| 02:50 | <MikeSmith> | Hixie: about "should be easier to check questions such as can a p tag contain a div" no ideas jump to mind |
| 02:51 | <MikeSmith> | but, well, the content model is a hyperlink that says "phrasing content" and if you click on that link you get http://www.whatwg.org/specs/web-apps/current-work/multipage/elements.html#phrasing-content |
| 02:52 | <MikeSmith> | which is a complete list of elements that are allowed as phrasing content |
| 02:52 | <MikeSmith> | so not sure how much easier you can make it than you already have |
| 02:53 | <Hixie> | yeah i dunno either |
| 02:54 | <MikeSmith> | Hixie: I guess instead of that hyperlink, the complete list of those elements could be repeated at each place where you have "content model: hyperlink" now. But I don't think most people would think that'd be in improvement. |
| 02:56 | <MikeSmith> | or maybe the problem is that whoever reported that instead wants to have the list of elements that are _not_ allowed. Maybe they don't understand that by not being listed, that means they're not allowed. |
| 02:56 | <MikeSmith> | But if so I don't think most people have that confusion problem. |
| 02:57 | <MikeSmith> | so anyway "p can't contain div" seems pretty clear from what you have now. About as clear as you can practically make it. |
| 02:57 | <Hixie> | yeah maybe |
| 02:58 | <Hixie> | one thing i'm thinking of doing is having popups appear when you hover over a link, essentially telling you what the link points to |
| 02:58 | <Hixie> | no idea how i'm gonna do that exactly |
| 02:58 | <Hixie> | but that could solve this |
| 02:58 | <MikeSmith> | ah yeah, that would be helpful |
| 02:59 | <MikeSmith> | for the content model links, popups with the list of elements for "phrasing content" and "flow content" would be good |
| 03:00 | <MikeSmith> | and also for whatever the other classes of content are that are referenced in the content model sections |
| 03:33 | MikeSmith | wonders what transitions Hixie is talking about |
| 03:43 | <Hixie> | load the multipage version of the spec |
| 03:43 | <Hixie> | or the dom spec |
| 03:48 | <Hixie> | MikeSmith: see them? |
| 04:48 | <MikeSmith> | Hixie: stepped out for a bit, looking now |
| 04:49 | <MikeSmith> | Hixie: page transitions? |
| 04:49 | <MikeSmith> | gradients at the end? |
| 04:56 | <JonathanNeal> | Anyone in here have experience with createShadowRoot? |
| 06:16 | <Hixie> | MikeSmith: move your cursor off the page for a bit |
| 06:17 | <MikeSmith> | Hixie: ok, trying now |
| 06:18 | <MikeSmith> | Hixie: ah the link thing? |
| 06:18 | <Hixie> | not just links, but yeah |
| 06:18 | <MikeSmith> | I thought that was a browser bug :) |
| 06:18 | <MikeSmith> | yeah |
| 06:18 | <MikeSmith> | but now that I know what it is I like it |
| 06:18 | <Hixie> | some people seem to want less colour |
| 06:18 | <Hixie> | so... |
| 06:19 | <MikeSmith> | ah yeah it's very slow transition |
| 06:19 | <Hixie> | personally i use the colour to figure out what's going on |
| 06:19 | <MikeSmith> | Hixie: bingo |
| 06:19 | <JonathanNeal> | These are the 4 “element query” techniques I’ve heard of. Am I missing others? https://gist.github.com/jonathantneal/d4e27ac9a65a426e6870 |
| 06:21 | <MikeSmith> | Hixie: yeah the color is there for genuine functional purposes. I think the people who don't like are people who don't actually use the spec, or don't need to. And for them it just hurts their purist aesthetic sensibilities about what a good document should look like in general, I guess |
| 06:21 | <Hixie> | yeah |
| 06:21 | <Hixie> | i also added a lot of whitespace, which i think works ok |
| 06:22 | <MikeSmith> | Hixie: yeah I think we can all agree that more whitespace is better |
| 06:22 | <Hixie> | within reason |
| 06:22 | <MikeSmith> | sure yeah |
