Options:
- # Session Start: Mon Feb 27 00:00:01 2012
- # Session Ident: #css
- # [01:03] * Joins: arno (arno@208.87.61.217)
- # [01:10] * Quits: arno (arno@208.87.61.217) (Quit: Leaving.)
- # [01:11] * Joins: jdaggett_ (jdaggett@202.221.217.73)
- # [01:15] * Joins: bradley (qw3birc@128.30.52.28)
- # [01:15] <bradley> hello
- # [01:26] * Quits: drublic (drublic@84.174.27.22) (Client exited)
- # [01:30] * Quits: bradley (qw3birc@128.30.52.28) (Quit: Page closed)
- # [02:21] * Joins: tantek (tantek@70.36.139.112)
- # [02:53] * Joins: miketaylr (miketaylr@68.203.0.108)
- # [03:44] * Quits: tantek (tantek@70.36.139.112) (Quit: tantek)
- # [03:59] * Quits: ed (ed@88.131.66.80) (Ping timeout)
- # [04:00] * Joins: ed (ed@88.131.66.80)
- # [05:30] * Joins: tantek (tantek@70.36.139.112)
- # [05:56] * Quits: miketaylr (miketaylr@68.203.0.108) (Quit: Leaving...)
- # [07:46] * Quits: nimbu (Adium@24.18.47.160) (Quit: Leaving.)
- # [09:07] * Quits: jdaggett_ (jdaggett@202.221.217.73) (Quit: jdaggett_)
- # [09:50] * Joins: drublic (drublic@80.152.180.137)
- # [10:10] * Quits: drublic (drublic@80.152.180.137) (Client exited)
- # [12:07] * Joins: drublic (drublic@80.152.180.137)
- # [12:45] * Quits: drublic (drublic@80.152.180.137) (Client exited)
- # [13:07] * Quits: jdaggett (jdaggett@180.235.8.216) (Quit: jdaggett)
- # [13:49] * Joins: drublic (drublic@80.152.180.137)
- # [14:21] * Joins: karl (karlcow@128.30.54.58)
- # [14:37] * Joins: florian (yaaic@84.215.51.223)
- # [14:37] * Quits: florian (yaaic@84.215.51.223) (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org)
- # [14:37] * Joins: florian (yaaic@84.215.51.223)
- # [14:37] * Quits: florian (yaaic@84.215.51.223) (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org)
- # [15:50] * Joins: nimbu (Adium@24.18.47.160)
- # [15:54] * Joins: miketaylr (miketaylr@68.203.0.108)
- # [16:04] * Quits: miketaylr (miketaylr@68.203.0.108) (Quit: Leaving...)
- # [16:09] * Joins: miketaylr (miketaylr@68.203.0.108)
- # [16:19] * Joins: danielfilho (danielfilh@187.31.77.7)
- # [16:40] * Joins: ksweeney (ksweeney@63.119.10.10)
- # [17:12] * Quits: drublic (drublic@80.152.180.137) (Connection reset by peer)
- # [17:49] * Joins: arno (arno@192.150.10.200)
- # [18:23] * Joins: Ms2ger (Ms2ger@91.181.60.137)
- # [18:39] * Quits: arno (arno@192.150.10.200) (Quit: Leaving.)
- # [19:19] * Quits: nimbu (Adium@24.18.47.160) (Quit: Leaving.)
- # [19:23] * Quits: SimonSapin (simon@82.232.219.95) (Ping timeout)
- # [19:37] * Quits: miketaylr (miketaylr@68.203.0.108) (Quit: dflk;adfslkj;alsiekfj;laiskdf)
- # [19:37] * Joins: jet (jet@159.63.23.38)
- # [19:38] * Joins: miketaylr (miketaylr@68.203.0.108)
- # [19:44] * Joins: nimbu (Adium@67.138.164.194)
- # [19:49] * Quits: tantek (tantek@70.36.139.112) (Quit: tantek)
- # [20:07] * Quits: jet (jet@159.63.23.38) (Quit: jet)
- # [20:37] * Joins: drublic (drublic@93.132.225.15)
- # [20:49] * miketaylr is now known as miketaylr|
- # [20:49] * miketaylr| is now known as miketaylr||
- # [20:49] * miketaylr|| is now known as miketaylr|||
- # [21:04] * Joins: jet (jet@159.63.23.38)
- # [21:13] * Quits: nimbu (Adium@67.138.164.194) (Quit: Leaving.)
- # [21:25] * Quits: Ms2ger (Ms2ger@91.181.60.137) (Quit: nn)
- # [21:44] * Quits: TabAtkins (tabatkins@216.239.45.4) (Quit: Lost terminal)
- # [21:52] * Quits: karl (karlcow@128.30.54.58) (Quit: This computer has gone to sleep)
- # [21:57] * Joins: TabAtkins_ (tabatkins@216.239.45.4)
- # [23:17] <fantasai> plinss: Is there a way to push changes to /sites/csswg.org/test without authenticating against every single subrepository?
- # [23:18] <plinss> every single? there should only be the one...
