uvos__ has quit [Ping timeout: 252 seconds]
akossh has quit [Quit: Leaving.]
<Wizzup> enyc: sorry for the delayed reply
<Wizzup> enyc: we'll get a big news post out soon
akossh has joined #maemo-leste
akossh has quit [Remote host closed the connection]
<Wizzup> freemangordon: I commented the if statement you linked and now it works well
<Wizzup> in osso-abook
<Wizzup> er
<Wizzup> the if statement in eds plugin
<Wizzup> but now osso-abook works well
<Wizzup> :)
mdz has quit [Ping timeout: 246 seconds]
joerg has quit [Ping timeout: 252 seconds]
joerg has joined #maemo-leste
slep has left #maemo-leste [#maemo-leste]
<freemangordon> Wizzup: but that is not a proper solution
sch has joined #maemo-leste
<freemangordon> hmm, your contact appeared with alias
<freemangordon> not as 'no name'
<enyc> Wizzup: I'd like to see "state of usability" "users perspective" in this
<tmlind> freemangordon: did you have some modem debug questions for me?
<tmlind> hmm i think there was some usb bootmode question but i don't see it in the history or irc logs
<freemangordon> perhaps, but I forgot :)
<freemangordon> ah, you already answered (how to debug regs)
<tmlind> ok nice :)
<freemangordon> yeah, but in general we discussed with uvos
<freemangordon> on how to properly implement modem audio in terms of kernel
<tmlind> freemangordon: so to get ofono upstreamed, changing motorola modem to use ell for n_gsm read/write is the way to go based on what was discussed years ago on the ofono list
<freemangordon> not in ofono terms
<tmlind> hmm?
<freemangordon> but in kernel
<freemangordon> now kernel driver is more or less hacky, no?
<freemangordon> it does stuff behind alsa's back
<tmlind> kernel mode n_gsm and userspace mode n_gsm pretty much the same way except for pm, so ofnono could be upstreamed no problem
<freemangordon> that's why we have the issue with disappearing sound etc
<tmlind> yeah the kernel driver solution seems to be to have callaudiod handle the mixer stuff
<freemangordon> no, sphone already does that
<freemangordon> (usiong ucm)
<freemangordon> *using
<freemangordon> so that's not an issue
<tmlind> well the issue is it should be done in a generic way, not just sphone
<freemangordon> but seems motmdm switching some frequencies on incommig call breaks hifi dai
<freemangordon> tmlind: sure, but that's not an issue
<tmlind> yeah that seems like a bug as hifi is on a different mcbsp instance compred to voice mcbsp
<freemangordon> yeah, so this bug I am talking about
<tmlind> so likely the bug is in trashing some cpcap registers
<freemangordon> mhm
fab_ has joined #maemo-leste
<freemangordon> Wizzup: ok, so I fixed the bug ion abook, somehow you should reset your tp addressbook to a proper state (like before the hack) to confirm the proper fix is working
<freemangordon> maybe deleting tp-eds cache db is enough
<freemangordon> or even removing the hack
fab_ has quit [Quit: fab_]
fab_ has joined #maemo-leste
fab_ has quit [Ping timeout: 256 seconds]
akossh has joined #maemo-leste
helslwed has joined #maemo-leste
pere_ has quit [Ping timeout: 256 seconds]
mdz has joined #maemo-leste
<Wizzup> freemangordon: I think just going offline/online is enough (well, and in my case restarting evolution processes)
<Wizzup> freemangordon: I think it's fine, I didn't try super hard to delete the dbs but I think the problem is well understood and if it occurs again I will let you know
<Wizzup> thanks
<Wizzup> freemangordon: just checked with the slack plugin as well, and now I do see the aliases/ids
<Wizzup> freemangordon: btw, what do I need to do to support letting conversations know when you click to perform an action on a contact
<Wizzup> does it just request a channel?
