[12:19] <thumper> lifeless: looks to me as if the library functions under the cover have changed
[12:19] <poolie> hello
[12:19] <thumper> lifeless: which is then calling different methods on the custom stream class
[12:19] <thumper> morning poolie
[12:20] <lifeless> thumper: ah so the standard doesn't really specify then? Whats the best canonical reference for doing custom streams 
[12:20] <lifeless> thumper: when I was doing that one the best docs I found where a powerpoint presentation from a c++ conference
[12:20] <thumper> lifeless: the standard only dictates the interface and behaviour
[12:21] <thumper> lifeless: "I/O streams and locales" by Angelica Langer and someone else or "C++ standard library" by Nico Josuttis
[12:21] <lifeless> thanks
[12:21] <thumper> I find Nico's book one of the best
[12:22] <thumper> and I suggest C++ people get it after Meyer's books
[12:22] <thumper> poolie: you back in AU now?
[12:22] <poolie> yes, i am
[12:22] <thumper> poolie: up for a call?
[12:23] <thumper> lifeless: also, if this is a change in behaviour in GCC, I'd expect the same behaviour in ubuntu with the same compiler
[12:24] <lifeless> indeed
[12:24] <lifeless> poolie is in demand today :)
[12:27] <poolie> mm, i expected to be
[12:27] <poolie> prices will rise accordingly :)
[12:29] <lifeless> from $0 to $0 ? :)
[12:31] <poolie> which reminds me, did you get an Economist yet?
[12:31] <lifeless> yes indeed - thank you!
[03:11] <superm1> apparently dogfood.launchpad.net is down right now?
[03:11] <superm1> is this intended
[03:11] <RAOF> It's been down for quite a while for me.
[03:11] <superm1> i've been out of town all weekend :)
[03:11] <superm1> then are ppa items building
[03:11] <superm1> if submitted to upload.launchpad.net?
[03:12] <superm1> er upload.dogfood.launchpad.net
[03:12] <Fujitsu> I doubt it.
[03:13] <superm1> Fujitsu, i ask because i got an email back from dogfood telling me that it was accepted, so wondered if it would still build
[03:14] <Fujitsu> OK, looks like it's just the web UI broken then. If one bit of Soyuz is working, the rest probably is too.
[03:16] <kiko-afk> hmmm, something's not looking good there
[03:16] <kiko-afk> cprov-ZzZ?
[03:19] <superm1> well without the web ui avail, do you know what the structure was to browse the contents of a ppa?
[03:19] <superm1> just directory listing and such
[03:20] <Fujitsu> ppa.dogfood.launchpad.net/yourusername
[03:21] <superm1> ah easy nough, thx Fujitsu 
[04:00] <mpt> Gooooooooooooooooood afternoon Launchpadders!
[04:13] <manchicken_> Does lauchpad have a place where I can put documentation about my project?
[04:22] <mpt> manchicken_, depends what kind of "documentation"
[04:22] <mpt> (that's quite a vague term!)
[04:23] <mpt> You can have informational specifications about the development process etc
[04:23] <mpt> To do that, register a blueprint, and change its "Implementation status" to "Informational"
[04:23] <mpt> (The specification itself isn't kept in Launchpad, though, but it will be eventually)
[04:24] <manchicken_> mpt: Like, a wiki, etc.
[04:24] <manchicken_> I've registered my own little project.
[04:24] <manchicken_> I'm wondering if I'm gonna have to deal with my own wiki.
[04:24] <mpt> manchicken_, a wiki is just a CMS, that doesn't tell me what sort of documentation it is :-)
[04:24] <mpt> Alternatively, if you mean help pages, you can register answers to common questions in the Answers app
[04:24] <manchicken_> For kubuntu stuff I do I just use the ubuntu wiki...
[04:24] <manchicken_> mpt: Library documentation.
[04:24] <mpt> Library documentation?
[04:25] <mpt> As in, how programmers can use a particular library?
[04:25] <manchicken_> Yeah.  Documenting an AJAX perl library I put up.
[04:25] <manchicken_> Yessir.
