[12:02] <bialix> ok
[12:04] <jml> Hello #bzr
[12:07] <lifeless> hello little island
[12:08] <bialix> luks: fixed and pushed to launchpad
[12:11] <jml> Has anything changed between 0.90 and 0.91 wrt options passed to SSH? (particularly NoHostAuthenticationToLocalhost)
[12:11] <lifeless> I don't believe so
[12:11] <mwhudson> ah
[12:11] <lifeless> however the port will have changed, because of the bug in spivs patch
[12:11] <lifeless> from default to 22-if-not-supplied
[12:12] <mwhudson> jml: https://bugs.edge.launchpad.net/bzr/+bug/146715
[12:12] <ubotu> Launchpad bug 146715 in bzr "bzr+ssh broken for non standard ports." [High,Triaged] 
[12:12] <mwhudson> launchpad.dev:5022 ftw
[12:20] <luks> bialix, thanks
[12:27] <bialix> luks, I don't understand how to right-align of push-button
[12:28] <bialix> there is exists some layot managers in Qt?
[12:28] <bialix> umm
[12:52] <Verterok> moin
[12:52] <lifeless> hola
[01:02] <Verterok> lifeless: Hi, time for one or two questions?
[01:02] <Verterok> I was wondering if the LogFormatter approach (like the one in bzrlib.log.LogFormatter) is the correct one for other commands like status, annotate so we can plug different output formats (like xml :) ).
[01:04] <Verterok> or maybe you (based on a more extensive knowledge of the internals) can give me a hint of an alternative approach?
[01:06] <lifeless> Verterok: sure
[01:06] <lifeless> so you want to use a process boundary right ?
[01:07] <lifeless> I think in that case that having some sort of filter for the output, that is selectable by your plugin code is good
[01:07] <lifeless> many of these things are already split into data generation/data presentation
[01:07] <lifeless> e.g. status
[01:08] <lifeless> status --short kicks in a different formatter
[01:08] <lifeless> in short - yes, that sounds fine to me, though the LogFormatter interface doesn't strike me as the easiest to use ;)
[01:10] <Verterok> right, I'm having a few problem with LogFormatter
[01:11] <Verterok> mostly because I don't have any context of the current revision to be logged only the revision itself
[01:14] <igc> morning all
[01:15] <Verterok> I'm stating to write a kind of stack for the log --xml, just to have some extra information about the context
[01:15] <Verterok> hi
[01:17] <lifeless> wb igc
[01:17] <igc> thanks lifeless
[01:18] <Verterok> also I recently added version, plugins and info command to the xmloutput plguin, and beuno added missing. we talk a bit about doing some kind of refactoring so we can avoid keep copying code from bzrlib and just plug to it.
[01:19] <Verterok> I just wonder if this kind of changes have any possibility of being accepted.
[01:37] <lifeless> Verterok: of course they do
[02:01] <spiv> Good morning.
[02:03] <i386> morning
[02:03] <i386> how was your weekend spiv?
[02:04] <spiv> Very pleasant.
[02:05] <i386> thus, feeling pretty haggard this morning
[02:20] <dfc> hello. i am having some trouble deciding what style of layout i would like for my repo. I think i want the nested-style or nested-tree.
[02:20] <dfc> i have a few "projects" that I use on all of my machines proj1-4 and a few that I only use on certain machines projA-C
[02:21] <dfc> is it best to do a init-repo basename ; bzr init basename/projBLAH
[02:21] <dfc> and only check out the branches that I need on each machine
[02:21] <dfc> versus just bzr init basename
[02:22] <dfc> where basename has many subdirs/braches with my proj directories?
[02:22] <Peng> Well don't do that.
[02:22] <dfc> what is the status of nestedtree support?
[02:22] <Peng> I don't know.
[02:22] <Peng> I *think* it's more or less supported, but not really exposed in the UI.
[02:25] <s|k> I had to kill a stalled commit and now I have a locked file
[02:25] <s|k> what are my options
[02:25] <lifeless> s|k: bzr break-lock
[02:25] <s|k> thanks
[02:27] <s|k> my commit keeps stalling
[02:28] <s|k>  Committing 1383/1386
[02:28] <s|k> it just stays there
[02:28] <lifeless> are you commiting a very big file
[02:28] <lifeless> like an iso or something?
