ajalkane | Anyway... QML was a pain for me at first. But once you learn the restrictions and how it does what it does, it becomes quite pleasing in many ways. | 00:00 |
---|---|---|
djszapi | too bad I need C++ code for many fundamental functionalities : | 00:00 |
djszapi | ;) | 00:00 |
djszapi | I do not see how it is pleasing to mess up with C++ code if you could add /one/ word in the qml bits. | 00:01 |
ajalkane | I don't think you need any C++ for what you want to do. If your anagram model is for example ListModel, it has all the functions you need already | 00:01 |
djszapi | not ListModel. | 00:02 |
djszapi | this is also sad, the repeater's model does not respect letter list :) | 00:02 |
ajalkane | What is it then? | 00:03 |
djszapi | I need to use QStringList for letter list (a.k.a. string) | 00:03 |
ajalkane | You need to create a subclassed version of that then. No way around it | 00:03 |
ajalkane | And provide some functions QML can access to it | 00:04 |
djszapi | subclassed version of what and why ? | 00:04 |
ajalkane | Because you need to access elements by index | 00:04 |
ajalkane | And for that you need to provide functions that can be called from QML | 00:05 |
djszapi | for not exploiting your time: I know how to do things C++ since you can do anything from there. My original question was more like about doing it nicely. | 00:06 |
djszapi | but apparently bugreport is the maximum I can do in that respect. | 00:06 |
ajalkane | indeed, if you want to use a C++ class as a model | 00:06 |
*** artemma has quit IRC | 00:07 | |
djszapi | on a side note: it is also not possible from qml to nicely extend or truncate the stringlist model either | 00:08 |
djszapi | so either adding a simple element in there, or just removing one. | 00:08 |
djszapi | anyway, thanks for your help. :D | 00:09 |
ajalkane | no. QStringList does not have QML accessible functions as far as I know | 00:09 |
djszapi | and that is pretty sad especially because they represent the repeater model with stringlists in the example code ... | 00:10 |
djszapi | I mean it is weird you cannot cooperate with such a basic type like a list in qml. | 00:11 |
*** artemma has joined #harmattan | 00:14 | |
ajalkane | Yes, I've been frustrated with that too | 00:16 |
ajalkane | Especially with QStringList | 00:16 |
ajalkane | You know, just wanting to show in a dialog choices that are populated by QStringList? Doesn't really work. | 00:17 |
djszapi | heh :D | 00:17 |
ajalkane | You're saving a lot of time by just going ahead and creating a QAbstractListModel that creates the functionality yourself | 00:17 |
djszapi | (I am not Finnish people are drunk on Friday though :D) | 00:18 |
djszapi | wondering* | 00:18 |
djszapi | yeah, sure ... I mean, I can do, just I need to get enough power to do that small task after the frustration and shock. | 00:18 |
*** risca has joined #harmattan | 00:19 | |
ajalkane | I know. I was flaggerbasted and tried all kinds of workarounds because I couldn't believe it. | 00:19 |
ajalkane | Besides, I think finnish people like to be drunk on friday and also saturday. Preferably during the week too :P | 00:20 |
djszapi | so the problem is that with the discussion above, I have two repeaters in a column, and for onClick I need to set one element in the other repeater at a given index. | 00:21 |
djszapi | (the same value like the one erased) | 00:22 |
ajalkane | adjszapi: you need a model that has the elements, and then you need a property in the element that sets a property that emits a signal. In a nutshell. | 00:22 |
ajalkane | THen the repeater automagically updates its value | 00:23 |
ajalkane | It feels overkill for a QString based model, but QString probably doesn't emit signals when its value is changed | 00:24 |
djszapi | Honestly, I do not see why I would need a signal apart from the general onClicked in qml. | 00:25 |
ajalkane | you need the signal if you want QML to notice that one of the values is changed | 00:25 |
djszapi | as far as I understand, I can store the currentindex in a property in currentIndex: 0; variable and pass the value of that to a C++ function ... | 00:25 |
ajalkane | otherwise it just keeps using the original value | 00:25 |
djszapi | and the method returns the altered model data, that is | 00:25 |
