/srv/irclogs.ubuntu.com/2011/06/22/#bzr.txt

=== medberry is now known as med_out
Kamping_Kaisercan i use bzr-buildpackage with only debian/ in revision control, or do i need to have upstream source there too?00:59
lifelessit can work either way01:04
Kamping_Kaiserperfect. i'll just keep debian/ around then.01:05
Kamping_Kaiserthanks :)01:05
AfCSay I've previously I did a01:08
AfC$ bzr branch lp:ubuntu/empathy natty01:08
AfCIt was "natty" at the time. Now I pull and suddenly it's "oneiric". I guess it did what I told it to, but I surely didn't get what I wanted; I thought I had a branch with natty's version in it01:08
AfCIs there a modification to the lp:ubuntu/package syntax to bind it to a specific release series?01:09
mwhudson_AfC: i think lp:ubuntu/natty/package works01:13
* AfC tries `bzr missing` with that01:14
AfCGrr. Why oh why is it transferring megs and megs of data to give `missing --line` when I already have the revisions present!01:16
AfCAnyway01:16
AfCmwhudson_: yes, that works. Thanks01:16
Kamping_Kaiserrelated to my last question, is it possble to copy the debian/ dir between repositories keeping history? i'd like to move it from packagename/debian into shared/packagename/debian so i don't have to carry all the source too01:26
=== mwhudson_ is now known as mwhudson
AfCIf I would like `bzr builddeb` to build both a source package and a .deb binary package, what options would I do?02:45
AfC$ bzr builddeb -S -- ...02:45
AfCseems not to build the .deb02:45
AfC(ok, fine)02:45
AfCbut what is the complimentary option?02:45
AfCOr should we not use `-S --` and instead be doing `-- -S -sa` or so?02:45
Kamping_Kaiser-S == source only02:46
AfC[the other problem I ran into is about 1 of 15 packages I've done lately didn't have the .orig.tar.gz available in target distro, so uploading to PPA bombed, whereas the rest of the time it all Just Worked™ leaving me quite gunshy]02:47
AfC[and thinking for good measure I should always be passing `-- sa`02:48
thumperhi bzr people06:35
thumperI have some code right now that works on working trees06:35
thumperhandling things like adding or editing files06:36
thumperI want to write some other code that does the same work, but works on a branch instead06:36
thumperhow do I get a tree representation of a branch that I can use like a working tree?06:36
thumperis a revision tree the way to go?06:37
lifelessyyou can use a revision tree + a transform06:50
lifelessthe lp translations export code does this, if you want to cargo cult06:50
thumperhmm...07:06
thumperlifeless: ta07:06
vilamaxb: about to restart the package importer, anything I should know ?07:39
maxbvila: for anything in particular?07:44
vilamaxb: well, no, that's the trick :) Anything that crosses your mind !07:44
maxbI was wondering if I need to fix my change to the path to lp_creds.txt first07:45
vilamaxb: p-i stopped07:45
vilamaxb: argh, landed ?07:45
maxbyes07:45
vilamaxb: wait, no need, your change is ok for jubany right ?07:45
vilaso you can do that later07:45
maxbI thought it was ok, but now I think I may have miscalculated the number of dirname calls07:46
vilamaxb: depends on local setups if I read it correctly it works for udd/branch but not for udd/bugs/branch and anyway you can't have both07:48
vilamaxb: the real solution is to have it in pkgimport.conf anyway07:48
maxbI am somewhat unconvinced about pkgimport.conf, unless it supports relative paths07:49
vilamaxb: waiting for GSA to update the script and then start the pi again07:49
vilamaxb: relative to what ?07:49
vilato the file itself ?07:49
maxbPossibly07:50
vilawow, 'somewhat' 'possible', you make it hard to satisfy :)07:50
maxbAnyway, I think jubany is now going to be looking for lp_creds.txt in the wrong place - /srv/package-import.canonical.com/new/scripts/ rather than /srv/package-import.canonical.com/new/ :-/07:50