[02:28] <s|k> I don't think so
[02:28] <lifeless> is this local or network
[02:28] <s|k> network
[02:28] <s|k> remote server
[02:29] <lifeless> sftp? bzr+ssh ?
[02:29] <s|k> sftp
[02:29] <lifeless> is there network activity happening?
[02:29] <s|k> no
[02:29] <s|k> it shouldn't take this long
[02:29] <s|k> the biggest thing in there is dojo
[02:29] <s|k> it's just text
[02:32] <lifeless> is there anything in ~/.bzr.log ?
[02:35] <s|k> hrm
[02:35] <s|k> I'll check
[02:38] <s|k> lifeless: has a a bunch of lines with what I added and then added revision_id and then KnitRepository and the final line is "fetch up to rev {..
[02:39] <s|k> nothing about an error or anything
[02:40] <s|k> the directory where it's supposed to go on the remote server is empty.
[02:41] <s|k> and it's still stalled
[02:41] <s|k> I'm going to set up svn and see if that works, oh well
[02:42] <lifeless> oh
[02:42] <lifeless> bye then
[02:42] <Odd_Bloke> :/
[02:53] <abentley> Peng: Nested trees are not currently supported.
[02:53] <Peng> Weren't they supported for a while?
[02:54] <abentley> No.
[02:54] <Peng> ....
[02:55] <Peng> Nested trees != subtrees?
[02:56] <abentley> Subtrees are part of the implementation of nested trees.
[02:56] <abentley> But subtrees aren't supported, either.
[02:56] <beuno> Verterok, hey, just uploaded the spanish doc branch: https://code.launchpad.net/~beuno/bzr/bzr.doc_es
[02:57] <beuno> I'm pretty sure uploadind the whole bzr branch was on overkill, but I'm lazy toda
[02:57] <beuno> *today
[02:57] <Peng> bzr upgrade --dirstate-with-subtree?
[02:57] <abentley> --dirstate-with-subtree is hidden because it's not supported.
[02:58] <Peng> What's the definition of "supported" we're using here?
[02:58] <abentley> It will make Bazaar act weird and fail in interesting new ways.
[02:58] <Peng> Haha, cool.
[02:58] <Peng> Okay, then, not supported. :)
[02:59] <Peng> So there are formats that aren't listed in 'bzr help upgrade'. Where are they listed?
[02:59] <abentley> Hidden formats are not listed anywhere.
[03:01] <lifeless> well
[03:01] <lifeless> in the code
[03:02] <lifeless> bisection is a lovely way to find interesting cache bugs
[03:02] <Peng> In the code is what I meant
[03:02] <Peng> grep register_metadir bzrdir.py
[03:02] <abentley> Ah.  I thought you were talking about help output.
[03:02] <Peng> Not really.
[03:02] <Peng> Help output, code, whatever.
[03:03] <Peng> I was just wondering what all of the secret hidden formats were. :)
[03:05] <poolie> good morning
[03:07] <igc> morning poolie
[03:07] <poolie> hello igc, welcome back
[03:07] <poolie> igc, quick call?
[03:07] <igc> thanks - great to be back
[03:07] <igc> sure
[03:39] <dfc> i would like to have a copy of an upstream svn repo in one of my shared repositories. do i use svn-import ?
[03:40] <Peng> bzr-svn?
[03:40] <poolie> dfc, either of those would work
[03:43] <dfc> i am confused about branches/co/pull and bzr-svn
[03:43] <dfc> i have set up my shared repo according to the reccomendations (i hope) such that they do not have working trees. if I do a bzr svn-import svnrepo mysharedrepo/importedsvnrepo
[03:44] <dfc> that would result in the importedsvnrepo being just like any other branch inside of my shared repo?
[03:57] <lifeless> finally, fixed bisection
[03:58] <lifeless> real lesson in the limits of tdd
[04:00] <poolie> mm?
[04:01] <lifeless> theres a theoretical hole with tdd, which is to have something sufficiently complex, and not decomposible, that you get what appears to be rock solid, fully tested, does everything code, that is actually completely broken
[04:01] <lifeless> I had this
[04:01] <lifeless> anyhow
[04:02] <lifeless> I now have bisecting indices that I think are solid
[04:02] <poolie> iow you test each step you make, but you miss some tests you should have added?