ajalkane | yes models emit signals when their data is changed | 00:26 |
ajalkane | but if you use QStringList as a model, it's not a real model | 00:26 |
ajalkane | Think of it as a static model, that should not be modified | 00:26 |
ajalkane | (this is from memory) | 00:27 |
ajalkane | I might be wrong, but I know for sure I had lotsa trouble trying to get QStringList working directly and just gave up on it. | 00:27 |
djszapi | yes, it is static, but it does not mean I need any special signal. It seems to me I can salvage without that. | 00:27 |
ajalkane | You can if you emit changed signals in QML itself | 00:28 |
ajalkane | you can declare signals in qml, and when you strange the qstringlist model a | 00:28 |
ajalkane | you emit a signal that's handled in your qml... you can do it that way too | 00:28 |
djszapi | I do not need to declare a new signal, just use the existing onClicked, right ? | 00:29 |
ajalkane | True | 00:29 |
djszapi | Thereby, emptying, and a simple YetAnotherRepeater.model: MyClassInCpp.MyMethod(currentIndex); | 00:32 |
djszapi | + value parameter for sure | 00:33 |
*** djszapi has left #harmattan | 00:36 | |
*** Venemo has quit IRC | 00:43 | |
*** ajalkane has quit IRC | 00:53 | |
*** NIN101 has quit IRC | 00:57 | |
*** NIN101 has joined #harmattan | 00:57 | |
trx | why did they make that so complicated i'll never know... | 01:01 |
*** norayr has joined #harmattan | 01:13 | |
*** rcg has quit IRC | 01:39 | |
*** xTs has quit IRC | 01:48 | |
*** admiral0_ has joined #harmattan | 01:52 | |
*** admiral0 has quit IRC | 01:55 | |
*** Natunen has quit IRC | 02:00 | |
*** djszapi has joined #harmattan | 02:02 | |
*** SqRt7744 has quit IRC | 02:02 | |
djszapi | Hi! Is it possible to set the background (landscape and portrait as well) individually for pages ? For instance, I would not like to use the default main page background for the settings page. | 02:03 |
djszapi | I do not see any "id" property for the platform style so that I should change it or so. | 02:04 |
*** zk8 has quit IRC | 02:09 | |
*** infobot has joined #harmattan | 02:11 | |
*** ChanServ sets mode: +v infobot | 02:11 | |
*** admiral0_ has quit IRC | 02:12 | |
*** Riussi has quit IRC | 02:18 | |
*** Riussi has joined #harmattan | 02:21 | |
artemma | code compiles fine on Mac/Desktop, but on Harmattan build I get undefined reference to `LanguageFiles::staticMetaObject' | 02:22 |
artemma | and undefined reference to `vtable for LanguageFiles' :/ | 02:22 |
artemma | Some linking problem probably, but why on Harm only.. | 02:23 |
djszapi | moc included in the source ? | 02:23 |
djszapi | and generated properly ? | 02:23 |
artemma | hmm, I don't know | 02:23 |
djszapi | Can you check out ? | 02:24 |
artemma | some moc_ files are in the build log | 02:24 |
artemma | but I don't really know how to check if they are generated correctly | 02:24 |
djszapi | they should be in the build tree.. | 02:25 |
artemma | I am using services from the common static library | 02:25 |
artemma | and it was working fine until I decided to create there an object for injection into QML | 02:26 |
djszapi | check your linkage options | 02:26 |
artemma | I am afraid your level of abstraction is too hight for me | 02:26 |
artemma | :) | 02:27 |
artemma | I don't know how to check these | 02:27 |
artemma | and what to look for | 02:27 |
djszapi | can you paste a build log ? | 02:27 |
artemma | though, hmm, if I check all PRE_TARGETDEPS.. | 02:28 |
artemma | build log - http://pastebin.com/ubANXREj | 02:29 |
artemma | aha, it is qmlRegisterType that is actually failing | 02:32 |
artemma | still some sort of a linking issue probably. It fails to register non-linked stuff maybe | 02:32 |
artemma | is it ok at all to qmlRegister types defined in a staticly linked library? | 02:32 |
artemma | funny thing is that on desktop everything's fine :/ | 02:34 |
artemma | hmm, could it actually be incorrect lib deployment? | 02:40 |
artemma | I mean engine's .a file should be put to some correct place, shouldn't it? | 02:40 |
artemma | aha, found moc files in build tree | 02:48 |
artemma | look ok | 02:48 |
artemma | shall I actually check .pro stuff? | 02:48 |
artemma | symbian: LIBS += -lswypeswitcherengine | 02:48 |
