IRC log of #maemo-ssu for Monday, 2012-11-12

*** rd_ has quit IRC00:00
*** MrPingu has joined #maemo-ssu00:00
*** toxarisswe has quit IRC00:22
*** MrPingu has quit IRC00:28
*** MrPingu has joined #maemo-ssu00:30
*** Pali has quit IRC00:41
*** Jade has quit IRC00:42
*** Jade has joined #maemo-ssu00:43
*** Jade has joined #maemo-ssu00:43
*** Jade has quit IRC00:46
*** Jade has joined #maemo-ssu00:46
*** Jade has joined #maemo-ssu00:46
*** Jade has quit IRC00:47
*** Jade has joined #maemo-ssu00:48
*** int_ua has quit IRC01:06
*** MrPingu has quit IRC01:08
*** nox- has quit IRC02:16
*** nox- has joined #maemo-ssu02:30
*** arcean has joined #maemo-ssu02:47
*** arcean_ has quit IRC02:48
*** arcean has quit IRC03:22
*** kolp has quit IRC03:33
*** M4rtinK has quit IRC04:10
*** nox- has quit IRC04:55
*** amiconn has quit IRC05:33
*** amiconn_ has joined #maemo-ssu05:33
*** amiconn_ is now known as amiconn05:33
*** DocScrutinizer05 has quit IRC06:04
*** DocScrutinizer05 has joined #maemo-ssu06:04
*** Jade has quit IRC06:53
*** Jade has joined #maemo-ssu07:07
*** Jade has joined #maemo-ssu07:07
*** Pali has joined #maemo-ssu08:41
*** Pali has quit IRC09:14
*** MrPingu has joined #maemo-ssu09:32
*** deepy has joined #maemo-ssu09:33
*** deepy is now known as deepy\SIGSEGV09:34
*** MrPingu has quit IRC10:12
kerioooh, there's a new hildon-desktop in -devel?10:36
keriomerlin1991: when will you make a new cssu-testing release?10:38
*** luf has joined #maemo-ssu10:48
*** kolp has joined #maemo-ssu11:05
*** M13 has joined #maemo-ssu12:25
*** arcean has joined #maemo-ssu12:26
*** lizardo has joined #maemo-ssu12:34
*** M4rtinK has joined #maemo-ssu12:45
*** gregoa has quit IRC12:45
*** gregoa has joined #maemo-ssu12:45
*** dafox has joined #maemo-ssu12:46
*** Mihanizat0r has joined #maemo-ssu12:50
*** M13 has quit IRC12:53
*** Mihanizat0r has quit IRC13:06
*** Pali has joined #maemo-ssu13:37
*** iDont has joined #maemo-ssu13:43
*** dafox has quit IRC13:54
*** dafox has joined #maemo-ssu13:59
*** dafox has quit IRC14:24
*** iDont has quit IRC14:29
*** dafox has joined #maemo-ssu14:32
*** dafox has quit IRC14:38
*** dafox has joined #maemo-ssu14:40
*** dhbiker has quit IRC14:42
*** dafox has quit IRC14:56
*** dafox has joined #maemo-ssu14:58
*** dhbiker has joined #maemo-ssu15:15
*** _ade_ has joined #maemo-ssu15:23
*** _ade_ has quit IRC15:24
*** _ade_ has joined #maemo-ssu15:40
freemangordon_ade_: hi15:40
_ade_hi15:40
_ade_freemangordon: does it mean a change in the alarm cookies15:40
freemangordonso, the point is, can some parameters be passed from alarmd15:41
freemangordonyeah, something like that15:41
freemangordonthough it is dbus call, with variable arguments AIUI15:41
freemangordonso no changes in alarmd should be needed15:41
freemangordonno idae if that can be done15:42
freemangordon*idea15:42
_ade_just extra (new) arguments in the cookie?15:42
freemangordonmaybe, NFC how you setup an alarm in worldclock :D:D:D15:42
freemangordonbut I guess so15:42
freemangordonfollowing KISS principle,I think snooze count and sound duration per alarm should be the only things to be aded15:43
_ade_freemangordon: I also asked you about per alarm sounds. You can specify them, but the default sound is always played. Is that a bug, or missing in alarmd/alarm-ui?15:44
freemangordonwe really don;t need dofferent sound per alarm, at least for now15:44
freemangordonhmm, strange, lemme look at the code15:44
freemangordonWTF?!?15:46
freemangordonhttps://gitorious.org/community-ssu/osso-systemui-alarm/blobs/master/osso-systemui-alarm.c#line65915:46