[04:03] <lifeless> you test each step you make
[04:03] <lifeless> but the steps combine so rapidly that there are missing interactions
[04:03] <lifeless> where you can't write small enough steps
[04:04] <poolie> strangely enough i was just thinking about that this morning
[04:04] <poolie> spiv, thanks for your promised fix for bug 146715
[04:04] <ubotu> Launchpad bug 146715 in bzr "bzr+ssh broken for non standard ports." [High,Triaged]  https://launchpad.net/bugs/146715
[04:08] <lifeless> blech, another glitch
[04:08] <lifeless> breakfast first, fix that after
[04:08] <lifeless> oh, btw, diff is 50% faster with bisection
[04:36] <spiv> poolie: I've sent it to the list now, it turned out to be essentially very easy (just don't set default_port for bzr+ssh:// or sftp://).
[04:45] <Verterok> beuno: ok, branching from it right now. which part of the doc are you translating right now? (to avoid working in the same thing)
[04:48] <abentley> poolie: Hi.  Your BB feature request: do you have a preference for lines vs kb?
[04:49] <poolie> abentley, if i was doing it i'd probably go by lines of the final diff
[04:49] <poolie> disregarding the packed section
[04:49] <abentley> Okay.
[04:49] <poolie> if that's easy
[04:49] <abentley> Hmm.  I'll see.  Sound really start hiding the packed section anyway.
[04:49] <abentley> s/Sound/Should
[04:53] <poolie> abentley, my review got error 503
[04:53] <poolie> maybe just because you were upgrading or tweaking it?
[04:53] <abentley> Yes, I was just upgrading it.
[04:54] <poolie> cool
[04:57] <poolie> spiv, re your patch
[04:57] <poolie> it looks ok
[04:57] <poolie> except i don't really see why ssh needs to be treated differently
[04:57] <poolie> with this in place, does it mean ssh://host/ != ssh://host:22/ ?
[04:58] <poolie> in a sense i suppose that's right - we don't know that they're the same
[04:58] <poolie> but maybe that is too pedantic
[04:58] <lifeless> to me
[04:58] <lifeless> default ports should be left None all the way down the stack
[04:58] <lifeless> thats kinda what default menas
[04:59] <poolie> that's what i was going to say too
[05:02] <abentley> poolie: did you want this number in the list of merges or for individual requests?
[05:09] <lifeless> it would be nice when looking at the list to be able to use it as a 'that will be easy to review' hint
[05:11] <beuno> Verterok, I've marked with TODO_ the ones that need translating, it's the latest, so you can grab pretty much anyone you want
[05:11] <beuno> I was going to aim for the guide next
[05:11] <abentley> Yeah, I'm guessing that was what he was looking for.
[05:11] <abentley> We'll see waht the performance is like.
[05:14] <Verterok> beuno: ok, I'll help with that, maybe server or plugins. anyway, I let you known when I start working on one of these
[05:14] <lifeless> abentley: you could cache it
[05:15] <beuno> Verterok, you rock  :D
[05:15] <abentley> Yeah.  I've just put it up without caching, and perf seems acceptable.
[05:16] <Verterok> beuno: this is going to be my first translation...let's see what happen :P
[05:16] <abentley> Caching would mean changing the DB schema, which I can do if necessary, but I'd rather not.
[05:16] <poolie> abentley, lifeless, yes, that's what i was wanting it for
[05:17] <abentley> poolie: It's up now.  You likey?
[05:17] <poolie> so if calculating the number of lines would be too slow but you can get the blob size direct from the db quickly
[05:17] <poolie> that would be ok with me
[05:18] <poolie> wow, sweet
[05:18] <abentley> Well, it's only hitting 24 bundles, so it seems alright for now.
[05:18] <abentley> I probably will cache it eventually.
[05:18] <dfc> i am all sorts of confuyed. i create a repo with no trees in /repo and i created a branch there /repo/project .
[05:19] <dfc> i then did bzr branch /repo/project ~/localcopy
[05:19] <dfc> i have committed changes to localcopy but when i try and branch from /repo/project to another computer/directory it says revision 0
[05:20] <poolie> dfc, you need to push if you want to get them back into the parent branch
[05:20] <dfc> poolie: push gives me an error
[05:21] <dfc> poolie: it says no push location
[05:21] <poolie> how about if you say
[05:21] <dfc> it knows where it was branched from.