<Wizzup> looks like it
<Wizzup> yeah so looks like I need to handle channel requests in handleChannels
uvos has joined #maemo-leste
<uvos> tmlind: your missing the real kernel issue
<uvos> the real issue is that atm that alsa has no idea that the modem is playing audio
<uvos> so it disables all the amps etc
<uvos> chaning the alsa controls via ucm makes the kernel reconfigure the audio tree for modem audio output but it dosent take effect
<uvos> as as far as the kernel knows nothing is playing so it just turns everything off
<uvos> we work around this using a hacky function that just powers up a buntch of cpcap devices
<uvos> this is the hack in the kernel
<uvos> i dont know any whay to solve this problem, besides implementing a fake dai that pretends to be playing audio when the modem is in a call and thats a hack too really
<uvos> note that in mainline linux voice calls apear to work (sortof) due to a silicon bug in cpcap
<uvos> when you power off all the amps, depending on the previous regiser state, the amp for the speakerphone will stay powered, even when the corrisponding bit in the register is cleared
<uvos> this makes the speakerphone (but nothing else, incl the mic) work in mainline linux, usually
fab_ has joined #maemo-leste
<Wizzup> uvos: btw I will get back to your sphone requests this week, splitting out the fixes into a separate pr and such
helslwed has quit [Read error: Connection reset by peer]
pere has joined #maemo-leste
helslwed has joined #maemo-leste
<Wizzup> enyc: would you like to contribute to that?
<Wizzup> arno11: btw I found a memleak in the conversations rtcom db reading code, so that might solve some issues, doubt it'd account for as much as you saw though
<Wizzup> freemangordon: the userActionTime in the channel request from osso-abook is like something from 1974
<Wizzup> string "org.freedesktop.Telepathy.ChannelRequest.UserActionTime"
<Wizzup> dict entry(
<Wizzup> variant int64 145964087
<Wizzup> )
<uvos> what makes that 1974?
helslwed has quit [Ping timeout: 256 seconds]
<uvos> Sat Aug 17 09:34:47 1974
<uvos> ah
<uvos> ok
fab_ has quit [Quit: fab_]
fab_ has joined #maemo-leste
helslwed has joined #maemo-leste
<tmlind> uvos: ok no idea how the tdm components should be done, also missing is bt to call and call capture
<tmlind> that's a project for some alsa developer, the comments i recall seeing was that it should use a custom alsa driver instead of the generic one
mdz has quit [Ping timeout: 256 seconds]
pere has quit [Ping timeout: 264 seconds]
arno11 has joined #maemo-leste
<arno11> Wizzup: ok, btw there is probably something really wrong with my main install (OMP, conversations, bitlbee, tp-haze, FB plugin, presence-ui are down after very last updates...)
<arno11> freemangordon: if you can confirm you don't have troubles like that on your N900 (kernel 6.1.67 -devel), i'll burn a fresh img
<Wizzup> arno11: what makes you think something is wrong, btw?
<Wizzup> what do you mean with 'are down' ?
Juest has quit [Ping timeout: 256 seconds]
<enyc> Wizzup: possibly yes tho i'd need a spare droid4 or something better to test with "in anger", not sure the state of my ol' n900 tbh
<enyc> Wizzup: certainly could qa any wording and explain unanswered points/questions
<arno11> Wizzup: down == not working anymore
<arno11> :(
<enyc> I've always been good at bringing up points on e.g. linuxmint test isos and bringing up edge-cases and user-pov
<Wizzup> arno11: hm, any obvious errors, ?
<Wizzup> enyc: it was just that you had a very specific request so it sounded like you had an idea on what towrite
<arno11> Wizzup: btw i had some troubles with xauth few weeks ago after playing with module removing (pm stuff)
<enyc> Wizzup: i have an idea of the methodology and the principle that sort of thing applies
<arno11> and then lot of things went wrong
<enyc> Wizzup: certainyl, be interested soon in dealing with testing PMos and updated Leste and unique-advantages and functionality in practie ...
<arno11> Wizzup: so to be sure there is no bias with that, it's probably better for me to start over
<Wizzup> hm
<arno11> ah my xauthority file is empty (?!)
<arno11> anyway i'll burn a new img
<Wizzup> weird
Juest has joined #maemo-leste
joerg has quit [Ping timeout: 252 seconds]
joerg has joined #maemo-leste
<arno11> Wizzup: currently dist-upgrading to devel from a fresh img
<arno11> working well
<arno11> increasing main partition size before the first boot seems to help
<arno11> 3GB by default is definitely too small now
<Wizzup> before first boot, really? hrmm
<Wizzup> well, yeah we can make it 4GB I guess.
<arno11> definitely a good idea. And once we'll have 16bpp and new ddx in fresh img (and 4GB partition), N900 should be usable on first boot, really :)
<Wizzup> I didn't notice the size to be an issue on first boot btw, how full is it on first boot?
