/srv/irclogs.ubuntu.com/2012/02/08/#ubuntuone.txt

nessitahola00:21
ralsinahola nessita00:21
ralsinago have a life! ;-)00:22
nessitaralsina: I'm "adelantando" from work from tom morning :-)00:22
nessitasome* work00:22
ralsinanessita: ok, just kidding00:22
nessitadobey: if you're around... any verdict on the packaging-dailies branch? I have a controlpanel branch that depends (for landing) in python-ubuntu-sso-client.tests :-)00:23
nessitaralsina: you with some minutes for a trivial?00:36
ralsinanessita: not right away, but I can do it tonight00:37
nessitaralsina: I'm having a test blocking in a branch, from a code that we no longer use, so I cut off "by the healthy option" and removed it00:37
nessitahttps://code.launchpad.net/~nataliabidart/ubuntu-sso-client/no-more-blocking/+merge/9195400:37
nessitaralsina: later is great, thanks00:37
nessitaralsina: if you agree, would you also globally approve when done?00:37
ralsinaif it's just a skip, I trust you enough to self-approve00:38
ralsinabut sure, I will do the global00:38
nessitaralsina: is not a skip, but the removal of a code we no longer use00:38
ralsinanessita: ok, will review00:38
nessitaack00:38
nessitawell, will have dinner now00:38
* nessita -> away00:39
mandelmorning!09:29
ralsinagood morning mandel!09:30
JamesTaitGood morning all! :D10:09
ryemorning10:37
psypher246hello all11:11
psypher246anyone here who can refresh anyone here who can assist with troubleshooting upload issues11:13
mandelpsypher246, rye is you man :)11:17
mandelralsina, hello!11:17
ryepsypher246, hello! What is the size of the file you are uploading?11:18
psypher246rye: 36000 files, all pictures all total about 8GB11:19
ryepsypher246, ok, what are the sympthoms?11:19
psypher246well i can't remember the commands anymore and lost my notes (somehiow tomboy sycned somthing and deleted all my new notes, but thats another story)11:20
psypher246anyway I want to know what is happening at this time11:20
psypher246I expect it to take long, but I want to see if it has imporved since I logged this bug11:20
psypher246https://bugs.launchpad.net/bugs/53957311:21
ubot4psypher246: Error: Bug #539573 not found.11:21
psypher246nothing is uploading at this time11:21
psypher246u1sdtool --waiting-content11:21
psypher246says11:21
psypher246Oops, an error ocurred:11:21
psypher246org.freedesktop.DBus.Error.NoReply:11:21
psypher246u1sdtool -s11:22
psypher246ubuntu one client says: file sync error. (org.freedesktop.DBus.Error.NoReply:)11:22
psypher246sercond time today11:22
psypher246I have to restart to make it work, yet i still don't kow if anything is actually happening11:23
psypher246rye: ok now filre sysnc is starting again according to u1client11:24
psypher246u1sdtool -s11:24
psypher246State: LOCAL_RESCAN11:25
psypher246    connection: With User With Network11:25
psypher246    description: doing local rescan11:25
psypher246    is_connected: False11:25
psypher246    is_error: False11:25
psypher246    is_online: False11:25
psypher246    queues: WORKING11:25
psypher246after scan starts, it doesn't start uploading11:25
psypher246i eman after scan stops11:25
ryepsypher246, what is the client version?11:25
ryepsypher246, and whet is the ubuntu version?11:26
psypher246precise11:26
ryepsypher246, metadata loading now takes 2-3 seconds11:26
psypher246what is the com,mand again something policy to see what u1 client ver?11:26
psypher246ok it has stoped scanning, little longer than 3 seconds about 2011:27
psypher246State: QUEUE_MANAGER11:27
psypher246    connection: With User With Network11:27
psypher246    description: processing the commands pool11:27
psypher246    is_connected: True11:27
psypher246    is_error: False11:27
psypher246    is_online: True11:27
psypher246    queues: WORKING11:27
psypher246but nothing is uploading11:27
ryepsypher246, metadata loading != local rescan, local rescan is a traversing of the folders to see whether anything has changed11:27
psypher246rye:  processing the commands pool = metadata loading?11:28
ryepsypher246, u1sdtool --waiting-content may be causing the dbus request to fail if it cannot be completed in 120 seconds11:28
psypher246rye: does that then kill the whole aplication?11:29
ryepsypher246, no, metadata loading is the initial phase before local rescan, where the client reads the database about all the files it knew11:29
psypher246oh sorry ok I get you now11:29
psypher246so if rescan is finished whats next?11:29
ryepsypher246, when i initially joined, i thought metadata loading and local rescan are the same thing.11:29
psypher246metadat, import old, local rescan import new?11:30
ryepsypher246, local rescan -> connecting -> queue_manager where it is processing the commands11:30
psypher246so when does it start uploading?11:30
ryepsypher246, at queue_manager, i.e. now. How do you see it is not uploading things?11:31
=== karni_ is now known as karni
ryealternatively how do i see they are uploading11:31
ryepsypher246, ah, run u1sdtool --current-transfers11:31
psypher246using nethogs I do not see anytraffci coming from python process11:31
ryepsypher246, please pastebinit since it will be a long list11:31
psypher246http://pastebin.com/twzPQZ6T excuse the filenames :) only folder i can find with thousands of small files11:32
ryepsypher246, ok, at this moment I need to ask you to archive the ~/.cache/ubuntuone/log folder and send it to ubuntuone-support@canonical.com - it does not start sending anything11:35
