00:03
<annevk-cloud>
waking up… yeah, that's basically what I got out of that RFC
00:05
<annevk-cloud>
you should be able to use the URL parser and do the scheme comparison and then if javascript pass scheme data, query, and fragment along for further processing
00:10
<Hixie_>
annevk: yeah, looking at url parser now
00:12
<Hixie_>
annevk: what does "If c is none of EOF code point, U+0009, U+000A, and U+000D, utf-8 percent encode c using the simple encode set, and append the result to url's scheme data." mean?
00:13
<Hixie_>
oh, i see, "simple encode set" isn't that simple
00:13
<Hixie_>
it's more like the "non-ascii encode set"
00:15
<Hixie_>
hm, is there some algorithm i can call that figures out if "fragment" is the empty string or null and does the right thing accordingly to reconcatenate scheme data, query, and fragment and then percent decode them?
00:23
<annevk-cloud>
Hixie_: i can write one
00:24
<annevk-cloud>
makes sense for other schemes too so should prolly be there
00:30
<annevk-cloud>
Fetch already treats javascript as network error fwiw
00:36
<Hixie_>
annevk-cloud: i can do it. how about an algorithm for percent-decode-utf-8-decode?
00:36
<Hixie_>
nothing like that yet either?
00:39
<annevk-cloud>
guess not
00:39
<annevk-cloud>
please file bugs and assume i'll fix them :)
00:39
<Hixie_>
nah, these are easy
00:39
<Hixie_>
i'm just lazy :-)
00:39
<Hixie_>
took me less time to write that than it did for you to answer :-)
00:40
<annevk-cloud>
well, these might make sense for other urls too
00:40
<annevk-cloud>
heh
00:42
<Hixie_>
does javascript strip leading BOMs automatically?
00:42
<Hixie_>
or do i need to do that for javascript: specifically?
00:44
<Hixie_>
hm, wow, <body onload="&#xFFFE;w('test')"> does work
00:44
<Hixie_>
so I guess I don't have to do it, javascript must be doing it
00:45
<annevk>
in JS a BOM is whitespace iirc
00:45
<annevk>
not in JSON though
00:47
<Hixie_>
ah
00:47
<Hixie_>
fun
01:24
<TabAtkins>
Anyone know if there's some overriding reason the SVGWG hasnt' started their meeting yet?
01:24
<TabAtkins>
It was supposed to start almost half an hour ago. :/
01:25
<MikeSmith>
TabAtkins: no idea but I guess it could be they decided to start late due to being out late last night
01:26
<TabAtkins>
Ah, didn't know they did. Plus, it looks like they don't have any agenda for the last 1.5 hours, so I guess starting an hour or so late is okay.
01:27
<MikeSmith>
I dunno if they were actually out late last night but I know there was a contigent that was last night and maybe included some of them
01:27
<MikeSmith>
anyway, I'll ping Doug I guess
01:27
<MikeSmith>
ah he's not on IRC
01:27
<heycam>
TabAtkins, yeah pretty much
01:27
<heycam>
TabAtkins, not everyone is here yet, but we'll start in a few minutes
01:27
<TabAtkins>
Okay, cool.
01:28
<TabAtkins>
Not a problem, I was just curious. ^_^
01:28
<TabAtkins>
I'm just drinking and videogaming right now anyway, so wtv
01:28
<annevk>
spotted a chair in the German beer thing earlier last night
01:29
<MikeSmith>
oh the other thing is the hotel apparently announced just yesterday that they unilaterally changed the check-out time for today to 9 am (from 12 noon)
01:29
<MikeSmith>
so there was a long line of people this morning trying to check out by 9am
01:29
<MikeSmith>
which line I personally avoided by just not bothering to try to check out at 9am
01:30
<MikeSmith>
since as far as I'm concerned my checkout time is still 12noon
01:30
<annevk>
+1
01:58
<Hixie_>
MikeSmith: you can't check out from the room tv?
02:00
<MikeSmith>
Hixie_: hah
02:00
<MikeSmith>
not here
02:00
<MikeSmith>
especially not at this hotel
02:04
<Hixie_>
i have never been so happy to have not gone to a tpac
02:04
<Hixie_>
between the horror stories about the location and hotel and the horror stories on w3cmemes... :-)
02:08
<MikeSmith>
this hotel is a much worse joke than some of worst jokes that went out from w3cmemes last night
02:12
<jgraham_>
MikeSmith: Don't hold back, tell us what you really think
02:14
<jgraham_>
To be fair the hotel did have some positive qualities. The showers were nice
02:14
<jgraham_>
OTOH the beds seemed to be made of granite
02:29
<sangwhan__>
i'm surprised nobody made a meme about the 9am check out nonsense
03:02
<TabAtkins>
MikeSmith: You in the Wuzhou Guest House?
03:03
<TabAtkins>
I didn't have a choice about my hotels - I literally could not book the Wuzhou. They had three different mechanisms to prevent me from entering required information into one of the forms.
03:25
<MikeSmith>
TabAtkins: yeah, Wuzhou Guest House
03:37
<annevk>
wait TabAtkins, you're here?
03:37
<TabAtkins>
No, but I had gotten a plane and hotel reservation before I had to cancel.
03:37
<annevk>
ah
03:38
<annevk>
So why does Google still not support 9am Toronto converted to London?