<arno11> really full lol
<Wizzup> ok
<Wizzup> well this is set per device, but we can change it for many devices in arm-sdk
<arno11> ok cool
<Wizzup> see 'boards: increase size' in arm-sdk
<Wizzup> 3.5GB, or 4GB?
<Wizzup> I am a little worried about extra 4GB because if someone somehow has a 4GB card it might not fit or something
<Wizzup> maybe we shouldn't care about such old, old cards
<arno11> ah indeed, could be a problem
<arno11> but yeah, anyway old small cards are definitely too slow :(
<arno11> lol i tried a phone call 'OOTB' running cmt_pulse as user but forgot that sphone portrait bug is still there by default
<arno11> maybe we need a specific n900 leste-config file for that ?
<arno11> (because there are just 2 parameters to change in sphone.ini)
<arno11> i mean just for N900
<Wizzup> yeah we probably do
<arno11> ok
<arno11> i'll try conversations stuff in few minutes and let you know if there is still weird stuff or not
<Wizzup> I'm still busy hacking away at conversations
<Wizzup> currently making osso_abook -> click on contact -> msg work
<Wizzup> so that it opens a window in conversations
<arno11> wow
<arno11> quite a big step forward
<Wizzup> ok, it works
<Wizzup> cool
<Wizzup> some of the code is spaghetti but we'll get it cleaned up a bit
<arno11> really nice :)
<arno11> argh, adding extras repo give me conflicting translations pkgs
<freemangordon> Wizzup: so, it seems the way to mark your own messages in group chat that come from another channel is through "scrollback", see https://telepathy.freedesktop.org/doc/telepathy-glib-0.24.x/TpMessage.html#tp-message-is-scrollback
slep has joined #maemo-leste
<Wizzup> yes, but I have never seen this set
<Wizzup> even for gabble where you get the last 5 messages
<Wizzup> bbia
<Wizzup> b
<freemangordon> you will see it with haze
<freemangordon> if you have FB account :)
<arno11> freemangordon: i need few minutes of your time plz
<freemangordon> just a minute to push the new purple-facebook
<arno11> ok thx
<arno11> btw in which repos do you push it ?
<freemangordon> testing
<freemangordon> maybe will push to non-testing soon
<arno11> ok
* dsc_ takes a looksie at conversations
<dsc_> new features nom nom
<dsc_> arno11: Re: memory usage, do you happen to be connected to rooms/chats that generate a lot of incoming messages?
<dsc_> i really doubt it's because of the messages but yeah
<arno11> dsc_: i'm currently reinstalling everything. got troubles with my install so not sure what was the issue with memory
<arno11> will have an answer soon
<dsc_> alright
<freemangordon> Wizzup: did you fix hildon-meta iphbd dependency?
<arno11> freemangordon: do you have still FB working on your n900 ?
<freemangordon> on my d4, yes
<arno11> because i still have the memory and cpu usage issue on a fresh install
<freemangordon> could you elaborate?
<freemangordon> like, which process is using the memory
<arno11> it depends, and really difficult to check because of the huge memory usage
<arno11> but mainly tp haze and conversations
<arno11> and it happens when i try to use purple-facebook
<freemangordon> yes, conversations seems to use lots of memory
<freemangordon> tp-haze not that much
<arno11> yes but 700MB of swap is not normal
<arno11> see previous msgs
<arno11> it was working well previously
<freemangordon> on d4 here mem usage is 306MB
<freemangordon> and no swap
<freemangordon> arno11: can't do much if cannot repro
<freemangordon> did you upgrade on latest purple-facebook and tp-haze?
<freemangordon> *to
<arno11> i tried almost every combinations :(
<arno11> still the same result
<freemangordon> also, what is this huge mem usage, per process
<freemangordon> maybe you have lots of FB contacts
<arno11> only one lol
<freemangordon> heh
<freemangordon> so, please provide some values
<arno11> ok
<arno11> 100% cpu usage instead of 10 (conversations / tp haze) and 250MB of RAM and 500MB of swap instead of 150 for both
<arno11> and after one minute, n900 is freezing
<freemangordon> wait, tp-haze is using 100% cpu?
<arno11> same issue using pidgin
<arno11> yes
<freemangordon> oh, same with pidgin
<freemangordon> did you try to delete pidgin log?
<arno11> yes, even worst
<freemangordon> how did you delete it?
