=== lifeless_ is now known as lifeless [11:08] hello guys, can someone please tell me if bzr qlog should be readily available when installing bzr or is it a third-party that i need to isntall? (Ubuntu 32bit 12.04 LTS) [11:13] Wiz_KeeD: Ought to be the, uh, qbzr package. I think. [11:13] i tried installing it with apt-get and i got The following packages have unmet dependencies: [11:13] qbzr : Depends: python-qt4 but it is not going to be installed [11:13] E: Unable to correct problems, you have held broken packages. [11:14] Sorry, apt is not my thing. [11:14] heh,thanks anyway man [11:51] Wiz_KeeD: there should be an apt command to fix your packages [11:52] tried several, none worked ychaouche [12:09] Hi #bzr. I want to start with bazaar, I just read the 5min tutorial and the getting started guide and I have a few questions : So suppose I have my code in code. I'd cd to code, then the getting started guide tells me to bzr init. bzr help init tells me that This creates and empty branche. How come I can create a branch without creating a repository first ? [12:10] Or is the repository created along with the branch and everything is put inside the .bzr directory ? [12:11] I have the feeling that source code (files) are mixed with the "revisions", i.e bzr files. They're all put inside the source code. Am I right ? [12:12] I'm coming from subversion where source code is checked out from a repository, where svn stores branches/revisions etc. [12:21] Please tell me if this is correct : http://pastie.org/pastes/5501937/text [12:28] You can do init without init-repo just fine. It'll just colocate and use an internal repo. [12:28] Which is a related, but not identical, construct to svn's repo. [12:37] fullermd: ok suppose I do init-repo to create a repo, then init to create a branch. Is this workflow correct ? [12:37] Assuming there's a "cd" in the middle, so the branch is under the repo. [12:37] But really, if you're just starting out working with bzr, I'd ignore it. [12:38] Using repos properly makes things a lot nicer, but it's not a semantic difference, and thinking about it at this point is likely to overcomplicate things. [12:39] You're thinking svn; in svn, the repo is an important (or _the_ important) boundary/unit. In bzr, it's completely irrelevant; the _branch_ is the important unit. [12:39] ah ok [12:40] thanks for this _important_ information [12:40] no pun intended [12:40] So one doesn't really need a repository right ? [12:41] Not an explicitly-created one, no. If you don't make one (with init-repo), each branch (made with init) will create and use an internal one. [12:41] fullermd: So I'll just create a normal directory with mkdir foo instead of bzr init-repo foo and put my branches there [12:41] All that a shared repo does is save you disk space/IO/etc by sharing storage between branches. Everything works the same either way. [12:42] ok, I'll do that in a second stage [12:42] And you can use reconfigure to switch things around to using a shared repo later, so I'd recommend ignoring it for the moment, just to cut down on the number of things you're thinking about. [12:42] ah cool [12:43] So, suppose I want to work on a dev branch, and from times to times, when features are ready, create a production branche. How do I create a production branche ? [12:43] `bzr branch dev production` [12:43] fine [12:44] let me try this [12:44] You may find some of the stuff in and around useful. Or maybe not. [12:44] Won't mean much until you use the commands for a little while, probably; doesn't help with that. But it does talk about the abstract pieces underneath that you're manipulating. [12:45] ok let's leave that for later. [12:45] ok I created a branche. How cool is that. [12:46] fullermd: so if I just copy everything inside a branch and give it to someone, it contains everything he needs to branch/merge and do stuff, right ? [12:47] Normally, you'd do that by having the branch somewhere they can read, and let them use it as the source for a `bzr branch` invocation of their own. [12:47] do I need to setup a webserver for that ? [12:47] or a bazar server ? [12:48] Most common is probably ssh access. Though that's more of a bother when people don't already have shared access to a system somewhere. [12:48] You can run over dumb http; it'll be slower than over a smart transport, though you probably won't notice the difference with small branches. [12:49] And there's the bzr:// server, which works OK for ad-hoc or global readability. It doesn't have any AAA support or hooks, so it won't suit a lot of cases. [13:24] bzr explorer : bzr: ERROR: No module named explorer.lib.commands http://pastie.org/5502113 [13:25] I'm using bzr 2.4.1 [15:42] Is it problematic if the parent branch is not where bzr thinks it is ? http://pastie.org/5502524 === yofel_ is now known as yofel [15:50] ychaouche: no, that shouldn't matter [15:50] ychaouche: it just means that it will do the wrong thing if you don't specify a branch explicitly to 'bzr pull' [15:50] jelmer: fine [15:50] Same for merging, right ? [15:50] should I merge back dev to prod. === _thumper_ is now known as thumper