| 06:23 | <Hixie> | it did add about 50 pages to the pdf version... |
| 06:25 | <MikeSmith> | Hixie: well anybody who's nuts enough to try to print out the whole PDF doesn't have anybody to blame but themselves |
| 06:26 | <Hixie> | hah |
| 06:32 | <MikeSmith> | hmm wchen adding Gecko support for <template> in application/xhtml+xml |
| 06:33 | <MikeSmith> | I wonder if other engines are doing that too |
| 06:34 | <MikeSmith> | I can't remember what the decision about that was... I vaguely recall it being "we don't need to make <template> work for XML" being what everybody agreed to at one point at least |
| 06:37 | <MikeSmith> | Hixie: oh I see you do actually cover it now in the "Parsing XHTML documents" section |
| 06:39 | <MikeSmith> | Hixie: action item: XML Core WG to add <template>-parsing steps to the errata for the stable TR draft of XML 1.0 Fifth Edition |
| 06:41 | Hixie | glares at MikeSmith |
| 07:41 | MikeSmith | stumbles across https://developer.mozilla.org/en-US/docs/Web/API/Navigator.requestWakeLock |
| 09:17 | <IZh> | Hi. It seems that something is changed in CSS file http://www.whatwg.org/style/specification. The prince PDF generator complains: warning: failed to parse all CSS rules. |
| 09:28 | <IZh> | Currently PDF document generation is aborted at any error or warning. I prefer to generate clean documents. ;-) But I can ignore this warning. |
| 09:30 | <IZh> | By the way, what CSS validator you are using? I tried http://www.css-validator.org/, and it complains with lots of warnings and errors. |
| 09:40 | <MikeSmith> | IZh: the CSS validator has some known issues. there are some CSS3 things it doesn't recognize. e.g., content leader |
| 09:41 | <IZh> | MikeSmith: Is there any other validators that can parse CSS3 correctly? |
| 09:44 | <MikeSmith> | IZh: maybe, but none that I know of. Myself I don't usually bother to validate my CSS unless I'm forced to |
| 10:01 | <IZh> | Hixie: In css rules for h1, h2, h3, h4, h5, h6 on line 44 "optimiseLegibility" -> "optimizeLegibility". |
| 10:03 | <IZh> | Also the block "DELETE ME" at the end is incorrectly commented because of nested comments that will produce parse errors. |
| 13:30 | <MikeSmith> | IZh: probably best to file bugs for those |
| 13:37 | <IZh> | MikeSmith: I've sent a letter to Hixie. By the way, I have bought one commercial validator, so soon there will be more bug reports. ;-) |
| 14:06 | <JonathanNeal> | Curious, if external SVGs can have media queries, does that mean sticking a lot of SVGs on the page is as taxing on the browser as sticking a lot of iframes on the page? |
| 15:11 | <annevk> | JonathanNeal: depends on how you embed them |
| 15:54 | <JonathanNeal> | annevk: is there a better way to embed them where they still respond to embedded media queries? |
| 16:47 | <JonathanNeal> | annevk: i was using <object> |
| 16:51 | <JonathanNeal> | Which is the defacto method I think people are using. |
| 18:24 | <gsnedders> | How is one meant to decide what URL schemes are safe to allow through a sanitizer? |
| 18:24 | <gsnedders> | Because, ergh. |
| 18:34 | <gsnedders> | Also, I hate the html5lib sanitizer. |
| 19:24 | <gsnedders> | jgraham: html5lib/html5lib-python#151 should be quick/easy to deal with, plz. |
| 19:40 | <IZh> | On main page of whatwg.org "lang" attribute is missing on <html>. |
| 19:47 | <tantek> | YAGNI |
| 20:55 | <gsnedders> | Wait, what? html5lib-tests includes tests for lone surrogates in the tokenizer tests, but not in the tree-construction tests? Um, okay. |
| 21:49 | <Hixie> | JonathanNeal: <object> for svg is in fact exactly equivalent to <iframe> under the hood, iirc |
| 21:50 | <JonathanNeal> | Hixie: KNEW IT! Thanks a ton. |
| 21:51 | <JonathanNeal> | Okay Hixie, the followup is: is it taxing on a page to have a bunch of SVGs loaded with <object>? |
| 21:51 | <Hixie> | no more so than having a bunch of <iframe>s |