ryepsypher246, once i get the logs i will get back to you here11:36
psypher246rye: thanks, seems to be going but at pathtic speeds11:39
psypher2461KB/s11:39
psypher246my backbone can support 30mbits/s11:40
psypher246rye: just tried going to the "Devices" tab and got this error: Value could not be retrieved. (IPCError: org.freedesktop.DBus.Error.NoReply: )11:40
psypher246don't quite know how many issues I am dealing with there11:40
psypher246here*11:40
psypher246tried again now, the devices opened but still that errror, was not longer than 120secodns, more like 30 seconds11:41
ryepsypher246, hm, how much memory does ubutuone-syncdaemon process use?11:47
psypher246151888411:50
psypher246rye 78%11:51
psypher246damn!11:51
ryeoh11:56
rye1.5G??11:56
psypher246yeah11:56
gatoxhi!12:02
ralsinahola gatox!12:03
ralsinagatox: small question. When you did the latest windows release, what release number did you give ir?12:03
ralsinagive it*12:04
gatoxralsina, let me check12:04
ralsinagatox: because on bzr, it still says 203, which is a problem :-)12:04
gatoxralsina, 2.99.312:06
ralsinagatox: that's the version. There is also a release number that has n dots12:06
ralsinano dots12:06
gatoxralsina, in the other file, ubuntuone_autoupdate  i have <version>2.99.3</version>12:07
ralsinagatox: ouch12:07
ralsinagatox: that one is supposed to be an integer12:07
gatoxahhhh i didn't know that........ sorry12:08
ralsinagatox: right now I am not sure I can make an autoupdater that will updatethis version :-(12:08
ralsinagatox: it's ok, it happens12:08
ralsinagatox: I will have to make some tests to see if I can fix it. No problem.12:09
gatoxralsina, let me know if i can do anything12:09
ralsinagatox: nah, you have enough on your plate :-)12:09
ralsinagatox: worst case, we need to make another release12:09
ryepsypher246, has u1sdtool --current-transfers progressed somewhere?12:16
psypher246i've actually moved all the files out of the folder and take this up again another day. thanks for the help. I will log a bug about this when I have a chance12:16
ryeralsina, psypher246 has 1Gb of SD usage with 36000 files which total to 8Gb12:16
rye1.Gb12:17
ralsinarye: memory usage?12:17
mandelralsina, gatox have you ever seen this message when running a small wt dialog: 'Gtk-ERROR **: GTK+ 2.x symbols detected. Using GTK+ 2.x and GTK+ 3 in the same process is not supported'12:22
ralsinamandel: no12:22
ryeralsina, yes12:22
gatoxmainerror, nop12:23
mandelgatox, due, he is going to hate you12:30
mandelgatox, you keep talking with him and he keeps ignoring you :)12:30
gatoxmandel, what?12:31
gatoxahhhhhhhhhhh12:32
gatoxjejejejee12:32
mandelgatox, that ;)12:32
gatoxgrrrrr completion!12:32
mandelgatox, trying to seduce completion? with your name and the 'grrrrr' looks like it hehehe12:33
ralsinarye: there is a linear memory usage pattern, 36K files is at the top end of what syncdaemon can handle reasonably12:33
gatoxmandel, jejejejeje12:33
mandelralsina, gatox are you using P?12:34
ralsinamandel: O12:35
gatoxmandel, no.... i'm going to change to P this weekend12:35
mandelnot even a vm.. :(12:35
gatoxmandel, sorry12:35
mandelrye, are you running P with the latests updates?12:35
mandelgatox, no worries12:35
mandelgatox, ralsina I talk with my ISP and apparently I have an static IP! so I'll be foward connection details to those of you that want to be able to use by windows vms12:36
ryemandel, yes, w/o nightlies thourh12:36
ryethough12:36
ralsinamandel: Ihave a nice window VM, but thanks anyway :-)12:36
mandelrye, let me give you a script to try for me, one min12:37
mandelralsina, oh, but I have one vm for EACH windows version :)12:38
mandelrye, can you try to execute this: http://paste.ubuntu.com/833835/12:42
mandelralsina, gatox can you find any reason why this should fail: http://paste.ubuntu.com/833835/12:42
ralsinamandel: hmmmm I would import QApplication from QtGui but other than that it looks ok to me12:43
ryemandel,  from ubuntu_sso import qt? I guess i am not that up-to-date12:44
mandelrye, grab the sso trunk and set the python path12:44
ralsinamandel: how does it fail?12:44
mandelrye, or just copy the script to the  sso trunk :12:45
mandelralsina, Gtk-ERROR **: GTK+ 2.x symbols detected. Using GTK+ 2.x and GTK+ 3 in the same process is not supported12:45
mandelTrace/breakpoint trap (core dumped)12:45
mandelralsina, I did change the import from Qt to QtGui, same issue12:45
ralsinamandel: well, it's something in the ubuntu_sso import12:45
ralsinamandel: if I comment that, it works12:45
mandelralsina, dah, it is, but wtf?12:46
mandelralsina, there should be no gtk there at all, right? gatox any idea?12:46
ralsinamandel: use python -v and grep for gtk?12:47
ryemandel, the same Gtk-ERROR message you printed above12:48
alecuhola all!12:49
gatoxmandel, let me check here12:50
mandelrye, ok12:50
mandelalecu, morning!12:50
mandelralsina, looking at that atm..12:50
ryehm, gi.overrides.Gtk?12:51
ryemandel, ubuntu_sso.main.glib imports from gi.repository import GLib, Gdk, Gtk12:52
mandelrye, yep, I just saw that too..12:53
ryeubuntu_sso.main.linux to be precise12:53
mandelWTF! why is that thing bringing the ubuntu_sso.main.glib.. I'm just importing the bloody qt module12:53
ralsinarye, mandel: looks like the main.linux has not been cleaned up to use the qt stuff12:54
