01:00
<zewt>
https://www.w3.org/Bugs/Public/show_bug.cgi?id=16285 if only someone had already written a clean spec documenting how dom events wo--oh, right
01:01
<zewt>
https://www.w3.org/Bugs/Public/show_bug.cgi?id=16285 if only someone had already written a clean spec documenting how dom events wo--oh, right
13:49
<smaug____>
um, nested event loops are so much fun :/
13:49
<smaug____>
um, nested event loops are so much fun :/
13:50
smaug____
wants to kill all the sync and nested event loop things from the web platform (window context)
13:50
smaug____
wants to kill all the sync and nested event loop things from the web platform (window context)
14:29
<smaug____>
alert, confirm, prompt. are there other sync/blocking dialogs
14:29
<smaug____>
alert, confirm, prompt. are there other sync/blocking dialogs
14:29
<zewt>
file pickers, usually
14:29
<smaug____>
ah
14:29
<smaug____>
I have those :)
14:30
<zewt>
file pickers, usually
14:30
<smaug____>
ah
14:30
<smaug____>
er
14:30
<smaug____>
I have those :)
14:30
<smaug____>
er
14:30
<smaug____>
hate even
14:30
<zewt>
i wonder how many sites would break if that was made non-modal; might confuse some stuff that calls input.click() or whatever it is
14:31
<zewt>
i wonder how many sites would break if that was made non-modal; might confuse some stuff that calls input.click() or whatever it is
14:31
<zewt>
i've wished those would at least be made tab-modal, like the other dialogs, but since they're OS-provided things that's hard to do, I think
14:31
<zewt>
i've wished those would at least be made tab-modal, like the other dialogs, but since they're OS-provided things that's hard to do, I think
14:32
<smaug____>
file picker doesn't have to be OS-provided though
14:32
<zewt>
it'd be annoying to users if it was nonstandard
14:32
<smaug____>
sure
14:32
<smaug____>
file picker doesn't have to be OS-provided though
14:32
<zewt>
it'd be annoying to users if it was nonstandard
14:32
<smaug____>
sure
14:32
<smaug____>
(expect in OSX anything is better than the default filepicker)
14:32
<smaug____>
except
14:32
<zewt>
oh yeah, onbeforeunload, technically
14:32
<smaug____>
I can't type today
14:32
<zewt>
not as bad as the others since it doesn't happen in the middle of scripts
14:33
<smaug____>
(expect in OSX anything is better than the default filepicker)
14:33
<smaug____>
except
14:33
<zewt>
oh yeah, onbeforeunload, technically
14:33
<smaug____>
I can't type today
14:33
<zewt>
not as bad as the others since it doesn't happen in the middle of scripts
14:33
<zewt>
so maybe it doesn't count
14:34
<zewt>
so maybe it doesn't count
14:34
<smaug____>
it does
14:35
<smaug____>
it does
14:35
<zewt>
depends on what you're asking for
14:35
<zewt>
depends on what you're asking for
14:35
<smaug____>
just doing something for mutationobservers...
14:36
<smaug____>
just doing something for mutationobservers...
14:36
<smaug____>
and this is very mozilla specific things..
14:36
<smaug____>
since all the alert() etc are also just DOM documents
14:36
<smaug____>
and this is very mozilla specific things..
14:36
<smaug____>
need to do so microtask hackery
14:36
<smaug____>
since all the alert() etc are also just DOM documents
14:36
<smaug____>
need to do so microtask hackery
14:36
<smaug____>
s/so/some/
14:37
<smaug____>
s/so/some/
14:37
<zewt>
maybe SSL nags in some cases?
14:38
<zewt>
maybe SSL nags in some cases?
14:38
<smaug____>
bah, I don't care about SSL :)
14:38
<smaug____>
bah, I don't care about SSL :)
14:39
<smaug____>
zewt: thanks. I had forgotten filepicker and beforeunload.
14:40
<smaug____>
zewt: thanks. I had forgotten filepicker and beforeunload.
16:21
<jwheare>
TabAtkins: hey, just read your http://www.xanthir.com/blog/b48H0 after having a chat with alex russel a few weeks ago re: constraint position layouts. do you know of any javascript implementations of your ideas or anything like them?
