? [01:21] spiv: that's all fine. [01:21] stub: https://chinstrap.ubuntu.com/~dsilvers/paste/file3WUgFg.html is what I found. [01:21] BjornT: I need the field in a single row, instead of the usual 'title', 'field', 'description' 'submit' button lines [01:21] SteveA: Cool. [01:22] stub: So long as "make check" in sourcecode/zope passes, I'm happy (which only needed two of the src/bugtracker ones, I believe). [01:22] carlos: well, you can use tal:define="widget nocall:view/foo_widget"... metal:use-macro="context/@@launchpad_widget_macros/launchpad_widget_row" [01:22] stub: Of course, it turns out "make check" doesn't pass unless you're in a timezone where January isn't in daylight savings... [01:23] stub: Anyway, that helps a lot. Thanks! [01:23] BjornT: what will do that? === mpt [n=mpt@203.109.220.214] has joined #launchpad [01:24] spiv: We run make check in some Indian timezone IIRC... [01:24] SteveA, I don't know what you mean [01:24] BjornT: I'm getting just one widget, that's not the problem. The problem is the way we render it [01:25] it uses three lines to represent a single widget and the submit button is in another line [01:25] BjornT: I need a single line to fit it with our current UI [01:25] carlos: ah, in that case it's better to do the layout yourself [01:25] SteveA, should "Items from last meeting" be renamed "Actions from last week" to match the section? [01:26] spiv: I've committed the rest of the __init__.py's anyway [01:26] BjornT: something like:

and add my own label? [01:27] carlos: yeah, exactly. since it's a choice widget you don't have to care about errors or whether it's required or not. [01:28] mpt: i guess there are two things going on here. 1. checking on the progress of things people said they'd do. 2. ensuring that items deferred for discussion in the next meeting are discussed in the next meeing [01:28] we need a section of the agenda for case 1 [01:28] for case 2, these appear as proposed agenda items [01:29] ok [01:29] BjornT: ok, cool [01:32] grr, the widget is inside a

tag [01:34] mpt: I think I will need some help from you to help the translation form UI layout [01:34] s/to help/to fix/ [01:36] ok [01:37] carlos: if necessary, it's not that hard to get rid of the

tag. [01:37] BjornT: how? [01:38] stub: Thanks! [01:38] carlos: you have to subclass SelectWidget and override _div() to return an empty string. [01:39] stub: Yeah, those tests passed, but I just notice they don't pass in Sydney's timezone :) [01:39] graaaaah [01:39] branch puller borken again [01:39] carlos: then, before you call setUpWidgets(), you do 'self.alternative_languagage_widget = CustomWidgetFactory(YourWidgetClass)' [01:40] jamesh: pretty please finish the patch to remove the supermirror-pull-list.txt page... [01:40] BjornT: ok, that would be a good solution [01:40] thanks === matsubara [n=matsubar@200-171-140-32.dsl.telesp.net.br] has joined #launchpad === SnkBite [n=SnkBite@212.25.63.35] has joined #launchpad [01:51] Meeting in 9 mins === bradb [n=bradb@modemcable092.66-130-66.mc.videotron.ca] has joined #launchpad === ddaa bumps the severity of bug 40383 [01:53] Malone bug 40383 in launchpad-bazaar "remove InternalHTTPLayer" [Normal,Confirmed] http://launchpad.net/bugs/40383 === benzai [n=zaheda@82-71-18-29.dsl.in-addr.zen.co.uk] has left #launchpad [] === ddaa uses admin superpowers to fix the nasty branch that had a space in its url [01:59] MEETING TIME [01:59] here! [01:59] welcome to this week's launchpad developer's meeting [01:59] who's here today? [01:59] me [01:59] me [01:59] me [01:59] me === stub belches [02:00] me [02:00] me [02:00] me [02:00] I'm around if needed for consultation, but I'm not officially here [02:00] ddaa: is there a name validation constraint missing there that allows the space? [02:00] me [02:00] me [02:00] stub: is that a thai salutation? [02:00] Iron chef is on [02:00] me [02:00] me [02:00] sabdfl: not sure, allowing spaces in URL is not a completely stupid idea [02:00] (though trailing spaces are almost certainly wrong) [02:01] is that everyone? [02:01] stub: Then you should say something like "bang a gong, we are on!" [02:01] ddaa, it's kinda not so good [02:01] jordi: ? [02:01] ok [02:01] == Agenda == [02:01] * Roll call [02:01] * Agenda [02:01] * Next meeting [02:01] * Activity reports [02:01] hi [02:01] * Items from last meeting [02:01] I'm here [02:01] * Launchpad oops milestone report [02:01] * Outstanding sysadmin requests [02:01] * Production and staging (stub) [02:01] * ---- [02:01] * Use of dapper bzr (kiko) [02:01] * cc experts onto bugs (steve) [02:01] * bzr and PQM performance (kiko) === sivang is here (still just observing) [02:01] * Rebalancing portlets (kiko) [02:01] * ---- [02:01] * Keep, Bag, Change === ..