gatoxmandel, i have this warning: /usr/lib/python2.7/dist-packages/gtk-2.0/gtk/__init__.py:127: RuntimeWarning: PyOS_InputHook is not available for interactive use of PyGTK12:54
gatox  set_interactive(1)12:54
ryemandel, USE_QT_MAINLOOP is not defined12:54
mandelgatox, that one is 'normal'12:54
ryemandel, if os.environ.get('USE_QT_MAINLOOP'): - qt, else glib12:54
mandelrye, where is that?12:55
ralsinamandel: what rye said. Because on linux sso supports gtk and qt12:55
gatoxmandel, ah yes.... you should use USE_QT_MAINLOOP=True when running12:55
ryemandel, ubuntu_sso.main.linux12:55
ryeand then your code works12:55
mandelgatox, rye I'd be happy with that except for one thing, why is ubuntu_sso.qt.gui brining ubuntu_sso.main, that is very very wrong12:56
mandelvery very very12:56
ralsinamandel: you are not supposed to just be importing this stuff, you know12:57
mandelralsina, so.. you want me to import main so that I show a little shitty dialog for creds? that does not make sense12:57
mandelralsina, I could move the creds dialog definition to another location, but where?12:58
ralsinamandel: I wonder why this is happening. __init__ does nothing12:58
mandelralsina, exactly, there is some very bad things happening in the import12:59
ryeralsina, from ubuntu_sso.main import linux12:59
ryeralsina, then linux imports glib12:59
mandelrye, yes, but that is in ubuntu_sso.main.__init__ which should not be imported at all12:59
ralsinaI don't get all those imports12:59
ryemandel, qt/controllers.py import main13:00
mandelme cago en la puta de oros!13:00
mandelrye, you are right13:00
ralsinaimport ubuntu_sso.qt for me imports only this: https://pastebin.canonical.com/59693/13:00
ryei suppose i should not learn that line in spanish ^13:00
ralsinarye: good idea :-)13:01
mandelrye, is not the worst one you can hear.. trust me13:01
ralsinamandel: so, it's a matter of untangling some imports on qt.gui13:02
mandelralsina, well, it seems that the controllers use main because BackendController uses main..13:02
ryemandel, me (forgot) unknown known known from diamond13:02
ralsinamandel: it's a trivial fix13:03
ralsinamandel: main is used in exactly *one* place, and can be removed from the top level import13:03
ralsinamandel: we may have to duplicate the definition of NO_OP as well13:04
ralsinamandel: or move it elsewhere13:04
mandelralsina, so, BackendController has an if condition in case the backend is None, I don't know why that has to be done, but that is the problem13:06
mandelgatox, any idea about that ^13:06
mandelgatox,  why does it happen?13:06
ralsinamandel: probably in case the backend fails to start13:06
gatoxmandel, ralsina everything about the controllers is changing..... so you shouldn't worry about that13:07
mandelgatox, except that atm is blocking my work, I have to do a work around and set the USE_QT_MAINLOOP which is horrible13:07
mandel:(13:07
gatoxmandel, why?? i thought everyone was using USE_QT_MAINLOOP13:08
mandelgatox, I can do that and will ignore the problems, but setting an env variable is a bad idea since the creds dialog is a diff process13:08
mandelgatox, if env variable changes, the dialog will crash with the above error13:09
mandelralsina, setSizeGripEnabled(False) plus layoutSizeContraint to FixedSize should stop a dialgo from being resiable, right?13:24
mandelgatox,  ^13:25
mandelralsina, gatox atm, I have those, I cannot resize but the mouse over the bottom left conner show the resize pointer, do I have to set the window flags too?13:25
gatoxmandel, fixedSize avoid resize13:25
mandelgatox, but I get the little <-> arrow in the mouse..13:26
gatoxmandel, i'm not sure if you can forcce a window to not show the arrows..... even if it isn't resizable.....13:29
mandelgatox, let me push the code and you will see what I mean..13:29
gatoxmandel, look for another not resizable window in the OS and check if it show that anywat13:30
gatoxanyeay13:30
gatoxahhhhh.... anyway13:30
gatoxmandel, first do that13:30
gatoxmandel, mmmm yes.... it shouldn't appear...... let me see the code13:31
mandelgatox, let me push it13:31
mandelgatox, here: lp:~mandel/ubuntu-sso-client/creds-dialog13:35
mandelgatox, and how I'm testing it atm: http://paste.ubuntu.com/833885/13:35
gatoxmandel, name of the file?13:35
mandelgatox, some manual work is done under ubuntu_sso/qt/gui.py the ui file is data/qt/proxy_credentials_dialog.ui13:36
gatoxmandel, you don't need this: self.setSizeGripEnabled(False)  (but just saying.....)13:37
mandelgatox, I was trying everything, I know is set in the .ui file13:39
mandelgatox, deperation coding style :)13:39
gatoxmandel, did you try to set the fixed size in the code.... just to check13:40
mandelgatox, I did not, you think that will matter?13:40
mandelgatox, did you try the code and show the pointer changing?13:40
gatoxmandel, on it13:42
gatoxmandel, yes.... i can see it.... let me try something13:43
mandelgatox, muchas gracias :)13:44
mandelgatox, is the first time I see this, when done in other os it works as expected13:44
mandelgatox, do you ming if I quicly go to have some food?13:47
gatoxmandel, no.... go13:47
mandelgatox, thx13:47
gatoxmandel, the problem is in the .ui13:49
mandeltell me tell me13:50
mandelgatox, I'll though I'll read it later :)13:50
* mandel runs to have lunch13:50
=== dduffey_afk is now known as dduffey