artemma | else:unix: LIBS += -L$$OUT_PWD/../../src/swypeswitcherengine/ -lswypeswitcherengine | 02:48 |
artemma | else: LIBS += -L$$OUT_PWD/../../src/swypeswitcherengine/ -lswypeswitcherengine | 02:48 |
*** Sazpaimon__ has quit IRC | 03:04 | |
*** artemma has quit IRC | 03:04 | |
*** artemma has joined #harmattan | 03:11 | |
*** arcean has quit IRC | 03:20 | |
*** NIN101 has quit IRC | 03:45 | |
*** pinheiro has quit IRC | 03:52 | |
*** Anssi138 has joined #harmattan | 03:59 | |
*** artemma has quit IRC | 04:03 | |
*** artemma has joined #harmattan | 04:09 | |
*** Sazpaimon__ has joined #harmattan | 04:10 | |
*** artemma has quit IRC | 04:14 | |
*** artemma has joined #harmattan | 04:20 | |
*** risca has quit IRC | 04:24 | |
*** risca has joined #harmattan | 04:24 | |
*** risca has quit IRC | 04:24 | |
*** artemma has quit IRC | 04:24 | |
*** M4rtinK has quit IRC | 04:27 | |
*** artemma has joined #harmattan | 04:31 | |
*** jluisn has joined #harmattan | 04:39 | |
*** jluisn has quit IRC | 05:05 | |
*** secyritas has quit IRC | 05:06 | |
*** secyritas has joined #harmattan | 05:06 | |
*** artemma has quit IRC | 05:13 | |
*** artemma has joined #harmattan | 05:19 | |
*** Natunen has joined #harmattan | 06:06 | |
*** Sazpaimon__ is now known as Sazpaimon | 06:39 | |
*** artemma has quit IRC | 06:43 | |
*** artemma has joined #harmattan | 06:50 | |
*** benares_98_ has joined #harmattan | 07:08 | |
*** benares_98 has quit IRC | 07:21 | |
*** benares_98_ is now known as benares_98 | 07:21 | |
*** decibyte_ has quit IRC | 07:55 | |
*** decibyte has joined #harmattan | 07:58 | |
*** coffetime has quit IRC | 08:18 | |
*** ZogG_lap1op has joined #harmattan | 08:37 | |
*** ZogG_laptop has quit IRC | 08:37 | |
*** hardaker has quit IRC | 08:44 | |
*** Anssi138 has quit IRC | 09:14 | |
*** benares_98 has quit IRC | 09:32 | |
*** diggy has joined #harmattan | 09:40 | |
*** lucido has quit IRC | 09:47 | |
*** beford has quit IRC | 10:09 | |
djszapi | Is it possible to set the background (landscape and portrait as well) individually for pages ? For instance, I would not like to use the default main page background for the settings page. I do not see any "id" property for the platform style so that I should change it or so. | 10:11 |
*** risca has joined #harmattan | 10:28 | |
*** diggy has quit IRC | 10:43 | |
*** chris_ has joined #harmattan | 10:47 | |
*** DocScrutinizer has quit IRC | 10:49 | |
*** DocScrutinizer has joined #harmattan | 10:49 | |
*** rcg has joined #harmattan | 10:50 | |
djszapi | hat is the best way of handling the event for the SelectionDialog when the user chooses an element out of it ? I do not see accepted signal for it in the documentation, like for the MultiSelectionDialog or just pure Dialog elements. | 10:58 |
*** chris_ has quit IRC | 11:11 | |
*** chris_ has joined #harmattan | 11:14 | |
frals | onSelectedIndexChanged | 11:21 |
frals | for page background you probably have to fill the page with an image/rectangle to do it | 11:23 |
*** zk8 has joined #harmattan | 11:43 | |
djszapi | frals: those pure qml elements are not aware of orientation change. | 11:47 |
*** xTs has joined #harmattan | 11:50 | |
*** admiral0 has joined #harmattan | 11:57 | |
*** risca has quit IRC | 12:02 | |
*** nix-cyrus has joined #harmattan | 12:02 | |
*** chris_ has quit IRC | 12:13 | |
frals | djszapi: well, you just make a binding to something that is aware | 12:22 |
djszapi | heh, that sounds rather dirty. | 12:23 |
djszapi | it should possibly be a simple property, just like the tools, and the client code can decide whether use the same background (again, just like in case toolbarlayouts) for various pages, or individual ones. | 12:25 |
*** NIN101 has joined #harmattan | 12:32 | |
*** SqRt7744 has joined #harmattan | 12:46 | |
*** faenil has joined #harmattan | 12:55 | |
*** trx has quit IRC | 13:16 | |
*** trx has joined #harmattan | 13:20 | |
Appiah | Anyway to debug a site with the browser? | 13:26 |
djszapi | Meh, the SelectionDialog does not accept stringlist model, just qlistmodel ... :) | 13:26 |
Appiah | Trying to see why a html5 site is not working | 13:26 |
Appiah | on the N9 | 13:26 |
faenil | o/ djszapi | 13:26 |
djszapi | Heya faenil :) | 13:26 |
