/srv/irclogs.ubuntu.com/2013/01/23/#ubuntu-desktop.txt

RAOFchrisccoulson: Why does Thunderbird make it so difficult to send plain-text email?00:15
TheMusoRAOF: How so? I've never had a problem sending plain text email...00:22
TheMusoMind you I always set things up such that I view plain text emails, and make sure my replies are sent as plain text only.00:22
desrtRAOF: you just have to find the right places to turn it off00:24
desrtRAOF: i agree -- it's pretty damn obnoxious00:24
desrtRAOF: consider, though, that mozilla is very very much about html, as a company00:24
desrtand it makes a bit of sense00:24
RAOFYeah.00:24
micahgRAOF: it's one preference00:29
RAOFmicahg: Where?00:30
micahgEdit -> preferences -> composition -> Send OPtions00:30
RAOFAnd add * to Plain Text Domains?00:31
micahghrm, that's not what I thought it was00:32
RAOFYeah, that's where I looked first, but it doesn't have a “Just send plain text, damnit” option.00:32
desrti think it's a per-account setting or something00:33
desrti remember it _not_ being easy to find00:33
desrtya... under account settings for each account00:34
desrtthere is a checkbox in "Composition and addressing"00:34
RAOFAh, yes. There we go.00:34
RAOFGood work on that dialog being > 768 pixels high, by the way.00:34
lifelessWIN00:35
TheMusoHTML in email really shoudl have been banned years ago.00:38
TheMusoIts a bitch when I get emails that don't have a plain text equivalent.00:39
TheMusoWhoops, meant pain, a little milder...00:39
desrtheh00:39
desrti'm infuriated enough when i get html-only email that i can't even imagine...00:40
lifelessTheMuso: bitch is fine :>00:45
TheMusoheh ok.00:49
lifelessTheMuso: if it was aimed at someone, it definitely wouldn't be00:50
lifelessTheMuso: but its very mild on the swearometer, and you are normally -so- polite ... :)00:50
desrtlifeless: i find people's swearometers are calibrated differently according to country00:51
lifelessdesrt: yeah00:53
RAOFMan, those fglrx-updates/nvidia01:39
RAOFMan, those fglrx-updates/nvidia-updates SRUs are a gnarly mass of unknown status.01:40
dufluRAOF: Unknown status?01:40
RAOFduflu: It's hard to determine what the status of the various pieces are. Have the updates been verified? Which packages are verified? What about Godot?01:41
dufluRAOF: So the usual... just trust us... this new version is better :)01:41
dufluOr are they being uploaded without any -proposed testing?01:42
RAOFOh, they're in -proposed now.01:43
RAOFI'm just trying to work out whether they should be move from -proposed to -updates.01:43
dufluRAOF: Mind-reading of graphics vendors is in your job description, no?01:44
RAOFOh, thsi is a problem partially of my own making. The *bugs* are a gnarly mass of confusion.01:45
RAOFOk. So the only thing preventing fglrx-updates from hitting precise-proposed is https://bugs.launchpad.net/ubuntu/+source/fglrx-installer-updates/+bug/1039397 not being verified.01:47
ubot2Ubuntu bug 1039397 in fglrx-installer-updates (Ubuntu) "[Dell Precision M6700] HDMI audio fails, HDMI/Displayport output has blank frame" [Medium,Incomplete]01:47
=== bschaefer_ is now known as bschaefer
=== rickspencer3_ is now known as rickspencer3
bryceRAOF, yeah it's quite a confusing mess.  Think the verification requirements are being overly strict?02:20
RAOFbryce: Not really. Two of the bugs linked in the SRU are not actually bugs, so can't be verified. That's my fault for letting the package into proposed.02:22
brycepart of the problem is the installer for the proprietary drivers involves a bit of playing russian roulette02:22
pittiGood morning05:29
highvoltagepitteah06:36
didrocksgood morning07:45
jibelgood morning08:00
didrockssalut jibel!08:02
pittihey didrocks, how are you?08:05
didrockspitti: still tired, but going in the right direction :)08:06
didrocksyourself?08:06
pittiI'm quite fine, thanks; making some nice progress on mocking hardware08:06
didrockssweet!08:06
pittigphoto2 --auto-detect already works, now working on actually mocking the data as well08:06
didrocksexcellent :)08:07
jibelhey didrocks08:07
