[00:28] i can't seem to get uploaded packages to show up on my PPA page [00:29] dput executes without any error messages [00:37] What exactly are you executing? [00:38] dput my-ppa pyrocket_0.5_source.changes [00:39] this is right after I run 'debuild -S -sa' [00:39] 'pyrocket' is the name of my package [00:57] kostmo: They should on't appear every 5 or 10 minutes. Do you get an email about it? [00:57] I never received an e-mail, though I expected one. I ran dput last night. [00:58] kostmo: Was your package appropriately signed, with a key associated with your Launchpad account? [01:02] it was. I went through the business of adding and verifying a GPG key to my Launchpad acct., and debuild did prompt me for my passphrase to sign the package. [01:02] nothing with "ppa" or "ubuntu" in my spam folder either [01:04] kostmo: To which directory on the FTP server were you uploading? [01:05] my .dput.cf file has a line that says: incoming = ~kostmo/ubuntu/ [01:06] OK, so that should be fine... [01:07] The last two lines that dput says are: (1) Successfully uploaded packages. , and (2) Not running dinstall. [01:07] Right, but dput doesn't do much - it merely uploads something to an FTP server. [01:08] Can you pastebin your unsigned .changes? It needs to be unsigned, as a signed one is dangerous. [01:08] There are only a couple of reasons for it to not email you. [01:10] ok, so I should run 'debuild -sa' without the '-S'? [01:10] No, you need the -S. [01:10] S is for source-only. [01:10] debuild -S -Sa -us -uc [01:10] (us == unsigned source, uc == unsigned .changes) [01:10] Er, -sa [01:11] Not -Sa [01:12] Format: 1.7 [01:12] Date: Sun, 17 Feb 2008 02:30:26 -0500 [01:12] Source: pyrocket [01:12] Architecture: source [01:12] Version: 0.5 [01:12] Distribution: unstable [01:12] Urgency: low [01:13] Maintainer: Karl Ostmo [01:13] Changed-By: Karl Ostmo [01:13] Changes: [01:13] pyrocket (0.5) unstable; urgency=low [01:13] . [01:13] * Initial release. [01:13] Files: [01:13] ff3937811f2941b997d769d7349bddd8 152 games optional pyrocket_0.5.dsc [01:13] d607386c1f38bf6b3e5123e177d6ff4b 61990 games optional pyrocket_0.5.tar.gz [01:14] !paste | kostmo [01:14] kostmo: pastebin is a service to post multiple-lined texts so you don't flood the channel. The Ubuntu pastebin is at http://paste.ubuntu.com (make sure you give us the URL for your paste - see also the channel topic) [01:14] That Changed-By is bogus. [01:14] That might be it. [01:14] Use a proper email address in the changelog [01:14] Though, hmm, I've seen uploads like that before. [01:14] Ah. [01:15] The distribution is wrong. [01:15] It needs to be an Ubuntu release. [01:15] hey guys, how to add a new distro to the page https://code.launchpad.net/distros ?? [01:15] eMxyzptlk, you can't [01:15] eMxyzptlk: Your best bet is probably to ask a question at the Launchpad Answers URL in the topic. [01:15] wgrant, ah ok [01:15] Thx [01:15] Only Launchpad Admins can do it. [01:16] Okay well I try to ask a question... [01:16] We already set up a project [01:16] for it on LP [01:16] wgrant: thanks, I'll change the Distribution line and see what happens [01:17] kostmo: Change it in the changelog. [01:18] should I change the word 'unstable' to 'hardy'? My ppa only lists repositories for 'intrepid' [01:19] kostmo: If you want it to go to Hardy, change it to hardy. [01:29] ok, I changed the distribution name to 'hardy' and the e-mail address from 'kostmo@lapster' to 'kostmo@gmail.com'. Waiting for a confirmation e-mail. Is it the e-mail that should take 5+ minutes, or just showing up in the PPA list? [01:32] Okay question added https://answers.launchpad.net/launchpad/+question/36900 [01:36] kostmo: They should occur simultaneously. [01:38] ok, 'cause I read in https://help.launchpad.net/PPAQuickStart that it could take up to twenty minutes... so I could be waiting that long for the email [01:39] kostmo: It could take up to 20 minutes for it to appear on ppa.launchpad.net. [01:40] right, but did you say that appearing there is simultaneous with the confirmation e-mail? [01:41] kostmo: Appearing on your PPA page, ie. https://launchpad.net/people/+me/+archive, is different from appearing on ppa.launchpad.net. [01:42] ok [01:43] looks like there's a tilde there rather than a +; https://launchpad.net/~kostmo/+archive [01:45] kostmo: /people/+me redirects to ~yourusername [01:47] ah ic that's a literal "me" -- i wasn't supposed to change it [01:49] Correct. [01:49] ok well anyway -- still no confirmation e-mail -- i suspect it would be here by now if it was going to work [01:49] kostmo: Right, we'll have to wait for somebody with access to the logs on the PPA server. [01:49] cprov: ^^ [01:53] another thing - I'm getting some warnings when I run "dbuild", such as "dpkg-source: warning: unknown information field 'Package'..." [01:54] though when I use "dpkg -b" to make my binary, it results in a perfectly installable .deb package [01:55] I'm concerned that these warnings might have something to do with this upload problem, though dput tells me that the upload has taken place [01:55] As I said earlier, dput does nothing but verify the signature on the .dsc and .changes, and upload them to an FTP server. [01:56] would this type of warning be something that this 'confirmation e-mail' reports back to me? Or will it silently fail? [01:58] I only know of two circumstances in which it will fail to even send an email - an invalid signature (or a signature by a key unknown to Launchpad), or an invalid maintainer address. [02:02] well, Launchpad says that I have an active key "1024D/93D6AA97", and I've been running "debuild -S -sa -k93D6AA97" [02:03] which prompts me for my passphrase, then "dput" later says 'gpg: Good signature from "Karl Ostmo "' [02:04] Right, so it's probably another problem. [02:06] alright... I can leave my xchat window open in case someone with PPA server log access drops in... but I have to take off in an hour [02:07] is there a board I can post on with this problem? [02:08] kostmo: You could ask at https://answers.launchpad.net/soyuz [02:12] kostmo: how are you building your source package, you changesfile is broken -> Unable to find mandatory field 'binary' in the changes file. [02:13] Aha, that will be from the dpkg-source 'Package' error. [02:14] kostmo: Can you please pastebin (not paste here) your debian/control? [02:14] I doubt `debuild -S` would pass thoughtthis control file w/o a warning [02:14] there were warnings for sure [02:14] cprov: Shouldn't it email in that case, as long as the sig is valid and there's a maintainer? [02:15] wgrant: changesfile is unparsable, since it lacks mandatory fields, we can't to much in such cases [02:16] cprov: OK, so this needs that feature where rejected uploads are displays somewhere on launchpad.net. [02:16] *displayed [02:16] wgrant: the plan for 3.0 is do to such basic checks before finishing the ftp/scp session [02:17] cprov: That would be even better. [02:17] wgrant: so dput will 'know [02:17] if something is horribly broken [02:17] ok, I used the pastebin thingy to upload my 'control' file [02:17] http://paste.ubuntu.com/21709/ [02:17] no binaries ? [02:18] Wow. That's broken. [02:18] Only one paragraph, and a Version field. [02:18] kostmo: You need another linefeed before the Package field, and you need to drop the Version field. [02:18] And your description needs to be split into 80-character lines. [02:19] And I would have thought that lintian would have picked all of this up. [02:19] yeah lintian complained too, but since dpkg -b worked, and since I didn't get any e-mail confirmation, I wasn't sure it was actually an issue [02:20] If lintian complains, it is an issue. [02:20] Unless you know why the error is there, and there's no way around it, lintian is always right. [02:21] ok. does dpkg use a different format for the 'control' file? Or is it just more robust? [02:21] Probably more robust. [02:21] Who knows why. [02:23] The control file that killed your PPA upload was a different one, however. [02:23] But generated from debian/control. [02:24] In my debian/control, dbuild still gives me the warning "unknown information field 'Package' in input data in general section of control info file", even after adding another linefeed [02:25] It needs to be "Source:" instead of "Package:" [02:25] ah, sorry, it is [02:27] james_w: it is what? [02:27] oh i get it [02:27] i think [02:29] OK, so when I add that newline in between the "" and "" lines, I get an error when I try to build the binary .deb file with "dpkg": "dpkg-deb: parse error, in file `pyrocket-0.4/DEBIAN/control' near line 5: [02:29] missing package name" [02:30] Pastebin it, please. [02:30] http://paste.ubuntu.com/21711/ [02:31] The description field is corrupt - continuations need to be indented by a space. [02:32] Here's a good example of mine: http://pastebin.com/f1c7db5a1 [02:32] ok... i think i'm seeing a trend of incompatability between dpkg -b and debuild [02:34] dpkg-deb: parse error, in file `pyrocket-0.4/DEBIAN/control' near line 16 package `pyrocket': [02:34] missing version [02:34] How are you trying to build it? [02:35] dpkg -b pyrocket-0.5 [02:35] I've never seen anybody do that before. [02:35] Use dpkg-buildpackage or debuild. [02:35] might have grabbed that from an old tutorial... [02:35] but it worked, so I stuck with it... until now [02:35] A decade old, maybe. [02:36] it also required capitalized DEBIAN, so I added a symbolic link of lowercase debian to the existing DEBIAN [02:39] I wonder if and where it would be appropriate to warn people away from my method (dpkg -b)? [02:54] I'm getting an error with debuild: [02:54] dpkg-genchanges >../pyrocket_0.5_i386.changes [02:54] dpkg-genchanges: failure: cannot read files list file: No such file or directory [02:54] dpkg-buildpackage: failure: dpkg-genchanges gave error exit status 2 [02:55] How can I make launchpad linkify an apt: link in the PPA description? [02:55] vadi2: filing a bug ;) It doesn't support it right now. [02:56] what's the "files list file"? [02:56] vadi2: does firefox supports apt:$url ? [02:57] cprov: yes, apt: works too [02:58] It seems a bug about this was already filed in january [02:59] vadi2: bug # ? [02:59] 179868 [03:04] vadi2: right [03:13] g2g, thanks for the direction wgrant, cprov. Will read up on some modern