djszapi | Appiah: as usual, the first step is to check out the console and syslog | 13:26 |
djszapi | and if grob has a specific log file, then that. | 13:27 |
*** arcean has joined #harmattan | 13:28 | |
djszapi | any simple way of putting, a dynamically generated, string list with unknown length in advance into a listmodel ? | 13:30 |
Appiah | ye I'll have to see what grob can do | 13:30 |
djszapi | this thingy: http://meegoharmattandev.blogspot.com/2011/07/populating-selectiondialog-from-c.html | 13:32 |
Appiah | can't find anything regarding grob | 13:34 |
djszapi | why would ? :) I did not highlight you, just continued the analization of my issue. ;-) | 13:35 |
Appiah | I mean | 13:35 |
Appiah | I looking for manpages and sites | 13:35 |
Appiah | and I come up with nothing | 13:35 |
djszapi | grob is the browser. | 13:37 |
Appiah | I know | 13:37 |
artemma | my yesterdays linking failure was cured by just deleting the build directory - apparently some old files weren\t deleted automatically.. | 13:47 |
artemma | shouldn't code that late again | 13:47 |
djszapi | artemma: do every single night :) | 13:56 |
*** jreznik has joined #harmattan | 13:58 | |
Appiah | http://www.developer.nokia.com/Community/Discussion/showthread.php?232223-web-browser-console well this is a start.. | 14:02 |
*** blueslee has joined #harmattan | 14:15 | |
*** M4rtinK has joined #harmattan | 14:20 | |
*** artemma has quit IRC | 14:24 | |
*** Stecchino_ has quit IRC | 14:28 | |
*** Stecchino has joined #harmattan | 14:29 | |
*** artemma has joined #harmattan | 14:30 | |
*** Smtih has joined #harmattan | 14:47 | |
*** blueslee has quit IRC | 14:53 | |
*** faenil has quit IRC | 14:54 | |
*** acidjunkie has quit IRC | 14:56 | |
*** acidjunkie has joined #harmattan | 14:58 | |
*** zk8 has quit IRC | 15:00 | |
*** jonni has quit IRC | 15:12 | |
*** ecloud has quit IRC | 15:14 | |
*** jonni has joined #harmattan | 15:26 | |
*** ecloud has joined #harmattan | 15:29 | |
*** artemma has quit IRC | 15:32 | |
*** leio_ has joined #harmattan | 15:45 | |
*** leio has quit IRC | 15:45 | |
*** lucido has joined #harmattan | 15:47 | |
*** tabasko has quit IRC | 16:00 | |
*** rantom_ has quit IRC | 16:00 | |
*** sandst1 has quit IRC | 16:00 | |
SqRt7744 | anyone interested in helping me port evopedia to harmattan? I could use some help... progress is already hosted on github. The C++ ListModel stuff is mostly done, now we need to do the interface in qml. | 16:00 |
*** zk8 has joined #harmattan | 16:00 | |
SqRt7744 | for me this program is at the top of the priority list, I don't know how anyone gets on without it. | 16:01 |
*** zk8 has quit IRC | 16:07 | |
*** auenf has quit IRC | 16:26 | |
*** faenil has joined #harmattan | 16:27 | |
*** auenf has joined #harmattan | 16:28 | |
* DocScrutinizer all of a sudden recalls openmoko wikireader | 16:34 | |
leinir | DocScrutinizer: *nods* Certainly their best product to date :) Something simple, which did one job, and did it extraordinarily well :) People actually /wanted/ to buy that one, not just for the freedom aspect, but because it was useful :) | 16:35 |
RST38h | Can't have that, can we? | 16:36 |
leinir | ;) | 16:36 |
DocScrutinizer | and the original design was rather nice (no wonder thanks to my massive participation), alas later on it went from clamshell with hw kbd to c-ts | 16:36 |
DocScrutinizer | I still wonder if the WR as sold now still has the footprint to retrofit a USB receptacle as I insisted it to have in original design | 16:38 |
DocScrutinizer | but that's life: you design a hw, then they mess it up, and finally you never touched the product with your own paws | 16:39 |
*** zz_gri is now known as gri | 16:40 | |
DocScrutinizer | RST38h: can't have what? | 16:40 |
*** gri is now known as zz_gri | 16:40 | |
*** artemma has joined #harmattan | 16:49 | |
*** hardaker has joined #harmattan | 17:00 | |
DocScrutinizer | RST38h: WR? it's available e.g, via CONRAD by a rebranding | 17:00 |
DocScrutinizer | http://www.conrad.de/ce/de/product/876441/MEDIONR-P89615-MD-83020-WIKIREADER | 17:02 |
DocScrutinizer | https://svn.internal.openmoko.org/trunk/wiki-reader/ haha, connection refused | 17:03 |
DocScrutinizer | who would've thought ;-P | 17:04 |