=== tkamppeter_ is now known as tkamppeter
seb128hey desktopers10:06
seb128pitti, hey, https://launchpad.net/ubuntu/+source/gconf/3.2.5-0ubuntu6 ... did you just forget to push or did you forgot to use the vcs?10:06
seb128pitti, if that's the second one don't bother, I can just merge it back there from the archive version10:07
pittiseb128: sorry, forgot VCS; I'll commit it now, for teaching myself :)10:07
seb128pitti, danke ;-)10:07
pittiseb128: done, sorry10:08
seb128pitti, no worry, thanks!10:09
=== vrruiz_ is now known as rvr_
=== Sweetsha1k is now known as Sweetshark
ogra_seb128, hmm, i have a slight problem with gsettings on the nexus ... we ship http://paste.ubuntu.com/1562339/ ... and in dconf-editor i can see "suspend" being the setting thats used, nontheless it gives me the interactive dialog *until* i once switch the setting in dconf-editor manually once11:26
ogra_any idea what that could be ?11:26
seb128ogra_, do you run glib-compile-schemas on rootfs/usr/share/glib-2.0/schemas/ubuntu-defaults-nexus7-powerbtn.override  to get gschemas.compiled updated?11:29
seb128ogra_, the glib triggers take care of that on the system when installing package, not sure that happens for your rootfs11:29
ogra_i thought dh does that automatically11:29
ogra_oh, rootfs is only a package path11:30
ogra_the binary package puts it in /11:30
ogra_(some weird setup of the ubuntu-defaults stuff)11:30
seb128ogra_, gsettings get org.gnome.settings-daemon.plugins.power button-power11:30
ogra_well, i changed it now, let me reset it first11:31
seb128what does that give when it's behaving buggy?11:31
seb128k11:31
ogra_well, dconf-editor showed the right value11:31
pittiseb128: ah, wrt. https://blueprints.launchpad.net/ubuntu/+spec/desktop-r-gnome-plans-review I wasn't sure whether the "update .." ones were DONE already, as we don't yet have their final versions11:31
ogra_("suspend")11:31
seb128pitti, well, we are on the unstable serie and will track it through version, I think it's good enough for the workitems, but feel free to change it back if they are useful for tracking for you11:32
pittiseb128: so you think they are enough for "get to current GNOME 3.7.x version", and the rest will just be part of the normal update cycle?11:32
seb128pitti, yes11:32
pittiseb128: I'm on d-devel-list, so through that I should get a reminder for new devel releases11:32
ogra_seb128, "set to default" in dconf-editor gets me "suspend" back in the value field and makes the entry non-bold ... though pressing the power button now gets me the dialog again11:33
pittiseb128: if you are also watching, please feel free to prod me if I forget11:33
seb128pitti, well, I'm watching http://people.canonical.com/~platform/desktop/desktop.html in any case ;-)11:33
seb128pitti, will do11:33
* pitti te donne une accolade11:33
* seb128 hugs pitti back11:33
seb128ogra_, if the gsettings command return the right value then there is probably a bug in gnome-session...11:34
ogra_ogra@nexus7:~$ gsettings get org.gnome.settings-daemon.plugins.power button-power11:34
ogra_'interactive'11:34
ogra_ogra@nexus7:~$11:34
ogra_aha11:34
ogra_funny, why does dconf show the other value11:34
ogra_could it be that i miss any quoting in the .override file ?11:35
pittiseb128: is https://blueprints.launchpad.net/ubuntu/+spec/desktop-r-gnome-fallback even relevant for this cycle still?11:36
pittiseb128: i. e. do we want to move the automounting code in this cycle to be prepared, or move it to free resources for nexus 7 and friends?11:37
seb128ogra_, dconf-editor is hackish, gsettings lacks the api it needs so it does guess work by parsing schemas itself11:37
seb128pitti, yes that spec is still revelant, and it would be good to clean/move the automount code, that's not the highest item on the list though11:38
seb128ogra_, try ='suspend' yes (with the '')11:39
ogra_no change11:42
ogra_even tried double quotes11:42
ogra_it is a list value, i wonder if it needs special treatment due to that11:42
ogra_hmm, no it says string11:44