[04:25] <mpt> Well, Launchpad doesn't have its own wiki
[04:26] <mpt> so probably you will need to set up your own at the moment
[04:28] <manchicken_> Okie dokie.  Thanks.
[04:38] <cprov-ZzZ> Fujitsu: https://dogfood.launchpad.net/ is back 
[04:44] <Fujitsu> cprov-ZzZ: Doesn't really affect me, but thanks!
[05:24] <c00i90wn> Hi, one little question, how do I disable the code tab for my project? 
[05:25] <Fujitsu> c00i90wn: You can't, at this time.
[05:26] <c00i90wn> Fujitsu: Oh I thought it was possible already since ubuntu, bazaar, etc. have their code tab disabled :)
[05:27] <Fujitsu> c00i90wn: bazaar doesn't, and Ubuntu is a distribution so doesn't have Code support at this time.
[05:27] <c00i90wn> Fujitsu: well, that's actually what I want :) I won't be using launchpad bzr, I prefer darcs :)
[05:28] <Fujitsu> Well, as far as I know there is no facility to disable it.
[05:28] <c00i90wn> Fujitsu: allright :) thanks :)
[05:29] <c00i90wn> and is there any plan to introduce other revision systems?
[05:35] <Fujitsu> c00i90wn: I don't think so, but I'm not a dev or anything.
[05:38] <c00i90wn> Fujitsu: ok, thanks again :)
[05:38] <Fujitsu> np
[06:09] <superm1> Well I appear to have spoken soon to be thinking ppa was fully functional still other than the GUI on dogfood
[06:09] <superm1> it accepts/rejects source packages, but nothing has built in the last 6 hours
[06:09] <Fujitsu> superm1: cprov said it's OK now, but perhaps the buildd queuer hasn't been turned back on.
[06:12] <superm1> Fujitsu, does something special need to be added to the source package to allow universe packages in the builds?
[06:12] <superm1> like a section: universe?
[06:13] <Fujitsu> If it build-depends on !main, it needs to be in that section.
[06:13] <superm1> what if it build depends on both universe and multiverse?
[06:13] <superm1> should it be in section: multiverse then?
[06:13] <jamesh> section: multiverse then
[06:13] <superm1> adding this to packages, will it break uploads to the normal archive?
[06:14] <jamesh> the idea is that all of ubuntu main can be built out of only packages in main, universe only with packages in main and universe, etc
[06:14] <superm1> right.
[06:15] <superm1> i've had uploads that didn't have this specified put into the archive, so i'm assuming that the archive admins would just place it in the appropriate location
[06:15] <jamesh> if you package depends on stuff in multiverse, then it probably belongs there too
[06:15] <superm1> in the archive it is already in multiverse indeed
[06:18] <Fujitsu> It doesn't matter for normal packages, because it's all controlled by overrides. Only in PPAs does the section need to be manually specified.
[06:19] <superm1> will it hurt in normal packages to put this there?
[06:19] <superm1> Also if the section was say "Section: graphics", it should be "Section: multiverse/graphics" then instead
[06:21] <Fujitsu> I don't know of any packages in the Ubuntu archive that have the component specified in the source, but it shouldn't hurt anything.
[06:21] <superm1> well i'd ideally like to be able to test the same debdiff in ppa prior to putting it into the archive
[06:22] <superm1> so it would make more sense at least for me to be able to do it this way
[06:22] <Fujitsu> In an ideal world, PPA would have overrides, but I've no idea if that's planned.
[06:22] <superm1> on ppa do the binaries need the same Section: multiverse/blah added, or is that then implied
[06:22] <superm1> by the source being in multiverse
[06:22] <Fujitsu> I think it's implied.
[06:23] <Fujitsu> But I don't know how it has been implemented.
[06:23] <superm1> well i'll experiment then
[06:23] <superm1> is that the correct way to list it though with "Section: multiverse/blah"?
[06:23] <Fujitsu> Yes.
[08:53] <carlos> morning
[03:53] <tumbleweed> is there any way to subscribe to a remote bug in launchpad, without f    *
[03:53] <tumbleweed> is there any way to subscribe to a remote bug in launchpad, without first reporting it to a launchpad-managed project?