djszapi | I am now trying to expose my own model from C++ to the selection dialog, but I seem to fail with that. I am using this code in the main.cpp: ctxt->setContextProperty("anagramCategoryModel", kanagramEngineHelper.anagramCategoryModel()); and then justmodel: anagramCategoryModel; inside the qml file. What is wrong about it ? qDebug() << kanagramEngineHelper.anagramCategoryModel()->stringList(); shows the strings, but the model is empty in qml | 17:23 |
*** auenf has quit IRC | 17:26 | |
*** auenf has joined #harmattan | 17:27 | |
*** blueslee has joined #harmattan | 17:32 | |
*** wook has joined #harmattan | 17:34 | |
wook | hi folks | 17:34 |
frals | iirc the selectiondialog is completely and utterly broken since they "fixed" some bug in 1.1 (or the release after that) | 17:34 |
wook | is there any tune up for camera module, to increase photo quality on device and enable 12mpx? | 17:35 |
djszapi | after that, but I imported the fix | 17:35 |
djszapi | frals: you might wanna read this: http://meegoharmattandev.blogspot.com/2011/07/populating-selectiondialog-from-c.html | 17:35 |
*** leio_ is now known as leio | 17:36 | |
djszapi | so I imported the Common- and SelectionDialog with the fix in my project, and according to the comment in there: Another option is to copy to your project "SelectionDialog.qml" (and "CommonDialog.qml" because it is not exported) as "MySelectionDialog.qml" and make this | 17:36 |
djszapi | small change. After that any model with a "name" role can be used in a MySelectionDialog. | 17:36 |
djszapi | and I made a QStringList derivative with a simple setRoleNames call in the ctor, and that is all. | 17:36 |
*** benares_98 has joined #harmattan | 17:37 | |
*** jabis_ has quit IRC | 17:42 | |
*** jabis_ has joined #harmattan | 17:42 | |
djszapi | frals: there is an interesting thing over here: "file:///usr/lib/qt4/imports/com/nokia/meego/SelectionDialog.qml: File not found" -> I moved that file to a backup location in order to make sure my code is using the imported SelectionModel which is put into the binary by using the resource system, but apparently: it cannot find that for some reason. This might be the culsprit for the aforementioned issue. | 17:46 |
*** arcean has quit IRC | 17:48 | |
*** arcean has joined #harmattan | 17:49 | |
*** jabis_ has quit IRC | 17:52 | |
*** jabis_ has joined #harmattan | 17:52 | |
*** wook has quit IRC | 17:55 | |
*** DocScrutinizer is now known as Doc | 17:57 | |
*** Doc is now known as DocScrutinizer | 17:58 | |
*** artemma has quit IRC | 17:59 | |
*** wook has joined #harmattan | 18:00 | |
djszapi | okay, at least MySelectionDialog works. :) | 18:00 |
*** wook is now known as Guest33949 | 18:00 | |
*** blueslee has quit IRC | 18:01 | |
djszapi | I do not understand why, but it at least works this way. I would like to understand the reason behind though. | 18:01 |
*** Guest33949 has quit IRC | 18:02 | |
*** vLassi has joined #harmattan | 18:07 | |
vLassi | well hello | 18:07 |
DocScrutinizer | frals: \o/ | 18:07 |
*** admiral0 has quit IRC | 18:10 | |
*** jabis_ has quit IRC | 18:11 | |
*** jabis_ has joined #harmattan | 18:11 | |
*** artemma has joined #harmattan | 18:11 | |
faenil | .... | 18:13 |
faenil | so, it comes out ContentX is also changed when you change contentWidth... | 18:13 |
faenil | thanks a ton Flickable...could have said that somewhere! | 18:13 |
djszapi | didn't you see that on the ui ? | 18:15 |
faenil | yes but I didn't know who was changing that back to 0 | 18:15 |
faenil | when I had bound that to other values! | 18:15 |
faenil | what's the point of setting it back each time to 0? -.- | 18:16 |
faenil | ruin my plans! | 18:16 |
faenil | I don't see why they must be related | 18:17 |
*** admiral0 has joined #harmattan | 18:22 | |
*** stroughtonsmith has joined #harmattan | 18:23 | |
*** auenf has quit IRC | 18:26 | |
*** artemma has quit IRC | 18:27 | |
*** jabis_ has quit IRC | 18:33 | |
*** jabis_ has joined #harmattan | 18:33 | |
*** artemma has joined #harmattan | 18:34 | |
faenil | djszapi: this way, you can't pinch an image, and at the same time let the flickable know that he's not looking at the left part of the image, but at the center part, because it will keep resetting contenX to 0 | 18:38 |
faenil | so it keeps thinking the center of the image (once zoomed) is the left edge of the image | 18:38 |