<arno11> i'm using a fresh install
<arno11> my main install is dead
<freemangordon> not much changes
<arno11> i know...already check
<freemangordon> after avatar sizes, there are 2 patches
<freemangordon> both should not make any harm
<arno11> ok
<freemangordon> could you start yp-haze from cmd line?
<freemangordon> *tp-haze
<arno11> (i'll try my first built to compare)
<arno11> ok
<freemangordon> PURPLE_VERBOSE_DEBUG=1 PURPLE_UNSAFE_DEBUG=1 PURPLE_DEBUG=1 HAZE_DEBUG=all HAZE_PERSIST=1 G_MESSAGES_DEBUG=all
<freemangordon> /usr/lib/telepathy/telepathy-haze
<freemangordon> you may want to stop conversations
<arno11> i can't stop it
<freemangordon> /usr/sbin/dsmetool -k '/usr/bin/conversations --background'
<freemangordon> yes, you can :)
<arno11> lol ok
<dsc_> killall -9 `pidof conversations`
<freemangordon> no
<freemangordon> you can;t kill dsme babysited process like that
<dsc_> ah
<arno11> freemangordon: the problem is that i can't easely and quickly test things
<freemangordon> why?
<freemangordon> just set status to offline
<arno11> now i get troubles to setup a FB account, even with mc-tool
<freemangordon> why do you use mc-tool?
<freemangordon> there is ui
<arno11> because currently not able to use the UI...
<freemangordon> ok, but why is tp-haze started if there is no account?
<arno11> i was talking about previous symptoms, when the device was half usable
<arno11> now it is really complicated, i can test but need time
<freemangordon> previous like? before writing new image?
<freemangordon> sure, no hurry
<freemangordon> just lmk when you have an account and you see unusual behaviour
<arno11> yes before using new image
<freemangordon> ok, but how it is with the new image?
<freemangordon> you don;t have account?
<freemangordon> please, enable extras-testing repo in HAM and install fb plugin from there
<arno11> yes ofc i have, that's not the issue
<freemangordon> I mean account set up
<freemangordon> after installing the plugin, set-up the account from there
<arno11> i'll try (if the device want lol)
<freemangordon> heh
<arno11> yes makes sense
<freemangordon> and LMK when you have it
<Wizzup> freemangordon: not yet @ dep
* freemangordon is afk, cooking
<freemangordon> Wizzup: ok
<arno11> freemangordon: thx man
<arno11> bbl
arno11 has quit [Quit: leaving]
pere has joined #maemo-leste
<Wizzup> freemangordon: so from abook I can now press 'Jabber IM' and that works, opens a window in conversations
<Wizzup> IRC IM does -not- work, it doesn't start conversations, it says internal error
<Wizzup> and before that, I had dino-im installed, and then it somehow started dino (which is not for irc)
mdz has joined #maemo-leste
<Wizzup> btw, shall we make --no-install-recommends the default for leste?
nela has quit [Ping timeout: 268 seconds]
<Wizzup> freemangordon: does it need a scheme handler or something?
pere has quit [Ping timeout: 252 seconds]
nela has joined #maemo-leste
<freemangordon> Wizzup: abook does fallback to xdg if tp gives error
<freemangordon> IRC is room handle, jabber is contact handle
<freemangordon> or?
<Wizzup> I made a contact
<Wizzup> called it 'wizzup irc'
<freemangordon> to irc?
<Wizzup> and gave it the irc nick 'wizzup'
<Wizzup> and clicked on 'irc im'
<freemangordon> how is that supposed to work?
<Wizzup> it works on fremantle
<freemangordon> hmm
<freemangordon> lemme find the code
<freemangordon> sec
<Wizzup> I need to go to dinner
<freemangordon> ok
<Wizzup> freemangordon: btw not super relevant but did you see this: 14:32 < Wizzup> freemangordon: the userActionTime in the channel request from osso-abook is like something from 1974
<Wizzup> It's not relevant, but it seems buggy
<freemangordon> yes, but I have no idea where that comes from
<freemangordon> ah, wait
<freemangordon> Wizzup: maybe make debug build of osso-abook and put bp in request_channel()
<freemangordon> that way we can see what's going on
<freemangordon> in the meanwhile I'll try to repro in the VM
<freemangordon> what account did you use when created IRC contact?
<freemangordon> Wizzup: I am unable to create irc contact, how did you do that?!?