tkamppeterseb128, there is an upstream fix for bug 1101213, can you apply it? Thanks.11:49
ubot2Launchpad bug 1101213 in gnome-control-center (Ubuntu) "G-c-c printing doesn't allow you to select a hp printer" [High,Triaged] https://launchpad.net/bugs/110121311:49
seb128tkamppeter, sure, thanks for looking to it!11:49
didrocksLaney: btw, the 2nd blog post should answer to some of your question about how to interact with them11:52
didrocksthem beeing the PS packages11:52
didrocksLaney: if you have any question, I think we should file a FAQ in the end11:52
didrocks(that I'll paste to a wiki page)11:52
Laneydidrocks: awesome, will read!11:53
LaneyI think I want a simple document that just says11:55
Laney - bzr branch <whatever>11:55
Laney - hack hack hack11:55
Laney - bzr commit --fixes=lp:foo; bzr push lp:~/...; bzr lp-propose ...11:55
didrocksLaney: can you add that to the desktop pad?11:55
Laney - set commit message on MP11:55
didrocksLaney: I'll grab the questions for doing the FAQ11:55
Laneyok I'll add my questions there11:56
didrocksthanks :)11:56
Laneyyubikeyyyyyyyyyy11:56
didrocksLaney: part 3 will be really on the daily release process, what's the bots are doing and so on11:56
ogra_seb128, argh ...11:59
seb128ogra_, ?11:59
ogra_so the file needs to end in .gschema.override, not in .override11:59
ogra_:P11:59
* ogra_ slaps forehead11:59
seb128lol11:59
seb128sure11:59
seb128sorry I didn't pay attention to the filename11:59
Laneydidrocks: what happened here: https://launchpadlibrarian.net/128756347/indicator-power_12.10.6daily13.01.17-0ubuntu1_12.10.6daily13.01.18-0ubuntu1.diff.gz - no explanation of the changes in the changelog?12:02
didrocksLaney: because nobody attached a bug to their branch, I'm keep telling them to do that, please join the effort :)12:03
Laneyinteresting12:03
xnoxLaney: i guess lp:indicator-power has commit messages.12:03
LaneyI'm sure it does12:03
didrocksright, but commit messages can be noisy and not really explanatory12:04
didrocksthat's why I decided to limit at bug reports, which is what we generally list for any releases12:04
Laneydoes it use the commit message you can set on MPs if they are set?12:04
seb128you also don't want an hundred commit messages listed in debian/changelog if there is an hundred commits12:04
xnox"g_type_init() is deprecated as of glib 2.35.4, so bump our glib minimum requirement and remove the g_type_init() call."12:04
xnoxis the commit message and whould be useful.12:04
didrocksxnox: in that case, yeah, but not the general rule12:04
LaneyI did a MP for something the other day and got told to set a commit message there12:05
seb128there is also the changelog spam effect12:05
xnoxdidrocks: what about using first line only? (git-dch style)12:05
Laneyshould those be generally useful?12:05
xnox* [-r161] g_type_init() is deprecated as of glib 2.35.4, so bump our glib minimum requirement and ...12:05
didrocksxnox: will cut in the middle a commit message, even worse12:05
didrocksxnox: they don't use that the "short/long summary"12:05
didrockss/that//12:05
xnoxclearly we should switch to git then12:05
seb128Laney, those are a requirement for the merger bot, not sure why anymore, something to do with the readability of the vcs logs I think12:05
* xnox hides12:05
didrocksLaney: and it's only on the unity stack, different teams are using different flows, other are just taking the description then12:06
Laneyseb128: Perhaps. I've never known what they are for really but it could be useful in this case12:06
didrocks(see part 2 ;))12:06
didrocksand I think this doesn't solve the changelog spam12:06
didrocksI think important changes needs to have a bug attached12:07
didrocksor being just specified manually in debian/changelog as part of the merge12:07
didrocksthis is more an educational thing than anything else12:07
didrockslarsu: btw, can you keep that in mind when reviewing/doing changes please? ^12:07
larsudidrocks, is a --fixes lpXXXX on the commit enough?12:09
larsuor associating the MR with the bug?12:09
didrockslarsu: either the one, or the other, or even just mentionning "this fixes bug #XXXX" any commit message of the changes12:09