[03:57] <tumbleweed> hmm, nobody?
[03:59] <mwhudson> tumbleweed: i don't think so
[04:00] <Odd_Bloke> tumbleweed: Why not subscribe in the upstream bug tracker?
[04:00] <tumbleweed> Odd_Bloke: it means creating accounts - I thought that's why launchpad *had* remote bug tracking...
[04:01] <Odd_Bloke> tumbleweed: Fair enough.  I think the upstream bug tracking is also there so that distributions can keep track of which bugs are in packages and which are upstream...
[04:02] <tumbleweed> Odd_Bloke: of course, but why not be more general?
[04:05] <tumbleweed> hmm, duh, remote bug tracking doesn't seem to do what I'm atfer, anyway :-(
[04:35] <mvo> is there a way to see OOPS-583D1520 and what causes it? I get this when I test the libapt apport package error launchpad uploading
[04:35] <ubotu> https://devpad.canonical.com/~jamesh/oops.cgi/583D1520
[04:40] <mvo> no luck when I tried it a second time, this time OOPS-583C1609 
[04:40] <ubotu> https://devpad.canonical.com/~jamesh/oops.cgi/583C1609
[04:41] <salgado> mvo, it takes some time (10 min, IIRC) for the oops logs to be synced to devpad
[04:41] <salgado> that may be the reason... usually we just have to wait a bit
[04:42] <mvo> aha, ok
[05:25] <ubotu> New bug: #130672 in soyuz "germinate output for Gobuntu" [Undecided,New]  https://launchpad.net/bugs/130672
[06:15] <mtaylor> does my - don't blog about it agreement mean I can't post links to my PPA until it's prod?
[06:16] <mtaylor> and another PPA question... is there a "good" way to push packages for more than one release other than making a source package for each one? 
[06:17] <mtaylor> like, if I had a package that should build from source properly on feisty and gutsy with no packaging changes, do I still have to make two branches and push two source packages? 
[06:44] <sabdfl> mtaylor: everything will get nuked before we move to production, iirc
[06:45] <sabdfl> and yes, i agree, it should be possible to target multiple series with a single upload
[07:31] <superm1> cprov-out, is there a way for PPA to include binaries of uploaded (and built) packages as possibilities for resolving build dependencies?
[07:31] <superm1> in future uploads/builds
[08:07] <moaiamorfo> hi all
[08:08] <moaiamorfo> I have a question: can I use Launchpad to translate a programming language manual?
[08:11] <moaiamorfo> no answers?
[08:14] <mwhudson> moaiamorfo: in what sense do you mean "can" ?
[08:15] <mwhudson> i don't think launchpad translations is set up to translate a book
[08:15] <moaiamorfo> mwhudson: I mean, I want to open a project under Launchpad to translate the Python reference from English to Italian
[08:16] <mwhudson> otoh, if you mean something like "can i use launchpad to host a bazaar branches containing the docbook source for the book" then, sure, so long as it's all open source
[08:16] <mwhudson> hm
[08:16] <mwhudson> well, you can do that, but why not just do it in a branch associated with python?
[08:17] <moaiamorfo> in what sense "with Python"?
[08:19] <mwhudson> 'bzr push sftp://bazaar.launchpad.net/~<you>/python/italian-reference
[08:21] <moaiamorfo> what the bazaar is?
[08:22] <mwhudson> the source control system that works with launchpad
[08:22] <mwhudson> let me back up
[08:23] <mwhudson> i'm not sure what you want to achieve, here
[08:25] <ubotu> New bug: #130706 in launchpad "Cannot Cancel Pending Membership Request" [Undecided,New]  https://launchpad.net/bugs/130706
[08:58] <kiko> moaiamorfo: is the python doc in pofile format?
