| antis | Hi, occasionally phone calls on FP2 enable the "on hold" mode during a call. Is this an issue to be fixed in the upcoming version? However with SF3 this behaviour has improved a lot and that's great! | 00:23 |
|---|---|---|
| *** feodoran_ is now known as feodoran | 03:14 | |
| *** feodoran_ is now known as feodoran | 12:21 | |
| tortoisedoc | lbt : got another problem (no log however as I triggered a new build) : /tmp partition seemed to be full, is that possible? | 12:35 |
| lbt | that would be OOM | 12:35 |
| tortoisedoc | lbt : oho | 12:36 |
| lbt | what package? | 12:36 |
| tortoisedoc | same rust package | 12:36 |
| tortoisedoc | https://build.merproject.org/package/show/home:tortoisedoc:branches:home:sfietkonstantin:sailfish:rust/rust-1.27.2 | 12:36 |
| tortoisedoc | i tried a df to see if i could see the size of /tmp but failed :P | 12:36 |
| tortoisedoc | (the build before that has the issue, in case you have access to it) | 12:36 |
| lbt | it's typically tmpfs | 12:36 |
| tortoisedoc | lbt : okay, are there hard limits? | 12:37 |
| lbt | yeah - but it's multi-Gb so I'm surprised | 12:37 |
| lbt | I wonder about power workers | 12:39 |
| tortoisedoc | lbt : yes me too; it seems however rustc can be quite demanding in terms of AST complexity (sourced from different rust devs); I don't know at this point if its more of a bug | 12:39 |
| tortoisedoc | power workers? | 12:39 |
| lbt | not sure they're setup any more... I'll have to look | 12:42 |
| * tortoisedoc lost | 13:03 | |
| tortoisedoc | whats the difference between a power worker and a non-power worker? | 13:03 |
| r0kk3rz | moar powaaa | 13:03 |
| dcaliste | jpetrell, pvuorela: I mentioned to chriadam some weeks ago having random issues with lipstick-ui-security. Once in a while, it gets stuck and password windows are not appearing anymore. | 13:22 |
| dcaliste | It's happening again, and also in gdb, I've got some traceback. | 13:22 |
| dcaliste | Are you aware of this problem, how can we proceed ? Since latest cbeta, it seems more easy to reproduce. Besides, thanks jpetrell for the display to always show the bottom part of the window, it's nice. | 13:23 |
| pvuorela | dcaliste: i'm not aware of much or any on that, but tracebacks can always help :) | 13:47 |
| dcaliste | pvuorela: yeh, but I don't install debug symbols and issue seems to be related to wayland and egl… | 13:48 |
| pvuorela | dcaliste: um, right. | 13:49 |
| dcaliste | pvuorela: see https://bitbucket.org/snippets/dcaliste/LeaArX | 13:50 |
| dcaliste | I've stopped lipstick-ui-security via systemctl | 13:51 |
| dcaliste | Them run one instance in gdb | 13:51 |
| dcaliste | With gpg2, I request a password. | 13:51 |
| dcaliste | The window open and I can get my password, but when the window disappears, the crash is reported in gdb. | 13:52 |
| dcaliste | At another moment, it didn"t crash but stop with the following warning: | 13:53 |
| dcaliste | [W] unknown:0 - The Wayland connection broke. Did the Wayland compositor die? | 13:53 |
| dcaliste | Using the same procedure as described before. | 13:53 |
| jpetrell | dcaliste: could you file a bug in internal bugzilla, add relevant info. and add denexter as cc | 14:01 |
| dcaliste | jpetrell: sorry, no I don't have any access to the internal bugzilla :( I can provide as many info to reproduce or that I gathered in the bitbucket snippet mentioned before though. | 14:02 |
| jpetrell | dcaliste: ah ok. well that works too | 14:02 |
| dcaliste | pvuorela, jpetrell, I've updated the snippet https://bitbucket.org/snippets/dcaliste/LeaArX with as much information as I have. The crash is not reproductible at will :( If you can ping denexter internally, that would be nice. | 14:22 |
| tortoisedoc_ | lbt : here we go the same OOM error :https://build.merproject.org/package/live_build_log/home:tortoisedoc:branches:home:sfietkonstantin:sailfish:rust/rust-1.27.2/sailfishos_sfiet_i486/i586 | 17:51 |