freemangordonhttps://gitorious.org/community-ssu/osso-systemui-alarm/blobs/master/osso-systemui-alarm.c#line66815:46
freemangordonwho the fuck wrote that code :(15:46
freemangordon_ade_: yeah, seems like bug15:47
*** arcean has quit IRC15:47
_ade_that would be the static alarm for the calender, the clock uses the one in gconf15:47
freemangordonBTW do we have UI to specify alarm sound?15:47
freemangordonboth use gconf setting15:47
freemangordonsee ^^^15:48
_ade_okay, but you can set a preferred alarmsound, and that will be used15:48
freemangordon_ade_: yes, but this is not sound per alarm, rather sound per alarm type15:49
_ade_Yes, it will always be the current one then. Alarmd gives the parameter to set a custom sound, but it has no effect15:50
freemangordon_ade_: yes, because alarmUI uses gconf setting, not the one in alarm event15:50
_ade_ah ok. Something else to adapt for you ;-)15:51
freemangordonbut lets leave sounds per alarm for now, it will be more complicated than snooze count/sound duration15:52
_ade_of course. I will have a look at the things you like to see and report back.15:52
freemangordon_ade_: if I edit /var/cache/alarm.d/... by hand, will it work to test?15:52
_ade_The alarmsoundfile? You can refer to a soundfile, but it has no effect15:53
freemangordon_ade_: not only that, but to add additional parameters15:53
_ade_Never tried to manually edit that...15:54
freemangordonok, i'll try15:54
*** Pali has quit IRC15:54
_ade_I always insert cookies15:54
freemangordonanyway, for now requirements are unclear, i'll ping you or post on the worldclock thread when have something for you15:55
_ade_thanks in advance!15:55
DocScrutinizer05WTF indeed, what a braindamaged concept is that behind alarms and alarmsounds?16:01
freemangordonDocScrutinizer05: I can't believe what that code (in its stock form) looks like16:03
DocScrutinizer05I looked, seen a hardcoded path in sourcecode inside instruction area (not even as a #define), threw up and closed it again16:04
freemangordonwell, that could be a result or REing, I am refactoring it right now16:06
DocScrutinizer05anyway keep in mind that alarmd cookies been broken in stock maemo iirc16:06
DocScrutinizer05you got a struct defining what's meant to be passed as "cookie" to alarmd during definition of an alarm, and returned by alarmd when alarm fires?16:08
DocScrutinizer05as of now16:08
DocScrutinizer05I'd gladly share my thoughts and suggestions how to augment it16:09
freemangordonDocScrutinizer05: this is not meant for stock Maemo, and cookie bug is fixed as for the last 2 CSSU-T releases16:09
DocScrutinizer05yes, I'm aware of all that16:09
_ade_freemangordon: what hardcoded paths are there, except gconf paths and default alarm (sound) files?16:11
DocScrutinizer05guys, either we stay compatible with brainfsckd 'old' original alarmclock/alarmd/systemui-alarm "API", or we define a new better one, preferably as future-proof and backwards compatible as possible16:13
DocScrutinizer05that's why I asked for a pointer to the struct of the 'cookie'16:14
freemangordon_ade_: /tmp/ACT_DEAD :D:D:D16:14
freemangordonDocScrutinizer05: naah, no change in API16:14
DocScrutinizer05aiui you already changed the API16:15
freemangordonAPI will remain the same, it allows waviable arguments to be passed along with alarm16:15
freemangordonDocScrutinizer05: no16:15
freemangordonwhich API?16:15
DocScrutinizer05freemangordon: your idea of my idea of what API includes/means is not correct16:16
freemangordonaah, ok16:16