didrockss/the one/this/12:09
didrocksI'm quite flexible to spot that :)12:10
didrocksyou don't want to see the regexp :p12:10
larsuhaha12:10
larsuokay :)12:10
didrocksthanks!12:11
Laneyhmmmmm12:11
larsudidrocks, np, I actually prefer having a bug reference myself12:11
larsuI just forget sometimes12:11
didrocksand you get the praise then!12:11
didrockswith your name :)12:11
didrocksin the changelog :p12:12
Laneydoes it already do one changelog entry per MP?12:12
didrocksone changelog entry per bug12:12
didrocksas you would find in any NEWS file12:12
Laneyack12:13
Laneyso it feels a bit heavy to make people file bugs when they can just JFDI a fix12:13
didrocksLaney: well, if they don't want to file bugs, they can themselve add it to debian/changelog as part of the MP12:14
didrocksif they feel it's important enough to get communicated12:14
Laneyah, that is allowed eh12:14
didrocksyep :)12:14
Laneywe get to my questions on the pad now12:14
Laneyso let's leave it here :P12:14
didrocksyep :)12:14
didrocksand if they already referenced the bug in the changelog12:14
didrocksthere is no duplication12:15
=== jhernandez is now known as jhernandez|lunch
=== MacSlow is now known as MacSlow|lunch
BigWhaleWhat rolling release?! Nooooo.... :>12:53
BigWhaleThis means less release parties :(12:53
ogra_just make them bug fixing parties then12:54
BigWhaleogra_, I have those on a weekly basis! ;)12:55
czajkowskilol13:00
chrisccoulsonhey seb128, how are you?13:19
davmor2BigWhale: What are you on about it means that you have to party everyday for the newly released stuff13:20
seb128chrisccoulson, hey, I'm good, how are you?13:24
chrisccoulsonseb128, yeah, not too bad thanks13:24
seb128chrisccoulson, just woke up? ;-)13:24
chrisccoulsoni wish :)13:24
chrisccoulsonseb128, i've got this for you http://paste.ubuntu.com/1562859/ ;)13:25
seb128chrisccoulson, will you be unhappy if I discard it?13:25
seb128chrisccoulson, I just noticed this morning, looking at version, that cairo 1.12.10 is out for a week and it includes that fix (with a stack of other ones)13:26
seb128chrisccoulson, I was going to just do the update13:26
chrisccoulsonseb128, oh, that's ok then :)13:26
seb128chrisccoulson, you can do the update if you are looking for launchpad karmas though ;-)13:26
chrisccoulsonseb128, yeah, i don't mind doing that13:27
seb128chrisccoulson, great, thanks13:27
seb128chrisccoulson, http://www.cairographics.org/releases/cairo-1.12.10.tar.xz13:28
chrisccoulsonthanks13:29
=== Ursinha_ is now known as Ursinha
chrisccoulsong_variant_parse doesn't return a floating reference, does it?13:51
larsuchrisccoulson, according to gvariant-parser.c, it does13:59
larsudocs could be a little more clear there13:59
=== jhernandez|lunch is now known as jhernandez
larsubut I think close to everything that gives you a variant gives you a floating ref14:00
larsudesrt, right?14:00
=== MacSlow|lunch is now known as MacSlow
czajkowskiSweetshark: you know most OSS are on G+ right ;)14:15
mlankhorstit's like a glossy magazine for geeks14:17
czajkowskiso now I get content on there, twitter, and my inbox, 3-4 times o the same story :) fun14:17
* czajkowski bounces her nexus 7 off a wall 14:17
mlankhorsttesting the gravity sensor? :P14:18
czajkowskimy on screen keyboard has vanished14:18
czajkowskiand I've tried a lot of things now to make it come back14:19
czajkowskinothing works and I want it for fosdem14:19
mlankhorstcan you attach usb?14:19
czajkowskimlankhorst: yup14:21
seb128czajkowski, is it activated in system settings, universal access, input tab?14:21
czajkowskihttp://ubuntuone.com/0S97eFdCGOGC1Dk6mEeFlZ14:22
mlankhorstthat's probably not going to work14:23
czajkowskiindeed14:23
czajkowskiit's not14:23
mitya57that's not Universal Access panel, that's Keyboard panel14:24
seb128czajkowski, that's not what I asked14:24
czajkowskihttp://ubuntuone.com/0qF9v89DMgQG7jEUGyxzOD14:24
czajkowskithere?14:24
seb128yeah14:24
czajkowskiseb128: sorry that was a paste from earlier on14:24
