] rickspencer3, it doesn't when you try to do that in Polish ;) [13:50] hehe [13:50] rickspencer3, we have three different versions ;) [13:51] nice [13:51] rickspencer3, singular, then 2-4, then 5-10 [13:51] rickspencer3, so no amount of concatenation of "minute" + "s" works for us :) [13:51] Saviq, Poles don't count past 10? [13:51] lol [13:51] rickspencer3, nah, it's just "lots" by then [13:51] :) [13:52] rickspencer3, here's the rule: [13:52] nplurals=3; plural=(n==1 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2); [13:52] ;) [13:53] rickspencer3, and if you want to look at them, here's the set for different languages http://docs.translatehouse.org/projects/localization-guide/en/latest/l10n/pluralforms.html [13:53] Saviq, I'll just use your suggestions [13:53] rickspencer3, sure :) [13:53] anyway, thanks for your help you guys [13:54] I just need to clean up the code and add a few features back, and I have a nicer version [13:54] :) [14:10] popey, try to pull from reminders-app trunk now: a bunch of branches have landed, see if it improves things [14:11] dpm: will do when I get back, need to go afk for a bit. need to finish dogfooding [14:11] ack === ara_ is now known as ara [15:11] hi nik90, around? Just a quick question: is there anything blocking the alarms feature in the clock app right now from the qt eds plugin side of things? [15:31] Can anyone give me some hints with Ubuntu.Content? I'm trying the sample in the docs and after selecting the photos the app freezes/goes blank without a reason [15:32] kenvandine, perhaps? ^ [15:33] randomcpp, i can try? [15:33] randomcpp, is this on the device or your desktop? [15:33] on device [15:33] on desktop Ubuntu.Content does nothing [15:33] and what kind of content are you requesting? [15:33] randomcpp, it can, if you run it right :) [15:33] but easier on the device [15:34] http://paste.ubuntu.com/6479511/ [15:34] ignore qtlocation [15:35] randomcpp, and how are you running your app? starting it from unity8? [15:35] ctrl+f12 from qtcreator :| [15:36] zsombi, do you know if that runs it with all the right APP_ID info? [15:36] or anyone from the sdk team? ^^ [15:37] randomcpp, for the content hub to work, it really relies on the environment click provides [15:37] we need to make sure that works when run from qtcreator [15:37] kenvandine: eventually Kaleo might know, it wasn't made by us, haven't checked what it requires to get it working... [15:37] but i haven't tried myself [15:37] ah [15:37] Kaleo, ^^ [15:37] i know he is a bit stuck atm :) [15:38] I try creating a click pkg with content_exchange policy [15:38] good [15:38] that is going to be more reliable [15:38] I cannot help now trusty laptop broken [15:38] imo qtcreator should do that to run with ctrl+f12 [15:39] Kaleo, i heard... sorry :/ [15:48] kenvandine, ok with "Install application on device" seems to work [15:53] how can I uninstall click pkgs now? :) [15:58] with pkcon [15:59] randomcpp, i don't recall exactly atm [15:59] pkcon remove APP_PKG_ID [15:59] maybe [16:03] :/ [16:28] kenvandine, do you knows how to read debug outputs when an app is running on device? [16:44] dpm: nothing from the clock side..just the qt eds plugin side of things [16:44] dpm: actually more specifically the indicator-datetime package [16:52] randomcpp, look in ~/.cache/upstart for logs [16:52] randomcpp, the log file will be application-click-APP_ID [16:53] here they are, thank you kenvandine :) [17:01] randomcpp, np [17:08] dpm: what is the copyright to be used in reminders-app? GPLv3 Canonical Ltd.? === greyback is now known as greyback|away [17:46] mzanetti, exactly [17:47] dpm: https://code.launchpad.net/~mzanetti/reminders-app/fix-copyright-headers/+merge/196757 [17:47] dpm: tbh I haven't really ever seen a bash script with a copyright header [17:48] :) [17:54] popey, mzanetti, could one of you review and top-approve if everything is ok? https://code.launchpad.net/~dpm/reminders-app/fix-dependencies/+merge/196696 [17:54] dpm: done. I was just waiting for jenkins to give green light [18:00] excellent, thanks! [18:11] dpm: I'm trying to reenable ssl. which lib did you install to get away with the unresolved symbol GENERAL_NAME_free? [18:12] mzanetti, I didn't install any particular library, I had to add explicit includes to libssl, which was a bit weird, since I think they're redundant. Let me check the commit where I added them to see where they were added [18:14] mzanetti, http://bazaar.launchpad.net/~dpm/reminders-app/evernote-api-plugin/revision/5 [18:14] dpm: hmm... that's in here but I still get the failure [18:20] mzanetti, I guess you've taken trunk and started by setting use_SSL to true ? [18:20] dpm: yep [18:20] * dpm tries that [18:20] dpm: in notesstore.cpp, not userstore.cpp [18:21] but shouldn't make a difference with this error [18:21] indeed, but notesstore.cpp is the one I was looking at in any case [18:23] hm, getting the same error now: ""Evernote": Cannot load library /home/dpm/dev/coreapps/build-reminders-app-Desktop-Debug/src/plugin/Evernote/libevernoteplugin.so: (/home/dpm/dev/coreapps/build-reminders-app-Desktop-Debug/src/plugin/Evernote/libevernoteplugin.so: undefined symbol: GENERAL_NAME_free)"" [18:43] mzanetti, this won't solve the "undefined symbol" error, but it might offer a way to getting SSL to work (right now we seem to be having the same "SSL_connect: certificate verify failed" as them) http://discussion.evernote.com/topic/39169-userstore-connecting-error-ssl-connect-certificate-verify-failed/?p=212785 [18:44] in any case, one thing that we seem to be missing is the loading of the trusted certificates: https://github.com/apache/thrift/blob/master/lib/cpp/README.SSL#L26 [18:45] I see [19:00] mzanetti, I've filed bug 1255234 with some more background after digging a bit deeper. I'm not an expert on SSL, but I think this might point us in the right direction [19:00] Launchpad bug 1255234 in Ubuntu Reminders app "[Evernote plugin] Enable SSL communication with the Evernote service" [High,Triaged] https://launchpad.net/bugs/1255234 [19:00] dpm: cool. I'm still stuck at the linking :/ [19:01] dpm: according to some folrum threads I think the onnection drops might go away once we use ssl [19:01] mzanetti, yeah, this bug with the undefined symbol is extremely annoying, I've no idea what to look at :/ [19:01] that sounds promising === salem_ is now known as _salem [20:58] why is http://developer.ubuntu.com/api/qml/sdk-1.0/Ubuntu.Content.ContentItem/#name-prop always empty? [20:59] url isn't empty, but it's an object without properties or functions [22:31] popey, ping xD [22:31] randomcpp: yo [22:32] that package I've uploaded doesn't work [22:32] oh? [22:32] or better, it works, but it has a bug [22:32] thats unfortunate ☻ [22:32] yeah I know..I'm wasting a lot of time because of it [22:33] because Ubuntu.Content copy the files I request in a .cache directory [22:33] but those file are deleted eventually [22:34] so I need to copy (move doesn't work) the contents to .local/share// dir [22:35] to have them accessible the next time [22:35] I'm thus creating a armhf chroot [22:35] and it takes ages to download all the packaged [22:35] packages* [22:45] popey, should I remove accounts and sharing features? [22:45] personally I would, I expect that infrastructure to change this cycle [22:45] ook [23:14] popey, do you have some knowledge of Ubuntu.Content [23:14] ? [23:14] nope [23:14] or at least how it works [23:15] because I can choose pictures, but then I can't do anything with them, [23:15] only show them in the app, as long as they are saved in .cache folder [23:20] ok goodnight popey :)