[00:13] <CIA-3> wubi: Agostino Russo * r101 trunk/ (8 files in 4 dirs):
[00:13] <CIA-3> wubi: * The full URL does not fit in the dialog, only show the filename
[00:13] <CIA-3> wubi: * Avoid division by zero when calculating bittorrent download progress
[00:13] <CIA-3> wubi: * Remove partially downloaded files before attempting a new download
[00:13] <CIA-3> wubi: * Cache ISO md5 calculations to avoid recalculating the md5 for the
[00:13] <CIA-3> wubi:  same file (LP: 342250)
[00:20] <CIA-3> wubi: Agostino Russo * r102 trunk/debian/changelog: Rearranged changelog
[01:01] <CIA-3> debian-installer: cjwatson * r1067 ubuntu/debian/changelog: releasing version 20081029ubuntu26
[01:32] <CIA-3> wubi: Agostino Russo * r103 trunk/ (4 files in 2 dirs): Ensure that the local ISO path is stored upon successful download
[01:33] <CIA-3> debian-installer: cjwatson * r1068 ubuntu/ (build/config/i386/netboot.cfg debian/changelog): Adjust netboot/386 inclusion to avoid breaking lpia.
[01:43] <xivulon> evand download manager should be ok in r103 both for bt and standard http
[01:43] <xivulon> hence I have updated isolist.ini (only for ubuntu.iso) to point to the beta
[01:48] <CIA-3> wubi: Agostino Russo * r104 trunk/ (data/isolist.ini debian/changelog): Point isolist.ini to beta release (only for Ubuntu ISO)
[02:06] <CIA-3> wubi: Agostino Russo * r105 trunk/ (data/isolist.ini debian/changelog):
[02:06] <CIA-3> wubi: * Had the wrong URL for beta in isolist.ini (thanks Colin Watson!)
[02:06] <CIA-3> wubi: * Added beta URL for Kubuntu
[02:06] <CIA-3> wubi: * Fixed Kubuntu package name and URLs
[05:16] <NCommander> good $time all
[05:33] <TheMuso> Hey NCommander
[05:33] <NCommander> Hey TheMuso!
[05:33] <NCommander> TheMuso, how goes it?
[05:33] <TheMuso> NCommander: not too bad thanks. Yourself?
[05:33] <NCommander> TheMuso, beating on the installer on a64 and ARM
[05:33] <TheMuso> ah ok
[05:34] <TheMuso> anyway, I'
[05:34] <NCommander> Reading the PS3 forums, seems the jaunty installer works
[05:34] <TheMuso> yeah
[05:34] <TheMuso> i'm about to run actually
[05:34] <NCommander> TheMuso, cya later
[07:06] <laxmi> hello
[07:06] <laxmi> I need some help
[07:08] <NCommander> Anyone around who can merge a d-i branch (to fix ia64; its been test built)
[07:08] <laxmi> all r busy?
[07:09] <NCommander> laxmi, just ask your question.
[07:09] <laxmi> I could not compile srilm in Ubuntu 8.4
[07:09] <laxmi> language modeling toolkit
[07:10] <NCommander> laxmi, wrong channel, this is for the devleopment of the Ubuntu installer, I recommend you try #ubuntu
[07:11] <laxmi> ok
=== shtylman_ is now known as shtylman
=== evand_ is now known as evand
[09:35] <cjwatson> NCommander: would be happy to, but neither lp:~mcasadevall/debian-installer/ia64-fix nor lp:~mcasadevall/debian-installer/ia64-correction shows up as having been pushed to yet, and besides I don't know which one to merge :-)
[09:36] <NCommander> cjwatson, I'm having issues pushing
[09:36] <NCommander> Lots of them
[09:36] <NCommander> Want a diff instead?
[09:36] <davmor2> Morning guys I'm having an issue with unr.  On the installer just before the partitioning section I get a little popup telling me that /dev/sdb is mount and needs to be unmounted.  The options I have on the box are go back and continue.  Clicking on continue loops back to the same dialogue and clicking on go back continues the install.  This is a bit confusing to say the least.