[05:21] <poolie> so to back up a bit
[05:21] <dfc> i guess i expected the push command to publish the changes to its closest ancestor
[05:21] <poolie> do you want ~/localcopy to be an independent branch, or just a checkout of the branch in your repository?
[05:22] <dfc> i gues i am struggling with that
[05:22] <dfc> in that i know i am supposed to understand the difference but i dont
[05:22] <lifeless> do you want them to be able to have different content
[05:23] <dfc> not really
[05:23] <dfc> the problem is one of the machines is a laptop that will not always have access to the parent branch so i do not understand how the checkout will work
[05:24] <dfc> so if i want them to be the same i do a checkout
[05:24] <dfc> what if i wanted an independent branch? how would i update the parent every now and then?
[05:26] <poolie> abentley, so what is the meaning of -r for bzr send now?
[05:27] <abentley> It controls the patch preview, and it controls how the merge is applied, but it does not control which revisions are included in the bundle.
[05:32] <poolie> so the merge is done as if you'd typed merge -r XXXX
[05:35] <ubotu> New bug: #150427 in bzr ""bzr send -r" needs help text" [Undecided,New]  https://launchpad.net/bugs/150427
[05:36] <abentley> poolie: Right.
[05:36] <abentley> Or -r X..Y, for that matter.
[06:26] <ubotu> New bug: #150438 in bzr "AssertionError: Could not find target parent in wt in wt4 _process_entry" [Undecided,New]  https://launchpad.net/bugs/150438
[06:29] <poolie> is there already a bug about hpss server-side hooks?
[06:31] <poolie> spiv, can you please answer daniel pittman re hpss hooks, and file an rfe bug if there isn't  one already?
[06:32] <poolie> hello jamesh
[06:33] <jamesh> hi poolie
[06:33] <poolie> reading ll's repo format patch
[06:56] <poolie> abentley, just what goes on the bb 'mine' page?
[06:56] <poolie> it's not just patches originated by me...
[07:08] <lifeless> named commit of a specific file
[07:08] <lifeless> real    0m17.163s
[07:08] <lifeless> user    0m15.861s
[07:08] <lifeless> sys     0m0.744s
[07:09] <lifeless> thats with bisection, which is now good I think, for the index layer, but the pack wrapper needs a tad more love
[07:09] <poolie> "
[07:09] <poolie> Subclasses do not *have* to call the constructor, so this is in fact nor
[07:09] <poolie> out of date."
[07:09] <poolie> ??
[07:09] <poolie> not out of date?
[10:25] <sandrot> what's the difference between base, tree and other when conflicts arise
[10:27] <sandrot> Hmm, BASE last revision, THIS, working tree, OTHER new merge
[10:28] <spiv> sandrot: BASE is the most recent common ancestor of the two branches.  So the changes between BASE and THIS are the changes in your branch, and the changes between BASE and OTHER are the changes to be merged with your changes.
[10:32] <sandrot> spiv: awesome
[10:33] <sandrot> a quick look at bzr help conflicts and bzr help resolve wasn't very helpful
[10:33] <sandrot> bzr help THIS or OTHER or BASE don't do much either
[10:33] <sandrot> think it should be added to the docs?
[10:34] <spiv> Yeah, it should be explained somewhere.
[10:34] <spiv> It's familiar to some people from other systems, but we shouldn't assume that everyone knows that.
[10:35] <sandrot> yeah, after a few diff's I got it and it made sense
[10:36] <sandrot> but it's f-ing late and at this hour I rely on bulleted docs =)
[11:19] <sabdfl> hey revisionistas
[11:19] <sabdfl> lifeless: i saw your landings over the w/e
[11:21] <sabdfl> is it possible now to do a bzr upgrade to pack format repos?
[11:22] <Peng> Has he said that people should yet?
[11:23] <Peng> It is experimental, after all.
[11:24] <igc> sabdfl: I think packs are still a few days away from being ready in bzr.dev IIUIC
[11:24] <sabdfl> ok
[11:59] <baijum> I was going through : http://bazaar-vcs.org/OneZeroDocumentation Do you have plan t release PDF versions of docs ?