<freemangordon> hmm, scratch that
<freemangordon> Wizzup: yet another bug, I am on it
xmn has joined #maemo-leste
<freemangordon> Wizzup: uvos: seems tp event timestamp is not epoch, but some x11 timestamp (or gdk, or...) see https://telepathy.freedesktop.org/doc/telepathy-glib-1/telepathy-glib-util.html#tp-user-action-time-from-x11
<Wizzup> freemangordon: the ones sent by converation internally are correct btw (2024)
pere has joined #maemo-leste
<freemangordon> are you sure they are?
<freemangordon> see the link ^^^
<Wizzup> back
<Wizzup> freemangordon: but it seems like you made it int64
<Wizzup> so that might be enough to fix it
<Wizzup> I will report
<Wizzup> freemangordon: now I get this: QDateTime(1975-07-08 13:31:28.000 PDT Qt::LocalTime)
<Wizzup> freemangordon: how should the irc dm work? should osso-abook request a channel with tp directly, or does it invoke conversations the app?
<freemangordon> irc dm?
<freemangordon> abook does tp_account_channel_request_new_text() and leaves to TP to do whatever needed
<freemangordon> still, what is 'irc dm'?
<Wizzup> so what I am seeing on my vm is that osso-abook is executing conversations
<Wizzup> not requesting a channel like it does for xmpp
<freemangordon> did you restart abook after the upgrade?
<Wizzup> irc dm = irc 1:1 message
<Wizzup> let me double check
<freemangordon> you have to kill the process
<Wizzup> ah, no, I grepped for abook
<Wizzup> lol
<Wizzup> not addressbook
<freemangordon> heh :)
<Wizzup> yes, now it works, ty
<Wizzup> it works for sip too
<Wizzup> great
<freemangordon> Wizzup: re timestamp - why do you think it is epoch?
<freemangordon> so, this is what abook does
<Wizzup> the x11 mention is there only to convert some x11 client event to a different time
<freemangordon> mhm
<Wizzup> gint64 tp_user_action_time_from_x11 (guint32 x11_time);
<freemangordon> see the commit ^^^
<Wizzup> I am looking at it
<freemangordon> x11_time is not epoch afaik
<Wizzup> I'm observing that it's still way off, compared to the other timestamps that I get
<freemangordon> but time since boot or somesuch
<Wizzup> let me find other doc
<freemangordon> please do, I was not able to
<Wizzup> freemangordon: well I can't figure out what the number that osso-abook passes would mean
<freemangordon> "Note that the time is dependant on the local environment and so is not necessarily a wall-clock time"
<Wizzup> like, what is 4 years off of 1970?
<Wizzup> conversations set the right time fwiw
<Wizzup> I nicely get 2024... from it
<Wizzup> so what is the local time then if it's 1974? or 4 years? or is it in milliseconds?
<freemangordon> where does it get it from?
<Wizzup> don't know, tp qt just does it
<freemangordon> ok, so, is this 32 or 64 bit epoch?
<Wizzup> https://docs.gtk.org/gtk3/func.get_current_event_time.html this says it's available since gtk3
<Wizzup> probably wrong
<freemangordon> no, it is avail in gtk2
<Wizzup> in any case, conversations ignores this timestamp
<Wizzup> so it doesn't matter, but the osso-abook one is wrong from my pov
<freemangordon> it should not :)
<Wizzup> what should it do with it?
<freemangordon> yes, I understand, but I don;t know which one is the proper one
<freemangordon> maybe we can check what empathy sends
<freemangordon> ttyl
<Wizzup> I think you can just send epoch
<Wizzup> and it'll be solved
fab_ has quit [Ping timeout: 246 seconds]
<freemangordon> will check what empathy is doing
<Wizzup> freemangordon: how could I add additional fields to a contact that aren't currently represented?
<Wizzup> for example, can you write a facebook contact id to a contact?
<freemangordon> how useful is that without fb account?
<freemangordon> if you have such account set-up, then facebook will appear as an option ;)
<freemangordon> there are still bugs in abook - it does not refresh properly when account is appended
<freemangordon> I will have to fix that
<Wizzup> hm?