- # [23:19] <plinss> but, yes, store your login credentials in a keyring
- # [23:19] <fantasai> I cloned
- # [23:19] <fantasai> https://hg.csswg.org/site/base
- # [23:19] <fantasai> and that required entering my password about 10 times
- # [23:19] <plinss> that's not the test repo
- # [23:19] <fantasai> no, that's the site repo
- # [23:20] <plinss> ah, missed your original path...
- # [23:20] <fantasai> which is what I need to modify
- # [23:20] <fantasai> so I tried to push a changeset from csswg.org/test
- # [23:20] <fantasai> but it seems to be asking me to authenticate against all 10 (or whatever) subrepos
- # [23:21] <fantasai> Is that necessary? Is there a way to just push to the one I'm modifying?
- # [23:21] <plinss> right, it'll do that.
- # [23:21] <plinss> the site repo is the root that contains all the subrepos for the installed apps
- # [23:21] <plinss> when you push to it, it'll auth against all the sub repos
- # [23:21] <fantasai> I see.
- # [23:22] <plinss> http://mercurial.selenic.com/wiki/KeyringExtension
- # [23:22] * Joins: arronei_ (arronei@131.107.0.94)
- # [23:22] <plinss> install the keyring and set your login name in your ~/.hgrc file
- # [23:22] <plinss> it'll never ask for a password again...
- # [23:23] * Joins: nimbu (Adium@24.18.47.160)
- # [23:24] * Quits: arronei (arronei@131.107.0.126) (Ping timeout)
- # [23:26] <plinss> fwiw, if you're making changes in one of the subrepos, you can just push there...
- # [23:26] * fantasai needs to modify htdocs
- # [23:27] <plinss> the index.html page I presume?
- # [23:27] * fantasai gets a parse error and gives up, puts this task in the "try again once the OS has been upgraded" bucket
- # [23:27] <fantasai> plinss: no, adding resources/testharnes.js
- # [23:27] <fantasai> plinss: http://lists.w3.org/Archives/Public/public-css-testsuite/2012Feb/0070.html
- # [23:28] <plinss> ah, right, on my todo list
- # [23:29] <plinss> that should probably be in a subrepo so it can rev independently of the entire site (and we can give other people access to it without giving them the whole web server)
- # [23:29] <plinss> I can set that up after lunch today if you want
- # [23:30] <fantasai> yay, it works!
- # [23:30] <fantasai> plinss: Up to you. Doesn't affect me, since I have access to the whole repo anyway :)
- # [23:31] <fantasai> plinss: we should spend some time sorting out the spec Mercurial migration, though
- # [23:31] <plinss> ok, I'll move it later
- # [23:31] <fantasai> plinss: not today, I told myself I'd work on testing today
- # [23:31] * fantasai has been putting this off
- # [23:31] <plinss> yes, that was what I was going to work on after lunch
- # [23:32] <plinss> I have a clone of the CVS repo and a conversion to hg on my machine
- # [23:32] <fantasai> plinss: did you check with sysreq about the checkout / hooks?
- # [23:32] <plinss> I also have the repo setup on csswg.org
- # [23:32] <plinss> yes, Mike can set up a hook to push to our mirror
- # [23:32] <fantasai> nice
- # [23:32] <fantasai> hmm
- # [23:32] <plinss> drafts.csswg.org is up too
- # [23:33] <fantasai> I wonder if we can check out into a checkout of the CVS repo
- # [23:33] <fantasai> and then push to dev.w3.org
- # [23:33] <plinss> I'm just looking at getting the CVS $date$ substitution working with hg
- # [23:33] <fantasai> Hm. I don't think that's too important.
- # [23:34] <fantasai> plinss: Given we auto-generate the dates now, we can just strip that from any specs that use it
- # [23:34] <fantasai> plinss: Bert's preprocessor will insert it into the generated spec
- # [23:34] <plinss> ok. it seemed to be all over the place
- # [23:34] <plinss> I just wanted functional paroty...
- # [23:34] <fantasai> plinss: heh. I don't think it's worth the trouble here :)
- # [23:34] <plinss> ok
- # [23:35] <plinss> I can always add it later
- # [23:35] <fantasai> plinss: Anything else you're concerned about?
- # [23:35] <plinss> (if someone screams)
- # [23:35] <plinss> not really, I was going to give my copy another once over, push it to our mirror, then ask you to double check it
- # [23:36] * Quits: drublic (drublic@93.132.225.15) (Connection reset by peer)
- # [23:36] <plinss> once you give the nod I'll push to dvcs.w3.org and we can shut down the CVS repo
- # [23:36] * Joins: drublic (drublic@93.132.225.15)
- # [23:36] <plinss> if you have time later I'll ping you after I push to the mirror
- # [23:37] <plinss> I just need to get some lunch first, learned the hard way not to make important system changes on low blood sugar :-)
- # [23:37] <fantasai> plinss: Ok. We also need to be ready to set up redirects and delete the files from CVS. So we should do the actual migration during a pre-announced period.