[09:37] <cjwatson> NCommander: sure
[09:38] <NCommander> cjwatson, its been happily test built :-)
[09:40] <NCommander> cjwatson, http://paste.ubuntu.com/133527/
[09:41] <evand> davmor2: what's /dev/sdb* mounted to, and why can it not be unmounted (do you have a shell open in the mountpoint)?
[09:42] <NCommander> cjwatson, I have no idea what's going on with sparc; I think something gone horribly horribly wrong with its kernel though ...
[09:42] <cjwatson> or it could be that the format has changed and it's no longer something you can validly un-gzip
[09:42] <NCommander> The SPARC kernel hasn't been tested at all since the last ports rebase
[09:43] <NCommander> ... probably hasn't been tested since Hardy.
[09:43] <cjwatson> come to that, i386 kernels aren't gunzippable
[09:44] <NCommander> I don't think its something fixable without a SPARC box someone can restart infront of them, one to test the kernel, and one to figure out how to fix d-i
[09:45] <cjwatson> ok, applying ia64 patch, thanks for that
[09:45] <davmor2> evand: it's the os it's netbook remix so it has jaunty unr on it
[09:45] <CIA-3> debian-installer: cjwatson * r1069 ubuntu/ (build/config/ia64.cfg debian/changelog): (log message trimmed)
[09:45] <CIA-3> debian-installer: * Fixed ia64 d-i builds by bumping image size
[09:45] <CIA-3> debian-installer:  - The build failure "Disk Full" was caused by the last linux-ports
[09:45] <CIA-3> debian-installer:  kernel upload due to the size of the kernel growing. d-i on ia64
[09:45] <CIA-3> debian-installer:  makes a large "floppy" disk image of a specified size (this was 16MB
[09:45] <CIA-3> debian-installer:  before hand). The new kernel plus the RAMdisk came out to be 17.1MB,
[09:45] <CIA-3> debian-installer:  causing the failure. The new size is 18MB to give the kernel some
[09:46] <davmor2> evand: Let me go and get it and start the install over and I'll tell you exactly what is said
[09:46]  * NCommander wonders if he was a little too --verbose on the changelog
[09:47] <evand> davmor2: I understand what it said, but there's a partition mounted that it cannot do anything about and I'm more curious about that
[09:47] <NCommander> cjwatson, is it possible to cross-build d-i (i.e. debuild -B -aarmel?)
[09:47] <NCommander> cjwatson, it looks like there is some support, but it failed to get build-deps ...
[09:52] <davmor2> evand: /dev/sdb is the usb stick with unr on.  The reason it can't unmount is the same reason you can't eject the cd while run live session I'm guessing
[09:52] <mcasadevall> cjwatson, sorry, internet hiccuped, I didn't see your reply (if any)
[09:52] <cjwatson> NCommander: no
[09:52] <mcasadevall> cjwatson, that's what I thought, thanks :-/
[09:52] <evand> davmor2: ok, so the only problem then is the labeling of the buttons I suppose.
[09:52] <cjwatson> I'm not sure whether it's a fundamental limitation or not. You'd certainly have to do some messy fiddling with apt
[09:53] <evand> "go back" is what you want to press, it's just not clear given the text
[09:53] <davmor2> evand: call go back ignore and it makes more sense
[09:53] <mcasadevall> cjwatson, No doubt. if only for ports.u.c/*.archive.u.c
[09:54] <evand> I'd really prefer it say "yes" and "no".
[09:54] <evand> "go back and ignore" implies that you're going back a step
[09:54] <davmor2> evand: then you need to word the text in the box as a question
[09:54] <evand> when you're actually just continuing
[09:54] <evand> it is...
[09:55] <evand>  The installer has detected that the following disks have mounted partitions:
[09:55] <evand>  .
[09:55] <evand>  ${DISKS}
[09:55] <evand>  .
[09:55] <evand>  Do you want the installer to try to unmount the partitions on these disks
[09:55] <evand>  before continuing?  If not, the installation will continue and the disks will
[09:55] <evand>  not be available as installation targets.
[09:55] <davmor2> evand: Yes sorry it's me read the last line and not the one before it :)
[09:56] <cjwatson> mcasadevall: no, I didn't mean that, I meant for cross-installation of packages
[09:57] <mcasadevall> cjwatson, oh ... yeah ... that would be an issue :-/
[09:57]  * mcasadevall is just getting his feet wet with d-i hacking
[09:57] <evand> probably a bit screwed here though, as I might be killed for uploading with new strings to be translated if I fix this.