[08:59] <moaiamorfo> mwhudson: uhm
[08:59] <moaiamorfo> kiko: I don't know
[09:06] <moaiamorfo> I just want a collaborative way to start the translation
[09:07] <kiko> moaiamorfo, if the doc were in pofile format it'd be fine to use rosetta. but I don't think it is, which means you'll need to manage it using another tool
[09:07] <kiko> mwhudson suggested using the bazaar revision control system
[09:08] <moaiamorfo> kiko: this one? http://en.wikipedia.org/wiki/Bazaar_%28software%29
[09:10] <moaiamorfo> is it like Subversion?
[09:10] <kiko> yes.
[10:30] <mpt> Gooooooooooooooooood morning Launchpadders!
[10:31] <ajmitch> morning mpt 
[10:32] <Odd_Bloke> Good morning mpt.
[10:49] <LaserJock> matsubara: around?
[10:50] <matsubara> LaserJock: Hi Jordan
[10:56] <LaserJock> matsubara: I'm not entirely sure what you're looking for in bug #130051
[10:56] <ubotu> Launchpad bug 130051 in malone "+text doesn't work for full bug URL" [Undecided,Incomplete]  https://launchpad.net/bugs/130051
[10:56] <LaserJock> matsubara: when I went to the bug I expected to be able to use +text
[10:56] <jhancock_> I was trying to push directly to launchpad using bzr and I got this error.
[10:56] <jhancock_> http://pastebin.com/d4e2e990e
[10:56] <jhancock_> and I followed this short and easy tutorial to set it up
[10:56] <jhancock_> https://svn.schooltool.org/trac/cando/wiki/LaunchpadTutorial
[10:57] <jhancock_> I might be doing something wrong
[10:57] <jhancock_> or the tutorial might do something wrong
[10:57] <jhancock_> I don't know
[11:00] <jhancock_> Can anyone help me?
[11:00] <matsubara> LaserJock: I just think the former URL in the report are more useful than the latter. the +text version of the bugtask url would show only the status of the given bugtask.
[11:00] <LaserJock> matsubara: but in the web UI you are sent to the later
[11:00] <LaserJock> matsubara: so you *have* to use bug.launchpad/bugs/#/+text
[11:00] <LaserJock> which I can't find documented anywhere, btw
[11:05] <matsubara> LaserJock: that "feature" wasn't supposed to last.
[11:05] <LaserJock> matsubara: I guess, but kiko keeps telling me it's the way to go ;-)
[11:07] <kiko> matsubara, and my question is always, why not?
[11:08] <matsubara> because we're supposed to have a nice xmlrpc interface instead of relying on parsing the +text page?
[11:15] <kiko> matsubara, what would the xmlrpc interface return to you, in that case?
[11:15] <matsubara> LaserJock: anyway, do you think the ubuntu/+source/gnome-session/+bug/26419/+text should display the same info as the bugs/26419/+text?
[11:15] <matsubara> kiko: anything that you would like to query about the bug or bugtask.
[11:15] <kiko> I think they can be the same, but I also think that more information could be added to that. 
[11:15] <kiko> if you're asking me :)
[11:15] <kiko> matsubara, and in what format would it be returned?
[11:15] <thumper> morning
[11:15] <matsubara> kiko: probably a nice xml
[11:15] <kiko> morning thumperoo
[11:15] <kiko> matsubara, see, this is where this argument falls down. where is xml better than rfc-822? 
[11:15] <kiko> where is xml even "nice"?
[11:15] <kiko> I find it harder to read
[11:15] <kiko> and it's harder to parse (you can no longer use cut-n-grep)
[11:15] <thumper> hi kiko
[11:16] <matsubara> maybe you're right.
[11:17] <matsubara> I just thought we'd not improve the +text page anymore...