gatoxmandel, i can't find which the problem is..... but removing the setup of the ui file in the code.... the window behave as expected14:02
Chipacadavidcalle: hi there. Where in launchpad is the remote scope? I branched lp:unity-lens-videos and that isn't it :)14:22
Chipacarats14:24
dobeyhrmm14:24
Chipacalooks like it's lp:~davidc3/unity-lens-videos/remote-videos14:24
nessitahello everyone!14:28
dobeyhola nessita14:28
nessitahola dobey, how is it going?14:28
gatoxnessita, hi14:28
dobeyeh, could be better14:29
nessitahola gatox, I'd guess you're off ac today? :-P14:29
gatoxnessita, jeje yes14:29
briancurtinhi nessita14:30
nessitahola briancurtin! how is it going?14:30
nessita(I owe you finishing the review... will do it today)14:31
dobeygwibber is being crashy :(14:31
nessitadobey: gah, when moving it to gi?14:31
dobeyor well, gwibber-accounts is being crashy; and not loading my accounts14:31
briancurtinnessita: i'm good, and whenever you get to the review is fine :)14:31
dobeynessita: yeah, but i'm pretty sure it's gnomekeyring that's causing it now14:31
dobeynessita: i just can't seem to get any good debug info from it though :(14:31
Chipacadavidcalle: hi there14:34
davidcalleChipaca, hey14:34
Chipacadavidcalle: do you have a few minutes? or would you rather it was later?14:34
davidcalleChipaca, I have them.14:34
Chipacadavidcalle: ok14:34
dobeyi have some for sale on ebay; but the auction is ending soon. so bid fast and high! :)14:35
dobeynessita: can you look at the error in https://launchpadlibrarian.net/92218315/buildlog_ubuntu-precise-amd64.ubuntuone-client_3.1%2Br1188-51~precise1_FAILEDTOBUILD.txt.gz let me know what you think?14:40
nessitadobey: yessir14:40
nessitadobey: never saw those before14:42
nessitadobey: but I have a suprise for you :-)14:42
nessitaralsina: shall we do the 1-1 after the standup?>\14:43
nessitadobey: enjoy!!! (all credits to the awesomeness of verterok) https://code.launchpad.net/~nataliabidart/ubuntuone-client/make-the-call-please/+merge/9204214:44
gatoxnessita, question..... the --help_text should replace the "forgotten password" text??14:45
nessitagatox: nopes14:45
nessitagatox: only the "subtitle" in the setup account screen14:45
gatoxahh ok14:45
dobeynessita: oh this fixes the non-workiness of sd?14:46
nessitadobey: YES14:46
dobeycome on launchpad. give me a diff already14:46
dobeyoh snap14:47
dobeysrsly14:47
Chipacagah14:47
Chipacasilly internets14:47
dobeynessita: ok, that is one thing that pyflakes definitely needs to bitch about :)14:48
nessitadobey: +114:48
dobeynessita: +1000000000000 on your branch.14:52
nessitaheh14:52
mandelgatox, ok, so it has to do with the .ui yet we don't know what..14:52
mandelgatox,I'll look in to it14:53
gatoxmandel, yes14:53
nessitastandup in 7 crow!d14:53
* dobey wonders what to do with 4096x1152 pixels exactly.14:57
gatoxme15:00
briancurtinme15:00
mandelme15:00
dobeymeh15:00
nessitame15:01
alecume15:02
nessitaralsina: standup?15:02
nessitagatox: go!15:02
gatoxDONE:15:02
gatoxLOT OF Refactoring done. Starting fixing the broken tests because of the refactoring.15:02
gatoxTODO:15:02
gatoxFinish with all the tests. Start refactoring the wizard.15:02
gatoxBLOCKED:15:02
gatoxNo15:02
ralsiname15:02
gatoxbriancurtin, go15:02
alecudobey, finally increase the font size?15:02
briancurtinDONE: i thought i fixed bug #820350 but it seems to have broken many other tests15:02
briancurtinTODO: figure out how i broke those other tests, try to propose the branch today15:02
briancurtinBLOCKED: none15:02
briancurtinNOTE: i have a doctor appointment this afternoon, will take a lunch/doc break and work a bit later on15:02
briancurtinNEXT: mandel15:02
ubot4Launchpad bug 820350 in ubuntuone-client "WIndows: either set_dir_readwrite and/or set_dir_readonly are not doing what they should (affects: 1) (heat: 7)" [High,In progress] https://launchpad.net/bugs/82035015:02
mandelDONE: Finished the support for user:password@domain. Done the new design of dialog.15:02
mandelTODO: Propose the user:password@domain support. Connect dialog with keyring. Connet dialog with webclient implementations.15:02
mandelBLOCK: no, yet waiting for input regarding texts in the dialogs.15:02
* mandel throws the ball too dobey15:02
dobeyλ DONE: fix the devtools release faux pas, bug #928317, gwibber pokery15:02
dobeyλ TODO: fix up nightlies packages, more gwibber pokery, twisted docs, move more code around, music store call15:02
ubot4Launchpad bug 928317 in rhythmbox-ubuntuone (Ubuntu) "Rhythmbox-UbuntuOne Plugin Crashes (affects: 2) (dups: 1) (heat: 14)" [Undecided,Fix released] https://launchpad.net/bugs/92831715:02
dobeyλ BLCK: gwibber is crashing in gnomekeyring.15:02
dobeynessita15:03
nessitaDONE: split ussoc nightlies in several binary packages, landed use-webclient for controlpanel, sprint with gatox, reviews15:03
nessitaTODO: make u1cp nightlies not depend on qt4reactor, provide u1cp-qt package, propose run-stuff-from-mainloop, finish reviews15:03
nessitaBLOCKED: no! SD is fixed! many thanks to guillo15:03
nessitaNEXT: alecu15:03
* alecu writting notes15:03
ralsinawhile alecu is writing his notes: DONE: cleaned up some code by david in the video lens, mgmt call, misc stuff, gave blood TODO: recover blood, reviews, 1-1s, misc. BLOCKED: nopes15:04
ralsinaNEXT alecu again15:04
alecuDONE: some reviews, researched DNS SRV feature in SD and if to use it when proxying15:05