[09:57] <cjwatson> evand: "go back" should imply "go back to the partitioner", no?
[09:58] <evand> cjwatson: sure, though either way it's not appropriate here.
[10:02] <evand> and unfortunately there isn't an existing string in ubiquity that is, as far as I can tell. :/
[10:05] <cjwatson> evand: from the behaviour davmor2 described above, shouldn't the buttons just be reversed?
[10:05] <cjwatson> if continue goes back and go back continues ...
[10:05] <cjwatson> oh, well, continue retries I suppose
[10:05] <evand> indeed
[10:05] <cjwatson> can't you get Yes and No from GTK?
[10:06] <evand> ah, duh.
[10:06] <evand> thanks
[10:06] <cjwatson> debian/imported-po/Makefile has runes for getting GTK translations into debconf
[10:06] <cjwatson> you just need to edit ubiquity.templates-imported appropriately
[10:07] <cjwatson> we should refresh those translations anyway
[10:09] <evand> ok
=== Nicke_ is now known as Nicke
[10:28] <cjwatson> evand: can you look at bug 325958?
[10:28] <ubottu> Launchpad bug 325958 in ubiquity "Jaunty Alpha 4: Ubiquity windows does not fit on 1024x600 screen" [High,Triaged] https://launchpad.net/bugs/325958
[10:30] <evand> will do
[10:42] <evand> hrm, msgmerge seems to be failing me here.
[10:42]  * evand digs
[10:53] <evand> ah, "Yes" is marked fuzzy.
[10:57] <evand> cjwatson: any objection to me removing --no-fuzzy from the arguments to msgattrib in imported-po/Makefile?
[10:57] <evand> A few other imported translations have also been marked fuzzy.
[10:57] <cjwatson> evand: I don't see the point. If it's fuzzy in GTK, we don't want it
[10:58] <evand> hrm, ok
[10:58] <cjwatson> the msgid for Yes in GTK is actually "_Yes"
[10:58] <cjwatson> you aren't running into that, are you?
[10:58] <cjwatson> likewise "_No"
[10:59] <cjwatson> (translations marked fuzzy aren't used, anyway)
[11:17] <juliux> hi
[11:17] <juliux> is is possible to get the log from the debian installer to an other host?
[11:17] <juliux> i mean the things which are displayed during the installation on the F4 terminal
[11:18] <cjwatson> juliux: yes, two ways
[11:18] <cjwatson> juliux: either go back to the main menu and select "save debug logs", from which you have various options including turning the installer into a temporary web server
[11:18] <cjwatson> juliux: or switch to alt-f2, enter 'anna-install openssh-client-udeb', and then you can use scp
[11:19] <juliux> thxs
[11:19] <juliux> but it isn't possible to do that via a boot option or with a preseed.cfg?
[11:19] <cjwatson> no
[11:19] <juliux> i want to do that automaticly
[11:19] <cjwatson> well, you could do it in a preseed/late_command
[11:20] <cjwatson> assuming you can deal with scp authentication somehow
[11:20] <juliux> ok
[11:20] <evand> cjwatson: I don't think so.  This is what I'm starting with: http://pastebin.ubuntu.com/133565/
[11:20] <cjwatson> 'd-i anna/choose_modules multiselect openssh-client-udeb' will get scp installed early on, and then you can use it from late_command
[11:21] <cjwatson> evand: ok, not seeing why that would be fuzzy. All languages or just some?
[11:23] <evand> most of the "_Yes" strings are marked as fuzzy in GTK.
[11:23] <evand> at least from what I can tell looking at the gtk source package using grep -B 1 "_Yes" po/*
[11:24] <evand> apologies for not fully understanding fuzzy before, I'm still slowly trying to wrap my head around gettext.