16:22
<jwheare>
TabAtkins: hey, just read your http://www.xanthir.com/blog/b48H0 after having a chat with alex russel a few weeks ago re: constraint position layouts. do you know of any javascript implementations of your ideas or anything like them?
16:40
Ms2ger
writes some JS tests
16:40
<Ms2ger>
Opera passes everything, foiled again
16:40
Ms2ger
writes some JS tests
16:40
<Ms2ger>
Opera passes everything, foiled again
16:42
<gsnedders>
Heheheh.
16:42
<gsnedders>
Go me/jgraham!
16:43
<gsnedders>
Heheheh.
16:43
<gsnedders>
Go me/jgraham!
16:43
<gsnedders>
Ms2ger: Are you testing hot-code behaviour or just cold-code?
16:43
<Ms2ger>
Cold
16:43
<gsnedders>
Ms2ger: Are you testing hot-code behaviour or just cold-code?
16:43
<Ms2ger>
Cold
16:44
<gsnedders>
Ms2ger: Ah, then you're unlikely to find anything *that* interesting.
16:44
<Ms2ger>
Mm
16:44
<Ms2ger>
Mm
16:44
<Ms2ger>
Relatedly
16:45
<Ms2ger>
Relatedly
16:45
<Ms2ger>
Does the spec define what Math.max({ valueOf: function() { print('a') } }, { valueOf: function() { print('b') } }) should print?
16:45
<gsnedders>
Yes.
16:46
<Ms2ger>
Does the spec define what Math.max({ valueOf: function() { print('a') } }, { valueOf: function() { print('b') } }) should print?
16:46
<gsnedders>
Yes.
16:46
<Ms2ger>
Do continue :)
16:46
<Ms2ger>
Do continue :)
16:47
<gsnedders>
Or at least I remember a discussion about this on es-discuss and I think the conclusion was yes. :)
16:47
<gsnedders>
Or at least I remember a discussion about this on es-discuss and I think the conclusion was yes. :)
16:49
<gsnedders>
Can't find anything, can't justify anything from spec.
16:50
<gsnedders>
Can't find anything, can't justify anything from spec.
16:50
<Ms2ger>
Me neither
16:51
<Ms2ger>
Me neither
16:51
Ms2ger
asserts 'ab'
16:51
Ms2ger
asserts 'ab'
16:52
<gsnedders>
Ms2ger: I believe everything does that
16:52
<gsnedders>
Ms2ger: I believe everything does that
16:52
<gsnedders>
Ms2ger: I'd quite like some tests for typed arrays testing JIT behaviour
16:52
<Ms2ger>
I'd like tests for all of HTML
16:52
<Ms2ger>
And a pony
16:53
<gsnedders>
Ms2ger: I'd quite like some tests for typed arrays testing JIT behaviour
16:53
<Ms2ger>
I'd like tests for all of HTML
16:53
<Ms2ger>
And a pony
16:53
<gsnedders>
left-to-right.
16:53
<gsnedders>
"Each of the following Math object functions applies the ToNumber abstract operator to each of its arguments (in left-to-right order if there is more than one)"
16:53
<gsnedders>
left-to-right.
16:53
<gsnedders>
"Each of the following Math object functions applies the ToNumber abstract operator to each of its arguments (in left-to-right order if there is more than one)"
16:53
<gsnedders>
15.8.2
16:54
<gsnedders>
15.8.2
16:54
<Ms2ger>
Excellent
16:54
<Ms2ger>
Excellent
16:54
<gsnedders>
I thought it was specified. :P
16:55
<gsnedders>
I thought it was specified. :P
16:55
<gsnedders>
Ms2ger: Also, if you want to find bugs in Opera, you better be testing ARM JIT. :P
16:55
<gsnedders>
Ms2ger: Also, if you want to find bugs in Opera, you better be testing ARM JIT. :P
16:55
<Ms2ger>
I don't care about ARM
16:56
<Ms2ger>
I don't care about ARM
16:57
<gsnedders>
Ms2ger: If you're testing ES, don't expect identical behaviour cross-platform. There are, unsurprisingly for compilers, backend specific bugs
16:57
<gsnedders>
Ms2ger: If you're testing ES, don't expect identical behaviour cross-platform. There are, unsurprisingly for compilers, backend specific bugs
16:58
<gsnedders>
Ms2ger: Also x86 v. x86_64
16:58
<gsnedders>
Ms2ger: Also x86 v. x86_64
16:58
<Philip`>
Are there e.g. SSE2 vs non-SSE2 paths in the backend?