<Wizzup> Ah, ok, so it does appear as an option
<freemangordon> yes
<freemangordon> but you have to have an account
<freemangordon> it is the same with IRC, for example
<freemangordon> if you don't have IRC account, then you cannot add IRC contact
<freemangordon> Wizzup: re timestamp - it should do whatever specs say
<Wizzup> 21:58 < freemangordon> but you have to have an account
<Wizzup> understood
<freemangordon> mhm
<Wizzup> my slack account just wasn't fully logged in
<Wizzup> that's why
<Wizzup> now it shows
<freemangordon> no, it is rather bug in abook not refreshing properly
<freemangordon> I don;'t think account shall be online in order to add contacts for that account
<freemangordon> but will test it properly soon
<Wizzup> the contacts also weren't showing
<Wizzup> but now they are
<Wizzup> I think it was a problem in haze this time
<Wizzup> 'slack im' also works from abook :)
<freemangordon> I think abook does not refresh properly
<freemangordon> Wizzup: why messages I send result in notification?
<Wizzup> let me see if I also see that
<freemangordon> hmm, not now
<freemangordon> but now yes
<Wizzup> I think I know how to fix it
<Wizzup> - if(isBackground && notificationsEnabled && !notificationMap.contains(msg->group_uid())) {
<Wizzup> + if(isBackground && notificationsEnabled && !notificationMap.contains(msg->group_uid()) && !msg->outgoing()) {
<freemangordon> sometimes it happens, sometimes not
<freemangordon> heh
<Wizzup> yeah it depends on whether the window is 'active' I think
<Wizzup> the notifications needs some more work
<freemangordon> ok
<freemangordon> nice progress though
<freemangordon> however, 100MB RES is not cool :)
<freemangordon> OMP uses 49, so it is not QT
<Wizzup> yes, I noticed that too
<Wizzup> it happens right on start
<freemangordon> hmm, osso-addressbook also uses too much mem for my taste
<freemangordon> I'll check if I can shave some of it
<Wizzup> I think one of the problems is that now the sphone messages will not show up anymore in conversations
arno11 has joined #maemo-leste
<Wizzup> because they don't have a group_uid
<Wizzup> but won't be a problem when conversations handles sms though
<freemangordon> what is this used for?
<Wizzup> what?
<freemangordon> group_uid
<Wizzup> everything :)
<Wizzup> it is the identifier for a specific chat
<freemangordon> ok, but what is the... ummm... purpose :)
<Wizzup> local_uid, remote_uid is not enough
<freemangordon> how's that?
<Wizzup> local_uid = you
<Wizzup> remote_uid = person that send or received a message
<freemangordon> local_id is not me
<Wizzup> if this happens in a channel, then how will you know?
<Wizzup> (you won't)
<Wizzup> this is why fremantle uses group_uid
<freemangordon> but rather the local account used
<freemangordon> ah, so it is not something new?
<Wizzup> and writes to it the persistent tp id
<Wizzup> no it is not new, it's what fremantle uses
<freemangordon> ok
<Wizzup> although they use a peculiar group_uid for sms
<Wizzup> (the last 6-7 digits of the phone number)
<Wizzup> I think the rtcom test ui even has an example of grouping by group uid
<freemangordon> ok
<freemangordon> I haven't looked at rtcom much
<arno11> sorry guys, can someone confirm to me the exact extras and extras-testing repo addresses (it makes me crazy with conflicting pkgs...)
<Wizzup> on leste?
<Wizzup> I don't know about extras-testing
<freemangordon> sec
<arno11> on leste yes
<arno11> thx
<freemangordon> arno11: keep in mind you shall use HAM to add the repo
<arno11> i know
<freemangordon> deb https://maedevu.maemo.org/extras chimaera-testing main contrib non-free
<arno11> ok tyvm
<freemangordon> the other one is already there (extras)
<freemangordon> deb https://maedevu.maemo.org/extras chimaera main contrib non-free
<arno11> yes for this one that's ok
<arno11> but it returns conflicting translations pkgs in apt btw
<arno11> that's pretty new
<freemangordon> weird
<freemangordon> dunno, Wizzup built some translations few days ago
<arno11> yes indeed
<arno11> ah
<Wizzup> wait, I did?
<freemangordon> didn't you?
<Wizzup> I thought you did
<freemangordon> no
<Wizzup> then I probably did
<freemangordon> you pastred some command in the irc
<Wizzup> btw, any ideas on how to debug conversations memory usage?
<Wizzup> valgrind?
<freemangordon> but I never executed it
<freemangordon> yes, valgrind
<freemangordon> Wizzup: we shall push the new translations anyway
<Wizzup> that is for what is freed or not, but will it also show where the most allocations come from>?
