00:00:03 <donri> yea i'm totally testing that right now
00:00:48 <donri> Repositories seem to be unrelated. Proceed? [yn]
00:01:02 <donri> let's see what happens!
00:01:13 <rlpowell> *boom*
00:02:17 <rlpowell> stepcut: Yeah, I think it would be better in the main happstack repo, myself.  FWIW.
00:03:00 <stepcut> possibly
00:03:15 <stepcut> unlike things such as acid-state and ixset, it is not useful with out other happstack stuff
00:03:48 <stepcut> what's so great about one giant repo though ?
00:03:56 <donri> i just think we should be consistent
00:04:01 <donri> either split up happstack or merge auth
00:04:06 <donri> ACTION OCD
00:04:22 <rlpowell> heh.
00:04:33 <rlpowell> 08-16:07 <     stepcut> what's so great about one giant repo though ? -- Well, it's rather too late for *that*.
00:04:46 <stepcut> what about happstack-yui?
00:04:58 <rlpowell> I'm only responding to the current state, which is that you've got this *huge* repo witth all these subdirs.
00:05:07 <rlpowell> Splitting them all out would be fine too.
00:05:10 <stepcut> it's never too late!
00:05:12 <rlpowell> As donri said, it's about consistency.
00:05:14 <rlpowell> Heheheh.
00:05:17 <donri> yea i'm planning to deprecate that in favor of yui-happstack and a bunch of other yui packages ;)
00:05:23 <donri> so in that case it's like web-routes-happstack etc
00:05:34 <rlpowell> What's yui?
00:05:40 <donri> js lib
00:05:43 <rlpowell> Ah.
00:05:49 <stepcut> what is the rule for deciding if something belongs in the happstack repository or not ?
00:06:05 <donri> stepcut: whether you're maintaining it and whether it's a happstack-* package :)
00:07:17 <stepcut> so, we should move ixset out then ?
00:07:25 <donri> we should
00:07:38 <donri> or rather, we should deprecate it in favor of data-store ;)
00:07:56 <stepcut> seems sensible
00:13:58 <donri> stepcut: DARCS POWAH http://hub.darcs.net/dag/happstack
00:17:06 <donri> also seems i managed to mess up hub with this
00:17:16 <donri> sm: http://hub.darcs.net/stepcut/happstack/patches 502 proxy error
00:17:49 <stepcut> >:(
00:19:03 <donri> the changelog becomes 4 pages of authenticate history leading back to 2011 and then bam two hours ago happstack-fay stuff
00:19:12 <donri> yay non-linear chronology
00:23:32 <stepcut> :)
00:24:06 <stepcut> well. We should figure out why hub is screwed up and confirm with the darcs people that this is not a really bad idea
00:24:13 <stepcut> but if everything checks out, I'll pull it
00:25:01 <rlpowell> stepcut: So, for auth to work, ControlV needs to know the URL it's being present at.
00:25:07 <donri> i imagine it could be a bad idea if they touch the same files, but don't think any files are the same here
00:25:10 <donri> by name
00:25:13 <rlpowell> Opinions on the right way to integrate that?  Command line argument?
00:25:48 <stepcut> what do you mean by 'present at' you mean the external hostname/port of the server?
00:26:10 <rlpowell> Yes.
00:26:15 <donri> i think he means the "approot" for web-routes
00:26:30 <rlpowell> "http://vrici.lojban.org:8080" is what I used when I tested it, for example.
00:26:45 <rlpowell> There's not really any way for the app to determine that, in the general case; it has to be told.
00:26:47 <stepcut> i usually put some defaults in the code (like localhost 8000) which can be overridden on the command line
00:27:01 <stepcut> because, as you say, there is no way for the app to know with out being told
00:27:02 <rlpowell> localhost si *guaranteed* not to work; this is for oauth.
00:27:03 <donri> i like to use dyre for app configuration
00:27:14 <stepcut> rlpowell: not oauth, but openid
00:27:26 <donri> and optparse-applicative for command line arguments
00:27:33 <rlpowell> *nod*  Same thing, though, yes?  It has to be accessible by the openid provider.
00:27:38 <rlpowell> WHich I guess you could run one locall.  :P
00:27:48 <stepcut> indeed
00:27:51 <rlpowell> donri: Yeah, but we're trying to *not* add things; this is a bare-bones sample app.
00:27:54 <donri> for controlv it might be best to just use the getopt stuff in the base package
00:28:00 <donri> yea
00:28:13 <stepcut> anyway, command-line options seem like the best approach
00:28:18 <rlpowell> Coo', thanks.