packaging procedures now :) === cprov is now known as cprov-away [04:47] Is edge meant to be OOPSing regularly (often while trying to authenticate my cookie, it appears) [04:47] *? [05:00] It's a study to see how determined people really are to use LP. :) [05:00] And somebody couldn't make up their mind on how to name table fields, I see. === asac_ is now known as asac [12:50] I clicked on a link to http://bazaar.launchpad.net/~lightyear/storm/twisted_integration_dfr_ref_multithread_fix/revision/204 and I got a page that said I should come here and say what happened [12:54] Odd. I only get a 404 error there (and not an LP-branded 404, but a "Powered by CherryPy 2.2.21 404) [12:54] Ah, but you've left :( [13:43] :P [13:43] persia: thats loggerhead === stgraber_ is now known as stgraber === barry__ is now known as barry === fta_ is now known as fta [17:37] hello, I have noticed that sometimes multiple people are working on the same bug at once unknowingly (say bugsquaders on new bugs) and one person may make changes, then the other submits their changes, and stomps over the first person [17:38] I was wondering if it might be possible to send along both the old and new value for each changed parameter, so launchpad could compare if the current value is what your submission thought was the old value, and if not, warn that it is changed since you loaded the page [17:46] The client only needs to send back the timestamp or ID of the last change, then the server can do the diffin.g [17:47] Bugzilla seems to be good about mid-air collisions. [17:48] Peng: yeah, those solutions would work. Granted sending back the old info would allow better feedback for the user, to see what has changed and have a screen to show what is in conflict and give choices [17:48] but, that would be a great improvement, just having a timestamp or id [17:50] No, what I mean is that the timestamp or ID is all that's needed for the server to be able to show all of that information. [17:51] Peng: I see, you may be right [17:51] * Peng wanders off. [17:51] Bye. :) [17:51] so, what is involved in having that done :) [17:52] Bribe the developers with muffins. :) [17:52] And anime. [17:52] * Peng wanders off. [17:52] * mrooney gets to work on anime-themed muffins === _neversfelde is now known as neversfelde [21:56] jelmer: are you there/ [21:56] Hi ianm_ [21:57] jelmer: I'd like to talk to you about https://launchpad.net/gruler [21:58] jelmer: that's you, no? [21:58] ianm_: ? I registered the trunk branch, yes - it's owned by the gnome-bzr-mirror team [21:58] jelmer: ok. well I'm the author of gruler [21:59] jelmer: it is now known as screenruler and development is happening in launchpad bzr [22:00] it's all quite confusing... there's gruler, screenruler, and screen-ruler [22:00] Ah, ok [22:00] https://launchpad.net/screen-ruler is the official page [22:00] ianm_: We've simply registered all branches that existed in GNOME svn [22:00] I would like to move that to https://launchpad.net/screenruler but someone just registered it [22:00] ianm_: I don't have access to any of these launchpad pages unfortunately (just registered that branch) [22:01] should I put in a Launchpad Answers request to remove it? [22:01] Yeah, that would probably be the easiest way to get rid of the gruler page [22:02] I've had the same problem with some other projects [22:04] there were various GTK+ GUI's registered for bzr that were later merged [22:04] who owns the https://launchpad.net/screenruler project? [22:05] it just has a link to my blog/project page [22:24] can a dev tell me who registered that project page (for MY project) so I can ask them if I can have it? [22:25] or is that info somewhere? [22:29] ianm_: https://edge.launchpad.net/~nitrofurano afaics [22:29] ianm_: but it doesn't seem to be owned/driven by anyone atm [22:30] https://answers.launchpad.net/launchpad/+question/36980 [22:30] ianm_: That should help [22:31] this project now has three launchpad pages [22:31] only one of them correct / owned by me [22:31] sigh :) [22:32] ianm_: you've filed a question, so it's trackable now. [22:55] hi LarstiQ [23:21] hi lifeless [23:33] hi LarstiQ [23:33] erm, lol, just did that [23:33] more caffeine [23:34] hi [23:38] is there any way to use filters on bugs so that it shows only bugs that haven't been fixed ( open bugs minus fix committed ones ) [23:43] I have a bug that was "marked for expiration 18 days ago" -- shouldn't launchpad janitor have made it invalid by now? [23:45] YokoZar, that was disabled. [23:46] cody-somerville_: oh...do I have to manually mark 50 expired bugs invalid now? [23:46] Preferably not. [23:46] Just leave them be unless they're *really* old [23:49] I'm pretty sure most are solved acutally (these are Wine bugs with applications, likely fixed in the newer Wine) [23:52] YokoZar: Maybe you could poke the submitters to retest with the newer wine (assuming you don't have the relevant applications to replicate)? [23:53] persia: That's what I do; when I make the poke I mark the bug incomplete. Some of these reports are now 90+ days old after having me asked that === cody-somerville_ is now known as cody-somerville