[12:01] <baijum> Creating PDF will not be much difficult, just: rst2latex.py and pdflatex would be enough. I have created such a PDF here for tutorial: http://bazaar-vcs.org/Documentation?action=AttachFile&do=get&target=tutorial.pdf
[12:02] <baijum> I used a LaTex style available here: http://svn.zope.org/grok/trunk/doc/style.tex?view=markup
[12:03] <baijum> May be you can also use: http://rst2a.com/api/
[12:34] <baijum> Hi sabdfl
[12:35] <baijum> sabdfl: got one minute ?
[12:52] <Ng> can I get bzr to show me a diff between a checked out directory and the parent (remote, http) branch?
[12:56] <spiv> Ng: bzr diff -r submit: uses the parent branch if there's no submit branch.
[12:56] <spiv> Ng: or possibly you want "bzr diff -r ancestor:URL"
[12:57] <Ng> spiv: ahh, cool, thanks :)
[12:59] <spiv> Ng: "bzr help revisionspec" summarises the available goodies :)
[12:59] <Ng> ta
[01:00] <sabdfl> baijum: sure
[01:01] <baijum> sabdfl, I was going through : http://bazaar-vcs.org/OneZeroDocumentation Do you have plan to release PDF versions of docs ?
[01:01] <baijum> Creating PDF will not be much difficult, just: rst2latex.py and pdflatex would be enough. I have created such a PDF here for tutorial: http://bazaar-vcs.org/Documentation?action=AttachFile&do=get&target=tutorial.pdf
[01:01] <baijum>  I used a LaTex style available here: http://svn.zope.org/grok/trunk/doc/style.tex?view=markup
[01:02] <baijum>  May be you can also use: http://rst2a.com/api/
[01:08] <spiv> baijum: ISTR plans for a PDF version were mentioned on the list at some stage, perhaps search the list archives?  Or perhaps post to the list about it.
[01:09] <baijum> spiv: if it is part of agenda for 1.0 release, that's fine.
[01:11] <spiv> baijum: I'm not sure off the top of my head.
[01:13] <sabdfl> baijum: looks good, i'm sure mrevell and others would appreciate a patch to spit out the pdf automatically
[01:16] <baijum> sabdfl: I will try to create a script to automate this
[01:19] <mrevell> baijum: that sounds good. If you have any other suggestions for the docs, feel free to ping or mail me. I'm working on them for at least the next few weeks.
[01:19] <mrevell> baijum: Thanks for looking at an automated PDF script!
[01:21] <baijum> mrevell, it is just two commands, so I wonder I should automate it or not
[01:21] <sabdfl> thanks baijum
[01:21] <baijum> np
[01:21] <baijum> mrevell, I have one more suggestion
[01:22] <baijum> For each document you can use .. contents:: directive
[01:22] <baijum> *without* .. sectnum:: directive
[01:22] <baijum> then
[01:22] <mrevell> baijum: I think that'd depend on how poolie_ and others wanted to manage the creation of such a PDF as part of the release process. DId you say you'd posted to the bazaar list about this?
[01:24] <baijum> I am not sure I will get enough time to dig more ...
[01:24] <mrevell> baijum: I'm looking in the mailing list logs now. It'd be great to get a discussion about this on the bazaar list.
[01:24] <mrevell> baijum: I can post to the list and quote the discussion here.
[01:24] <baijum> ok
[01:25] <baijum> while generating tex source: use --use-latex-toc option
[01:26] <baijum> this will give you a good table of contents
[01:26] <baijum> you will be also required run "pdflatex" command two time to get a proper TOC
[01:27] <baijum> see this: example http://svn.zope.org/grok/trunk/doc/grok2pdf.sh?view=markup
[01:27] <baijum> that's all I wanted to say.
[01:28] <baijum> and if you want permission to use: zope style for latex, I can arrange that
[01:28] <baijum> http://svn.zope.org/grok/trunk/doc/style.tex?view=markup
[01:28] <mrevell> baijum: Ah I see, well thanks for the input. I'll get a discussion going on the bazaar development list as I think it's a great idea.