freemangordonyou mean between worldclock/alarmUI?16:16
DocScrutinizer05'variable arguments' never is exactly true16:16
freemangordonthat could be kept bacward-=compatible, no worries16:17
freemangordon*backward16:17
DocScrutinizer05since you always have a few interfaces which have to agree upon *some* structure in those 'variable arguments'16:17
freemangordonthis is not the case AIUI, as parameters are passed with dbus call16:17
*** luf has quit IRC16:17
DocScrutinizer05so what?16:18
freemangordonand ayatemui iterates through them and calls alarmUI plugin with GArray of parameters16:18
freemangordon*systemui16:18
DocScrutinizer05either you expect those arguments in same format somewhere sometime, or they are cruft16:18
_ade_As long as the stock clock ignores additional arguments I see no issue16:18
freemangordonyes16:18
freemangordonactually it is not like that16:19
freemangordonalarmUI gets alarm event cookie16:19
freemangordonand calls libalarm for the other stuff16:19
DocScrutinizer05guys, are you suggesting "I see no issue" so nobody else may look at it?16:19
freemangordonDocScrutinizer05: come on, libalarm and alarmUI source code is on gitorious, libalarm API is well defined16:20
_ade_But the worldclock will read the cookies at startup, that's what I am talking about.16:20
freemangordonlibalarm API allows you to pass alarm event attribute16:20
freemangordon*attributes16:20
freemangordonNAMED16:21
DocScrutinizer05aha, so yes, you basically say exactly that. "we got it sorted, if you wanna know what we're talking about, go the fuck serching for stuff yourself"16:21
freemangordonDocScrutinizer05: what makes you think I know libalarm API so I can explain it to you?16:21
freemangordonwhat I know I've lerant it the last week while I was REing alarmUI16:21
freemangordon*learnt16:22
DocScrutinizer05who mentioned libalarm?16:23
freemangordonme16:24
freemangordonas that is the way worldclock and alarmUI communicate16:25
_ade_freemangordon: did you get the point from my last remark, or am I still missing the point?16:26
freemangordon_ade_: yes16:26
_ade_the first or last part ;-)16:27
freemangordonif it is for that one "_ade_> As long as the stock clock ignores additional arguments I see no issue"16:27
DocScrutinizer05seems you're more happy with me looking at your final work and then start bashing the flaws I might find, rather than my offer to contribute early in the process. Oh well16:27
freemangordonDocScrutinizer05: come on, there is still no "process" I am still cleaning up the code and finding whatever bugs I introduced during REing16:28
freemangordonThe "process" will start once replacement does what stock one do without the bugs16:29
_ade_freemangordon: is the ALARM_EVENT_BOOT flag related to alarmUI (see http://talk.maemo.org/showpost.php?p=1145748&postcount=146 for an explanation of why I ask)? I do not see it in the code right away, so I guess not.16:42
freemangordon_ade_: seems alarm ui uses some "hidded" API reserved for systemui16:46
_ade_freemangordon: okay, thanks16:47
freemangordonjust a secon, i'll show you the code16:47
freemangordonhmm, seems I misunderstood your cuestion, but anyway, seems that raising the most significant bit when respondind to alarm event while in act_dead leads to boot16:49
freemangordon*question16:49
freemangordonhere https://gitorious.org/community-ssu/osso-systemui-alarm/blobs/master/osso-systemui-alarm.c#line104316:49
freemangordonoh shit16:51
freemangordonhttps://gitorious.org/community-ssu/osso-systemui-alarm/blobs/master/osso-systemui-alarm.c#line58816:51