| attah | Are there some good play/pause, progress etc qml widgets that one could steal from somewhere? | 18:57 |
| attah | Is it impossible to inline-define an object with multiple properties in a function call? | 19:46 |
| attah | and/or how do i pass my delegated item to the next page? | 19:47 |
| tortoisedoc_ | attah : what do you mean with "in a function call"? As a factory? | 19:51 |
| attah | constructed with just curlys inside a something() | 19:51 |
| attah | it seems comma is the key | 19:52 |
| attah | d'oh | 19:52 |
| attah | i was going for semicolon between properties | 19:52 |
| tortoisedoc_ | oh heh :) | 19:53 |
| attah | consistency is overrated apparently :) | 19:53 |
| attah | any ideas on the other question? i have my BackgroundItem{id: delegate}, how would i pass that to the pagestacker? | 19:54 |
| attah | i want a handful of its attributes really.. | 19:55 |
| tortoisedoc_ | hmhm I might be wrong, but arent qml objects global until their parents disappear | 19:56 |
| tortoisedoc_ | coderus : *summons* | 19:56 |
| attah | mayhaps... so how do i *refer* to it | 19:57 |
| tortoisedoc_ | actually, refcounts in qml *should* keep them alive | 19:57 |
| tortoisedoc_ | okkam's razor : with the id? | 19:57 |
| attah | well, it's "delegate"... i have the impression that's special | 19:58 |
| tortoisedoc_ | okay | 19:59 |
| tortoisedoc_ | http://doc-snapshots.qt.io/qt5-5.11/qtquick-modelviewsdata-modelview.html <- this doesnt mention any cross-qml references (tl;dr;) | 19:59 |
| tortoisedoc_ | (at least not at first sight) | 19:59 |
| tortoisedoc_ | attah : out of curiosity, why would you like to refer the same delegate from different components (pages)? | 20:00 |
| attah | i really wanted to pass all parameters from whatever list item i click down to the next page | 20:01 |
| attah | not too picky about how... but typing them out feels inconvenient | 20:01 |
| tortoisedoc_ | hmm store them in a json object on a js backend store? | 20:02 |
| tortoisedoc_ | you now, the usual MVC voodoo :) | 20:02 |
| tortoisedoc_ | *know | 20:02 |
| attah | meh :P | 20:02 |
| tortoisedoc_ | :D | 20:02 |
| tortoisedoc_ | why have it simple, whe you can make it complex | 20:03 |
| tortoisedoc_ | but keep in mind, less is more! | 20:03 |
| attah | i'm on a semi-involuntary Android detox | 20:03 |
| tortoisedoc_ | ah, a padawan on the way to the bright side of the force, i see! | 20:03 |
| attah | or i should say Alien Dalvik detox | 20:04 |
| tortoisedoc_ | hmmm | 20:04 |
| attah | i really only need a few apps, and i can write at least 2 of them myself | 20:04 |
| tortoisedoc_ | ive always wondered if qml was qt's biggest achievemnt or the biggest disappointment | 20:05 |
| tortoisedoc_ | on one way, its a great tech | 20:05 |
| tortoisedoc_ | on the other, its fragmentation at it's purest | 20:05 |
| attah | it desperately needs a new programming style | 20:05 |
| tortoisedoc_ | Jobs got it better IMO | 20:05 |
| attah | i have never owned an Android phone privately... first smartphone was a N9 | 20:06 |
| coderus | tortoisedoc_: sup? | 20:06 |
| tortoisedoc_ | by clearly separating structure and behaviour | 20:06 |
| attah | Elm for QML would be neato /methinks | 20:06 |
| tortoisedoc_ | coderus : see above attah's question about how to create an object from a function | 20:06 |
| tortoisedoc_ | qml is much clearer than the bs xml crap you need to write down in ios | 20:07 |
| attah | remaining question: can i pass my whole delegated list item to the next page? | 20:07 |
| tortoisedoc_ | but hear hear, they wanted more so they forged javascript (= control) in the view | 20:07 |
| coderus | if you pageStack.push(page, {prop1: someobject.somevalue}) this value be copied | 20:07 |
| coderus | attah: you cant copy item, it is passed by pointer | 20:08 |