alecuTODO: more proxy work in SD; compile a bug list for FF15:05
alecuBLOCKED: no15:05
nessitaany comments anyone?15:07
ralsinanope15:07
ralsinaEOM15:07
ralsinanessita: really quick 1-1?15:07
nessitaralsina: sure, can I have 5 minutes to have some water?15:08
ralsinanessita: of course15:08
dobeynessita: no water for you! you go bread line! :)15:09
nessitaralsina: loggin in15:13
mandelgatox, FWI I fixed it :)15:15
gatoxmandel, how?15:15
nessitadobey: no15:30
nessitaoops15:30
nessitadobey: no tarmac for u1client? (branch is not landing)15:31
dobeybah15:31
dobeyblah15:32
dobeynessita: UnicodeDecodeError :(15:33
dobeynessita: i wonder what is leaving such a file around15:33
dobeynessita: hopefully it will land next time15:35
ralsinanessita: I am shocked, but yes, you can embed a QWizard inside of another widget15:36
nessitaralsina: NICE! (not)15:36
nessitadobey: thanks!15:36
ralsinanessita: example ugliness: https://pastebin.canonical.com/59722/15:37
nessitaralsina: I *may* usa that (no promises though)15:37
ralsinanessita: just bringing you the info ;-)15:38
nessita;-)15:38
nessitaralsina: but you're not using a stackwidget there15:38
ralsinanessita: shouldn't matter15:39
nessitaok15:39
ralsinanessita: if you can embed in one place, you can embed everywhere15:39
ralsinanessita: the whole wizard would be a single page of the stack, of course15:39
nessitaralsina: of course!15:39
nessitadobey: will change u1cp packaging dailes to source depend on the python-ussoc.tests package... any advice?15:40
nessita(I mean, anything I should do before that? like update tarmac something?)15:41
dobeynessita: no that sounds fine15:41
alecuYo troll, I'll embed a wizard within your wizard so you can enchant while you are throwing spellz15:47
nessitadobey: will the new package python-ussoc.tests be installed in tarmac? or shall I do that?15:55
dobeynessita: i already did it15:56
nessitadobey: you rock!15:56
mandelgatox, the xml was setting the resize grib :P16:00
mandelgatox,  a stupid mistake hehe16:00
gatoxmandel, :P16:00
dobeynessita: branch merge fail again :(16:01
dobeynessita: also this appeared: Command appears to be hung. There has been no output for 900 seconds. Sending SIGTERM.16:02
nessitadobey: uh?16:04
nessitadobey: smells like tarmac instance is "dirty"?16:04
dobeynessita: seems something is wonky with tests; and seems like something is causing a unicode filename to get left around16:05
nessitadobey: gah16:05
dobeynessita: doubtful, since i removed the tree to have a clean one, before the run that hung for 900 seconds16:06
nessitadobey: there is a test suite that specifically creates an invalid filename... perhaps that's busting the env?16:06
dobeynessita: it may be the same hang that happened in the nightlies build on amd64 though16:06
nessitadobey: True16:07
dobeynessita: indeed. though it should also be removing the file in tearDown; :-/16:07
nessitadobey: let me confirm that, JIC16:07
nessitadobey: apparently all invalid files and dirs are being removed16:11
nessitawith addCleanups, even16:11
nessita(which is much better)16:11
dobeynessita: right; need to debug it more16:12
mandelnessita, gatox I'm writing some tests for the creds dialog in sso, I've notice that the views tests use the normal unittest.TestCase, do you think is a problem if I use the trial one? I want to be able to use patch :)16:12
nessitamandel: what do you mean with the "view tests"? from what project?16:14
mandelnessita, ubuntu_sso/qt/tests/test_qt_views.py16:15
nessitamandel: ah... I would advice use a reference what we have in controlpanel16:15
nessitamandel: ubuntuone/controlpanel/gui/qt/tests16:15
* mandel looks16:16
nessitamandel: you will see there how we use trial's TestCase and also some other helpers that are great for testing UIs16:16
nessitamandel: but the answer to your question is yes, we can certainly use the trial's TestCase16:17
nessitamandel: those qt view tests are yours, right?16:17
mandelnessita, yes, bzr blame says that :)16:18
nessitaheh16:18
nessitamandel: perhaps at the time you considered something to use unittest's testcase?16:18
mandelnessita, uh.. and mocker.. shit shit shit16:19
nessitamandel: NO MOCKER PLEASE :-)16:19
mandelnessita, I did consider it at that point in time.. seems to be an old file, I'll fix that16:19
nessitamandel: hum, sorry, what would you fix? removing mocker from the old tests?16:20
mandelnessita, yes, should not be a a lot of work16:20
nessitamandel: please don't, gatox is refactoring all that code from top to bottom16:20
mandelnessita, ok, then I'll add my test and touch nothing16:20
nessitamandel: sounds great16:20
nessitadobey: so tarmac will not land u1cp branches either, no?16:22
dobeynessita: not at the moment it seems16:23
nessitaright16:23
nessitathanks!16:23
dobeywhich is quite odd16:23
nessitaodd? I thought you had "stopped" it16:23
dobeystopped? no16:23
nessitaso, it will not land u1cp because he does not want to? who does he think he is?16:23
dobeyUnicodeDecodeError stopped it16:23
nessitabu16:24
dobeyand it's odd, because it landed a branch yesterday. and one on monday, both from facundo16:24
dobeynessita: however, the UnicodeDecodeError issue means it won't land branches for anything which comes after ubuntuone-client, in the list of branches it checks (whichs seems to be some arbitrary order, thanks to the brillians of ConfigParser)16:26