[02:06] <gabe> hi all
[02:06] <gabe> i have a checkout
[02:07] <gabe> bzr st shows files removed and   'kind changed' and unknow
[02:07] <gabe> but these operations were performed from another checkout
[02:07] <gabe> how do I get this checkout to reflect the new structre? bzr update   doesn't do this :(
[03:12] <corporate_cookie> I'm having a bit of trouble installing paramiko ..python dose not seam to recognize its out there and thus BZR is saying Unable to import paramiko (required for sftp support): No module named Crypto.Util.randpool
[03:13] <corporate_cookie> however Crypto.Util.randpool is our there ...I can see it
[04:57] <vila> corporate_cookie: try
[04:57] <vila> python
[04:57] <vila> import paramiko
[04:57] <vila> what does he say ?
[04:58] <corporate_cookie> vila ImportError: No module named paramiko
[04:58] <corporate_cookie> this is the problem
[04:58] <corporate_cookie> i cant get python to recognize paramiko
[04:58] <vila> so that's a problem in your installation, if python can't find paramiko, neither can bzr
[04:59] <vila> what's your os ?
[05:00] <corporate_cookie> Red Hat Enterprise Server 4
[05:01] <vila> ouch, python 2.3 ?
[05:01] <corporate_cookie> python -V shows i'm running python 2.4.4
[05:01] <corporate_cookie> is there a way to verify python is installed correctly ?
[05:02] <vila> python -v should show you some interesting paths
[05:03] <vila> or even better
[05:03] <vila> python
[05:03] <vila> import sys; print sys.path
[05:03] <vila> so that you have the complete list. paramiko must be in one of them to be found
[05:04] <vila> how did you install it ?
[05:04] <corporate_cookie> I made an rpm from src
[05:05] <vila> hmm, can't remember the rpm command that lists the installed files
[05:06] <corporate_cookie> vila import sys; print sys.path gives me
[05:06] <corporate_cookie> '', '/usr/local/lib/python24.zip', '/usr/local/lib/python2.4', '/usr/local/lib/python2.4/plat-linux2', '/usr/local/lib/python2.4/lib-tk', '/usr/local/lib/python2.4/lib-dynload', '/usr/local/lib/python2.4/site-packages'
[05:06] <corporate_cookie> ...which is missing paramiko
[05:06] <corporate_cookie> is there a way to edit this path ?
[05:06] <vila> no, paramiko must be *inside* one of this to be found
[05:07] <vila> or alternatively you can use PYTHONPATH
[05:08] <vila> ha, what says rpm -q paramiko -l
[05:13] <corporate_cookie> vila allas, it tells me package paramiko is not installed
[05:13] <corporate_cookie> when i do apt-get install paramiko ...it tells me I have the latest version
[05:14] <vila> err, you said you made an rpm, you didn't install it ?
[05:14] <corporate_cookie> i made a python rpm ..i installed paramiko through apt (from the DAG repo)
[05:14] <corporate_cookie> i did install the python2.4 rom
[05:14] <corporate_cookie> er rpm
[05:15] <vila> so paramiko should be installed for for python apt-get thinks is the right one, which may not be your python-2.4.4.
[05:15] <corporate_cookie> ah i see
[05:15] <vila> s/for for/for which/
[05:16] <corporate_cookie> vila:  what  dose s/for for/for which/ do ?
[05:17] <vila> :) substitute "for for" in my previous sentence by "for which"
[05:19] <vila> a workaround is to create $HOME/lib/python, define a symlink there pointing to where paramiko is installed and add a PYTHONPATH=$HOME/lib/python in your profile
[05:19] <vila> paramiko seems to be pure python so that should work
[05:22] <corporate_cookie> vila:  what dose s/for which/for which/ do ?
[05:22] <corporate_cookie> (pardon my denseness )
[05:22] <vila> I said: "so paramiko should be installed for for python apt-get thinks is the right one, which may not be your python-2.4.4."
[05:23] <vila> I wanted to say: "so paramiko should be installed for which python apt-get thinks is the right one, which may not be your python-2.4.4."
[05:24] <vila> s/for for/for which/ is the sed command to apply to my sentence to fix it, it's just a shortcut to tell people I have made a typo
[05:24] <corporate_cookie> sweet golly ... i got ya :" )
[07:19] <corporate_cookie> vila:  I installed paramiko via src  with python2.4 setup install --prefix= /usr/local/lib/python2.4/site-packages/paramiko
[07:19] <corporate_cookie> paramiko (so far as the readout states) was able to install and byte compile correctly
[07:20] <corporate_cookie> yet python is still not seeing it ... any ideas ?