16:58
<Philip`>
(or NEON vs non-NEON or whatever)
16:59
<Philip`>
Are there e.g. SSE2 vs non-SSE2 paths in the backend?
16:59
<Philip`>
(or NEON vs non-NEON or whatever)
17:00
<gsnedders>
Philip`: SSE2 v. non-SSE2 on x86 (not on x86-64, as SSE2 exists on x86-64 by definition)
17:01
<gsnedders>
Philip`: SSE2 v. non-SSE2 on x86 (not on x86-64, as SSE2 exists on x86-64 by definition)
17:01
<gsnedders>
Philip`: There's also other SSE stuff used, more recently
17:01
<gsnedders>
Philip`: Dunno about NEON
17:01
<gsnedders>
Philip`: There's also other SSE stuff used, more recently
17:01
<gsnedders>
Philip`: Dunno about NEON
17:05
<gsnedders>
Philip`: (though we ship with the non-SSE2 JIT disabled, because it's buggy and nobody has ever bothered to fix it)
17:05
<gsnedders>
Philip`: (though we ship with the non-SSE2 JIT disabled, because it's buggy and nobody has ever bothered to fix it)
17:13
<Ms2ger>
gsnedders, I've got failures in all my tests in at least two of {Fx, IE, Chrome}, I'm not going to bother with Opera-on-ARM :)
17:14
<Ms2ger>
gsnedders, I've got failures in all my tests in at least two of {Fx, IE, Chrome}, I'm not going to bother with Opera-on-ARM :)
17:15
<gsnedders>
Ms2ger: Heh. Testing IE10 along-side latest versions of the others?
17:16
<gsnedders>
Ms2ger: Heh. Testing IE10 along-side latest versions of the others?
17:16
<Ms2ger>
Yeah
17:16
<Ms2ger>
I should check if Chrome is the latest, actually
17:16
<Ms2ger>
Yeah
17:16
<Ms2ger>
I should check if Chrome is the latest, actually
17:16
<Ms2ger>
18.0.1025.100 beta, I guess that's pretty out of date
17:17
<Ms2ger>
18.0.1025.100 beta, I guess that's pretty out of date
17:17
<gsnedders>
19 is Dev now
17:18
<gsnedders>
19 is Dev now
17:21
<Ms2ger>
Opera 11.60 beta, how old is that?
17:21
<Ms2ger>
Opera 11.60 beta, how old is that?
17:22
<gsnedders>
Six months?
17:22
<gsnedders>
Six months?
17:23
<gsnedders>
Nah, only four.
17:23
Ms2ger
tries to upgrade
17:24
<gsnedders>
Nah, only four.
17:24
Ms2ger
tries to upgrade
17:25
gsnedders
wonders if the beta doesn't autoupdate
17:25
gsnedders
wonders if the beta doesn't autoupdate
17:25
<Ms2ger>
11.61?
17:25
<Ms2ger>
Hmm, that's not beta, I gues
17:25
<Ms2ger>
s
17:26
<Ms2ger>
11.61?
17:26
<Ms2ger>
Hmm, that's not beta, I gues
17:26
<Ms2ger>
s
17:27
<gsnedders>
Ms2ger: Indeed.
17:27
<gsnedders>
Ms2ger: But we should autoupdate from beta to stable (and ideally back to beta, if there becomes another)
17:28
<gsnedders>
Ms2ger: Indeed.
17:28
<gsnedders>
Ms2ger: But we should autoupdate from beta to stable (and ideally back to beta, if there becomes another)
17:28
<gsnedders>
Opera Next should stick with Next, which is typically latest snapshot with a slight delay.
17:29
<gsnedders>
Opera Next should stick with Next, which is typically latest snapshot with a slight delay.
17:30
<Ms2ger>
Oh, so 11.61 is what I want to be on?
17:31
<Ms2ger>
Oh, so 11.61 is what I want to be on?
17:31
<gsnedders>
Either get Opera Next which is 12 Alpha, or 11.61 if you want stable.
17:31
<gsnedders>
Either get Opera Next which is 12 Alpha, or 11.61 if you want stable.