vilawhat's wrong with conf file ?07:50
=== hunger_ is now known as hunger
vilamaxb: are you sure about that ? losa is ready to start it again, should I hold him ?07:51
maxbYes, sure :-(  Probably best to just put in a symlink for now, to not block on admin actions07:53
maxbi.e. cd /srv/package-import.canonical.com/new/scripts/ && ln -s ../lp_creds.txt07:53
vilamaxb: confirmed07:53
vilamaxb: done07:54
vilamaxb: re-started07:56
maxbthanks08:00
maxbcan you requeue --all-of-type bzr ?08:01
vilamaxb: and stopped08:01
maxboh08:01
vilamaxb: and started again08:01
maxb?08:01
vilamass-import is bogus for 'restart' it doesn't delete STOPFILE, I'll file a bug and fix08:02
maxboh, heh08:02
vilamaxb: requeue done08:02
maxbI have also filed a MP to minimally undo my path thinko08:02
vilamaxb: cool, could you mention that we need to delete the symlink when this lands ?08:05
maxbs/need to/can/, but yes08:07
* vila nods08:08
vilamaxb: just making sure someone don't run into it and spend time wondering why it's there08:08
maxbHmm, there are a bunch of spurious failures - I suggest requeue_package.py --auto linux-restricted-modules haskell-utf8-string language-pack-kn language-pack-sq-base08:09
vilamaxb: old spurious or new spurious ?08:10
vilamaxb: wow, nice to the failure split !08:11
vilamaxb: requeue done08:13
vilarhaaa lovely... finally I can type just 'categorise_failures.py'08:14
maxboh?08:15
vilayeah, fixit.sh set the right env vars....08:15
maxbSo, 465 failures... now we just wait to see how many of the currently retrying ones succeed :-)08:16
vilamaxb: should we proceed with bug #795703 then ?  :D08:17
ubot5Launchpad bug 795703 in Ubuntu Distributed Development "jubany fixups relating to misplaced tags (AssertionError:<module>:main:find_unimported_versions:check)" [Undecided,New] https://launchpad.net/bugs/79570308:17
maxbYes - shall I sort the instructions for the new fixit syntax now?08:21
vilamaxb: as you will, I'm doing it right now, we can compare notes (I don't mind a double-check there ;)08:22
vilas/will/wish/08:22
vilas/as you wish/yes, please do/08:22
vilaor something ;)08:22
vilahmm, right, UI mistake, the package should come first, that's what we are acting upon (a GUI could display the relevant packaging branches or even tags, allow selection then action)08:26
vilawe'll see to it later08:26
maxbdone, posted to the bug.08:28
maxbAnd I wish launchpad could be asked not to line-wrap for displa08:28
maxby08:28
maxbit all comes out right in copy/paste, though08:28
vilaat least copy/pastin works...08:28
vilahehe08:28
maxbawesome, the failure count is going to settle at at most 469 already08:29
maxbwhich is a nice step down from the 485 we were at when all this LP mess started08:30
vilamaxb: yeah, well done08:32
=== maxb changed the topic of #bzr to: Bazaar version control <http://bazaar.canonical.com> | try https://answers.launchpad.net/bzr for more help | http://irclogs.ubuntu.com/ | Patch pilot: spiv | UDD failures: 465
vilamaxb: http://paste.ubuntu.com/630689/ ok ?08:36
maxbah, oops. yes08:38
vilano problem, I had one too :) double checks are good :)08:42
vilamaxb: procedding08:44
vilaa bit slow08:45
vilaof course establishing ssh connection for each command doesn't help08:45
maxbRight, so that's A through E.08:46
maxbI should have a look at the rest of the alphabet :-)08:47
vilaowww08:47
viladone08:47
vilamaxb: and report posted on comment08:48
maxbHmm, I need to revisit bug 494481 one of these days08:49
ubot5Launchpad bug 494481 in Ubuntu Distributed Development "Too easy for people to not use merge-upstream" [High,Triaged] https://launchpad.net/bugs/49448108:49