seb128czajkowski, do you have a keyboard or a ssh access to type commands?14:24
czajkowskicant get up the keyboard at all14:24
seb128can you ssh to it or plug an usb keyboard for debugging?14:25
czajkowskiit's plugged in now.14:25
seb128czajkowski, can you see if onboard is running (ps ax | grep onboard)14:26
Sweetsharkczajkowski: echo chambers are fun ;)14:32
czajkowskinot really :)14:32
czajkowskiseb128: thank you14:32
Sweetsharkczajkowski: but one can test on G+, what works and then carry that to the outside world ...14:33
seb128czajkowski, yw, good luck14:33
bcurtiswxI have  a weird issue with online accounts. It says I have an AIM account that requires authorization once I start empathy, but online accounts has no record of this AIM account. I'm searching for a bad config file somewhere, but not exactly sure where to look. Any ideas?14:34
seb128bcurtiswx, run empathy-accounts14:34
seb128bcurtiswx, I guess it's an account you configured in empathy itself by then, out of the integrated online accounts14:35
bcurtiswxseb128, looks to be the case. thx14:35
seb128yw ;-)14:35
bcurtiswxif i wanted to issue a command 30 seconds after I lock the screen and have it quit after i return. what would I use ?14:40
=== m_conley_away is now known as m_conley
chrisccoulsonlarsu, doesn't http://git.gnome.org/browse/glib/tree/glib/gvariant-parser.c#n2371 make sure that the GVariant returned is not floating?14:58
larsuchrisccoulson, oh, nice catch. I must have read over that, sorry :)14:59
chrisccoulsonlarsu, i think ted must have assumed it returned a floating reference too15:00
chrisccoulsonhttp://bazaar.launchpad.net/~dbusmenu-team/libdbusmenu/trunk.13.04/view/head:/libdbusmenu-glib/server.c#L106115:00
chrisccoulsonand http://bazaar.launchpad.net/~dbusmenu-team/libdbusmenu/trunk.13.04/view/head:/libdbusmenu-glib/menuitem.c#L165915:00
chrisccoulsonin fact, everywhere it uses g_variant_parse ;)15:00
desrtlarsu: no15:01
desrtlarsu: only constructors15:01
larsuchrisccoulson, yes :(15:01
larsudesrt, thanks15:01
desrtlarsu: and you have to pretend that g_variant_builder_end() is called g_variant_new_from_builder() :)15:01
larsuhaha15:02
chrisccoulsonfantastic :)15:02
chrisccoulsondoes anyone have some rope i could borrow for a few moments?15:02
larsudesrt, another question: g_main_context_pending seems to return FALSE if there are no immediate sources to be dispatched. Is there a function to find out whether there are *any* sources?15:03
bcurtiswxdental floss OK ?15:03
desrtlarsu: you have to poll...15:03
larsu:-/15:03
desrtso like prepare/query/poll/check15:03
larsuthis makes writing a loop that stops when no events are attached very hard15:03
desrtheh15:04
* larsu could use that for his tests15:04
desrtyup :)15:04
larsudesrt, care to add a function that does this?15:04
desrtwe talked about adding this to gmaincontext15:04
desrtbut you don't want it15:04
larsuwhy?15:04
desrtbecause 'pending' is a very difficult term15:04
desrtis a timeout that will fire in 1 second from now pending?15:04
larsuwell, I just spend 20 minutes finding out "pending" is not what I thought it meant15:04
larsuYES.15:05
desrtokay15:05
desrthow about a pending async result that will fire 1 second from now?15:05
larsuyes15:05
desrtthat's the problem15:05
larsubecause you don't know about these?15:05
desrtasync results don't become GSources until they are (immediately) ready to dispatch15:05
larsuright15:05
desrtso that made us think maybe we should have like g_main_context_hold() as a sort of special "don't exit now!" ref15:06
desrtthen the whole thing just started feeling too much like GApplication15:06
larsuhm, and that's a bad thing?15:07
desrtmeh15:07
desrtit would require changes everywhere15:07
desrteveryone who may at some point add an idle would have to call this hold()15:07
desrti guess GTask and GSimpleAsyncResult could be updated to do that...15:08
desrtbut this is starting to get out of hand, wouldn't you agree?15:08
desrthttps://bugzilla.gnome.org/show_bug.cgi?id=690126 is the bug btw15:08
ubot2Gnome bug 690126 in gdbus "Do not quit mainloop while there are pending GSources" [Normal,Assigned]15:08
