| 05:09 | <heycam> | when CSS folks come up with properties that take lengths and percentages, what guidelines do you use for determine whether to have the computed value be the equivalent absolute value, or whether to instead leave the computed value as specified? |
| 05:10 | <tantek> | heycam, I think you meant to ask that in irc://irc.w3.org:6665/css ;) |
| 05:10 | <heycam> | tantek, good idea :) |
| 05:10 | <tantek> | :) |
| 10:37 | <jgraham> | heycam|away: So I should assume that SVG scripts are supposed to fire all the same events as HTML scripts, for example? |
| 10:38 | <Ms2ger> | Or that <script for event> has the same result? |
| 10:38 | <Ms2ger> | (Hint: it doesn't) |
| 10:46 | <jgraham> | Well I hadn't got as far as thinking about that particular unpleasantness yet |
| 11:00 | <Ms2ger> | hober, http://i.qkme.me/3pcrk0.jpg |
| 17:50 | <annevk> | <picture> was first mentioned on the WHATWG list by hsivonen |
| 17:50 | <annevk> | #randomfacts |
| 17:50 | <annevk> | http://lists.w3.org/Archives/Public/public-whatwg-archive/2012Jan/0093.html |
| 17:51 | <annevk> | (found by a search for "<picture>", similar elements have almost certainly been proposed in the past) |
| 17:53 | <miketaylr> | annevk: beat you, http://lists.w3.org/Archives/Public/wai-xtech/2007Jun/0080.html |
| 17:53 | <miketaylr> | oh not whatwg... :( |
| 17:53 | miketaylr | quits |
| 17:54 | <annevk> | miketaylr: heh nice |
| 18:16 | <WeirdAl> | annevk, Ms2ger: question on DOM4 spec about attribute list... is there a defined ordering to how these attributes are stored? |
| 18:16 | <Ms2ger> | Yes |
| 18:17 | <WeirdAl> | Where can I find that? |
| 18:17 | <Ms2ger> | Hixie refuses to spec what it is for parser-inserted elements, though |
| 18:17 | <WeirdAl> | heh |
| 18:19 | <WeirdAl> | well, I'm thinking of the case: var x = document.documentElement; x.setAttributeNS("http://www.w3.org/xlink";, "xlink:href", "#foo"); x.setAttribute("lang", "en"); x.attributes[0] // what should that return? |
| 18:19 | <annevk> | Ms2ger: I talked to Hixie about that, what he needs is setting a bunch of attributes at once |
| 18:20 | <annevk> | WeirdAl: xlink:href I think |
| 18:20 | <Ms2ger> | Yes, first-set |
| 18:21 | <WeirdAl> | again, where's that specified? :) |
| 18:21 | <annevk> | WeirdAl: setAttriubte just appends to the list |
| 18:21 | <annevk> | WeirdAl: and append means to the end of the list |
| 18:22 | <annevk> | Ms2ger: in particular for when you have an element that triggers some kind of algorithm on an attribute being set or on being inserted |
| 18:22 | WeirdAl | found it |
| 18:22 | <annevk> | Ms2ger: all the attributes associated with it need to be available right away and therefore setAttribute cannot be used |
| 18:23 | <Ms2ger> | That's his claim, yes |
| 18:23 | <Ms2ger> | Gecko manages to implement that with sequential setAttributes, though |
| 18:24 | <annevk> | interesting |
| 18:24 | <annevk> | in what order does it call setAttribute? |
| 18:25 | <WeirdAl> | hmm, this is going to make things interesting for my own DOM design. |
| 18:25 | <Ms2ger> | Still reverse source order, IIRC |
| 18:26 | <Ms2ger> | And the element is told when the parser is done adding attributes |
| 18:28 | <annevk> | Ms2ger: aaah |
| 18:30 | WeirdAl | frowns: DOM4 is not a nice spec to implement |
| 18:31 | <Ms2ger> | The web is not a nice thing to spec |
| 18:31 | <WeirdAl> | true |
| 18:32 | <WeirdAl> | and besides, I'm only going for limited DOM4 support in my implementation, so maybe this is one of those things I just don't need to be that precise on right away |
| 18:34 | <annevk> | WeirdAl: is it just the somewhat weird requirements or does the spec read weird? |
| 18:34 | <WeirdAl> | neither, really |
| 18:35 | <annevk> | man, Breaking Bad does not come back until July |
| 18:35 | <WeirdAl> | I'm just starting an implementation, and if I go full-bore compliance, I'll be spending far too long on it |
| 18:36 | <annevk> | what language? |
| 18:36 | <WeirdAl> | right now it's "get something resembling the DOM working" |
| 18:36 | <WeirdAl> | in pure JavaScript |
| 18:36 | <WeirdAl> | and yes, I have my reasons |
| 18:36 | <annevk> | you know there is one already? |
| 18:37 | <WeirdAl> | if you mean envjs or dom.js, I'm aware of those projects, and I don't think either one will meet my needs |
| 18:37 | <annevk> | k |
| 18:37 | <WeirdAl> | (which really sucks, but oh well) |
| 18:38 | <annevk> | if you find something in the spec that can be done better be sure to let us know |
| 18:39 | <WeirdAl> | imho, the best thing you, Aryeh and Ms2ger can do is start porting the W3C DOM tests for DOM4 |
| 18:39 | <WeirdAl> | checking the tests against what the spec says |
| 18:40 | <Ms2ger> | I've done the first three or so, a while back :) |
| 18:40 | WeirdAl | blinks |
| 18:42 | <WeirdAl> | are they hosted somewhere? |
| 18:43 | <Ms2ger> | w3c-test.org/webapps/DOMCore/tests/submissions/Ms2ger |
| 18:43 | <Ms2ger> | If I didn't typo that |
| 18:44 | <WeirdAl> | I'll have to bookmark that :) |
| 18:47 | <Ms2ger> | You may be able to reuse dom.js's testing glue |
| 18:47 | <WeirdAl> | I'll need to look |
| 18:49 | WeirdAl | is just trying to figure out a good way to implement attributes in his funny shadow content/undo history model |
| 18:52 | <WeirdAl> | I think I've got it though |
| 19:44 | <Ms2ger> | gsnedders, hey, any chance for a patch within a year of your last comment in http://code.google.com/p/html5lib/issues/detail?id=157 ? :) |
| 19:45 | <gsnedders> | Ms2ger: Bah! |
| 19:45 | gsnedders | wonders what happened to the code he wrote for that… |
| 19:45 | gsnedders | wonders also why he's suddenly feeling sick |
| 19:48 | <Ms2ger> | Heh, sorry |
| 19:49 | <gsnedders> | (Literally around the second you asked.) |
| 20:03 | <gsnedders> | Ms2ger: Amusingly, I was just thinking about html5lib on PyPy. Really want to get the large-scale changes in first that sort out implicit unicode conversion |
| 23:44 | <rniwa> | Hixie, abarth: are you guys there? |
| 23:44 | <rniwa> | it appears that there is a slight bug in the spec with respect to parser-inserted flag |