freemangordonDocScrutinizer05: you meant hardcoded like that?16:51
_ade_Is this address are a result of REing, and would it have been some kind of label in the original source? Sorry if it is a stupid question :-)16:53
freemangordon_ade_: address?16:53
freemangordonyou mean file path?16:54
_ade_well, called 0x80000000u an address16:54
freemangordonnoo, this is not address16:54
freemangordonsee here:https://gitorious.org/community-ssu/osso-systemui-alarm/blobs/master/osso-systemui-alarm.c#line16616:55
freemangordon_ade_: I didn't find that defined in the headers, seems like some kind of internal API flag16:57
freemangordonbut it is a result of REing16:57
_ade_okay, somewhat to complicated for me. Anyway, I will keep in mind that ALARM_EVENT_BOOT just doesn't work, and I should stick to ACT_DEAD16:58
freemangordonbut we can translate it like #define ALARM_EVENT_ACK_BOOT 1<<3116:59
freemangordonor similar, codle still needs some cleanup16:59
freemangordon*code16:59
_ade_If you could restore functionalty to what it should do, I am always in favour17:00
freemangordonwell, so far it seems to work as it should17:00
freemangordonit woke me up this morning :D:D:D17:01
*** mickname has quit IRC17:01
_ade_You did read my link to TMO? That is what I am reffering to regarding ALARM_EVENT_BOOT and not working as it should.17:02
freemangordondid not read, just take a quick look17:03
freemangordonwill do now17:03
freemangordonyeah, got it. though those flags are unrelated to alarmUI17:04
_ade_okay, that's what I thought. The discussion got a whole other twist was my impression17:05
freemangordonyeah17:06
freemangordonok, going to put that in -devel17:08
*** arcean has joined #maemo-ssu17:15
freemangordonarcean, _ade_: alarUI replacement is in -devel, please install the version from there17:20
freemangordon*alarmUI17:20
_ade_freemangordon: can you upload to extras-devel atm?17:21
freemangordon_ade_: CSSU-devel :P17:21
freemangordonthis is #maemo-ssu :D:D:D17:21
_ade_oh, yeah, just thought of that when I replied ;-)17:21
_ade_It is hosted by Merlin1991, so no issues over there17:24
*** arcean has quit IRC17:40
*** _ade_ has left #maemo-ssu18:07
*** dhbiker has quit IRC18:21
*** arcean has joined #maemo-ssu18:26
*** _ade_ has joined #maemo-ssu18:52
_ade_freemangordon: about the bug report we discussed yesterday (https://bugs.maemo.org/show_bug.cgi?id=12661): as far as I can see it is already linked to CSSU, with subcomponent H-D18:56
povbotBug 12661: Popup menu fails with X Error while switching portrait/landscape18:56
arcean_ade_, "First annoyance is that it flips back to landscape18:58
arcean(why?)" is already fixed for non-Qt based apps18:58
_ade_arcean: great, a first step ;-)18:59
_ade_arcean: just started FAPMAN, it rotates back to landscape at startup ?19:00
arceanthat's normal with current cssu-testing19:00
_ade_the fix still needs to arrive then?19:01
arceanyes, and I have no idea when it will be :P19:01
_ade_It might help me in some cases regarding my issue19:02
arceanQt uses its own rotation mechanism, which doesn't work well with hildon-desktop19:03
*** Pali has joined #maemo-ssu19:18
*** arcean has quit IRC19:22
*** dhbiker has joined #maemo-ssu19:22
*** arcean has joined #maemo-ssu19:24
*** mickname has joined #maemo-ssu19:48
*** arcean has quit IRC20:01
*** arcean has joined #maemo-ssu20:08
kerioQt sucks20:14
kerioit doesn't follow my keyboard bindings :(20:15