| 21:51 | <JonathanNeal> | And is that taxing? |
| 21:51 | <Hixie> | more so than having inline svg, probably, though |
| 21:51 | <JonathanNeal> | (when I said “KNEW IT” I meant in the sense of “ah good, the data suggested this”) |
| 21:52 | <Hixie> | iframes are as taxing as a new tab, basically. well, not quite as much. |
| 21:52 | <Hixie> | i would treat a Document as expensive, in theory |
| 21:52 | <Hixie> | in practice... |
| 21:52 | <Hixie> | what do i know |
| 21:53 | <Hixie> | the doc i publish is 1000+ pages |
| 21:53 | <JonathanNeal> | A document or a tab is pretty expensive, no? Or are they as expensive as createDocumentFragment? |
| 21:53 | <Hixie> | document fragments are (hopefully) way cheaper than whole documents, which are way cheaper than document+window pairs |
| 21:54 | <JonathanNeal> | Is <object> cheaper than document+window? Same with <iframe>? |
| 21:54 | <Hixie> | object or iframe = document+window |
| 21:55 | <JonathanNeal> | ouch. |
| 21:55 | <Hixie> | i should say, object when rendered using a nested browsing context = iframe = document+window |
| 21:56 | <caitp> | do any user agents actually have a concept of browsing contexts as they are in the web apps specs |
| 21:56 | <JonathanNeal> | The <object data="path/to/some.svg"> method was recommended for svgs + art direction. I never thought about them being mini iframes scattered throughtout the page. |
| 21:57 | <Hixie> | caitp: the spec is written to match what browsers do, so, in theory, at least, yes... i'm sure there's differences though |
| 21:58 | <caitp> | I saw a bug on gecko that involved a bunch of hacks to implement a subset of it as specified, and as far as I can tell there's no proper analog in blink, so I'm not sure if that's actually the case |
| 21:58 | <caitp> | unless opera/webkit/ie are doing it all |
| 21:59 | <Hixie> | how does what the browsers have differ from the spec? (and could you phrase your answer in the form of a bug report at http://whatwg.org/newbug :-) ) |
| 22:00 | <caitp> | well, it's not somethig that's necessarily worth fussing about, I think it's not unmanageable, it's just that none of it is really organized in an easy way to simplify doing browsing context work |
| 22:01 | <caitp> | like you have a bunch of things that do part of it, but not all of it |
| 22:02 | <JonathanNeal> | Hixie: would you say technique #4 is the most expensive? https://gist.github.com/jonathantneal/d4e27ac9a65a426e6870 |
| 22:02 | <JonathanNeal> | re: our conversation about <iframe>. |
| 22:58 | <Hixie> | JonathanNeal: i have no idea. measure it. :-) |
| 22:59 | <JonathanNeal> | You kind, terrible, wise person. |
| 23:04 | <gsnedders> | In totally unrelated news, trying to decide what laptop to buy is absolutely aggrevating. |
| 23:13 | <JonathanNeal> | gsnedders: can’t decide which macbook is best? |
| 23:14 | <gsnedders> | I really just want to get something not that expensive to use as a desktop replacement while travelling till I graduate |
| 23:14 | <gsnedders> | And the MBPs fail the "not too expensive" part :) |
| 23:15 | <tantek> | gsnedders, MBA11 fully loaded. |
| 23:15 | <gsnedders> | I really want something with a bigger screen. Like, I literally am only going to be using it while at my parents and while on holiday, pretty much |
| 23:16 | <gsnedders> | Because around uni I typically just use tablet or computers in the lab… and I scarcely have much in way of reason to even be around campus next year |
| 23:20 | <gsnedders> | I mean, I'm still using an ancient MBP 17". But it's still mostly fine. Could do with a bit more RAM, an SSD, and a faster processor… But for how much I use it, it is basically fine… |
| 23:21 | <gsnedders> | I could go through the horrible process of replacing the HDD with an SSD, which is pretty hellish as it's not a user-replacable part. |
