| 08:44 | <Ms2ger`> | krijn, down again? |
| 11:19 | <krijnh> | Ms2ger`: am I/ |
| 11:19 | <krijnh> | Ah |
| 11:21 | <krijnh> | And back up! |
| 11:23 | <Ms2ger`> | Yep, thanks |
| 16:46 | <dglazkov> | good morning, Whatwg! |
| 17:44 | <jgraham> | Is stringification of HTML Elements actually specified these days? I suppose it is… |
| 17:44 | <jgraham> | Anyway I am all in favour of deleting these Microsoft tests since they are so hopelessly inefficient |
| 18:05 | <gsnedders> | jgraham: Which stringification? |
| 18:18 | <jgraham> | [object HTMLElement] |
| 18:24 | <gsnedders> | jgraham: That's been spec'd for ages by virtue of WebIDL, no? |
| 18:25 | <gsnedders> | jgraham: Also, when you have time, look through the open html5lib-python PRs? |
| 18:27 | <jgraham> | gsnedders: Probably |
| 18:28 | <jgraham> | gsnedders: I started looking at the PR about using setuptools, but I wasn't sure why it's an improvement |
| 18:32 | <gsnedders> | jgraham: It allows you to do stuff like pip install html5lib[test] and get html5lib and all test dependencies installed |
| 18:33 | <gsnedders> | I'm perfectly fine with it in principle, just as my review comments say, there's plenty wrong with the patch… |
| 18:33 | <gsnedders> | jgraham: (I'm esp. interested in PRs I'm the author of and cannot review myself, inevitably.) |
| 18:34 | <gsnedders> | jgraham: #95, and opinions on what to do with html5lib-tests, more than anything |
| 18:34 | <gsnedders> | I mean, I could just make a unilateral decision here. :) |
| 18:35 | <jgraham> | gsnedders: Oh, I didn't see that you had some comments already |
| 18:36 | <gsnedders> | jgraham: On GitHub, given Critic is broken from force pushed commits. |
| 18:37 | <gsnedders> | jgraham: I wonder if it's worthwhile dealing with force pushed commits by adding in a merge (with theirs strategy) commit to keep it identical (in terms of content, if not in terms of commit log). |
| 18:38 | <jgraham> | gsnedders: Critic "supports" rebasing forced pushes now. Although it doesn't really do anything useful in this case |
| 18:38 | <gsnedders> | jgraham: Oh, my comments are attached to the commit that used to be the top of the branch |
| 18:39 | <gsnedders> | jgraham: Does Critic send emails yet if they haven't logged in there? |
| 18:39 | <gsnedders> | jgraham: Because if not, it's of limited use. :| |
| 18:39 | <jgraham> | gsnedders: So I can't actually see them from the github UI without hunting for that commit? |
| 18:39 | <gsnedders> | jgraham: Yup! |
| 18:39 | <gsnedders> | The one thing that hasn't been dealt with is splitting the commit into two commits. |
| 18:39 | <jgraham> | gsnedders: No, critic only emails people that actually have accounts |
| 18:40 | <jgraham> | Which means logging in |
| 18:40 | <gsnedders> | jgraham: Which means lots of contributors never get notified about review comments. Which is /really/ annoying. |
| 18:40 | <jgraham> | "lots of contributers" |
| 18:40 | <jgraham> | lol |
| 18:40 | <gsnedders> | jgraham: There's a surprising number of PRs which come from people who come and give one, then are never seen again. |
| 18:40 | <jgraham> | I typically add a "comments are on critic" comment |
| 18:42 | <jgraham> | Anyway the github thing of hiding all comments when someone does a history rewrite, and encouraging people to do history rewrites is more annoying |
| 18:45 | <gsnedders> | jgraham: So how does Critic deal with rebasing stuff? |
| 18:49 | <jgraham> | gsnedders: Well I thought that given an old head H, a new head H' and a common base B it would construct the diffs B-H and B-H' and then diff those diffs to work out what changes had been introduced by the rebase. That would be represented as an equivalent merge commit. |
| 18:51 | <jgraham> | Which I guess is what it's doing, but it doesn't work too well if there is a move and a squash, because the changes from the move become part of the merge commit |
| 18:51 | <jgraham> | Hard to know how to deal with that really |
| 18:55 | <gsnedders> | With #95, the big question is what to do with tests, really |
| 19:08 | <JonathanNeal> | Where is a good place to ask people about DOM Promises. |
| 19:09 | <Domenic_> | JonathanNeal: nowhere, because DOM Promises do not exist ;). It's ES6 promises now. |
| 19:09 | <Domenic_> | #promises is a good channel though |
| 19:11 | <JonathanNeal> | Oh, good to know. |
| 19:11 | <JonathanNeal> | Thanks Domenic_. |
| 19:23 | <Domenic_> | Anyone know how to move pages in MDN? |
| 20:06 | <SimonSapin> | Domenic_: IIRC that’s a missing feature in MDN |
| 20:07 | <SimonSapin> | try #mdn on irc.mozilla.org |
| 20:09 | <ondras> | Domenic_: ? |
| 20:10 | <Domenic_> | SimonSapin: I found how to create redirects, which seems good enough |
| 20:10 | <Domenic_> | ondras: ??? |
| 20:10 | <ondras> | Domenic_: I have a promises-related question |
| 20:10 | <ondras> | Domenic_: is there some common nomenclature for the case when you want to "glue" two promises together? |
| 20:10 | <Domenic_> | What does that mean? |
| 20:11 | <ondras> | Domenic_: as in, you have P1 and P2 and want to chain them as in "P1.then(P2)" |
| 20:11 | <ondras> | or, more specifically |
| 20:11 | <ondras> | P1.then(P2.fulfill.bind(P2), P2.reject.bind(P2)) |
| 20:11 | <ondras> | Domenic_: I would like to know whether there is some "standardized" method to do so |
| 20:11 | <ondras> | as in, P1.chain(P2) |
| 20:11 | <ondras> | or something. |
| 20:11 | <Domenic_> | P1.then(() => P2) |
| 20:12 | <ondras> | yes. but without the explicit anonymous function :) |
| 20:12 | <Domenic_> | well, nope. Some libraries like Q have P1.thenResolve(P2) |
| 20:12 | <ondras> | (which happens to be a bit more verbose in JS interpreters without arrow notation :-)) |
| 20:13 | <ondras> | Domenic_: okay. I will probably make a method for my impl, was just looking for a suitable name |
| 20:13 | <ondras> | I happen to run into this use case rather frequently... |
| 22:01 | <anon__> | What's the best way to resize a site with CSS? Like I've got a column of content that's been positioned absolutely but if it extends past the page you can't scroll too it |
| 22:16 | <hauleth> | hi |