*** pinheiro has joined #harmattan | 18:43 | |
*** artemma has quit IRC | 18:44 | |
*** vladest_ has joined #harmattan | 18:45 | |
*** vladest has quit IRC | 18:45 | |
*** vladest_ is now known as vladest | 18:46 | |
*** artemma has joined #harmattan | 18:50 | |
*** arcean has quit IRC | 18:52 | |
*** Shadikka has quit IRC | 18:53 | |
M4rtinK | thp: ping :) | 18:54 |
*** M4rtinK has quit IRC | 19:01 | |
*** artemma has quit IRC | 19:03 | |
*** tabasko has joined #harmattan | 19:03 | |
*** Smtih has quit IRC | 19:03 | |
faenil | how can I stop Flickable from resetting contentX to 0 when contentWidth changes? apart from modifying the sourcecode of flickable :) | 19:08 |
*** artemma has joined #harmattan | 19:09 | |
*** Shadikka has joined #harmattan | 19:13 | |
*** artemma has quit IRC | 19:13 | |
*** M4rtinK has joined #harmattan | 19:16 | |
*** artemma has joined #harmattan | 19:20 | |
*** Shadikka has quit IRC | 19:23 | |
*** Shadikka has joined #harmattan | 19:23 | |
*** artemma has quit IRC | 19:26 | |
*** artemma has joined #harmattan | 19:32 | |
*** faenil has quit IRC | 19:34 | |
thp | M4rtinK: pong | 19:42 |
*** fignew has quit IRC | 19:44 | |
*** faenil has joined #harmattan | 19:44 | |
*** fignew has joined #harmattan | 19:46 | |
*** admiral0 has quit IRC | 19:49 | |
*** admiral0 has joined #harmattan | 19:51 | |
*** l32606 has joined #harmattan | 19:51 | |
l32606 | hi. when i use the dbus-monitor command on N9, i can only get the signal type output. but i can get all types of msg including method_call on N900. | 19:56 |
*** admiral0 has quit IRC | 19:56 | |
l32606 | does anybody have ideas how can i get the method_call message? Thanks. | 19:57 |
*** nix-cyrus has left #harmattan | 20:01 | |
*** diggy has joined #harmattan | 20:01 | |
*** benares_98 has left #harmattan | 20:06 | |
*** l32606 has quit IRC | 20:07 | |
*** eeanm has joined #harmattan | 20:14 | |
eeanm | hey is there a package of common qml imports for stuff like writing files? | 20:17 |
djszapi | what do you mean ? | 20:18 |
*** eeanm has quit IRC | 20:21 | |
*** Shadikka has quit IRC | 20:24 | |
*** vladest has quit IRC | 20:25 | |
*** faenil has quit IRC | 20:26 | |
*** vladest has joined #harmattan | 20:27 | |
*** hardaker has quit IRC | 20:27 | |
*** hardaker has joined #harmattan | 20:28 | |
*** admiral0 has joined #harmattan | 20:31 | |
*** hardaker2 has joined #harmattan | 20:32 | |
*** SqRt7744 has quit IRC | 20:32 | |
*** hardaker has quit IRC | 20:33 | |
*** fuz_ has quit IRC | 20:36 | |
*** risca has joined #harmattan | 20:40 | |
*** hardaker has joined #harmattan | 20:40 | |
*** fuz_ has joined #harmattan | 20:41 | |
*** hardaker2 has quit IRC | 20:42 | |
*** Shadikka has joined #harmattan | 20:58 | |
*** tabasko has quit IRC | 21:00 | |
*** Shadikka has quit IRC | 21:04 | |
*** oberling has joined #harmattan | 21:06 | |
*** faenil has joined #harmattan | 21:17 | |
*** vLassi has quit IRC | 21:30 | |
M4rtinK | thp: the Touchpad PyGame guide is done: http://modrana.org/trac/wiki/pygame_touchpad | 21:39 |
*** beford has joined #harmattan | 21:47 | |
*** beford has joined #harmattan | 21:47 | |
*** Anssi138 has joined #harmattan | 21:53 | |
*** Anssi138 has quit IRC | 21:58 | |
*** Shadikka has joined #harmattan | 22:03 | |
*** tabasko has joined #harmattan | 22:10 | |
*** tabasko has joined #harmattan | 22:11 | |
*** Anssi138 has joined #harmattan | 22:17 | |
*** tabasko has quit IRC | 22:18 | |
*** khertan has quit IRC | 22:20 | |
*** jreznik has quit IRC | 22:21 | |
*** tabasko has joined #harmattan | 22:22 | |
djszapi | Is it possible to dynamically change the text value of a Text element for instance in a repeater, but from outside the repeater ? I would like to set it for instance to "" in the beginning, but for an event outside, I would like to switch it to "modelData". Was trying to do something like: text: isInit ? "" : modelData; but if I change the isInit boolean property value outside, the text does not change. :/ | 22:23 |
djszapi | (of course it is viable in C++, but that is behind the point of the question since I am interested in doing the operation in QML.) | 22:25 |