[11:29] <cjwatson> looks like crap translations; check that it's ok in de and fr
[11:31] <evand> Unfortunately not:
[11:31] <evand> evan@bunny:/tmp/gtk+2.0-2.16.0/po$ grep -B 1 _Yes {de,fr}.po
[11:31] <evand> de.po-#, fuzzy
[11:31] <evand> de.po:msgid "_Yes"
[11:31] <evand> --
[11:31] <evand> fr.po-#, fuzzy
[11:31] <evand> fr.po:msgid "_Yes"
[11:38]  * cjwatson tries
[11:39] <cjwatson> but, um, that doesn't match the gtk+2.0 tree I have here
[11:39] <cjwatson> if nothing else, mine has msgctxt lines too
[11:39] <evand> hrm.
[11:39] <cjwatson> #: gtk/gtkstock.c:409
[11:39] <cjwatson> msgctxt "Stock label"
[11:39] <cjwatson> msgid "_Yes"
[11:39] <cjwatson> msgstr "_Ja"
[11:40] <cjwatson> though it does seem to show up as fuzzy here, hmm
[11:40] <cjwatson> maybe it's precisely due to the contexts
[11:47] <cjwatson> yeah, it's the lack of msgctxt lines in the .pot generated by po-debconf
[11:48]  * cjwatson pokes
[11:56] <cjwatson> got it, I think
[11:57] <evand> nice
[11:57]  * mcasadevall has the start of the iMX51 d-i port :-)
[11:58] <mcasadevall> I'm in a good mood now
=== mcasadevall is now known as NCommander
[12:02] <cjwatson> evand: a number of translations do disappear, though - I think they haven't been brought up to date for the new context system
[12:03] <thekorn> I don't know if you are working on bug 325958, but I think I understand why this happens, the timezone widget and the partition widget are requesting a minimum size
[12:03] <ubottu> Launchpad bug 325958 in ubiquity "Jaunty Alpha 4: Ubiquity windows does not fit on 1024x600 screen" [High,Triaged] https://launchpad.net/bugs/325958
[12:05] <thekorn> and the body of all other views is using the same size as a minimum, this also explains why the border around the first step is bigger than 16px
[12:05] <evand> ah, I bet the timezone widget is requesting too large a minimum size.  I was concerned with losing accuracy for the aspect ratio by making it too small (as we're dealing with a fair number of problems with accurate representation on the map), but if I have to change it, so be it
[12:05] <cjwatson> evand: we've lost some translations due to that, but no worse than GTK itself, AFAICT; we'll need to remember to update from 2.16.1 when it's released
[12:05] <CIA-3> ubiquity: cjwatson * r3115 ubiquity/ (73 files in 3 dirs):
[12:05] <CIA-3> ubiquity: Update for new message context system in GTK+ 2.15.1, and update
[12:05] <CIA-3> ubiquity: imported translations from gtk+2.0 2.16.0-1ubuntu2.
[12:06] <evand> cjwatson: ok, will do.  Thanks a bunch for your help with that.
[12:06] <cjwatson> note moderately foul hack in debian/imported-po/add-contexts
[12:13] <evand> heh, indeed
[12:28] <evand> ever so slightly concerned about a separate home option being in this list: https://wiki.ubuntu.com/GoogleSoC2009/Ideas
[12:28] <evand> wasn't the consensus that this is a bad idea as we cannot make a reasonable guess
[12:29] <cjwatson> sigh, it's impossible to reject ideas
[12:29] <cjwatson> they just keep coming back
[12:30] <cjwatson> I'll talk to jorge
[12:30] <evand> ok
[12:30] <evand> thanks
[12:37] <CIA-3> ubiquity: evand * r3116 ubiquity/ (45 files in 3 dirs): Make the partman/unmount_active dialog a yes or no question.
[12:59] <NCommander> cjwatson, I'm working on adding imx51 to d-i, but I ran into a snag. WHile I can build netboot initramfs's, I can't build cdrom ones. I'm going to hazard a guess and say armel lacks the necessary code to do it (arm has it). Any ideas on how to extend d-i to build an armel cdrom image?
[13:00] <cjwatson> should be just like the other architectures. what's the failure?
[13:00] <NCommander> cjwatson, make build_imx51_cdrom does nothing. I see it in the list of targets however.