00:28:37 <rlpowell> Should I pull happstack-authenticate, or wait for the other stuff y'all are doing?
00:29:04 <stepcut> well. I dunno :)
00:29:13 <rlpowell> Heh.
00:29:29 <stepcut> in theory you should be able to pull happstack-authentication and then somehow push that patch to donris repo?
00:29:42 <donri> on the bright side, i *think* darcs should be able to commute patches for us nicely here so you can continue to work against the happstack-authenticate repo?
00:29:55 <donri> if not, i can easily re-do the merger
00:30:28 <donri> tldr pretend i haven't done anything yet
00:33:58 <rlpowell> 'k.
00:37:48 <rlpowell> After all that, it looks like whatever changes I made to h-a have already been integrated.  -___-  Sorry.
00:38:58 <rlpowell> Ah, wait, maybe not; diff -r and diff do differente things on directories it turns out.  :)
00:40:06 <rlpowell> Wow, I made a *lot* of changes.
00:41:38 <rlpowell> 400+ lines.
00:42:08 <donri> or maybe you made changes to an earlier version of hs-auth and you're seeing negative diffs for upstream
00:44:00 <rlpowell> donri: Nope, I kept around the version I modified.
00:44:08 <rlpowell> THis is a direct compare to the pre-modified copy.
00:44:09 <donri> ah, good :)
00:44:23 <rlpowell> Aaah, that was why!
00:44:33 <rlpowell> So that the auths you had already authed wouldn't show up in the list.
00:44:45 <rlpowell> The data structures simply weren't preparedfor that; it didn't store enough info.
00:50:23 <rlpowell> stepcut: ^^ Just to be aware, when you're like "holy crap this is a huge patch!", what's going on.
01:03:35 <stepcut> show up in what list?
01:05:20 <rlpowell> stepcut: The list of possible auth methods.
01:05:29 <rlpowell> We actually talked about this back in the day; you agreed it was cool. :)
01:05:41 <stepcut> i did ? :(
01:05:52 <rlpowell> It'll be like "would you like to log in with your google, or facebook, or whatever?", even though you'vea lready logged in with google.
01:06:05 <rlpowell> You did; this was *months* ago.
01:06:09 <stepcut> well.. I have, like, 5 google accounts
01:06:29 <stepcut> was I drinking at the time ?
01:06:34 <rlpowell> *snrk*
01:06:37 <rlpowell> Dunno.
01:07:04 <rlpowell> You would want to log in with more than one?
01:07:30 <stepcut> why not?
01:08:02 <rlpowell> I did'nt know it was possible to be logged in to more than one google account in the same browser.
01:08:24 <stepcut> you have to enable multiple signon
01:08:52 <rlpowell> *nod*
01:08:53 <rlpowell> 'k.
01:08:55 <rlpowell> Well, nevermind then. :P
01:09:43 <stepcut> speaking of drinking.. time to go do some
02:52:37 <kxra> stepcut: heya, how goes!
02:52:46 <kxra> i got disconnected a while ago asking about the blog
02:52:51 <kxra> finally got everything working again
03:33:25 <donri> @tell Palmik probably obvious, but actually ixset doesn't have them. would be nice to have Foldable and Traversable for Store.
03:33:25 <lambdabot> Consider it noted.
04:31:08 <kxra> stepcut: have you considered using Simple Defects as a bug tracker?
06:58:36 <stepkut> kxra: bugs? What bugs?
07:41:18 <kxra> stepcut: hah
08:23:14 <kxra> stepcut: have a prediction on the new clckwrks release?
08:23:19 <kxra> i'm ready to test!
16:57:02 <sm> hey all
16:57:24 <sm> you probably figured out already, but your repo is fine, the proxy error is a darcs hub problem
16:58:45 <sm> the branches page can get slow, and requests are currently limited to 10s each, and snap time out does not always work so we are seeing the ugly apache timeout :(
18:13:53 <stepcut> urgh
18:13:58 <stepcut> as long as the data is ok :)
18:18:14 <donri> seems heffalump said it was ok, but that you might get conflicts. i didn't.
18:32:01 <donri> heh yuiconf: "Douglas Crockford will be giving a special evening keynote entitled Monads & Gonads."
18:42:06 <adnam_> that's humorous
21:07:52 <stepkut> donri: any progress on Generics + SafeCopy?
21:08:17 <donri> nope, i'm pretty much stuck until someone can help me
21:08:31 <stepkut> >:(
21:08:31 <donri> it works, but is rather inefficient
21:09:09 <stepkut> perhaps we should post it somewhere and see if anyone is interested in taking it to the next step?
21:09:19 <donri> it's on my hub