[topic/#launchpad:SteveA] : https://launchpad.net/ | developer meeting: Thu 118May, 1200UTC (wiki:MeetingAgenda) | launchpad-users@lists.canonical.com (wiki:MailingLists) | Channel logs: http://tinyurl.com/72w39 [02:01] * Three sentences [02:02] [02:02] next meeting: same time next week? [02:02] ddaa: talk to me later, I have a branch with a widget that strips leading and trailing whitespace. [02:02] 5 [02:02] 4 === benzai [n=zaheda@82-71-18-29.dsl.in-addr.zen.co.uk] has joined #launchpad [02:02] 3 [02:02] 2 [02:02] 1 [02:02] done [02:02] activity reports [02:02] i'm still the suck, having produced none recently. jamesh and i talked about this this morning. [02:02] how's everyone else doing? [02:02] uptodate as usual [02:02] up to date [02:02] up to date [02:02] I'm up to date. [02:02] i'm up to date [02:02] up to date, not as usual :) [02:02] I am good [02:02] up to date [02:02] i'm up to date [02:02] I sent some in, not much detail yet because I'm not doing much that's useful outside my own head yet [02:02] I suck [02:02] well done jordi [02:02] up to date [02:02] SteveA, cprov calls and sends in an excuse -- network is temporarily down at his pad [02:02] thanks malcolm [02:03] kiko: ok, thanks [02:03] one day behind [02:03] mpt: okay. don't let it spoil a great record [02:03] thanks to ddaa, carlos, bradb, spiv [02:03] BjornT, kiko, stub, matsubara, and salgado [02:03] ... up to date [02:03] also [02:03] cool [02:03] thx steve [02:03] so, just james and me [02:04] oy [02:04] we discussed some ideas this morning, which have been discussed before [02:04] kiko: let's chat about it later [02:04] * Items from last meeting [02:04] * '''bradb''' to make an RT ticket on killing malone-users@, and tell SteveA the RT number [02:04] k === bradb did that [02:04] * '''carlos''' to meet either kiko or SteveA about scheduling [02:04] that's done [02:05] * '''cprov''' to sort out the librarian bug [02:05] we did it [02:05] * '''jamesh and spiv''' to send activity summaries [02:05] SteveA, he did. [02:05] I've sent mine. [02:05] we went over it at length and wrote a number of changes to avoid the issue [02:05] fixed the sync tool [02:05] and I've been trying to get some mdz time to sort out the packages [02:05] but they are only 3 [02:05] so it is not ABD [02:06] jamesh: did you send a summary? [02:06] * '''mpt''' to land text addition to oops page about subscribing to the list [02:06] * '''mpt''' to get `matthew.thomas@canonical` working and registered on Launchpad [02:06] done [02:06] * '''SteveA to talk with jamesh and spiv''' about the activity reports of all three [02:06] and done https://launchpad.net/people/mpt/+portlet-emails [02:06] jamesh and i talked. didn't talk with spiv, except in general [02:06] i think that wraps up items from last week [02:06] SteveA: no I didn't. I'll fix it up now [02:07] thanks jamesh [02:07] * Launchpad oops milestone report [02:07] matsubara, take the floor! [02:07] https://launchpad.net/products/launchpad/+bug/43802 and there's nobody assigned to it. [02:07] Kiko, Steve any news about the other two time outs? bugs 2497 and 35850 [02:07] Malone bug 35850 in launchpad "TeamParticipation caching not working" [Major,In progress] http://launchpad.net/bugs/35850 [02:07] matsubara, stub landed a fix for the latter yesterday, though he has the bad habit of not including bug numbers in his merges. [02:07] bug 43802 is the $distro/+builds pages [02:08] stub did a patch for the TeamParticipation caching problem [02:08] And the SQL Out of order is reported as bug 