03:38
<annevk>
"9AM Toronto converted to London" as query that is
03:40
<SimonSapin>
annevk: well, that may be ambiguous, depending on the date, with DST and all
03:40
<annevk>
SimonSapin: same day would be fine
03:40
<annevk>
SimonSapin: maybe with some warnings around dates where it changes
03:52
<zewt>
http://www.wolframalpha.com/input/?i=9am+toronto+in+london+time
14:10
<hallvors1>
annevk-cloud: are you around?
16:50
<Hixie_>
anyone able to translate https://www.w3.org/Bugs/Public/show_bug.cgi?id=23387 ?
16:57
<GPHemsley>
What's this I hear about WebRTC not supporting conference calls?
16:59
<GPHemsley>
Hixie_: With a name like "Gotta Getmedat", are you sure he's not just yanking your chain?
17:02
<Hixie_>
GPHemsley: i've no idea
17:03
GPHemsley
didn't bother to read any of the words beyond that.
17:03
<Hixie_>
GPHemsley: if they are, it's not a very good yanking, since i've just no idea what it means
17:03
Hixie_
ignores names in bugs
17:03
<GPHemsley>
ah
17:03
<Hixie_>
(which is why sometimes i wontfix bugs i filed...)
17:03
<GPHemsley>
one of those table-reading things, I guess ;)
17:03
<Hixie_>
table-reading?
17:08
<GPHemsley>
which headers are read first
17:46
<zcorpan>
anyone have a bright idea on how to detect that a navigation is *not* happening, in a reliable way that doesn't use timeouts?
17:47
<zcorpan>
i want to test e.g. <area>.click() and see that it navigates an iframe to the right place, but i don't want to wait for the test to time out if there was no navigation
17:48
<zcorpan>
so there are various things the spec does on navigation, like aborting fetches, so maybe i can insert an <img> to the document just before i do the click() and if the img receives a load event, that means that the navigation failed. but i'm not confident that that isn't racy
18:09
<zcorpan>
or open a websocket and see if it gets an open event
18:10
<zcorpan>
but the cleanup steps happen async while the fetch abort steps should be sync per spec, i think
18:20
<gsnedders>
I can't think of any way of doing that. You can probably abuse the details of the document loading steps, but nobody implements them so probably not that useful.
18:20
<gsnedders>
Well, unless you care about testing Zombie-Presto.
18:22
<gsnedders>
Huh, I'm getting www.dropbox.com failing cert validation; somehow got a cert for *.opendns.com. he hell.
18:25
<zcorpan>
yeah it would be a bit easier to reason about what will happen if the spec was actually implemented :-| it looks like blink does what i think should happen for <img>, but gecko doesn't abort the img load early
18:26
<zcorpan>
if i attempt a new nagivation immediately after click() (using window.location), then that navigation is ignored in gecko but is not ignored in blink
18:26
<zcorpan>
but don't worry, the web has proven this is interoperable
18:29
<gsnedders>
Oh. So without HTTPS Everywhere, dropbox loads to OpenDNS block page.
18:29
<gsnedders>
I just want to download a small file. ;_;
18:32
<gsnedders>
Well, I guess that's not unreasonable given the on-train wifi often gets 1KB/s, or less.
18:37
<TabAtkins>
Oh good god, FAA finally updated its regs about small electronics.
18:37
<TabAtkins>
I'd heard rumbles, but just got an email from my main carrier with the details.
18:37
<TabAtkins>
Fucking finally.
18:38
<ebollens>
Yup they finally did the no duh thing
18:39
<ebollens>
Now it's just on the carriers to implement the changes.
18:40
<TabAtkins>
Right. US is properly updated now, at least on its mainline flights.
18:40
<TabAtkins>
And that's all I care about.
19:54
<dglazkov>
mrbkap: got a sec?
19:55
<dglazkov>
mrbkap: https://www.w3.org/Bugs/Public/show_bug.cgi?id=23778 penny for your thought
19:57
<mrbkap>
dglazkov: looking.
19:57
<mrbkap>
dglazkov: I was just about to ping you!
20:00
<mrbkap>
dglazkov: I have to go to lunch. I'll comment in the bug.
20:00
<dglazkov>
mrbkap: lunches are important. I probably need to do that too.
20:02
<mrbkap>
dglazkov: What should the result of https://pastebin.mozilla.org/3611161 be?
20:09
<dglazkov>
mrbkap: it should throw NotSupportedError when trying to register
20:10
<dglazkov>
mrbkap: http://www.w3.org/TR/custom-elements/#dfn-custom-element-constructor-generation
20:10
<dglazkov>
mrbkap: step 1
20:11
<Ms2ger>
tr/, eh
20:44
<xxtjaxx>
http://imagebin.org/index.php?mode=image&id=277192 .o(Why am I doing this to me...)
23:42
<jorendorff>
Domenic_: has anyone suggested Promise.prototype.finally(cb) yet?
23:43
<Domenic_>
jorendorff: haha yeah, I and others want it. https://github.com/domenic/promises-unwrapping/issues/18
23:43
<jorendorff>
all this talk of "lifetimes" made me think of it
23:43
<Domenic_>
If you can think of a use case for it in the loaders spec, that'd make the argument easier :)
23:45
<jorendorff>
good grief, there are already some nice use cases in that bug
23:45
<jorendorff>
well, two anyway: event listeners and testing cleanup
23:46
<Domenic_>
yeah, but there's nothing like "we need it for another part of the spec" to advance something to es6 status, as we know from september :)
23:46
<jorendorff>
:)
23:46
<jorendorff>
well, i'm afraid i don't think we need it for anything, but i'll keep my eyes open :)