07:34
<annevk>
hayato: if you're still awake, I updated https://github.com/whatwg/dom/pull/281 with a commit that address a couple more nits and fixes some bikeshed issues
07:34
<annevk>
hayato: I think it's ready to land now, if you agree
07:36
<hayato>
annevk: I am awake. :) Thank you for addressing nits! I appreciate if you land it.
10:07
<jgraham>
If I want to say that some value is the value of the getter for the Document interface's title attribute for some specific document instance (i.e. the value that would be returned by document.title before any user modifications), how do I say that?
10:09
<annevk>
jgraham: if it's in a spec you'd probably abstract document.title in HTML to invoke some kind of algorithm you can share
10:10
<annevk>
You can talk about initial value of various things too, but it gets icky
10:12
<jgraham>
annevk: It's in a spec. I didn't really intend to have to edit HTML to make this work :)
10:15
<jgraham>
annevk: Do you have an example of the hacky way?
10:16
<annevk>
jgraham: XMLHttpRequest has "Let JSON object be the result of invoking the initial value of the parse property of the JSON object, with JSON text as its only argument." but that's also a little questionable since it assumes you got the pure JSON object somehow
10:17
<annevk>
Fetch uses similar language
10:18
<jgraham>
Thanks
10:19
<jgraham>
I think that will be enough better than the wrongness it will replace for now
13:34
<nox>
jgraham: There is some similar wording for the MathML integration points.
13:35
<nox>
'A MathML annotation-xml element whose start tag token had an attribute with the name "encoding" whose value was an ASCII case-insensitive match for the string "text/html"'
13:35
<nox>
"whose start tag token had", because even if the encoding is changed afterwards it still counts as HTML for the parser.
13:36
<nox>
jgraham: "i.e. the value that would be returned by document.title before any user modifications" even DOM modifications?
13:41
<jgraham>
nox: No, I mean before someone overrides the getter
13:41
<nox>
Oh.
13:41
<jgraham>
Anyway I think what I write is OK
13:47
<annevk>
jochen__: around? Any idea about https://github.com/whatwg/fetch/issues/323?
13:55
<daleharvey>
is there any consensus on how cancelling fetch requests is going to work?
13:56
<daleharvey>
would really like to switch to https://github.com/github/fetch instead of using some home brew xhr lib, but we need to abort requests
13:56
<boogyman>
what's the use-case?
13:57
<daleharvey>
the use case for aborting requests?
13:59
<annevk>
daleharvey: not yet
13:59
<annevk>
daleharvey: https://github.com/domenic/cancelable-promise has the latest
14:00
<daleharvey>
cool thanks
17:27
<annevk>
Anyone else have opinions on how we format enums? https://github.com/whatwg/storage/pull/36#discussion_r71914761
17:32
<Domenic>
I've always preferred `"test"` but HTML uses "`test`" pervasively so I guess I prefer that...
17:36
<annevk>
Interesting, would you also prefer `\`test\`` for byte sequences?
17:36
<annevk>
(not sure if I got the escaping correct)
17:44
<Domenic>
hmm probably
17:44
<Domenic>
It's not a strong preference
17:44
<Domenic>
I think it's just based on programming where <code>5</code> is a value and so is <code>"string"</code>
17:45
<Domenic>
Like when you do var x = ...
17:45
<Domenic>
I think of whatever replaces the ... as the value
17:45
<annevk>
Isn't string the value and " the syntax to denote it?
17:45
<Domenic>
I can see that point of view too
17:46
<annevk>
I guess I'm okay with both too, bit invested in one variant, but oh well
17:46
<Domenic>
Yeah I mean I mainly don't want to change HTML...
17:55
<annevk>
jyasskin: I gotta go for a bit, might have some time to integrate it on Sunday during travel
17:55
<jyasskin>
annevk: 'k. I'll see if I can find a better way to concisely create a WebIDL value from spec text.
17:55
<annevk>
jyasskin: your attention to detail is much appreciated
17:56
<jyasskin>
:)
20:40
<smaug____>
does github have a way to mark a bug to be a duplicate of another one?
20:40
<smaug____>
s/bug/issue/
20:49
<jyasskin>
smaug____: As far as I can see, you just have to comment to that effect and close the duplicate.