[11:34] <Cimi> davidbarth: I would move the deadline for the RGBA style property support in murrine after 27th august
[11:35] <Cimi> before the 27th, I hope cody will send me patches for the widget ordering
[11:35] <Cimi> and the resize gripper
[11:46] <davidbarth> Cimi: hi
[11:50] <davidbarth> Cimi: the argb style prop s not critical, but what's blocking to choose the name (as it seems the rest of the code is here already)?
[11:50] <davidbarth> Cimi: widget ordering, how critical is that for the theme? cody was not really keen on changing that at this point in the release
[11:58] <Cimi> davidbarth: it is not critical
[11:58] <Cimi> it's just better :)
[11:58] <Cimi> with that change, the theme will definitely look better
[12:00] <klattimer> davidbarth: do you know of any possible dbusmenu bugs which might cause strange race conditions?
[13:13] <davidbarth> klattimer: hi
[13:13] <klattimer> hey
[13:13] <davidbarth> what kind of side-effect are you getting?
[13:14] <klattimer> well, if the ibus icon is passed through libindicator into the notification area it works, if it's in the indicator-applet it doesn't
[13:14] <klattimer> throwing a dbus error which looks like a race condition
[13:15] <davidbarth> passed thru lib as opposed to the indicator? i don't follow you
[13:15] <klattimer> well, if you don't have the indicator applet on the panel
[13:15] <klattimer> it puts an icon in the status area instead
[13:15] <davidbarth> ah ok
[13:16] <davidbarth> and? (re-reading)
[13:16] <klattimer> so the only difference being the existence in the indicator applet or the notification area, one works, one doesn't
[13:16] <davidbarth> so libindicator is not properly propagating the icon
[13:17] <davidbarth> the old systray works, not the new indicator one, right?
[13:17] <klattimer> let me explain again
[13:17] <klattimer> starting a patched ibus-daemon without an indicator applet in the panel, puts an icon in the systray (albeit incorrect, but that's a bug in the way icon picking works)
[13:18] <davidbarth> ok
[13:18] <davidbarth> normal fallback mode for libindicator
[13:18] <klattimer> starting ibus-daemon with an indicator applet, will cause the indicator to throw a dbus error on selecting the input method engine
[13:18] <davidbarth> k
[13:18] <klattimer> you can see the bug report here; https://bugs.launchpad.net/ubuntu/+source/ibus/+bug/564034
[13:18] <klattimer> comment #38 has the dbus message in it
[13:18] <davidbarth> checkng
[13:19] <klattimer> when I tested this on friday things were working
[13:19] <klattimer> so an update between then and now has caused the problem
[13:21] <davidbarth> klattimer: and i guess that breaks around line 201 in __set_im_icon
[13:21] <klattimer> nope
[13:22] <klattimer> around 520 in __im_menu_item_activated_cb
[13:22] <davidbarth> ah really?
[13:22] <klattimer> the set_im_icon stuff works perfectly
[13:22] <klattimer> yep... when calling focus_ic.set_engine(engine)
[13:24] <davidbarth> i don't see the connection between the engine and anything gtk that could trigger a signal and force libindicator to push a change across the bus
[13:26] <klattimer> davidbarth: neither do I
[13:26] <klattimer> so, I'm currently thinking that it's some kind of dbusmenu slowness or another race condition
[13:26] <davidbarth> hmm, there seems to be a reference to engine.icon and engine.longname somewhere in this panel.py module
[13:27] <davidbarth> right, set_im_icon(engine.icon)
[13:27] <davidbarth> so even if it's not pushed immediately, it's the link
[13:27] <davidbarth> the callback somehow triggers a signal that sends an update across dbus
[13:28] <davidbarth> but why does it break, no ideas so far
[13:28] <klattimer> but why would it generate an error in dbus saying there's no engine
[13:28] <klattimer> it makes no sense
[13:28] <klattimer> and also, why would it work fine in the systray but not in the indicator-applet
[13:29] <davidbarth> the log of changes is here: https://code.edge.launchpad.net/~indicator-applet-developers/indicator-application/trunk
[13:29] <davidbarth> because of a change in API
[13:29] <davidbarth> the change broke the ABI
[13:29] <davidbarth> if you have elements that are not packaged you're relying on, you'd better make distclean and re-autogen.sh
[13:30] <klattimer> right I'll give that a go
[13:30] <davidbarth> that's the most probable reason to me so far
[13:31] <davidbarth> klattimer: btw, do you have everything you need for the rest of the keyb patches? like icons?