| coderus | as pointer | 20:08 |
| attah | well, can i point back to it? | 20:09 |
| attah | no? | 20:09 |
| coderus | attah: well, sure, you can interact with it until it dies :) | 20:10 |
| coderus | if you push, not replace, it should still alive | 20:10 |
| attah | interesting :) but how to refer to it? | 20:10 |
| coderus | attah: mmh? | 20:11 |
| coderus | property var ? | 20:11 |
| coderus | it always works :) | 20:11 |
| attah | and there is nothing special with it being named delegate? | 20:12 |
| coderus | attah: what you talking about? | 20:12 |
| coderus | you should pass item by id | 20:12 |
| attah | i have a BackgroundItem{id:delegate} in my list | 20:13 |
| coderus | yep | 20:13 |
| attah | if i click that i push a new page, on which i want to handle many of the attributes on said BackgroundItem | 20:13 |
| coderus | good | 20:14 |
| tortoisedoc_ | grrr | 20:15 |
| tortoisedoc_ | is there a reason the ld.so is stripped in mer? | 20:15 |
| attah | still not working as i would have thought | 20:16 |
| tortoisedoc_ | this ld.so is more of a burden that a good tool really :S | 20:16 |
| coderus | attah: why? | 20:16 |
| * tortoisedoc_ throws ld.so in the trashcan | 20:16 | |
| attah | maybe the properties are not *on the BackgroundItem* but loose somehow | 20:16 |
| attah | the model makes them | 20:17 |
| coderus | attah: what props you looking for? | 20:17 |
| coderus | if you want model, then pass model | 20:17 |
| tortoisedoc_ | coderus : how much do you know of sb2? | 20:18 |
| coderus | tortoisedoc_: less than i want to :) | 20:18 |
| tortoisedoc_ | :D | 20:18 |
| attah | but how do i get *this* list item? | 20:18 |
| tortoisedoc_ | mal : do you know if an unstripped version of ld.so is available somewhere+ | 20:18 |
| coderus | attah: pass both item and model :) | 20:19 |
| coderus | attah: tell me what you want | 20:19 |
| attah | i'm trying! | 20:19 |
| attah | i ahev say, variables title and url accessible (and different) in all list items | 20:20 |
| coderus | pageStack.push(page, {item: delegate, model: model}) | 20:20 |
| attah | i want to bulk-pass them on | 20:20 |
| attah | but the stiff that arrives from what i bound delegate to doesn't work | 20:20 |
| coderus | attah: show me how you calling pagestack push and how you using it on that page | 20:21 |
| attah | https://pastebin.com/keQWuVA8 | 20:22 |
| coderus | pass model | 20:23 |
| coderus | pageStack.push(Qt.resolvedUrl("PlayPage.qml"), {model: model}) | 20:23 |
| coderus | text: model.title | 20:23 |
| attah | Unable to assign [undefined] to QString | 20:26 |
| coderus | attah: you did mexactly?:) | 20:26 |
| attah | model: programs.model | 20:27 |
| coderus | no | 20:27 |
| attah | pageStack.push(Qt.resolvedUrl("PlayPage.qml"), | 20:27 |
| attah | {model: programs.model}); | 20:27 |
| coderus | exactly model: model | 20:27 |
| coderus | :) | 20:27 |
| attah | text: model.title | 20:27 |
| attah | (on the other page) | 20:27 |
| attah | property var model | 20:28 |
| attah | ^defined as so | 20:28 |
| coderus | please do exactly pageStack.push(Qt.resolvedUrl("PlayPage.qml"), {model: model}) | 20:28 |
| attah | wow, that's weird | 20:29 |
| attah | ...and it works | 20:29 |
| coderus | :) | 20:29 |
| attah | who bound that? xD | 20:29 |
| coderus | inside delagete item "model" is a context propery pointing to current item model data | 20:30 |
| coderus | you can omit it, and delegate is trying to get undefined values from that "model" property | 20:30 |
| attah | nteresting | 20:31 |
| attah | this needs a programming paradigm overhaul | 20:31 |
| attah | but the core concepts seems to check out nicely... kinda like git's user interface | 20:31 |
| attah | coderus: thanks btw :) | 20:38 |
| coderus | 😎 | 20:51 |
| tortoisedoc_ | has anyone succeeded in using valgrind in mer sb2? | 21:04 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!