larsudesrt, agreed. It seemed like an easy thing at first :)15:15
desrtlarsu: did you do a lesson yet? :)15:24
larsudesrt, no :(15:24
desrtdude15:24
desrtyou're going to be in toronto in two weeks and unable to speak the local language15:25
larsuthere's just so much other stuff I need to take care of15:25
desrtlessons are quick :p15:25
larsuI do have the tab open, does that count?15:25
larsufor something?15:25
desrthahahah15:26
desrtyou TODO list by tab, hm?15:26
larsunot usually, only in this case15:26
larsubecause I actually meant to do it :)15:26
desrtwhenever i see ken he always says "haven't heard from lars yet..."15:27
larsuhe already know me?15:28
larsu*knows15:28
desrthe knows you will be sending lessons soon :)15:28
larsuheh15:28
didrocksgrrr, on 3G16:11
=== alex3f is now known as alexef
=== alexef is now known as alex3f
dobeyseb128: can i get you or someone to look at my patch on https://bugs.launchpad.net/ubuntu/+source/gnome-keyring/+bug/859600 please?17:07
ubot2Ubuntu bug 859600 in gnome-keyring (Ubuntu Precise) "Please convert gnome-keyring to multiarch" [High,In progress]17:07
seb128dobey, will have a look17:16
dobeyseb128: thanks17:21
xnoxdobey: i think last time around this was attempted, the use case was to have the library co-installed and the problem was that the /etc/ conffile encodes multiarch path.17:27
xnoxwhich means that conf file is no longer the same across all architectures and thus failing to co-install.17:27
xnoxhas this been addressed?17:27
* xnox can't figure that bit out just by looking at the patch.17:28
* xnox didn't build both amd64/i386 & try to co-install them yet.17:28
dobeyxnox: which conf file?17:30
xnoxlet me dig into it. 1sec.17:31
dobeyxnox: i have them co-installed on my machine from my ppa right now.17:31
xnoxawesome =)17:31
dobeyxnox: the p11kit conf file only has the module name in it17:31
xnoxhmm..17:31
dobey/etc/pkcs11/modules/gnome-keyring.module that file, that is17:32
xnoxdobey: why is the patch against oneiric and not raring?!17:40
dobeyxnox: eh?17:41
dobey+gnome-keyring (3.6.2-0ubuntu2) raring; urgency=low17:41
xnoxdobey: i downloaded not your patch...17:41
dobeyxnox: you're not looking at *my* patch :)17:41
xnoxi was looking at your patch, but my clipboard had the other url in it...17:42
dobeyoh17:42
xnox=) all is fine now.17:42
dobeythanks17:43
seb128chrisccoulson, still working on that cairo update?18:17
seb128chrisccoulson, if you do, please include http://cgit.freedesktop.org/cairo/commit/?id=fa4f48cccb6c7f4e1afb2ff4b98b906b7d8d4afc , it fixes a regression from the new version18:18
seb128c.f bug #110131618:18
ubot2Launchpad bug 1101316 in cairo (Ubuntu) "Evince switches to wrong page while scrolling in continuous mode" [High,Invalid] https://launchpad.net/bugs/110131618:18
LIDHHello, I have Ubuntu 12.10 and a POS system EBN X-950 with touchscreen (EgalaxyTouch according to the manual), so, i tried $lsusb and it doesn't list the touchscreen controller. If I do a screen /dev/ttyS[0-4] can't get any input from the touchscreen. Already did $ modprobe -r usbtouchscreen and still doesn't detect, any ideas what's the problem?18:36
dobeyLIDH: #ubuntu is the general help channel18:43
LIDHok18:43
LIDHthank you but i've already post there, and no one helped me :s18:43
dobeywell be patient18:45
dobeyif you're impatient, nobody will want to help you :)18:46
sarnoldLIDH: check 'dmesg' output for information, you might get lucky18:47
robruxnox, thanks for fixing bug 1099589 ;-)19:21
ubot2Launchpad bug 1099589 in devscripts (Ubuntu) "wrap-and-sort crashed with TypeError in save(): must be str, not bytes" [Critical,Fix released] https://launchpad.net/bugs/109958919:21
xnoxrobru: np =)19:21
chrisccoulsonhi seb128. sorry, was away for dinner and exercise20:13
chrisccoulsonyeah, i'll include that change too20:13
seb128chrisccoulson, no worry, thanks ;-)20:14
seb128chrisccoulson, I hope for you it's "exercise and dinner", not good to go to cycle after eating ;-)20:15