[13:31] <klattimer> davidbarth: the existing icons are being used atm
[13:31] <klattimer> I've found that I need to produce a very long list of icons
[13:31] <klattimer> if we're going to theme them
[13:32] <klattimer> a good start would be the m17n icon list I already sent you
[13:32] <davidbarth> right
[13:32] <davidbarth> but design has it too, need to check when they can get back to you
[13:33] <klattimer> k
[13:33] <klattimer> I'm starting to think that ibus is going to miss the release of maverick
[13:33] <klattimer> the code is unpredictable a lot of the time :/
[13:33] <davidbarth> also, i wanted to mention a new change request for g-p-m, to update some strings and logic for displaying the time remaining for a battery
[13:34] <davidbarth> that's "Change the battery strings that are too long in the menu: TODO" in https://blueprints.edge.launchpad.net/ubuntu/+spec/dx-m-indicator-power
[13:35] <davidbarth> and the strings definitions are at: https://wiki.ubuntu.com/BatteryStatusMenu#Items
[13:37] <klattimer> k
[13:37] <klattimer> is there a bug report?
[13:40] <davidbarth> klattimer: i'm looking for it, but the gpm bug list is, errr, impressive
[13:42] <davidbarth> mpt: ping? i'm almost sure you had a bug to change the menu strings to something more compact for the power indicator (referring to the "strech limo" of status strings or something)
[13:42] <klattimer> oh great... I've somehow managed to lose my filesystem entirely :/
[13:42] <klattimer> thankfully it's a vm
[13:42] <davidbarth> klattimer: kernel issue?
[13:43] <klattimer> no the disk strangely went into read only mode, so I rebooted the vm
[13:43] <klattimer> and now apparently there is no /sbin/init
[13:43] <klattimer> :/
[13:43] <klattimer> whole FS is gone
[13:43] <klattimer> :/
[13:43] <mpt> davidbarth, I don't remember there being a bug report about it.
[13:44] <davidbarth> mpt: ok, i'll open one for that then; at least getting that string properly fixed before it's too late
[13:44] <davidbarth> mpt taking https://wiki.ubuntu.com/BatteryStatusMenu#Items as the reference
[13:51] <davidbarth> Cimi: i've pinged ronoc, asking him to join this channel
[13:52] <davidbarth> Cimi: part of the sound menu widgets do come from the IDO lib (lp:ido), like the scrub bar
[13:54] <Cimi> I would like to add small rounded corners for the cover art
[13:54] <Cimi> and maybe
[13:54] <Cimi> fixing the "player"
[14:08] <ronoc> Cimi: hi
[14:09] <ronoc> i hear you are interested in working onthe sound menu
[14:09] <Cimi> yeah
[14:09] <Cimi> I found some bugs
[14:10] <ronoc> Cimi, yes it is a work in progress,  todays release should fix some of those
[14:10] <ronoc> which bugs in particular
[14:10] <Cimi> for example
[14:10] <Cimi> I guess below the playe
[14:10] <Cimi> r
[14:11] <Cimi> there's a squared area
[14:11] <Cimi> filled with the menu bg color
[14:11] <davidbarth> klattimer: https://bugs.launchpad.net/ubuntu/+source/gnome-power-manager/+bug/619816
[14:11] <klattimer> davidbarth: yeah I saw that come in
[14:11] <ronoc> cimi: fixed in this release 0.4.0
[14:11] <Cimi> ronoc: that does not work with the new menus we have in maverick
[14:11] <Cimi> ok
[14:11] <Cimi> maybe I should just need to wait
[14:12] <davidbarth> klattimer: watch out, there are string changes, but i guess there is also some of the algorithm to check to see if the time intervals (<12h, >, calculating) can match the different strings
[14:12] <davidbarth> klattimer: they make sense from a user point of view, but that may not be the code/HW reality; i haven't checked yet
[14:12] <klattimer> k
[14:13] <ronoc> Cimi: I think it would be best since I have not released in 3 weeks because i was waiting for mpris2 to stabilize, most UI quirks have been resolved. Nice to haves can come next once all functional stuff is resolved. The next week or two will be working on integration bugs with the clients who are at present moving over to mpris2
[14:15] <Cimi> great
[14:15] <Cimi> will wait then
[14:15] <davidbarth> Cimi: so i think you should work from the indicator-sound source branch today if you want to adapt the theme
[14:15] <davidbarth> Cimi: or wait for tomorrow's release
[14:16] <Cimi> wait tomorrow :)
[14:16] <davidbarth> Cimi: if you're into indicators, the datetime one would need some love to get the calendar styled better i think
[14:17] <davidbarth> Cimi: ie probably removing the surrounding frame to integrate it seamlessly into the rest of the menu
[14:17] <davidbarth> Cimi: i think only the month/year header should be decorated, whereas the rest should not
[14:17] <davidbarth> Cimi: but you may want to double check that with chaotic first
[14:18] <davidbarth> Cimi: or compare with his latest mockups
[14:18] <chaotic> davidbarth: I'll upload the latest mockup for Cimi
[14:19] <Cimi> yeah
[14:19] <davidbarth> chaotic: cool thanks; you're on all fronts ;)
[14:19] <chaotic> davidbarth: ;)
[14:19] <chaotic> Cimi: done
[14:21] <Cimi> maybe just removing the bigger fram
[14:21] <Cimi> e
[14:21] <Cimi> the header is useful, because it is clickable
[14:22] <Cimi> but maybe it can be fixed
[14:24] <MDC1> is it possible to create a ppa directly from a launchpad bzr now?