*** timo^ has joined #maemo-ssu20:19
*** _ade_ has left #maemo-ssu20:21
*** arcean has quit IRC21:00
*** arcean has joined #maemo-ssu21:02
*** luf has joined #maemo-ssu21:04
*** M4rtinK has quit IRC21:16
*** arcean has quit IRC21:23
*** iDont has joined #maemo-ssu21:27
*** arcean has joined #maemo-ssu21:27
*** infobot has quit IRC21:38
*** macmaN has quit IRC21:39
*** macmaN has joined #maemo-ssu21:43
*** dhbiker has quit IRC21:45
*** M4rtinK has joined #maemo-ssu21:48
*** arcean has quit IRC21:48
*** arcean has joined #maemo-ssu21:49
*** timo^ has quit IRC22:02
*** iDont has quit IRC22:07
*** arcean has quit IRC22:19
*** Sc0rpius has quit IRC22:21
*** Sc0rpius has joined #maemo-ssu22:26
*** arcean has joined #maemo-ssu22:29
*** _rd has joined #maemo-ssu22:38
merlin1991new -testing pre test incomin22:39
merlin1991g22:39
freemangordonlol22:40
kerioyay22:41
freemangordonmerlin1991, Pali: would you look at alarmUI source code to check if it needs further cleanup/refactoring? When you have some spare time of course :)22:41
merlin1991comeon "lol"? it is hard work to type dpkg-buildpackage -rfakeroot -us -uc in the terminal ;)22:42
freemangordonmerlin1991: I know22:42
*** arcean has quit IRC22:42
freemangordontrust me :D22:42
freemangordonmerlin1991: -I.git is missing, check your packages :P22:43
merlin1991tbh I don't care about -I.git on the smaller packages22:43
freemangordon-sa too22:44
merlin1991on qt though I DO care :D22:44
merlin1991hm what's -sa?22:44
freemangordonafaik it should be "dpkg-buildpackage -rfakeroot -sa -us -uc -I.git" :P22:44
freemangordonfull source upload22:44
merlin1991our packges are all "native" in their debian/source descriptions22:45
merlin1991thus full source anyway22:45
freemangordonit includes .orig.tar.gz in .changes22:45
freemangordon^^^22:45
*** infobot has joined #maemo-ssu22:45
*** ChanServ sets mode: +v infobot22:45
merlin1991hm I built obexd already22:47
merlin1991well that just leaves ke-recv22:47
merlin1991I should create a build-cssu command that wraps that dpkg-buildpackage call :D22:52
*** _rd has quit IRC22:53
*** dhbiker has joined #maemo-ssu22:55
*** arcean has joined #maemo-ssu22:55
*** lizardo has quit IRC22:58
*** _ade_ has joined #maemo-ssu23:03
merlin1991DocScrutinizer05: got some time to test the new -testing?23:08
DocScrutinizer05possibly23:08
_ade_arcean: adapted worldclock so it cannot start multiple occurrences the date/time applet in the worldclock-  and main screen23:08
DocScrutinizer05if you got some time for me to find some time for that23:08
merlin1991timeception?23:09
arcean_ade_, thanks :)23:09
merlin1991arcean: are you running a clean -testing or -thumb?23:13
arceantesting + devel combo :P23:13
merlin1991that's fine, wanna test the new -testing? :D23:14
arceanwhy not :)23:14
freemangordonmerlin1991: those with -thumb should be able to test it too23:15
freemangordonafter all we fixed the versioning23:15
merlin1991freemangordon: but if possible I'd like to test without sideffects ;23:15
merlin1991hm found the first bug myself23:16
freemangordoncare to share?23:16
merlin1991just a path mistake in palis patch for ke-recv23:17
merlin1991Pali: your patch to ke-recv seems to work23:39
Palinice :-) (I did not tested them yet)23:39
*** _rd has joined #maemo-ssu23:40
*** Pali has quit IRC23:41
merlin1991oaky arcean, install http://cdnm.at/~christian/maemo/cssu/testing-testing-enabler_0.1_all.deb and then run ham :)23:42
merlin1991changelog can be found here (since the wiki i non editable) http://gitorious.org/community-ssu/mp-fremantle-community-pr/blobs/master/debian/changelog23:43
keriodafuq is that23:45
merlin1991kerio: hm?23:45
keriooh, HAM domain information23:46