[13:00] <cjwatson> I mean, you need to write a configuration file obviously
[13:01] <cjwatson> have a look at other cdrom.cfg files
[13:01] <cjwatson> and see build/README to understand the build system
[13:01] <NCommander> Oh
[13:01] <NCommander> Thanks
[13:02] <NCommander> No, I'm an idiot, I forgot to add a cdrom.cfg file :-/
[13:16] <CIA-3> ubiquity: evand * r3117 ubiquity/ (debian/changelog ubiquity/frontend/gtk_ui.py): Sort the list of regions on the timezone page (LP: #344334).
[14:43] <juliux> evand: i just read your log from openweek regarding to ubuntu unattended installation and howto create a pressed.cfg, is there some more to read? perhaps some troubleshooting pages?
[14:45] <evand> juliux: There's the installation guide (apt-get install installation-guide-i386; firefox /usr/share/doc/installation-guide-i386/en/index.html)
[14:45] <evand> what problem are you having?
[14:46] <juliux> if i restart my installed system and press any key on my keyboard i only get some cryptic symbols
[14:46] <juliux> so i am unabled to log in
[14:46] <juliux> everything else is working well atm
[14:47] <cjwatson> I bet I know
[14:47] <cjwatson> boot in rescue mode and look in /etc/default/console-setup
[14:47] <cjwatson> also, show us your preseed file, whose keyboard configuration I bet is wrong :-)
[14:47] <juliux> mom
[14:47] <davmor2> evand: kubuntu is still using grey for new system
[14:48] <evand> davmor2: ok
[14:48] <cjwatson> see also bug 328078
[14:48] <ubottu> Launchpad bug 328078 in console-setup "surprising keyboard choice if not preseeded" [Medium,Triaged] https://launchpad.net/bugs/328078
[14:48] <evand> shtylman: ^ Would you mind taking a look at that?
[14:48] <juliux> cjwatson: http://ubuntu.juliux.de/preseed.cfg that is my preseed.cfg
[14:48] <juliux> cjwatson: i get some errors during the installtion regarding to console-setup
[14:48] <cjwatson> juliux: what is your kernel command line?
[14:49] <davmor2> evand: I know it's not the end of the world but it still looks wrong :)
[14:49] <evand> sure
[14:49] <juliux> http://paste.ubuntuusers.de/394511/
[14:49] <cjwatson> annoying noise but not relevant
[14:49] <juliux> cjwatson: http://paste.ubuntuusers.de/394512/
[14:50] <cjwatson> juliux: seems to be cut off at "console-"
[14:50] <cjwatson> just paste the "append" line here
[14:51] <juliux> append tasks=standard vga=normal pkgsel/language-pack-patterns= pkgsel/install-language-support=false initrd=ubuntu-installer/i386/initrd.gz locale=en_GB console-setup/layoutcode=en_GB netcfg/wireless_wep= netcfg/choose_interface=eth1 netcfg/get_hostname=ubuntu preseed/url=http://172.23.43.2/preseed.cfg --
[14:51] <cjwatson> that's not a valid layoutcode
[14:51] <cjwatson> en_GB is a locale not a keyboard layout
[14:51] <cjwatson> console-setup/layoutcode=gb
[14:51] <juliux> ahh
[14:52] <juliux> so for germany it would be just de?
[14:52] <cjwatson> ideally console-setup would throw a visible error about this ...
[14:52] <cjwatson> right
[14:52] <cjwatson> (as in your preseed file)
[14:52] <cjwatson> it's the same as the X keyboard layout names
[14:53] <juliux> thxs
[14:53] <juliux> i will give it a try
[14:57]  * mcasadevall sighs
[14:57] <mcasadevall> Is the cdrom d-i type supposed to generate a ramdisk?
[15:01] <juliux> cjwatson: thxs, that was my mistake
[15:06] <cjwatson> mcasadevall: yes
[15:07]  * mcasadevall found that out, it failed to build because I'm an idiot :-)
[15:07] <mcasadevall> now I need to find the one thing I never have in my apartment; a blank CD ...
[15:22] <shtylman> evand: will do
[15:24] <evand> thanks
[15:45] <juliux> evand: cjwatson thank you very much, it is working well now
[15:45] <cjwatson> excellent
[15:45] <juliux> i will write something in german about it;)
[15:52] <davmor2> evand: you know the yes no for the partition unmount earlier can the same be done for password to weak dialogue?