44032 and assigned to Stuart. [02:08] Malone bug 44032 in launchpad "Out of order SQL queries triggering foreign key constraints" [Normal,Unconfirmed] http://launchpad.net/bugs/44032 [02:08] matsubara, I can fix the former with some prejoining, but I've been very busy. [02:08] Bug https://launchpad.net/products/malone/+bug/39674 (Attachments of private bugreports are public) is a security concern raised by ddaa this week on the list. There's nobody assigned to it. [02:08] Malone bug 39674 in malone "Attachments of private bugreports are public" [Normal,Confirmed] [02:09] matsubara: probably a good idea not to announce that in a public meeting [02:09] matsubara, yeah, well, it requires some spiv time and I believe some architecting. [02:09] SteveA: should I make the bug private them? [02:09] kiko: Keybuk and infinity are really keen for 43802 to be fixed, they were complaining about it a several hours ago. [02:09] matsubara: if the bug is public, then it's already public [02:10] spiv, I... I can't do anything about it this week. you can add a prejoin to the page if you like, though [02:10] They aren't really public - if you know the URL then the bug information has already been exposed. [02:10] It's been public since last month [02:10] okay, so it's not a problem then [02:10] It isn't really much of a security issue. It could be improved, but I don't think it is a priority. [02:10] kiko: I don't have time this week either, unfortunately. [02:10] And I would like to know if is there any policy about changing pagetests to the new format? [02:11] It's a problem if people are talking about attachments on IRC on the assumption that they're private because they're attached to private bugs [02:11] (And if someone deems it a priority, I think we are better off obfuscating URLs rather than making the librarian understand auth in the short term. Long term, we want the librarian to hook into the auth systems) [02:11] mpt: Valid point. [02:11] mpt++ there's a habit here of linking to private data (e.g. the nopaste) [02:11] the URLs already contain unguessable parts, don't they? [02:11] matsubara: we should change pagetests to the new format if you're working on them [02:11] or do you mean not including the URLs in emails? === bradb has been converting PTs to testbrowser as he goes [02:12] matsubara: particularly if you're debugging a failure. it can be easier to rewrite in the new way than debug a complex diff [02:12] SteveA: Is it ok to mix the two kinds of tests in the same file? [02:12] yes, the nopaste URLs are a good analogy [02:12] and yes, as jamesh points out, email transmission is not secured either [02:12] jamesh: Attachments are slightly guessable, assuming you know that the bug exists and can guess appropriate filenames [02:13] matsubara: yes it is fine [02:13] stub: don't the URLs include the library file alias ID too? === aa_ [n=ali@host81-153-28-158.range81-153.btcentralplus.com] has joined #launchpad [02:13] jamesh: Yes, but that is trivially brute forced [02:13] jamesh: It's a sequentially allocated integer, so extremely guessable. [02:14] i'd like us to make security-related bugs on launchpad private, in general [02:14] it is easier to make a private bug public than to try and hide a public bug [02:14] and we should take the time to consider the effects of the bug. [02:15] the effects of a security bug can be surprising, considering all the systems it may be involved with [02:15] The bug was reported as public by an Ubuntu developer, iirc [02:15] as mpt points out in this case, exchanging URLs on irc is an issue [02:15] It was a day or two before any of us looked at it [02:15] mpt: and the reporter didn't mark it as "security related" ? [02:16] It was reported before Malone had a "security related" flag [02:16] It was next touched three weeks later, when matsubara confirmed it [02:16] https://launchpad.net/products/malone/+bug/39674/+activity [02:16] Malone bug 39674 in malone "Attachments of private bugreports are public" [Normal,Confirmed] === cprov [n=cprov@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [02:17] i see [02:17] so we could reasonably expect a