04:44
<TabAtkins>
Urg, why did nobody tell me DOM was 9000 lines long? This is gonna take a while.
04:44
<TabAtkins>
It speeds up as I get through it, but it still took about 3 hours to do 20%.
05:38
<Domenic>
If we told you, you might be scared off! :P
05:38
<Domenic>
It's easy to forget about treewalker stuff...
06:41
<TabAtkins>
Domenic: Luckily most of the work is search/replacing linking syntax, so as I work through it more and more of the links have already been replaced, and I can just plow through fixing the occasional things.
06:42
<TabAtkins>
I'm not going to kill myself switching it over to Markdown headings and lists like I did for CSSOM View.
06:42
<Domenic>
Heh
09:39
<hemanth>
http://dev.w3.org/ is down? Or is it just me?
09:41
<annevk>
hemanth: you
09:45
<hemanth>
:(
09:46
<hemanth>
::spelling-error nice selector :D
10:11
<jsx>
How can I start a CSS list counter from a negative value?
15:20
<TabAtkins>
jsx: same way you start it at any particular value - supply a starting value in counter-reset when you establish the counter.
15:21
<TabAtkins>
(Usually you want the starting value to be one less than the first <li> value, since counter-increment happens before display.)
15:41
<jsx>
TabAtkins: thanks! that will do. I wanted to add an example of using the negative descriptor of @counter-style
19:22
<pyfisch>
hello. are there any HTTP status code smaller than 100 or greater 999 out there? Because I am contributing to a HTTP library and we failed a W3C that used a status code bigger than 599 as some webapps may do. We can not (or do not want to) support all integers, we want to limit the number of status codes as much as possible.
19:23
<Ms2ger>
It's a 16-bit field, no?
19:24
<pyfisch>
Ms2ger: no it an enum.
19:25
<pyfisch>
actually it is hyper http lib written in rust.
19:25
<Ms2ger>
I know hyper :)
19:26
<pyfisch>
Ms2ger: yes your name sounded familiar. :-D and you contribute to rust-lang ;-)
19:26
<Ms2ger>
Rarely, mostly Servo :)
19:26
<Ms2ger>
Anyway, XHR has a 16-bit attribute for it
19:28
<pyfisch>
ok servo, they asked us to change it :-)
19:28
<Ms2ger>
Can't find anything that restricts it to that range, though...
19:28
<pyfisch>
so 0 would be a valid status code?
19:28
<Ms2ger>
Yeah
19:28
<pyfisch>
but I hope it is unsigned? o_O
19:28
<Ms2ger>
For some value of valid :)
19:29
<Ms2ger>
Yes
19:29
<Ms2ger>
HTTP1.1 seems to claim it's three-digit
19:31
<pyfisch>
and what about HTTP/2?
19:33
<pyfisch>
looks like also 3 digit https://http2.github.io/http2-spec/#HttpResponse it refers [RFC7231]
19:34
<Ms2ger>
Yep, seems that way
19:34
<pyfisch>
so 42 would be a not very much valid status code
19:35
<Ms2ger>
Hmm
19:35
<Ms2ger>
Maybe 042 would, though
19:36
<pyfisch>
yes 042 would be
19:40
Ms2ger
makes a note to write tests
19:43
<pyfisch>
Ms2ger: https://github.com/hyperium/hyper/issues/177
20:01
<pyfisch>
Ms2ger: both firefox and chromium deal with 42 and 1024 as codes but only chromium understands 2147483647, firefox thinks it is 200. You will have to do much work until Firefox is as good as Chrome again :-P
20:04
<Ms2ger>
Ha
20:10
<pyfisch>
Ms2ger: but you really have nice integer underflow for firefox HTTP/1.0 -5 UNKNOWN results in 65531 ;-)
20:10
<pyfisch>
this has not even chromium
20:12
<Ms2ger>
Woops...
20:33
<pyfisch>
bye
22:16
<JakeA>
Could a friendly Mozillian triage this https://bugzilla.mozilla.org/show_bug.cgi?id=1125667 - SVGs served from blob urls break
22:16
<JakeA>
I think it could be the url resolver, not sur
22:16
<JakeA>
e