robrumterry, if you get a sec I'd like to ask you some questions about that mp you sent me21:00
mterryrobru, the deja-dup one?  sure21:00
robrumterry, yeah, so, what steps should I take to ensure that your change is working21:00
robru?21:00
mterryrobru, well make check is always good, but I only fixed the existing tests to work with the code change, I didn't add any new tests for it, because I didn't know how to fake partitions without root access21:04
mterryrobru, so I personally verified it manually.   The trick is to do a backup from a partition that is different than the partition /tmp is on21:04
mterryrobru, then run "G_MESSAGES_DEBUG=all deja-dup --backup" and notice the arguments we pass to duplicity.  There should be a --tempdir with an appropriate (same partition as source files) directory21:05
robrumterry, yeah, that's what i meant. so my home directory is different than tmp21:05
mterryrobru, so deja-dup should pass ~/.cache/deja-dup/tmp as tempdir to duplicity21:05
robrumterry, oh, btw, I started using deja-dup for my actual, real $HOME backups. Will that command interfere with the backups I have going?21:06
robrushould i make a dummy user for this? ;-)21:06
mterryrobru, it will use your normal settings to make a backup yeah.  But it won't (shouldn't) interfere with them, just start a new backup21:08
robrumterry, so it'll try to backup all 500GBs then? maybe I will make a dummy user ;-)21:08
mterryrobru, you also might want to double confirm that ~/.cache/deja-dup/tmp is empty after the run21:08
mterryrobru, well, it will do an incremental or whatever, but if you haven't made a first backup yet, then yah.  dummy or change your backup settings momentarily21:09
robrumterry, ok, great. I'll try it out shortly. thanks21:09
robrumterry, I'm getting a weird xmllint error on 'make check'21:49
mterryxmllint?21:49
mterrypaste me21:50
robrumterry, http://paste.ubuntu.com/1564351/ does this mean anything to you?21:50
robruprobably a missing dep or something simple21:50
mterryrobru, oh, I bet the gnome-docs automake stuff is missing a dep on the 'all' target for the 'check' target21:50
robrubut I'm not sure21:50
mterryrobru, run make && make check21:50
robruahhh21:51
robruin unrelated news, today I found a little netbook of mine that's been in storage since November. It had raring installed on it... now I'm just installing all 905 updates ;-)21:52
robrumterry, ok, xmllint is passing, tests are going ok so far, just have to do the manual checks and then I'll approve your branch21:53
mterryrobru, cool.  Thanks!21:54
mterryrobru, heh, only 905?!21:54
robrumterry, well, it was put in storage *late* in november ;-)21:56
robruso just 905 updates for december/jan21:56
mterry:)21:57
robrumterry, ok, so if I see '--tempdir=/home/robru/.cache/deja-dup/tmp' in the debugging output, then that for sure means it worked? ~/.cache/deja-dup/tmp directory was never used prior to this branch?22:07
mterryrobru, right22:08
mterryrobru, nor was --tempdir either22:08
robrumterry, k, great. just gonna peruse the diff briefly then approve ;-)22:09
mterryrobru, I just noticed something myself looking at the diff22:17
robruoh?22:17
mterryrobru, I changed DuplicityInstance.start from normal to async.  And thus dropped it's 'throw Error' line22:17
robru... ok?22:17
mterryrobru, but, when I called it from DuplicityJob before, I would bubble up the error message via show_error().  In this new diff, the error message gets swallowed by DuplicityInstance and the user will end up just seeing a generic "Unknown error" message22:18
mterryrobru,  let me fix that real quick22:18
robruok22:18
robrumterry, oh, I like the way you allow this tempdir logic to be overridden by an env variable. That's thinking ahead ;-)22:22
mterryrobru, I fixed the issue I found locally, but I decided to write a test for this case for the future.  But I have to run.  So I'll update tomorrow22:31
robrumterry, ok22:31
mterryrobru, thanks for the review!22:34
robrumterry, yw, have a good night!22:34
=== m_conley is now known as m_conley_away

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!