reporter to use the security-related tag next time [02:17] cprov: I'm late, sorry [02:17] thank you for your oops report, matsubara. have all your issues been dealt with? [02:17] hi cprov. kiko explained [02:18] yes. [02:18] I don't understand the point of separate "security-related" and "private" flags, though [02:18] I'm up to date (activities) [02:18] but I suppose other people have discussed that [02:18] mpt: put it on the list for our next voice call. [02:18] mpt, yeah, we have -- bradb is changing that somewhat. [02:18] thanks matsubara [02:18] * Outstanding sysadmin requests [02:18] I haven't changed anything though! [02:18] 6 [02:18] 5 [02:18] 4 [02:18] mpt: Launchpad has private bugs when we need to detail too much database schema or code. Launchpad has security bugs when we have a security problem. [02:18] 3 [02:18] 2 [02:18] 1 [02:18] pushsftp log file access (#8372) [02:18] 2073 [02:19] would be nice eventually [02:19] RT 2073? [02:19] yes [02:19] spiv: RT-8372? [02:19] SteveA: yep. [02:19] because it would let people do usability tests on staging without having to bat away alerts all the time [02:19] mpt: what is RT2073 about ? [02:20] staging.ubuntu.com -> staging.launchpad.net with a proper certificate, mainly [02:20] mpt: this is underway [02:20] spiv: noted about RT-8372 [02:20] how will staging.launchpad.net interact with us having multiple *.launchpad.net domains? [02:21] mpt: That issue is probably irrelevant. All staging domains have recently been moved to *.staging.launchpad.net (and some new ones for blueprint and other projects) [02:21] jamesh: foo.staging.launchpad.net [02:21] SteveA: I was thinking cookie and SSL cert wise [02:21] huh! [02:21] mpt: Although Konqueror and IE6 are bitching about the certificate due to the depth of the domain names we are using [02:21] i suppose we must be careful about the domain attached to cookies on staging. [02:22] Mmm... yes... [02:22] perhaps use a different cookie name for staging [02:22] a *.launchpad.net cookie used for auth will be passed to *.staging.launchpad.net [02:22] Sounds good. [02:22] I'll file a bug [02:22] i think sharing the cookie among launchpad.net and blueprint.launchpad.net is a feature [02:23] thanks jamesh. good observation. [02:23] stub: please subscribe me to the bug [02:23] * Production and staging (stub) [02:23] Staging is being boring, which is good. [02:23] Production rollout got missed this week due to bzr taking way to long to branch on Tuesday and not wanting downtime on Wednesday for the Ubuntu Hug day. [02:23] I can roll updates out tonight if anyone is going to cry - otherwise I'll skip this weeks rollout and do an update next Tuesday of HEAD-as-of-now(ish). [02:23] stub: mark requested a rollout this week [02:24] stub, yeah, do it tonight. [02:24] I have the report hot [02:24] Of any particular feature or patch? [02:24] all the pending ones :-) [02:24] stub: yes, let me check... [02:24] rev 1851 === ddaa would like the new bzr _not_ to be rolled out until bug 44183 and bug 44182 are fixed [02:24] Malone bug 44183 in launchpad-bazaar "supermirror branch puller does not preserve source branch format" [Major,Confirmed] http://launchpad.net/bugs/44183 [02:24] Malone bug 44182 in launchpad-bazaar "supermirror branch puller leaves empty branch when initial mirroring fails" [Major,Confirmed] http://launchpad.net/bugs/44182 [02:24] it will help a lot to translate dapper [02:25] though it's not a critical issue [02:25] ddaa: Can I just not update the supermirror code - keep it running on current code? [02:25] I don't think I saw any DB patches land that would bite it... [02:26] stub: it's okay, the new bzr has not yet landed AFAIK [02:26] lifeless/spiv just hold it back until rollout [02:27] stub: what's your decision? [02:27] stub: I mean, please rollout supermirror stuff [02:27] Well... since I don't know what Mark actually wanted to land, I'll roll out 3538 or there abouts and cross my fingers === AlinuxSOS [n=AlinuxOS@d83-176-98-164.cust.tele2.it] has joined #launchpad [02:28] I'll need to trawl the commits lists to determine the exact revision [02:28] stub, including 3541 :-) [02:28] stub: there were db patches in my bzrsync landing that tighten some constraints. That might cause problems if the bzrsync code is not updated [02:28] mpt: Noted. [02:29] HOWEVER! If the bzr tagging takes too long again, it might be tomorrow [02:29] stub: The revision number I asked you is one of marks' landing [02:29] stub, there's 3540, 3541, 3545, 3547 and 3548 though [02:29] which all fix bugs [02:29] i.e. the branch scanner will need to be rolled out, or it will break [02:29] you could also include 3546 if you are feeling brave [02:29] kiko: Noted [02:29] let's move on [02:29] kiko: I landed that - no problems ;) [02:30] we can refine the patch list later [02:30] * Use of dapper bzr (kiko) [02:30] okay [02:30] can you all make sure right now that you are using bzr from dapper? [02:30] I believe that is 0.8~200604291148-0ubuntu2 [02:30] bzr (bazaar-ng) 0.8rc1 [02:30] so ending 1148-0ubuntu2 [02:31] there is a 0.8 package in dapper but it conflicts with the current bzrtools === ddaa removes his symlink to the local bzr.dev checkout [02:31] current bzrtools package, that is [02:31] mpt, you need to use dpkg -l bzr to check. [02:31] mpt: apt-cache policy bzr [02:31] Yes, I didn't upgrade today because it wanted to remove bzrtools, and I didn't want to do that [02:31] jamesh, yeah, bzrtools I have is from 0407. [02:31] Bug 44160 [02:31] Malone bug 44160 in bzrtools "bzrtools removed on bzr upgrade and cannot be reinstalled" [Normal,Unconfirmed] http://launchpad.net/bugs/44160 === bradb is running 0.8~200605051825-0ubuntu1 [02:31] same as bradb [02:31] use: apt-cache policy bzr, and check that the marked one is from dapper/universe [02:32] huh [02:32] Why is it in universe? [02:32] ok, offtopic, sorry [02:32] I'm on dapper, but my bzr is just 0.8-0ubuntu1 === carlos is using the same version as bradb due the bug #44160 [02:32] Malone bug 44160 in bzrtools "bzrtools removed on bzr upgrade and cannot be reinstalled" [Normal,Unconfirmed] http://launchpad.net/bugs/44160 [02:32] MalcolmCleaton: use apt-cache policy bzr, or dpkg -l bzr [02:32] MalcolmCleaton: does that still report that version? [02:32] MalcolmCleaton: that's a good one [02:33] Yes [02:33] oh, I think that's the newer version. [02:33] is latest version [02:33] just did apt-get update, and apt-cache policy tells me the only thing it knows is 0.8-0ubuntu1 [02:33] cool [02:33] but you lack the bzrtools package then [02:33] not if he did an apt-get update [02:34] (or upgrade) [02:34] hmmm [02:34] okay, so a mess [02:34] let me ask this again next week [02:34] ddaa: why not rollout ? [02:34] the bzr packagers should get the bzrtools thing sorted [02:35] bradb: bzrtools conflicts with that bzr version, if he has it, bzrtools is not installed [02:35] lifeless: will cause knit branches to be published at weave [02:35] kiko: may i move on? [02:35] MeetingAction: kiko to talk to bzrtools and bzr packagers to sort out this nonsense. [02:35] lifeless: all of them [02:35] yes [02:35] * cc experts onto bugs (steve) [02:35] carlos: yeah, but upgrade doesn't remove packages. only dist-upgrade would. [02:35] after meeting I guess [02:35] so, we're getting too much bug traffic for various people such as stu and me to keep up with it all [02:35] i haven't been keeping up for *ages* [02:35] bradb: and you will not get that new package either ;-) [02:36] so, i propose that we proactively cc stuart onto bugs where you think he might be interested. same for me. same for other people who specialize in a particular area in launchpad. [02:36] it is easy to unsubscribe yourself if you find you're subscribed to a bug you're not interested in [02:37] kiko, matsubara: does that sound okay with you? === carlos is not following the bug mailing list either [02:37] SteveA: yes. [02:37] cool, thanks [02:37] * bzr and PQM performance (kiko) [02:37] SteveA, how do you filter these bugs? [02:37] okay, n/m === bradb attempts to, but it would eat at least a couple days to catch up. otherwise it's at least a couple hours per day to keep up with. [02:37] I'm subscribed to launchpad-bugs@, so if anyone (or any checkbox) subscribes me to any bug I will