<freemangordon> no
<Wizzup> because I don't think 100MB on start is a memory leak issue
<freemangordon> I don;t think there is an easy way to track that
<freemangordon> I would start with commenting objects creation
<freemangordon> until you get a drop :)
<Wizzup> unless it's in abook, tpqt, or some other place :)
<freemangordon> abook itslef uses about 40MB
<Wizzup> in-process?
<freemangordon> osso-addressbook
<Wizzup> ok
<freemangordon> home applet uses about 30MB
<freemangordon> so the lib itself should not be that big
<freemangordon> maybe 15MB or somesuch
<Wizzup> ok
<Wizzup> it's not nothing, but the question is where the other ram goes
<freemangordon> hmm, actually osso-addressbook uses 36MB
<freemangordon> wtym? it uses GTK and TP and EDS
<Wizzup> I mean in conversations
<freemangordon> ah
<Wizzup> like i'm sure there are probably some leaks to fix now in my tp code, but none of that should add up to 100MB at the start
<freemangordon> ok, could it be that conversations load all the theme?
<freemangordon> instead of using sapwood?
<Wizzup> I was thinking if it was the addition of the theme code, yes
<freemangordon> or, it could be js engine
<Wizzup> it wasn't this much before
<Wizzup> so I don't think it is js engine
<freemangordon> bisect?
<freemangordon> Wizzup: BTW, conversations does not properly identify group chat messages
<freemangordon> like, message comes from a particular contact handle, but, in group channel
<freemangordon> conversations show it like it is IM message from that particular contact
<freemangordon> instead of using contact handle to get the alias to show in the UI
<freemangordon> I tested with empathy and it behaves properly
<freemangordon> just FYI
<freemangordon> anyway, going to have some sleep, night!
mdz has quit [Ping timeout: 268 seconds]
<Wizzup> freemangordon: this is with haze?
<Wizzup> freemangordon: because this works fine with say telepathy-gabble
<Wizzup> 22:51 < freemangordon> conversations show it like it is IM message from that particular contact
<Wizzup> with 0.6.0 ?
<Wizzup> fwiw I reverted the theming code and there is no big difference
<Wizzup> wait I think there might be more
<arno11> freemangordon: i have FB working well with bitlbee-FB but purple-FB is still not working: tp-haze debug returns: cancelling connection, releasing a socket, destroying socket, request performed without succes
<arno11> i can send you tomorrow full details
<Wizzup> arno11: do you get channels in bitlbee-fb ?
<Wizzup> in conversations I mean
<Wizzup> assuming you configured an irc account in conversations
<arno11> atm i can't but sure i'll do it tomorrow
<arno11> that's a good way to compare
<Wizzup> conversations 0.6.0 doesn't have ui to join channels yet
<Wizzup> but I think if you get added to a channel by bitlbee it should work
<arno11> yes pobably
<arno11> *probably
<Wizzup> freemangordon: seems to be 6.5 in stable, but now we also link abook, gtk, ...
<Wizzup> memory usage percent I meant
<dsc_> the theming feature only reads some hildon themefiles files, parses the color codes, and passes them to QML
<dsc_> it would not be something that introduces more memory
<dsc_> without theming the QML would also have color codes defined :P
<Wizzup> yeah it does not I checked
<dsc_> i should look into Qt profiling tools
<dsc_> will do tomorrow...
<uvos> i mean qml is also just a pig in terms of memory, since it has to load the interpreter
<dsc_> that should go into IFDEF DEBUG
<dsc_> (related to startup time, not memory)
arno11 has left #maemo-leste [#maemo-leste]
arno11 has joined #maemo-leste
bencoh_ has joined #maemo-leste
Amnesia_ has joined #maemo-leste
parazyd has quit [Ping timeout: 260 seconds]
parazyd has joined #maemo-leste
arno11 has left #maemo-leste [#maemo-leste]
bencoh has quit [Ping timeout: 268 seconds]
Amnesia has quit [Ping timeout: 268 seconds]
bencoh_ has quit [Changing host]
bencoh_ has joined #maemo-leste
bencoh_ is now known as bencoh
uvos has quit [Read error: Connection reset by peer]
xmn has quit [Quit: ZZZzzz…]
xmn has joined #maemo-leste
l_bratch has quit [Ping timeout: 276 seconds]