- # [23:37] <plinss> yes
- # [23:37] * Parts: ksweeney (ksweeney@63.119.10.10)
- # [23:37] <fantasai> plinss: maybe Thursday
- # [23:37] <fantasai> plinss: Oh, but we also need to make sure the Mercurial docs are all set up!
- # [23:37] <plinss> that long?
- # [23:38] <fantasai> plinss: And that Tantek's QAed the process of setting up hg and checking in a change
- # [23:38] <fantasai> plinss: Including one where we trigger a conflict.
- # [23:38] <plinss> I think he has hg set up and the docs (for OSX at least) for that
- # [23:39] <fantasai> plinss: I'm less concerned about installing mercurial than using it in a real environment
- # [23:39] <plinss> sure
- # [23:39] <fantasai> plinss: where other people are checking into the repo
- # [23:39] <fantasai> plinss: between your pushes
- # [23:40] <plinss> you think we need more than is at: http://wiki.csswg.org/tools/hg ?
- # [23:40] <fantasai> plinss: yes
- # [23:40] <fantasai> plinss: There's a mention of all the relevant commands, but there isn't a good explanation of workflow
- # [23:40] <fantasai> plinss: And rebase isn't covered
- # [23:41] <fantasai> plinss: The docs need to be workflow-oriented, or they won't make very much sense.
- # [23:41] <plinss> ok
- # [23:41] <plinss> I'll work on that later today
- # [23:41] <fantasai> plinss: Like, it should say
- # [23:41] <fantasai> "Getting Started" - everything about setting up, i.e. link to install docs, config files, and cloning
- # [23:41] <fantasai> "Making a Change"
- # [23:42] <fantasai> updating your tree, making the change, checking the diff, pushing it to the repo, resolving conflicts
- # [23:43] <fantasai> that's just a rough idea..
- # [23:43] <fantasai> but there should basically be two sets of commends to run: stuff before I make edits, and stuff after I make edits
- # [23:43] <fantasai> if the stuff happens to be 5 commands, bunch them into one series. Handle exceptions (stuff going wrong) after that.
- # [23:44] * Joins: karl (karlcow@128.30.54.58)
- # [23:44] <fantasai> Splitting out the commands and learning more about what they do is Advanced Topics
- # [23:44] <fantasai> The basic thing is
- # [23:45] <fantasai> "How do I update my tree" and "How do I push my changes"
- # [23:45] <fantasai> Those two need to be dead simple.
- # [23:45] <plinss> I don't want to get into advanced topics in our docs of hg, there's plenty of that around
- # [23:45] <plinss> we should just focus on basics of using our repo
- # [23:46] <fantasai> I consider 'hg incoming' to be an Advanced Topic
- # [23:46] <fantasai> and that's on the wiki right now
- # [23:46] <fantasai> I also consider 'hg revert' to be an Advanced Topic
- # [23:46] <plinss> ok
- # [23:47] <fantasai> basic topics are "How do I update my tree", "How do I review my changes", and "How do I check in my changes"
- # [23:47] <fantasai> Anything *you* consider and advanced topic probably doesn't belong on the wiki at all :)
- # [23:47] <plinss> :-P
- # [23:48] <plinss> how many spec editors do we really have that don't get the basics of VCS?
- # [23:48] <fantasai> Things like offline commits, hg revert, and other handy tools should be in a separate section.
- # [23:48] <fantasai> from Basic Topics
- # [23:48] <fantasai> plinss: People understand the concept of a VCS. That's different from learning how to use it.
- # [23:49] <fantasai> s/it/a specific one/
- # [23:49] <fantasai> plinss: Also, a significant number of our spec editors don't use the command line
- # [23:49] <plinss> really?
- # [23:49] <fantasai> yes really
- # [23:49] <plinss> scary
- # [23:49] <fantasai> like, all the ones at Microsoft IIRC
- # [23:49] <plinss> they have TortoiseHG...
- # [23:50] <fantasai> That's not the command line, now, is it :)
- # [23:50] <plinss> same operations, just point and click
- # [23:50] <fantasai> My point stands.
- # [23:50] <plinss> ok
- # [23:52] <plinss> while I'm not opposed to good documentation, part of me wonders if we're over thinking this, I'd think spec editors would be more DVCS savvy than test authors, and what we had for the test repo was OK then...
- # [23:55] <plinss> IMO there's enough documentation explaining Mercurial already on the web, our docs should only focus on the specific issues of working with our repo and should point to other existing docs
- # [23:56] * fantasai looks on the test wiki for instructions on using HG and fails to find them... :/
- # [23:56] <fantasai> plinss: We had a guide to migrating from Subversion.
- # [23:56] <plinss> http://wiki.csswg.org/tools/hg#more-information
- # [23:57] <fantasai> plinss: Too much detail
- # [23:58] <fantasai> plinss: We need a 2-screenful "this is how you make an edit". Otherwise it's frustrating.
- # [23:58] * Joins: jdaggett (jdaggett@180.235.8.216)
- # [23:59] <plinss> I hear you, and I'll work on it this afternoon. I just don't want to go crazy with hg docs.
- # Session Close: Tue Feb 28 00:00:00 2012
The end :)