[14:26] <Cimi> chaotic: did you see the latest work I've done for the sound indicator? :)
[14:26] <Cimi> look at the slider
[14:26] <Cimi> ;)
[14:26] <Cimi> also, I improved the rendering of disabled states for slider (mainly) and a touch at buttons too
[14:27] <chaotic> Cimi: yes, it's looking much better - is that slider big enought o grab ;)
[14:30] <Cimi> davidbarth: the code for the calendar widget is in ido?
[14:30] <Cimi> chaotic: on my netbook I can grab it
[14:31] <Cimi> chaotic: the lower slider is small
[14:31] <chaotic> Cimi: it is
[14:31] <Cimi> but you should speak with ronoc
[14:32] <chaotic> Cimi: it's a scrubber - so I wonder if it should look a little different - I had intended that one to be round
[14:57] <davidbarth> Cimi: it is, yes; idocalendar or something
[15:00] <Cimi> yeah improved btw
[15:09] <Cimi> ok fixed the calendar for both themes
[16:40] <klattimer> tedg: it seems my absolute filename patch to libindicator isn't released into the repos yet, can you tell me when I can expect it?
[16:49] <tedg> klattimer, Hmm, I believe that it is...
[16:50] <klattimer> tedg: I'm upto date on the latest libindicator and testing the ibus patch once again
[16:50] <klattimer> and all I'm getting is the missing-image
[16:50] <tedg> klattimer, Oh, no.  So what happened is that cause another bug, which they then distro patched it to break things.
[16:50] <klattimer> tedg: in ibus?
[16:50] <klattimer> I saw there was a distro patch in ibus which reversed some of my changes and broke others
[16:51] <tedg> klattimer, No, in libindicator -- basically there was an error case that hit a g_error now and shutdown unity :)
[16:51] <klattimer> oh
[16:51] <klattimer> that's not great
[16:52] <tedg> Turns out people don't like it when their window manager shuts down :)
[16:52] <klattimer> no, I can imagine it sticks in their craw a little
[16:52] <tedg> So I have a patch for that, we should release it.
[16:52] <klattimer> cool, if you can do that, then I want to push my ibus packages to a ppa
[16:52] <klattimer> so I can get them tested quickly
[16:54] <tedg> kenvandine, Can we update the distro patch to be my branch?
[16:54] <tedg> kenvandine, Do you have time for some libindicator packaging?
[16:55] <kenvandine> tedg, sure
[16:55] <tedg> kenvandine, Do you want a release, or just update the distro patch?
[16:55] <kenvandine> a patch is fine
[16:55] <kenvandine> or... a release if it means no release tomorrow :)
[16:56] <kenvandine> could kill two birds
[16:56] <kenvandine> if there is no more changes due
[16:56] <tedg> I can't imagine anymore...
[16:56] <tedg> So let's just make a release.
[16:56] <kenvandine> great
[17:02] <klattimer> kenvandine: I need to sign a package but it's assuming someone elses email address
[17:03] <klattimer> what do I need to change in order to make it work with my key?
[17:04] <kenvandine> set DEBEMAIL env variable
[17:05] <kenvandine> before doing the dch
[17:05] <kenvandine> to get it right in the changelog, which is what it looks for when signing
[17:05] <kenvandine> or
[17:05] <klattimer> dch?
[17:05] <kenvandine> debsign -k<key>
[17:05] <kenvandine> dch creates a new version, in debian/changelog
[17:05] <klattimer> oh ok
[17:06] <kenvandine> debsign -k<key> blah.sources
[17:06]  * klattimer does that stuff manually for some reason 
[17:06] <kenvandine> to sign the package for upload, if it wasn't signed when created
[17:06] <kenvandine> use dch -i :)
[17:06] <klattimer> k
[17:06] <klattimer> I will in future
[17:06] <kenvandine> dch -i creates a new version
[17:06] <kenvandine> dch
[17:06] <kenvandine> edits
[17:17] <Cimi> iainfarrell: I've found a bug in your blog :-P
[17:17] <iainfarrell> hey cimi
[17:17] <iainfarrell> thanks :)
[17:17] <Cimi> ehehe
[17:17] <iainfarrell> in design or hungfu?
[17:17] <Cimi> look at the "team page"
[17:17] <Cimi> :D
[17:17] <Cimi> Thirteen heads are
[17:17] <Cimi> better than one.
[17:18] <Cimi> but there are fourteen heads there :P
[17:18] <iainfarrell> heh!
[17:18] <Cimi> unless your head does not count! ahah