nessita:_/16:26
dobeynessita: i just set your branch back to needs review for now16:27
nessitaack16:27
dobeyand cleaned up from the breakage, so other projects should be able to land16:29
dobeynessita: i really need to get some lunch, but we can debug more after that if you like16:29
nessitadobey: sure, let me know16:29
dobeyk16:29
dobeybbiab16:29
nessitaalecu: just FYI, yesterday I landed a branch that removes the whole thread_execute code and tests suite16:34
nessitaalecu: since is no longer used and it was hanging in some test runs16:34
ralsinanessita: about that branch. I did the approve, and tarmac bounced it with the tests passing. No idea why.16:35
nessitaralsina: yeah, it was a about a gdbus error... I re approved and landed16:35
nessita(late last night)16:35
ralsinanessita: ok16:36
alecunessita, very nice, cool.16:39
nessitamandel: so, I finally have the branch ready for spawnning processes from the sso service. Would you be able to review it? https://code.launchpad.net/~nataliabidart/ubuntu-sso-client/run-stuff-from-mainloop/+merge/8995616:39
mandelnessita, sure, let me some mins to finish some of my code, context switch, you know :)16:40
nessitamandel: yes, let me confirm I merged trunk...16:40
=== alecu is now known as alecu_lunch_n_er
nessitamandel: indeed, trunk was not merged, pushing that now....16:41
nessitapushed to revno 86116:41
mandelnessita, ack16:42
mandelnessita, ralsina FYI we have just asked robert and roberta to take a look at the current dialog impl and get the signed off asap17:06
ralsinamandel: ack17:06
nessitamandel: was that by email?17:06
mandelnessita, yes it was, lisettte sent it17:06
nessitamandel: would you please add me (and ralsina if he's not there) to the email?17:06
mandelnessita, sure, I'll fwd it (and will add u1 links incase I don't get to attach things)17:07
nessitamandel: thanks!17:07
gatoxwow! i forgot to eat....17:07
* gatox lunch17:07
lisetttemandel: i can add to the thread so you get all the replies?17:07
mandellisettte, much better, please :)17:08
nessitalisettte: can you re-send the attachments? I don't see any...17:18
lisetttenessita: will forward you the original mail17:19
nessitathanks!17:19
ralsinanessita: they are there, it's something on your mail client17:19
lisetttenessita: yes, i just checked and they should be in there, but will send you them as attachments17:20
nessitaralsina: I use thunderbird17:20
lisetttenessita: so do i :)17:20
ralsinanessita: me too17:20
ralsinanessita: they are inline in the message17:20
lisetttenessita: images are inline; do  you have text only?17:20
nessitayes, text only17:21
nessitawho would use not text-only email theses days?!?!? :-D17:21
* nessita is hard core17:21
nessitaralsina, lisettte: this is what I see: http://ubuntuone.com/0xoQBqEic5AOyZSShqgH3817:22
lisetttenessita: looks like text only17:22
nessitalisettte: yeah, I always set my email clients to use text only17:23
nessitalisettte: can you attach without putting them inline?17:23
lisetttenessita: i will from now on17:23
lisetttenessita: just sent you them anyway17:24
nessitalisettte: what's the english term for the opposite of "text only" email?17:24
lisetttenessita: html?17:24
nessitaah17:24
ralsinanessita: for future reference, it's something in your thunderbird configuration: http://www.flickr.com/photos/ralsina/6842045653/in/photostream17:24
lisetttenessita: rich text?17:24
mandelemail?17:24
mandel:P17:24
lisettte:)17:24
nessitaralsina: I explicitely set my email to be text only, I *hate* html email17:25
ralsinanessita: ok, but then you don't see attachments and others have to resend them 3 times...17:25
nessitadid not image that will hide attachments17:25
ralsinanessita: we know now :-)17:25
nessitaralsina: is there any benefit from sending attachs as inline (I'm wondering, philosophical question)17:26
ralsinanessita: easier to explain things. Hey, look at this <image> and this other thing <image>17:26
nessitaralsina: hum... but... from my POV (biased, of course), geeks will usually handle text only email, no?17:27
nessitaI guess is a philosophical question equivalent to top posting vs bottom posting :-)17:28
ralsinanessita: well,I am a geek. I used to be one, at least :-) YMMV.17:28
ralsinanessita: pretty much17:28
nessitaralsina: *used* is the key word there :-P17:28
ralsinanessita: ouch.17:28
ralsinanessita: ;-)17:28
nessitaralsina: I guess I would have expected my email client say at least "there are images that are not being shown" or something like that17:29
ralsinajust checked, kmail shows inline images also as attachments. Looks like a thunderbird thing17:29
ralsinanessita: maybe you have to check/uncheck view / display attachments inline17:31
nessitalooking for the setting17:31
ralsinanessita: nah, doesn't work. If you set body as plain text, it never shows inlined attachments at all. Even claims the message has no attachments.17:32
ralsinanessita: you can switch to simple html for one of these messages when you notice "hey, there should be something here", though17:33
nessitaralsina: yeah, I can't find a setting though17:33
ralsinanessita: not in settings, in view -> message body17:33
nessitaralsina: I meant a setting to show inlined as attachments17:34
lisetttenessita: next time i will do a ascii mock for you ;)17:34
ralsinanessita: there isn't any. It doesn't even show the clip for "this message has attachments"17:34