[11:17] <LaserJock> well
[11:17] <LaserJock> if you're using xml for queries then it makes sense I suppose
[11:18] <kiko> xmlrpc is not xml, though
[11:18] <kiko> there is nothing xml-ish that the programmer needs to deal with
[11:19] <kiko> in a way xmlrpc is poorly named because it underlines the format, and not the service
[11:19] <LaserJock> I thought you still used xml to do queries, but admittedly I know close to nothing about xmlrpc
[11:19] <kiko> no
[11:20] <LaserJock> these +text pages are nice for data mining LP
[11:20] <LaserJock> but I'd like to see a doc on what pages have +text pages
[11:22] <kiko> LaserJock, matsubara: http://docs.python.org/lib/xmlrpc-client-example.html
[11:22] <kiko> >>> server = ServerProxy("http://betty.userland.com")
[11:22] <ajmitch> xml-rpc & python can be done nicely
[11:22] <ajmitch> afaik it's been on the todo list for launchpad for awhile
[11:22] <matsubara> LaserJock: there're only 2 pages with +text versions: +bug/XXX/+text and project/+bugs-text
[11:23] <LaserJock> isn't there also a text for search results?
[11:23] <LaserJock> dang it, my memory sucks
[11:24] <LaserJock> ajmitch: is there anything we gain with xml-rpc over plain text pages?
[11:24] <matsubara> I believe it's the +bugs-text one, which returns all bugs for the given project/source package.
[11:24] <LaserJock> oh right, sorry, doh
[11:24] <ajmitch> LaserJock: that xml-rpc is meant to be used for callign remote procedures
[11:25] <ajmitch> so not much
[11:25] <LaserJock> If we have a good email+plain text interface do we need xml-rpc?
[11:25] <ajmitch> it does mean you wouldn't have to build query strings up yourself for searches
[11:27] <ajmitch> the ServerProxy that kiko was talking about earlier means that you can treat things as a normal python object in your code & get results from methods
[11:27] <kiko> >>> print server.examples.getStateName(41)
[11:43] <kiko> South Dakota
[11:43] <kiko> no xml! :)
[11:43] <kiko> right
[11:43] <kiko> bugs-text actually does query IIRC
[11:43] <LaserJock> ok, now I understand matsubara's question better
[11:43] <LaserJock> bug 26419 can be either https://bugs.launchpad.net/gnome-session/+bug/26419 or https://bugs.launchpad.net/ubuntu/+source/gnome-session/+bug/26419
[11:43] <ubotu> Launchpad bug 26419 in gnome-session "gnome-session hangs when "lo" is not correctly configured" [Medium,Triaged]   - Assigned to Sebastien Bacher (seb128)
[11:43] <ubotu> Launchpad bug 26419 in gnome-session "gnome-session hangs when "lo" is not correctly configured" [Medium,Triaged]  https://launchpad.net/bugs/26419 - Assigned to Sebastien Bacher (seb128)
[11:43] <LaserJock> so the question would be what data would +text have for the individual tasks
[11:43] <LaserJock> well, that's confusing
[11:43] <kiko> right
[11:43] <kiko> I was just saying that +bugs-text right now accepts GET parameters
[11:43] <kiko> I'm not arguing against an RPC interface!
[11:43] <LaserJock> but how the heck do you guys decide where launchpad.net/bugs/# goes?
[11:43] <LaserJock> that was my problem
[11:45] <kiko> the redirect? I think it goes to the first task
[11:45] <LaserJock> right
[11:45] <LaserJock> so in a practical sense
[11:46] <LaserJock> I go to the bug in my browser and expect to be able to put +text on the end
[11:46] <LaserJock> but instead I need to make a new URL
[11:46] <matsubara> LaserJock: sorry if I wasn't clear.
[11:47] <kiko> the content should be identical
[11:47] <LaserJock> well, I'm just having an idiot day I guess. I didn't realize there was a URL for *each* task
[11:47] <LaserJock> I thought it was all one report/URL
[11:48] <LaserJock> now I get what you mean by "context bugtask"
[11:50] <LaserJock> you guys don't happen to have a glossary do you? ;-)
[12:09] <kiko> heh
[12:09] <kiko> ajmitch, I am in favor of "bugtask" 100% but I'm just a voice in the crows it seems
[12:14] <ajmitch> kiko: took me awhile to understand bugtasks but I'm certainly not opposed to calling them that :)
[12:14] <kiko> ajmitch, the problem is if you /don't/ call them bugtasks, what do you call them? :)
[12:14] <ajmitch> 'bug entrails'