*** khertan has joined #harmattan | 22:33 | |
*** SqRt7744 has joined #harmattan | 22:38 | |
*** coffetime has joined #harmattan | 22:39 | |
*** mcfrisk has quit IRC | 22:40 | |
*** aergus has joined #harmattan | 22:40 | |
*** aergus has left #harmattan | 22:42 | |
*** Sazpaimon_ has joined #harmattan | 22:44 | |
thp | M4rtinK: thanks, i've already linked it :) good work! | 22:45 |
*** Sazpaimon has quit IRC | 22:46 | |
*** SqRt7744 has quit IRC | 22:46 | |
M4rtinK | thp: oh, nice ! :) | 22:46 |
M4rtinK | hope the guide would be useful to someone :) | 22:47 |
M4rtinK | thp: also, any plans for a PySide port ? :) | 22:47 |
M4rtinK | the launcher already uses Qt 4.7&QML AFAIK :) | 22:48 |
*** nid0 has joined #harmattan | 22:48 | |
*** Sazpaimon__ has joined #harmattan | 22:51 | |
*** Sazpaimon_ has quit IRC | 22:51 | |
djszapi | ok, I was somehow lame. My boolean idea worked. I made a mistake inside the onClicked signal handling by typing ":" instead of "=" for the assignment. :/ | 22:52 |
*** decibyte has quit IRC | 22:58 | |
faenil | djszapi: okay, coz that sounded strange, should have worked :) | 23:01 |
faenil | glad it did | 23:01 |
djszapi | that is the only way of interacting between repeaters for instance because dynamic id is a no-go. For sure, C++ is always a fallback alternative. | 23:02 |
faenil | :) | 23:02 |
djszapi | so my only left issue is this: http://meegoharmattandev.blogspot.com/2011/07/populating-selectiondialog-from-c.html | 23:02 |
djszapi | see my long comment in a reply to the guy .. | 23:03 |
faenil | ouch :) I don't think I can help with that :) | 23:03 |
artemma | what's wrong with having an alias to property of a QML object implemented in C++? Works on Desktop, but on Harmattan I get Invalid alias location | 23:03 |
*** decibyte has joined #harmattan | 23:04 | |
CissWit | Hi there, i had the 232 repository set up in my souces.list.d but it returnes a 404 error. | 23:04 |
CissWit | I tried to change it with the 005, that is configured in a friend's n9, and it now proposes more upgrades | 23:05 |
CissWit | do you know if it is safe to upgrade ? | 23:05 |
CissWit | is there some kind of incompatibilities between versions ? | 23:05 |
djszapi | yes there are for sure. | 23:05 |
artemma | Hmm. What could be a reason for a QML property (of a c++ object) be accessible on desktop, but not on Harmattan? :? | 23:06 |
CissWit | there are hardware differences ? | 23:06 |
djszapi | CissWit: you wanna take a look at some community page speaking about backups. | 23:06 |
djszapi | might* | 23:07 |
*** auenf has joined #harmattan | 23:07 | |
* artemma found the yet another linking problem, that was solved by deleting the build tree. Why doesn't clean clean everything? | 23:08 | |
djszapi | artemma: I guess you use qmake ? | 23:09 |
artemma | djszapi: I was trying to clean from QtCreator (and tried its "Run qmake" too) | 23:09 |
djszapi | I am sorry, but I cannot help with qmake. | 23:10 |
*** artemma has quit IRC | 23:13 | |
CissWit | djszapi: to backup before trying or to find information about country codes and compatibility ? | 23:16 |
*** hardaker2 has joined #harmattan | 23:16 | |
djszapi | both | 23:17 |
*** hardaker has quit IRC | 23:18 | |
*** artemma has joined #harmattan | 23:19 | |
thp | M4rtinK: hmm not really ;) probably depends on the availability of qt first. after that, we can talk about pyside ;) | 23:24 |
*** leinir has quit IRC | 23:24 | |
artemma | Is there any was for an app to write to files in /usr/share owned by root? | 23:25 |
artemma | I got tired editing swype settings manually, half-wrote an app for that and.. swype settings file is owned by root | 23:26 |
artemma | maybe some cool aegis manifest could request the needed right? Or is writing to root-owned files totally forbidden even in /usr/share? | 23:28 |
djszapi | it is a fairly simple task to achieve.. | 23:29 |
djszapi | you might wanna check out the output of the following command in devel shell: "accli -I", and if not clear how to write a manifest file, look up the platform security documentation. | 23:29 |
djszapi | it really is just few lines. | 23:30 |
*** vladest_ has joined #harmattan | 23:32 | |
*** vladest has quit IRC | 23:34 | |
*** vladest_ is now known as vladest | 23:34 | |
artemma | thinking.. | 23:35 |