nessitalisettte: that's my girl!17:34
nessita:-P17:35
nessitaok, lunchtime for me17:35
ralsinanessita: looks like a bug to me17:35
nessitayeap17:35
nessitabrb!17:35
dobeynessita: trying  your branch another time, to try and get some deug info if it fails again17:50
mandelnessita, why looking for twisted like this: result = 'twisted.internet.reactor' in sys.modules can you just try to import twisted?18:01
* mandel is curious18:01
=== m_conley_away is now known as m_conley
mandelnessita, also, it would be nice to have a way to let the spawner not to stderr to /dev/null if we are debugging (I don't know how hard is that) but if it is a lot of work, ignore me :)18:05
mandelI could be wrong but passing standard_error twith true will get the fd of the stderr which could be useful18:06
dobeynessita: just my luck18:10
dobeynessita: it merged18:11
nessitadobey: yey! (?)18:11
mandelnessita, and something similar can be done with the others, like using setStandardErrorFile to set the fd for it, but as I said, is just an idea to make debugging easier18:11
nessitamandel: for the first question, I don't want to import a reactor is there isn't one already there18:11
dobeynessita: yeah, it's merged which is good, but i still have no idea what broke :(18:11
nessitamandel: and no, for now we ignore stdout and stderr from the spawned process. The process should do proper logging to debug...18:12
mandelnessita, I meant, import twisted if twisted is there, there is at least a few reactors18:12
mandelnessita, about stderr, ok18:12
nessitamandel: perhaps I'm missing your question, sorry... but that method do not want to import anything, just see if is already imported18:13
mandelnessita, ah, ok18:13
dobeymandel, nessita: you could have an env var to not add the DEV_NULL flags, if the env var exists18:13
nessitadobey: right, but we also have spawners in qt and tx, and we need to handle those in all18:13
nessitadobey: which is out of the scope for now...18:13
nessitadobey: since we're building the programs to spawn, we'll make them log properly18:14
dobey*shrug*18:14
dobeyi have enough problems to deal with right now anyway :)18:14
nessitadobey: ;-)18:17
mandelnessita, I'm talking without knowing, but does getProcessOutputAndValue raise exception eventhough it returns a deferred?18:18
nessitamandel: yes, on windows... :-.\18:18
nessita:-/18:18
mandelnessita, mierda de windows..18:18
nessitamandel: I noticed that while running tests18:18
mandelnessita, at least you can understand why I asked ;)18:18
nessita;-)18:19
mandelnessita, in 'target = os.path.join(path, program + EXE_EXT)' what happens if I passed the program to be cmd.exe?18:24
mandelnessita, that is, the extension was passed18:25
nessitamandel: hum... will not work. I can tweak the code if you want... though the idea is to use the twisted tx only from tests18:26
mandelnessita, ah.. I think you can just check if it is present and append when needed, do it at your discretion18:27
mandelnessita, but if it is not much work better, in case we ever use it18:27
nessitamandel: will tweak that18:27
dobeynessita: hrmm. looks like python-ubuntu-sso-client needs to depend on python-httplib2 (>= 0.7.2) perhaps18:29
nessitahum... a ver....18:30
nessitaubuntu_sso/utils/webclient/common.py:21:from httplib2 import iri2uri18:30
nessitayes18:30
dobey:)18:30
nessitadobey: let me add that, I also need to add a new python package that will land woth the branch mandel is reviweing18:30
dobeynessita: a new binary package, or a new dependency?18:31
nessitadobey: a new python package that needs to be installed in python-ussoc (ubuntu_sso.utils.runner)18:31
dobeyah ok18:31
nessitadobey: and then we can re-build nighgtlies when my branch lands18:31
mandelnessita, questions, once your branch has landed, the ui won't be executed in a diff process, right? so on widows, which main loop are we going to be using? I guess that we not longer need the qtreactor, so we will be using just twisted, is that correct?18:33
nessitadobey: I should also add that new dep to build-depends, right?18:33
nessitamandel: not yet, next branch18:34
nessitamandel: sorry, let me give you a better answer18:34
nessitamandel: the ui will not be executed in a diff process yet. When it does (incoming branch), yes, the sso service will only need a plain twisted reactor18:34
mandelnessita, I was going to ask more hehe18:34
nessitamandel: but the sso UI (which is now a new process) will still use the at4reactor18:35
dobeynessita: in sso? yes i think so.18:35
nessitamandel: FYI, EXE_EXT change pushed18:35
mandelnessita, so, in that case, why did you write a Qt process 'spawner' since sso on windows will not use qt at all18:35
mandelor I'm getting confused :(18:35
nessitamandel: it will use in in linu18:35
nessitalinux*18:35
nessitamandel: if qt is available in linux, we'll use that18:36
dobeynessita: generally speaking, anything the binary packages depend on, should probably be in build-depends (or build-depends-indep) as well, for tests to work.18:36
nessitadobey: my thought exactly, but wanted to confirm it with you (perhaps I was missing something I never dealt with)18:37
mandelnessita, ahhh so we prefer to use qt over glib on linux, is that right?18:37
nessitamandel: yessir18:37
mandelnessita, ha, I had no idea we were going to Qt in sso O.o18:37
