[00:00] I've spent a few minutes trying to work out where file uploads are handled in lazr.restful [00:03] file_content=Bytes(constraint=productrelease_file_size_constraint), [00:03] So it'll be in the Bytes handler [00:03] Support was only added relatively late, AFAICR [00:10] wgrant: ... which I can't find in lazr.restful [00:11] class BytesFieldMarshaller(SimpleFieldMarshaller): [00:13] (is it just me, or does lazr.restful have marshall/unmarshall around the wrong way?) [00:17] wgrant: I can't see anything in either BytesFieldMarshaller or SimpleFieldMarshaller that would call _decode [00:17] StevenK: No, so you'll have to trace through to see how it happens [00:17] And why it doesn't happen for normal file uploads [00:24] wgrant: Hmmm, pdb.set_trace() inside IProductRelease.addReleaseFile() is probably far too late [00:25] StevenK: Probably, but it might vaguely hint where it could get the data [00:26] I'd rather break in the Bytes (un)marshaller [00:29] wallyworld___: https://lh6.googleusercontent.com/-awustg4A1HQ/UQ1pZ49aYjI/AAAAAAAB0_k/ZXhib0JPDXQ/s728/YoomaShreddedBedText.jpg [00:30] what you saying 'bout me? [00:52] * StevenK stabs this script [00:54] Unknown consumer (System-wide: Ubuntu (undermined)). [00:55] undermined? :) [00:56] mwhudson: Matches my current naming scheme [01:16] wgrant: How can I debug Unknown consumer? [01:16] StevenK: Probably means your gnome-keyring has a token that your Local DB no longer has [01:17] Assuming it's not production [01:17] It's dev [01:17] Right [01:17] Poke around with seahorse [01:17] And delete the key that is no longer valid [01:18] Right, deleted two of them [01:18] The authorization page: [01:18] ... [01:18] Success! [01:18] Great [01:19] Hmm, but my pdb did not fire [01:21] Oh [01:21] I edited the wrong version [01:23] Hah [01:33] wgrant: http://pastebin.ubuntu.com/1606599/ [01:34] I can't see the tarball content being called in _decode, but the GPG sig is [01:35] StevenK: You might have to look through the code to see how lazr.restful gets its values, and how Zope forms do it differently. [01:35] Because Zope forms clearly don't use the decoded version [01:39] Hmmm [03:20] * StevenK stabs lazr.restful for being obstreperous. [04:30] wgrant: QA could be fun [04:34] Quite [04:38] StevenK: SPRB and TTB are populated on DF, BPB is about 1/3 done... [04:38] Wow [04:38] Nice [04:39] It's been going for about 3 hoursish, though [04:39] Or 2 [04:39] Something like that [04:58] Bleh [04:58] launchpadlib, forget your auth token! [05:42] wgrant: Is DF done? [05:43] No [05:43] f | 2112345 [05:43] t | 1721777 [05:44] Almost half way [06:12] wgrant: Put me out of my misery trying to work where lazr.restful gets it values from? [06:26] StevenK: ResourceOperation.validate [06:27] (I just stuck a pdb.set_trace() in addReleaseFile, called it with lp-shell, and looked through the traceback to see what would be parsing args) [06:29] Bleh, I did that too, but didn't see ResourceOperation there [06:29] It wasn't [06:29] Well, I guess it was [06:29] But that method wasn't [06:32] Unknown consumer again! And Seahorse is unhelpful [06:33] You probably nuked your DB [06:33] Or it expired [06:36] I did nuke my DB [06:36] Your denorm columns not existing required a make schema [06:38] Or a database/schema/upgrade.py :) [06:39] I will usually just run make schema === Nigel_ is now known as G [07:30] StevenK: https://code.launchpad.net/~wgrant/launchpad/flatten-bfj-5-app-cleanup/+merge/146349 is a big bit boring and mostly red diff [07:30] And contains two conflicts [07:31] Pretend they're not there, and I'll fix them as I merge the pipe :) [07:32] 183>>>>>>> MERGE-SOURCE [07:32] Haha, looks like they're earlier up the chain [07:33] wgrant: PackageBuildMixin does not die yet? [07:33] StevenK: It never dies [07:33] It's a mixin [07:33] Used for functionality that's shared between BPB and SPRB [07:33] Like uploads and such [07:34] Possibly rename it? Or you don't care enough? [07:34] Hmm? [07:34] It's a mixin for package builds [07:34] So PackageBuildMixin seems like an apt name. [07:34] Fairy nuff [07:34] Did you have an alternate suggestion? [07:37] No, was just trying get the name 'PackageBuild' killed everywhere [07:39] wgrant: Oh, when does IPackageBuild die? [07:41] StevenK: It doesn't [07:41] It's an interface for package builds [07:44] wgrant: So IBinaryPackageBuild will also implement IPackageBuild ? [07:44] StevenK: It in fact inherits from it [07:45] BinaryPackageBuild implements it, with the help of PackageBuildMixin [07:45] Right [07:45] wgrant: r=me [07:45] Thanks [08:22] Has anyone had a chance to look at https://code.launchpad.net/~cjwatson/launchpad/bpph-phase/+merge/144154 ? I have another piece of work to do that touches some of the same code, so it would be nice to get this out of the way. [08:24] cjwatson: Oops, sorry [08:24] Should have time to look at that tomorrow [08:24] Great, thanks [08:29] StevenK: Can you QA 16468 tonight? [08:29] Should be on qas shortly [08:31] Is now, apparently [08:33] wgrant: Done [08:34] StevenK: Thanks [08:34] We shall garbo it up on prod tonight [08:41] wgrant: All gree [08:41] *green [08:46] good morning === _mup__ is now known as _mup_ === gary_poster|away is now known as gary_poster === Ursinha-afk is now known as Ursinha === teknico_away is now known as teknico === salgado is now known as salgado-lunch === salgado-lunch is now known as salgado === slank_away is now known as slank [15:05] * mpt rediscovers bug 533044 [15:05] <_mup_> Bug #533044: Resummarizing bug report doesn't change page title < https://launchpad.net/bugs/533044 > === teknico_away_ is now known as teknico === teknico_ is now known as teknico === deryck is now known as deryck[lunch] === deryck[lunch] is now known as deryck === gary_poster is now known as gary_poster|away === gary_poster|away is now known as gary_poster