[15:53] <evand> davmor2: I don't think it would make sense in that context
[15:53] <evand> as it's not a yes or no question
[15:53] <evand> oh, I guess it sort of is, given the title
[15:53] <davmor2> evand: true I just re-read the question :)
[15:53] <evand> but I'm still not convinced
[15:53] <evand> ok
[15:55] <davmor2> evand: I think the trouble is when your used to seeing it you abbreviate what it says rather than actually reading it :)  So the dialogue becomes use weak password :)
[15:57] <evand> so regarding the ubiquity window being 546px tall, it's being caused by the user setup page
[15:57] <evand> which we discussed at the sprint, but it had since fallen out of my head
[15:57] <evand> unfortunately I don't think there's much we can do there
[15:57] <evand> we're already using up most of the space
[15:58] <evand> other than the obvious packing it in a scrolled window
[15:58] <evand> davmor2: heh
[15:59] <evand> or perhaps some slight violations of the HIG
[15:59] <evand> that wont bring us anywhere near 400px though
[16:00] <evand> err nevermind on that last point, I now see that the bug report says 600px
[16:05] <evand> I've updated the bug with my thoughts
[16:12] <davmor2> cjwatson: evand: has the live system gone back a few days or something?   Only ubiquity just crashed on 64live with the NameError in remove_extra that I thought got fixed?
[16:17] <davmor2> Oh yes boat loads of updates
[16:17] <davmor2> :)
[16:23] <evand> davmor2: what's the output of ubiquity --version?
[16:23] <davmor2> 1.11.17
[16:24] <davmor2> there is a ubiquity update in the repos along with 256mb of others
[16:25] <cjwatson> I can't think of anything that would cause the livefs to ever go backwards
[16:27] <davmor2> cjwatson: it might not of I've not tested 64bit for a few days so the fix is in 32bit.  But looking at the manifest 64bit is 16th and 32bit is the 17th
[16:28] <evand> davmor2: are you sure you're not accidentally using the wrong CD?
[16:28] <davmor2> evand: I've only got single images of each that are rsynced daily
[16:28] <cjwatson> maybe it's just a livefs build failure
[16:28] <cjwatson> http://people.ubuntu.com/~ubuntu-archive/livefs-build-logs/
[16:29] <cjwatson>   brasero: Conflicts: nautilus-cd-burner but 2.24.0-0ubuntu1 is to be installed
[16:30] <davmor2> cjwatson: yes 32bit 17th 64bit 16th
[16:30] <cjwatson>   libbrasero-media0: Conflicts: nautilus-cd-burner but 2.24.0-0ubuntu1 is to be installed
[16:32] <davmor2> :(
[16:33] <mcasadevall> hey cjwatson, when you get a few minutes, I have a couple of d-i questions I'd like to ask. I've managed to get the cdrom image to start. But there is no subarchitecture defined according to log causing the installation to fail
[16:33] <cjwatson> mcasadevall: I imagine libdebian-installer needs to be educated
[16:33] <mcasadevall> Oh fun.
[16:34] <mcasadevall> Secondly, how does cdrom-detector probe for the CD-ROM drive, it failed to detect it; I had to manually mount my pendisk /dev/cdrom
[16:34] <cjwatson> see list-devices in debian-installer-utils, which monkeys about in /sys/block
[16:35] <cjwatson> davmor2: diagnosed on #ubuntu-devel, seb128 said he'll fix it
[16:35] <mcasadevall> /sys/block?
[16:35] <mcasadevall> Oh
[16:36]  * mcasadevall bets that missing from the kernel
[16:37] <davmor2> cjwatson: Sweet! any chance of a re-spin once it's fixed or will this be a test it tomorrow issue?
[16:38] <cjwatson> davmor2: test it tomorrow at this point I think
[16:38] <cjwatson> mcasadevall: *blink* really?