[07:20] <vila> ok, so back to:
[07:20] <vila> python
[07:20] <vila> import paramiko
[07:20] <vila> message crossed on the wire, hold on
[07:20] <corporate_cookie> ok
[07:20] <vila> what did setup install said ?
[07:22] <corporate_cookie> byte-compiling /usr/local/lib/python2.4/ ...
[07:22] <corporate_cookie> i also get copying build/lib/paramiko/config.py -> /usr/local/lib/python2.4/site-packages/paramiko
[07:23] <corporate_cookie> ..it copies each lib
[07:23] <vila> Good.
[07:23] <vila> ls -l  /usr/local/lib/python2.4/site-packages/paramiko
[07:24] <corporate_cookie> this shows me all the py and pyc files
[07:24] <vila> try again:
[07:24] <vila> python2.4
[07:24] <vila> import paramiko
[07:25] <corporate_cookie> yet python2.4 .... import paramiko still yeilds an import error
[07:25] <vila> import sys; print sys.path
[07:25] <corporate_cookie> '', '/usr/local/lib/python24.zip', '/usr/local/lib/python2.4', '/usr/local/lib/python2.4/plat-linux2', '/usr/local/lib/python2.4/lib-tk', '/usr/local/lib/python2.4/lib-dynload', '/usr/local/lib/python2.4/site-packages'
[07:25] <corporate_cookie> hum
[07:25] <vila> asme here ;-)
[07:25] <vila> s/asme/same/
[07:26] <vila> hold on !
[07:26] <vila> --prefix= /usr/local/lib/python2.4/site-packages/paramiko is wrong
[07:26] <vila> should be
[07:26] <vila> --prefix= /usr/local/lib/python2.4/site-packages/
[07:27] <vila> ls -l  /usr/local/lib/python2.4/site-packages/paramiko should have shown you paramiko only or you used completion and didn't realize their was two paramiko at the end ?
[07:28] <corporate_cookie> ls -l  /usr/local/lib/python2.4/site-packages/ shows me bzrlib, lib, and paramiko
[07:29] <vila> and ls -l  /usr/local/lib/python2.4/site-packages/paramiko ?
[07:29] <corporate_cookie> ls -l  /usr/local/lib/python2.4/site-packages/paramiko shows me whats in the   /usr/local/lib/python2.4/site-packages/paramiko folder
[07:29] <corporate_cookie> which are the py's and pyc's
[07:29] <vila> python
[07:29] <vila> import bzrlib
[07:29] <vila> ?
[07:30] <corporate_cookie> no errors
[07:31] <vila> print bzrlib.__file__
[07:32] <corporate_cookie>  usr/local/lib/python2.4/site-packages/bzrlib/__init__.pyc
[07:33] <vila> hmm, so paramiko is present but refuses to import, should be missing some dependency
[07:33] <vila> import pycrypto
[07:34] <vila> http://www.lag.net/paramiko/
[07:34] <vila> points to
[07:34] <vila> http://www.amk.ca/python/code/crypto.html
[07:34] <vila> install pycrypto, rinse, repeat
[07:34] <corporate_cookie> i have pycrypto
[07:35] <corporate_cookie> lol : )
[07:35] <vila> sorry, you lost me there :-)
[07:35] <vila> import pycrypto raises no errors ?
[07:36] <vila> you said: "ls -l  /usr/local/lib/python2.4/site-packages/ shows me bzrlib, lib, and paramiko"
[07:36] <corporate_cookie> ah ha
[07:36] <corporate_cookie> it dose yeild an error
[07:36] <corporate_cookie> damn you apt ...it installed pycrypto in python2.3
[07:36] <vila> :-)
[07:37] <vila> of course, that's the python *it* knows :)
[07:38] <fullermd> See, if you were using a BSD system, you wouldn't have these problems   ;)
[07:38] <vila> fullermd: I was just about to say that :-)
[07:39] <fullermd> Great minds think alike.
[07:39] <vila> Your majesty is too good...
[07:40] <fullermd> Ah well.  Another time, perhaps.