vilamaxb: pfew, that took long overall (sorry for that) but I think it's a nice step in the right direction08:50
vilamaxb: feel free to update the topic ;)08:52
maxbI'll be waiting on that to check that they *actually* succeed :-)[5~08:53
vilashouldnt' be long now, 0 outstanding jobs08:54
vilajam, Riddel: jelmer being sick, spiv offline, poolie sleeping, sprint next week, should we drop the standup ?09:02
jamvila: I'm fine either way09:03
RiddellI don't mind09:04
maxbHmm, it's looking quite achievable to drive the Ubuntu main UDD failure count below 100 in the next fortnight09:04
Riddellwas going to try and convince people to review my gpg stuff09:04
Riddelldo we know what we're doing at the sprint next week?09:07
vilamaxb: what about bug #796751, should it be marked invalid ?09:13
ubot5Launchpad bug 796751 in Bazaar "TreeTransform _override_conflicts checking is not sensitive to symlinks" [Medium,New] https://launchpad.net/bugs/79675109:13
=== maxb changed the topic of #bzr to: Bazaar version control <http://bazaar.canonical.com> | try https://answers.launchpad.net/bzr for more help | http://irclogs.ubuntu.com/ | Patch pilot: spiv | UDD failures: 455
vila\o/09:14
maxbYes, I think it should09:14
maxbdone09:16
vilahehe, done too :)09:17
vilamaxb: boom http://webnumbr.com/ubuntu-package-import-failures.from%282011-06-16%29 :)09:25
vilamaxb: setting bug #795703 on the assumption that you'll post more there09:27
ubot5Launchpad bug 795703 in Ubuntu Distributed Development "jubany fixups relating to misplaced tags (AssertionError:<module>:main:find_unimported_versions:check) [package names A through E]" [Undecided,In progress] https://launchpad.net/bugs/79570309:27
vilamaxb: setting it 'im progress' that is...09:27
maxbhah, nice graph09:32
maxbvila: I've landed the lp_creds.txt path change, if you want to deploy and then remove the symlink09:32
vilamaxb: done, waiting a bit to delete the symlink (just because)09:33
vilamaxb: ha right, just got your mails about the symlink, good, we're on the same page09:36
Kamping_Kaiseris there a way to make bzr only show files below `pwd` that have been modified? (like svn would). i keep trying to copy+paste the paths it tells me are modified and realsing i'm not in the repo root09:57
vilaKamping_Kaiser: not that I know of but I'm pretty sure there is a bug about that10:00
Kamping_Kaiservila: i'll have a look on lP, cheer10:02
Kamping_Kaisers10:02
Kamping_Kaisercan i do something in bzr if i want to redirect http attempts to access foo/packages/ so its foo/packges-version instead? would i have to do it using the webservers rewriting, or can i use bzr somehow?10:43
a7ndrewThere's an incomplete sentence at http://doc.bazaar.canonical.com/latest/en/mini-tutorial/#starting-a-new-project where would be the best place to report that?12:17
Riddella7ndrew: bugs.launchpad.net/bzr  or if you can work out how to fix it we can just commit the fix12:27
Riddellwell I know how to fix it, just need to think about what text12:28
a7ndrewI'm not quite sure what the author meant to say to finish off that sentence.12:50
=== zyga-food is now known as zyga
Riddella7ndrew: well report a bug and we'll get to it12:57
=== med_out is now known as medberry
johnfHow would it be possible for me to have a shared respository which also seems to have revisions in it? i.e. it has shared sub branches but if I run bzr log I also have revisions14:48
johnfThis repo was originally created in 2006 so it is quite old14:48
briandealwisjohnf: you can have a branch coexist with a repository.  In fact a standalone branch is exactly that14:53
briandealwisI remember being puzzled by this too14:53
=== oubiwann` is now known as oubiwann
=== joey` is now known as joey
pooliehi all16:24