[16:38] <cjwatson> that's a lame kernel you've got there :)
[16:38] <davmor2> cjwatson: np's
[16:39] <mcasadevall> cjwatson, a lot of things you'd except were missing in version one
[16:39] <mcasadevall> Like USB support compiled in on an SoC with only USB ports to access HIDs :-)
[16:39] <mcasadevall> er, compiled as a module
[16:40] <cjwatson> right, but sysfs is pretty vital nowadays to EVERYTHING
[16:42] <mcasadevall> cjwatson, no, I think its just /sys/block thats MIA, but I'll have a more indepth look
[16:42] <mcasadevall> ... *sighs*
[16:42] <mcasadevall> I can't use a stacked branch it seems ...
[16:42] <mcasadevall> bzr: ERROR: RemoteRepository(bzr+ssh://bazaar.launchpad.net/%7Ecjwatson/debian-installer/main/.bzr/)
[16:42] <mcasadevall> is not compatible with
[16:42] <mcasadevall> KnitPackRepository('bzr+ssh://bazaar.launchpad.net/%7Emcasadevall/debian-installer/imx51-support/.bzr/repository/')
[16:42] <mcasadevall> :-P
[16:43] <cjwatson> the d-i branch is awkward due to bzr-svn I'm afraid
[16:43] <cjwatson> not very much I can do about that in practical terms until they unify the formats
[16:43]  * mcasadevall never had an issue with stacked branches and bzr-svn
[16:44] <cjwatson> you might be able to use 'bzr init-repo --1.6.1-rich-root' or something
[16:44] <mcasadevall> cjwatson, this repo was made bzr pull .
[16:44] <mcasadevall> well, its pushing
[16:44] <cjwatson> oh, sorry, don't know. ask #bzr?
[16:44] <mcasadevall> Probably finish in a week.
[17:43] <mcasadevall> cjwatson, so seemingly the reason stacking does not work with d-i is because the development focus is on the SVN branch. Changing the focus to the ~ubuntu-core-dev/debian-installer/ubuntu would fix the stacking, ergo, pushes would take seconds vs. hours
[17:44] <cjwatson> mm, right, but that would be wrong since it isn't the upstream development branch
[17:44] <cjwatson> can't you nominate a stacking branch rather than using the default?
[17:44] <mcasadevall> not when using launchpad.
[17:44] <cjwatson> I think this is an LP bug
[17:44] <cjwatson> it should be possible to have more than one level of stacking
[17:45] <mcasadevall> cjwatson, its an LP bug, bazaar supports this just file.
[17:45] <cjwatson> I don't want to declare the Ubuntu branch to be lp:debian-installer - it simply isn't
[17:45] <mcasadevall> No, that I get
[17:45] <cjwatson> so sorry, it'll just have to take time :( or you can send patches
[17:45] <mcasadevall> what's the bzr equivelent of git format-patches?
[17:45] <mcasadevall> :-)
[17:45] <cjwatson> bzr send
[17:45]  * mcasadevall has an incredibly pathetic upstream pipe
[17:46] <mcasadevall> Like, it makes the Berlin sprint look fast in comparsion ...
[17:47] <cjwatson> I have 448kbps up, on paper. In practice I don't think I get that
[17:47] <cjwatson> (kilobits, not kilobytes)
[17:47] <mcasadevall> You win.
[17:47] <mcasadevall> 368kilobits
[17:47] <cjwatson> still, same order
[17:47] <mcasadevall> cjwatson, yeah, but your physically closer to the DC :-)
=== robbiew1 is now known as robbiew
=== mpt_ is now known as mpt
=== fader|lunch is now known as fader
[21:36] <CIA-3> casper: TheMuso * r595 trunk/ (3 files in 3 dirs): (log message trimmed)
[21:36] <CIA-3> casper: * scripts/casper-bottom/30accessibility && ubiquity-hooks/30accessibility:
[21:36] <CIA-3> casper:  - When the blindness accessibility profile is chosen, create the .orca
[21:36] <CIA-3> casper:  directory in the live user's home directory. This seems to allow orca
[21:36] <CIA-3> casper:  to properly reload and use settings in the live session without needing
[21:36] <CIA-3> casper:  to be restarted. This is only needed for the blindness profile, due to
[21:36] <CIA-3> casper:  the other profiles that use orca already creating the .orca directory.
[21:37] <CIA-3> casper: TheMuso * r596 trunk/debian/changelog: releasing version 1.165