growl at them [02:37] um, would there be any interest in me releasing this bug-reporter separately from the application I have embedded it in? [02:38] am I correct in saying that PQM is at the moment being as slow as molasses? [02:38] and that the root cause of this is bzr merge slowness? [02:38] and if so, is there a plan to fix that? and if so, is there an ETA? [02:38] aa_: please ask after the launchpad meeting. it will finish at 4.45 sharp [02:38] um, 1245 UTC [02:38] kiko: As far as I can tell, yes for the cause. [02:38] slow yes. I can generate some hard stats on precisely why, but I have lots of indications that it is reweaving being triggered still [02:39] the plan is to migrate to knits. [02:39] SteveA: oh, okie dokie [02:39] lifeless: please communicate the plan and projected dates to kiko [02:39] lifeless, indication but not proof? I'm in search of hard facts [02:39] bzr 0.8 is out. bzrtools 0.8 is out, so I need to finish off testing the tool chain (nearly done) [02:39] * Rebalancing portlets (kiko) [02:39] we have 5 minutes of meeting left === kiko sighs [02:39] and then do the documentation and we can upgrade [02:39] so i'm aggressively moving on [02:39] done [02:40] * Rebalancing portlets (kiko) [02:40] mark has requested we take a look at our existing pages [02:40] after the recent UI work with the new hierarhy and javascript menu bling... [02:41] we removed the "site map" portlet [02:41] the fact that navigation has moved from the left to the right means that many pages do not have RHS portlets [02:41] yep [02:41] some of the ones without RHS portlets could probably be switched to 2 column layout === ddaa cannot help but find it a good thing [02:41] e.g. the bzr branch info page [02:41] mark wants us to make sure that 3-col pages look "balanced" in terms of having a fair balance of LHS and RHS portlets [02:42] ddaa, it violates the current UI guidelines, and that is a higher-priority problem. if you want to fix the UI guidelines, pick that battle. === beyond [n=beyond@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [02:42] jamesh: in those cases, we should consider whether there is any information that could be helpfully presented in the RHS, or whether a 2 col layout is really better for that page. [02:42] jamesh, 2-col is only supposed to be used if there is an actual issue with page content flodding into RHS portlets. [02:42] jamesh: please consult with kiko or me in these cases [02:42] kiko: I know it's a lost battle right now, and I have other things to attract sabdfl's wrath [02:43] I am not particularly in favor of the current UI guidelines, but I am in favor of us following them until they are changed. [02:43] (IOW I feel your pain, but you need to work with what we have) [02:43] more UI discussion for after this meeting, if necessary please [02:43] kiko: sure. I brought up the branch info page because it is displaying revision log messages as one column in a table [02:43] * Keep Bag Change (defered) [02:43] * Three Sentences [02:43] DONE: schooltool merge/import, rosetta email processing, promotion of official rosetta usage for Gaim (ongoing) [02:43] jamesh, right [02:43] TODO: Clear the import queue, several pending emails [02:43] DONE: Various production odds and sods. Drive by bug fixes. [02:43] TODO: Searching g inprovements (Bug 29227), try to investigate the-row-i-just-inserted-isn't-there errors (Bug 44032) [02:43] BLOCKED: No. [02:43] Malone bug 29227 in malone "Searching for "pmu" doesn't find "/dev/pmu"" [Critical,In progress] http://launchpad.net/bugs/29227 [02:43] DONE: Taught the mirror prober how to probe cdimage mirrors and did a few other fixes on it, code review, shipit fixes and discussion and some random bug fixes [02:43] DONE: started coming up to speed [02:43] TODO: Get shipit reviewed and land it, finish changes on the mirror prober and get it through review. [02:43] BLOCKED: No [02:43] Malone bug 44032 in launchpad "Out of order SQL queries triggering foreign key constraints" [Normal,Unconfirmed] http://launchpad.net/bugs/44032 [02:44] DONE: oops report analysis, bug triage, support requests, fixed oops bugs. [02:44]