pooliejohnf, i think we now warn if you try to create that but it may have been possible in the past16:24
poolieor, obviously, yo ucould just move the branch directory16:25
Riddellgood $timeOfDay poolie16:28
Riddellpoolie: I don't notice any warning if i run init-repo then init16:28
=== zyga is now known as zyga-food
pooliehello there17:04
pooliehm17:04
=== deryck is now known as deryck[lunch]
* maxb marves at bug 80077117:31
ubot5Launchpad bug 800771 in Bazaar "pull -overwrite overwrites not only files but all information about commits" [Undecided,New] https://launchpad.net/bugs/80077117:31
maxb*marvels17:31
=== deryck[lunch] is now known as deryck
=== deryck is now known as deryck[lunch]
fullermdImpressive.17:53
ScottKNice.17:54
pommevertehello18:24
pommeverteI have a question (that may sound a little stupid). But I have a project that is going to be setup for a few clients. Each client will have specific (personal) code implementations. I was thinking of branching my project for each client but then how would I go about correcting, say.. a bug accross all branches?18:28
pommeverteor am I not going the right way with this?18:28
pommeverteI'm kind of looking for a way to make patches18:29
pommeverteis there maybe a way to shelve changes and then share them with another diverging branch? Or is there maybe a way of merging a single commit and dealing with conflicts?18:31
pommeverteAny help would be appreciated I couldn't seem to find anything about this in the documentation18:32
jelmerpommeverte: do you mean like "bzr merge -c" ?18:32
pommevertehmmm what does the -c do?18:34
pommeverte(looking through man)18:34
jelmermerges just the changes in one particular revision18:35
pommevertehmm well yes that sounds exactly like it. Completely overlooked it since I checked the (outdated) man online18:40
pommevertethanks18:40
pommevertedo you think this is the best way of going about it?18:40
jelmerI think so18:43
pommeverteok perfect. Thank you very much18:44
=== deryck[lunch] is now known as deryck
=== yofel_ is now known as yofel
=== zyga-food is now known as zyga
pooliehi jelmer20:25
jelmerhi poolie20:25
pooliehow's your week?20:44
jelmerapart from the sick day, pretty good20:48
jelmerI did some more Launchpad work this week, and had a look at adding support for multiple upstream tarballs in the package importer.20:50
poolieoh, right20:50
pooliei hope you're feeling better20:50
jelmerI am, thanks :)20:50
=== yofel_ is now known as yofel
etenilHey there22:19
maxbHello22:21
etenilI've made a small bzr repo on my server. My workflow is to create branches for each feature I work on and then merge them into trunk and delete the now useless branch. I've discovered I can easily push new branches up to my small server, but I haven't found yet how to delete them.22:21
etenilCould you give me a hand?22:22
maxbThere is not a command within bzr itself for deleting branches. People usually just rm -rf them22:23
etenilah22:23
etenil:/22:23
etenilthe thing is, I'm not working alone, and I'm restricting the ssh access of the other devs to running bzr22:24
etenilis there really no way to accomplish this but rm -rf?22:24
maxbbzrtools provides "bzr zap --branch" - i've never used it though, so I don't know if it works remotely22:25
etenilwell that's a start anyway. I'll give it a try22:26
etenilthanks maxb22:26
etenilmmh22:27
etenilthat doesn't seem like the right thing.22:27
etenilI guess I'm good for posting a feature request :)22:27
__yhvh__hey I built in a recently downloaded repo, did make clean and make distclean, bzr status shows loads of files changed22:59
__yhvh__I also rm'ed the whole thing and d/led again fresh but the modified still show up,23:00
__yhvh__is this info stored outside the repo dir?23:00

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