| 23:22 | <Hixie> | what do you use a laptop for? |
| 23:26 | <JonathanNeal> | gsnedders: 8GB of RAM => $40, 200GB SSHD => $110. |
| 23:26 | <gsnedders> | A mixture of doing random stuff on the web and email (and, ergh, Flash on OS X's performance is bad enough 1080p video on YouTube doesn't play back entirely smoothly…) and whatever assignments uni throws at me, mostly. It's the video playback and compile/test times that make me want to upgrade it, really. |
| 23:26 | <gsnedders> | JonathanNeal: Chipset only support 6GB of RAM. |
| 23:27 | <JonathanNeal> | I did that to my wife’s 2009 MBP and it sings. Of course, back then $110 couldn’t get you 200GB of solid state. |
| 23:27 | <gsnedders> | JonathanNeal: And putting an SSD in it is *hard*. |
| 23:27 | <gsnedders> | JonathanNeal: This is a mid-07 MBP :) |
| 23:27 | <JonathanNeal> | Yea, before the body change. |
| 23:27 | <gsnedders> | Indeed. The Unibodies have user-accessible drives, don't they? |
| 23:43 | <Hixie> | gsnedders: sounds like a chromebook would be sufficient |
| 23:44 | <gsnedders> | Hixie: That's fine except for the doing random uni assignments bit. Which next year is basically going to be my project. It's nice being able to have spec open next to code. :) |
| 23:45 | <Hixie> | install ubuntu on a chromebook? :-) |
| 23:45 | <Hixie> | (https://github.com/dnschneid/crouton) |
| 23:46 | <gsnedders> | I more meant having the screen space to do so :) |
| 23:46 | <Hixie> | you need bigger than 12"? |
| 23:47 | <Hixie> | make that 14" |
| 23:48 | <Hixie> | ( http://www.google.com/intl/en/chrome/devices/chromebooks.html#hp-cb-14 ) |
| 23:48 | <gsnedders> | For something I'm using as a work computer, yeah, I'd rather. But I'm probably just silly from having a 30" monitor at home :) |
| 23:49 | <Hixie> | if you want a laptop with bigger than 14" screen, then yeah, mbp. |
| 23:49 | <Hixie> | or just get an external monitor... |
| 23:51 | <gsnedders> | mhmm, yeah, maybe that would be the best idea |
| 23:51 | <Huvet> | Hi, I'm using version 0.999 of html5lib for python and get this error: |
| 23:51 | <Huvet> | python -c "import html5lib; html5lib.parse('', treebuilder='lxml')" |
| 23:51 | <Huvet> | ValueError: All strings must be XML compatible: Unicode or ASCII, no NULL bytes or control characters |
| 23:51 | <gsnedders> | Yes, known bug. |
| 23:52 | <Huvet> | is this a bug I should report or is this known and fixed? |
| 23:52 | <gsnedders> | Neither. Known and unfixed. :) |
| 23:52 | <gsnedders> | "Soon". |
| 23:52 | <Huvet> | thanks, will hack around then, all entities below a certain number? |
| 23:53 | <gsnedders> | See ihatexml |
| 23:53 | <Huvet> | if you want it seen in the wild it's on this page: https://celebrity.yahoo.com/the-insider/watch-zach-galifianakis-interview-bradley-224209765.html |
| 23:53 | <gsnedders> | And it's not just entities, lxml never sees it as an entity |
| 23:53 | <Huvet> | ok! |
| 23:54 | <gsnedders> | Like, "\x03" would trigger the exact same case |
| 23:55 | <gsnedders> | Huvet: Essentially, we should be checking it matches the Char production in XML 1.0 (Fourth Edition) |
| 23:56 | <Huvet> | that's way over my head, but I feel happy that you know how to fix it :) |
| 23:57 | <Huvet> | I'll just regexp it away and be happy that stuff doesn't break |
| 23:57 | <Huvet> | I'm a simple person like that |
| 23:57 | <gsnedders> | Also pretty much all that code is broken for anything outside of the BMP (i.e., anything above U+FFFF). So I was rather planning on rewriting it all. |
| 23:57 | <Hixie> | presumably another tree builder would fix it? |
| 23:57 | <gsnedders> | Also what Hixie just said :) |
| 23:57 | <Huvet> | yes, not using lxml threebuilder works |
| 23:58 | <Huvet> | but I'm using cssselect (https://pypi.python.org/pypi/cssselect) and it depends on lxml |