keriomeh23:46
arceanok, HAM is running and melting my cpu :)23:46
merlin1991and pretty please don't run into "no update here" or "dependency hell" :D23:46
keriomerlin1991: is the priority for that repo the same as the other cssu repos?23:47
keriootherwise, it's going to be hell to go back for the peeps using HAM only23:47
kerioand yes, i said peeps23:47
merlin1991kerio: this is not ment for the peeps23:47
merlin1991but rather for like 3 people to test a release before we release it :D23:48
DocScrutinizer05kerio's concern is valid though23:48
DocScrutinizer05would it 'recover' from test by simply enabling the 'normal' repos again?23:49
DocScrutinizer05or would any tester have to rollback/reflash?23:50
merlin1991this is rollback reflash material23:50
DocScrutinizer05...to join community of normal mortals23:50
merlin1991you can "recover" your ham by simply removing the package (even inside ham)23:50
DocScrutinizer05ok23:50
merlin1991but that sill would leave the upgraded mp on your device23:51
DocScrutinizer05yep23:51
DocScrutinizer05which wouldn't update 'normally' on next CSSU release23:51
kerioit appears to be the same as community-devel23:51
merlin1991DocScrutinizer05: well mext cssu relaese would be 100% the same packages23:52
DocScrutinizer05I could run this on a virgin device23:52
keriothis, of course, ignoring the fact that it might just *not work* because HAM also does checking by gpg key and thus it could put every package in one priority basket23:52
merlin1991and the release after that would work as usual again23:52
kerios/it/the priority separation/23:52
infobotkerio meant: this, of course, ignoring the fact that the priority separation might just *not work* because HAM also does checking by gpg key and thus it could put every package in one priority basket23:52
merlin1991since as soon as you remove the enabler the "domain" the package comes from has no priority, thus any package from the "main" cssu repo has higher priority23:52
keriomerlin1991: hm, that's a good point23:53
merlin1991the only secnario in which you have to roll back is when shit blows up :D23:53
kerioDocScrutinizer05: "easy" way to fix, considering that people that will install that are "l33t", would be to enable redpill mode and disable the domain checking23:54
DocScrutinizer05well, anyway I could test run this on a virgin N900 only. No way to nuke my daily phone with it23:54
merlin1991but it shouldn't (obviously I ran it on my device already)23:54
DocScrutinizer05so I could either run a direct install of new testingtesting, or install recent official cssu-t then upgrade23:55
keriomerlin1991: why did you password the repo, btw?23:56
merlin19912nd23:56
merlin1991kerio: so that random peeps with 0 knowledge don't get in23:56
merlin1991since this is meant to break stuff in a limited circle where we can minimze damages23:57
DocScrutinizer05hmm password?23:57
* kerio kinda wants to paste the url here23:58
merlin1991DocScrutinizer05:  well the repo resides here: http://maemo.merlin1991.at/cssu/future/23:58
merlin1991it uses basic_auth which apt supports, the enabler adds user and pw to the sources.list23:58
DocScrutinizer05aah23:58
arceanmerlin1991, so far everything works ;)23:59
DocScrutinizer05ok then, has to wait another 2 or 3 hours23:59
DocScrutinizer05dinner first23:59
*** user_ has joined #maemo-ssu23:59
merlin1991kerio: what do you think of the url + user + pass combinarion? :D23:59
keriomerlin1991: "the future happens now"?23:59
merlin1991yep23:59

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!