dobeywell, we're going to have a gtk3 control panel on friday, right nessita? ;)18:37
nessitadobey: *on my next freaky friday* which will surely not be the next friday before feature freeze :-D18:38
mandelnessita, I can do the gtk creds dialog for you :)18:42
mandelis a boring boring dialog..18:42
nessitamandel: no no not yet :-)18:42
mandelnessita, the only thing I'm worried is about this: http://developer.qt.nokia.com/doc/qt-4.8/qprocess.html#exitStatus but if you tell me we are not using qt for windows for spawning process I'm happy with the branch :)18:43
nessitamandel: I'm not good reading windows specifics... would you translate that for me? ie, what concern raise that for you?18:45
mandelnessita, only that if a process kill or use process from outside (task manager) qt things it terminated ok18:45
mandelnessita, sorry, that is not even eng..18:45
mandelis later here..18:45
nessitamandel: another chance!18:46
mandelnessita, ok, so if a process from outside sso kills the ui process, qt will think everything went ok, that is, return code 018:46
mandelnessita, but it is such a remote thing that we should not pay attention to that :)18:47
mandelnessita, check my review comment with the +118:47
nessitamandel: I will, thanks!18:47
mandeland with that, I go to walk the dog a frezze my balls :)18:47
mandelnessita, is a nice comment, no info what so ever hehe18:47
nessitamandel: hehehe :-)18:49
mandelall, c you tom!18:49
ralsinanessita, gatox: need any reviews? If not, I am taking a 15 minute break18:49
gatoxmandel, bye18:49
nessitaralsina: not for now, thanks!18:49
nessitabye mandel18:49
gatoxralsina, nopp....18:49
briancurtinbe back later - lunch/doctor18:49
ralsinanessita, gatox: ok, see you in a little bit18:49
nessitabriancurtin: good luck!18:49
gatox@ping18:50
ubot4pong18:50
dobeyhrmm, need to disable some more nightlies builds18:54
dobeynessita: u1cp will only pass tests on oneiric now as well, right?19:01
dobeyerr19:01
dobeys/oneiric/precise/19:01
nessitadobey: ah, yes19:01
nessitadobey: I guess that after FF I can build a different script for < P19:01
nessita(for every project)19:01
dobeywe can probably use the same script, with some magic19:02
dobeyassuming other stuff works right19:02
nessitadobey: yeah, that too19:02
dobeystuff that uses gtk is going to be a problem.19:04
nessitadobey: even if we run them isolated?19:04
dobeynessita: what do you mean isolated?19:08
nessitadobey: u1trial -t some_regex_that_will_match_a_few or u1trial a_specific_module19:08
dobeynessita: but what good is having a package that the user can install, but doesn't work?19:09
nessitadobey: you're mixing running tests with working! :-D19:10
dobeyno, i'm mixing nightlies builds with nightlies builds :P19:10
duanedesignnessita: is this from a non ascii file name? https://pastebin.canonical.com/59786/20:37
nessitalooking20:38
dobeyduanedesign: yes20:38
dobeyduanedesign: well, unicode username20:39
nessitaduanedesign: that looks like the user home has non ascii chars, which we support since last version (but not before)20:39
duanedesignok, so user needs to update20:39
duanedesignthank you20:39
duanedesignnessita: 2.0.3 is the last version?20:40
nessitaduanedesign: nopes20:40
nessitaralsina: would you please remind me and duanedesign the latest windows release number?20:41
duanedesignlooks like their is a 2.99.3 uploaded a couple days ago20:42
dobeyduanedesign: don't look at the +download page on launchpad20:50
duanedesigndobey: i was looking at https://one.ubuntu.com/windows/20:52
dobeyoh20:52
briancurtinback. only took 2 hours... :/21:03
nessitadobey: is the u1client nightlies build upload failing because there isn't a new revno in trunk?21:03
nessitabriancurtin: congrats! (?)21:03
dobeynessita: hrmm?21:04
nessitatoo cryptic?21:04
nessitadobey: why don't you read my mind already?!?!?21:04
nessitahttps://code.launchpad.net/~ubuntuone/+archive/nightlies/+recipebuild/172318 :-)21:04
dobeynessita: probably; is there not a new version in trunk?21:04
nessitadobey: no... I requested a new build since sso generated a new build21:04
nessitabut then a new upload will no go thru21:05
nessitabecause the version clashes21:05
dobeyah, is new sso built?21:05
dobeynessita: that sso fixes the httplib2 dep?21:05
nessitayeap! https://code.launchpad.net/~ubuntuone-hackers/+recipe/sso-dailies21:05
nessitait should21:06
dobeynessita: you can go to the faild builds of u1client in the nightlies, and tell it to retry the build21:06
dobeyinstead of trying to upload a new copy21:06
nessitaaahhhh, nice, did not know that21:06
ralsinanessita: 2.99.321:09
nessitaralsina: hola!21:09
nessitaduanedesign: 2.99.3, as you found out21:09
ralsinaI will probably have to do *another* 2.99.3 to fix a bug in the autoupdater :-(21:10
ralsinaOr, of course, a 2.99.421:10
dobeyif you make a 2.99.4 that's not actually 2.99.4, i will be very unhappy21:17
dobeyyou can however, make a 2.99.3.1, if you really have to :)21:18
ryehe he, i just downloaded 20000 photos from Ubuntu One, it took 4 hours21:19
nessitaok, gotta go to pilates class21:45
nessitawill be back later tonight, to finish this branch21:45
gatoxEOD for me! bye people22:09
alecubye gatox22:09
briancurtinbye gatox22:09
dobeylater all22:26
briancurtinhave a good evening dobey22:35
=== m_conley is now known as m_conley_away
=== gatox is now known as gatox_swim

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