artemma | djszapi: do u mean that any app can just request the write to overwriting root-owned files? | 23:35 |
*** vladest_ has joined #harmattan | 23:35 | |
artemma | djszapi: and that would pass the ovi store QA? | 23:36 |
djszapi | yes, any app can request it. As I said, check out my scenario above. About QA: dunno, I am not responsible for OVI. | 23:36 |
artemma | Heh, then I guess I just have to try, submit and see what happens | 23:37 |
djszapi | well if you do not use abuse it ... | 23:37 |
artemma | never used accli before | 23:37 |
artemma | djszapi: how can they know if I am abusing it or not | 23:38 |
*** rantom_ has joined #harmattan | 23:38 | |
artemma | I am certainly going to change swype settings file only | 23:38 |
artemma | but even that could potentially be buggy | 23:38 |
*** vladest has quit IRC | 23:38 | |
djszapi | also, make sure you do not brick the device. | 23:38 |
* artemma went googling for aegis tutorials | 23:38 | |
*** vladest_ is now known as vladest | 23:38 | |
djszapi | I do not think the aegis tutorials I wrote are in public yet. | 23:39 |
djszapi | maybe with the next release :P | 23:39 |
* artemma trying accli | 23:39 | |
artemma | I am not sure what set of credentials accli -I shows | 23:41 |
djszapi | available. | 23:41 |
artemma | you mean all credentials possible? | 23:42 |
artemma | aha, then I can examine the list and see if there's somethign that looks like rooted files access | 23:42 |
djszapi | I could give you those few lines you need to write, but then you do not learn anything ... | 23:42 |
artemma | djszapi: it's ok, now I am more into learning mode | 23:43 |
artemma | over midnight is my usual time for copy-pasting recipes :) | 23:43 |
artemma | still 15 mins till midnight in Finland | 23:43 |
djszapi | which city in Finland ? /me Helsinki | 23:43 |
artemma | Tampere | 23:44 |
artemma | reading about aegis credentials | 23:44 |
artemma | is it really so that I can request an app to have all the powers of root? | 23:45 |
artemma | sounds too powerful | 23:45 |
artemma | using <credential name="UID::root" /> | 23:45 |
artemma | well, root uid | 23:45 |
artemma | or in root case, GRP::root is probably better | 23:46 |
Aard | uid 0 != powerful with aegis | 23:46 |
djszapi | root is not a real unix root concept | 23:46 |
artemma | ah, indeed, I keep forgetting that | 23:46 |
artemma | yet with GRP:root I should be able to do whatever with all the root-owned files on the device, shouldn't I? | 23:47 |
djszapi | you should not. | 23:47 |
artemma | then I am missing the point probably | 23:47 |
artemma | I am following these docs http://harmattan-dev.nokia.com/docs/library/html/guide/html/Developer_Library_Developing_for_Harmattan_Harmattan_security_Security_guide_Using_Harmattan_security_features.html | 23:48 |
artemma | Requesting specific user accounts and groups is the only part I found that seems to be relevant | 23:48 |
djszapi | it is not that simple as you think. | 23:49 |
djszapi | also, requesting root for your binary is plain wrong. | 23:49 |
djszapi | since you will /not/ be able to access user thingy anymore | 23:49 |
artemma | oups, it won't work. Manifest requires user used to be created in the same manifest | 23:49 |
djszapi | what you need to actually achieve is to separate the variety of functionalities into different executables. | 23:49 |
artemma | oh, so it's not like adding user, it's like replacing user | 23:50 |
djszapi | no | 23:50 |
djszapi | it is like running the executable with that user or group id | 23:50 |
*** artemma has quit IRC | 23:54 | |
*** vladest_ has joined #harmattan | 23:55 | |
marsje | I'm trying to make my very 1st app of my life | 23:57 |
*** gabriel9 has joined #harmattan | 23:57 | |
*** vladest has quit IRC | 23:57 | |
*** vladest_ is now known as vladest | 23:57 | |
marsje | I created a new project in Qt Creator and try to run it | 23:57 |
marsje | I get this message now: "Could not connect to the in-process QML debugger:" | 23:57 |
marsje | The application is not set up for QML/JS debugging. | 23:57 |
marsje | what am I doing wrong? | 23:58 |
djszapi | you use QtCreator ;-) | 23:58 |
marsje | yes.... | 23:58 |
*** artemma has joined #harmattan | 23:59 | |
beford | QtCreator ftw | 23:59 |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!