[07:41] <vila> sure :)
[07:41] <NfNit|oop> fullermd: How so?  (re: BSD)
[07:43] <fullermd> Oh, I was mostly just prodding around with a sharp stick.  Mixing package-manager-controller and manual stuff on any system gets you in trouble.
[07:45] <vila> Words of wisdom :)
[07:46] <vila> I try to keep $HOME/[bin|lib|rest of the zoo]  as empty as possible :)
[07:47] <vila> The real fun begins when you need to test some code on several platforms sharing some sources from nfs mounted volumes while using different versions of required libraries
[07:47] <fullermd> Variant symlinks ftw!
[07:48] <vila> Additional bonus if the said platforms use different OSes and different processor endianess (too easy otherwise)
[07:48] <fullermd> At least one of which should be a non-Von Neumann architecture   ;)
[07:49] <vila> variants symlinks do not work for *source* (think python or perl) :-)
[07:50] <vila> You make me wonders about installing a *BSD on my old Bebox :-)
[07:50] <corporate_cookie> vila:  looks like your pycrypto fix worked ..thanks for the help and patience!
[07:51] <vila> corporate_cookie: you're welcome, happy hacking with bzr :)
[07:51] <corporate_cookie> thanks : )
[07:51] <fullermd> Bebox??  Man, I always _thought_ you were nuts...
[07:52] <vila> hehe, I know there is a BSD port somewhere, but I never had the courage to try it...
[07:53] <vila> ok, ok, Have to go :)
[09:36] <ubotu> New bug: #150682 in bzr "must handle IOError exceptions gracefully" [Undecided,New]  https://launchpad.net/bugs/150682
[10:21] <AnMaster> is there any way I can hide specific changes from bzr missing and prevent them from getting added in a bzr merge without parameters
[10:21] <AnMaster> like blacklist changesets from getting merged
[10:22] <AnMaster> would be a very useful feature
[10:22] <AnMaster> can't find anything like it though
[10:22] <AnMaster> iirc svnmerge got it so can't see why bzr don't
[10:26] <bialix> bzr call this cherrypicking
[10:27] <bialix> but bzr has not full support of cherrypicking
[10:28] <bialix> because bzr use DAG to manipulate history there is no simple way to juggle with revisions -- otherwise it breaks DAG
[10:43] <AnMaster> bialix, ah
[10:43] <AnMaster> sucks
[10:44] <bialix> hm
[10:44] <AnMaster> what version control support it
[10:44] <AnMaster> because I will change from bzr now, it is not an alternative for this project
[10:45] <bialix> try darcs ten
[10:45] <bialix> then
[10:45] <AnMaster> what about git?
[10:45] <bialix> it's only RCS I know that allows you to juggle with patches
[10:46] <bialix> git -- is not my kind of tools because I'm working on windows
[10:46] <AnMaster> well I'm on *BSD and Linux
[10:46] <bialix> it's your way, that's my way
[10:51] <erikgz> is there an easy way to do bzr over ssh?
[10:52] <LeoNerd> Well there's sftp for an easy start, which is an SSH subsystem
[10:53] <AnMaster> now this is odd. when I use merge with -r argument it no longer shows that as merged...
[10:53] <AnMaster> so well cherrypicking in bzr is totally missing
[10:53] <AnMaster> not nice if you try to maintain stable and trunk
[10:53] <AnMaster> and merge history won't be nice even if/when that feature is added
[10:53] <AnMaster> ..........
[10:55] <bialix> record for cherrypicking is missing
[10:56] <AnMaster> exactly
[10:56] <AnMaster> when will that get added?
[10:57] <bialix> may be in 2008
[10:57] <AnMaster> when in 2008?
[10:57] <bialix> I don't know
[10:57] <AnMaster> I see
[10:57] <bialix> bzr has very long list of features nice to have
[10:58] <bialix> but main goal for 1.0 release is performance
[10:59] <bialix> https://blueprints.launchpad.net/bzr/
[10:59] <bialix> https://blueprints.launchpad.net/bzr/+spec/bzr-cpick-data
[11:00] <AnMaster> bialix, so for now I will recommend people I know to not use bzr
[11:00] <AnMaster> thanks for that
[11:01] <AnMaster> I like bzr but no cherrypicking makes it almost useless
[11:01] <AnMaster> night
[11:02] <bialix> someone always will be unhappy