00:00
<gsnedders>
jgraham: https://github.com/html5lib/html5lib-python/pull/95 — opinion needed as to what to do for tests for the serializer
00:18
<gsnedders>
hsivonen, MikeSmith: https://github.com/html5lib/html5lib-tests/issues/33
00:18
<gsnedders>
annevk: https://github.com/html5lib/html5lib-tests/issues/31
00:18
<annevk>
gsnedders: I think jgraham included those
00:20
<gsnedders>
annevk: https://github.com/html5lib/html5lib-tests/commit/a0e73c60dd788b18f07d9e2c412779c1759185df claims otherwise :)
00:30
<annevk>
seven years ago
00:36
<annevk>
odinho: replied, basically I'm not sure what's up
00:56
<tantek>
Fernandos - it's pointless trying to reason about semantic storage without a specific end user use-case.
00:57
<tantek>
so with that Fernandos, what's the problem you're trying to solve?
01:13
<gsnedders>
annevk: Hey, I never claimed you'd have *recent* knowledge. :)
09:23
<Ms2ger>
TIL: jgraham can't write separated (seperated), available (avaliable and avalible), or initialize (initalize)
09:43
<SimonSapin>
Ms2ger: https://twitter.com/alex_gaynor/status/414814132718678016
09:48
<MikeSmith>
Ms2ger: jgraham put those in there to test you
10:27
<FireFly>
och tja
10:27
<FireFly>
det var nân fest efter disputationen med kakor och sânt tror jag
10:27
<FireFly>
er
10:27
<FireFly>
Oops
10:47
<Ms2ger>
MikeSmith, clearly
12:42
<smola>
annevk: hello, ping?
12:42
<annevk>
smola: waddup?
12:44
<smola>
annevk: int he URL spec, it seems that a URL formed only by a scheme ("http") is parsed to an empty URL ("")
12:44
<smola>
is that the expected behaviour? am I reading wrong?
12:44
<annevk>
smola: it would return failure I think
12:44
<smola>
in the parser, scheme state:
12:44
<smola>
Otherwise, if c is the EOF code point, terminate this algorithm.
12:45
<smola>
but no parse error, which is only given if there is a non-alphanum char
12:45
<annevk>
smola: that only happens if there's a state override
12:45
<annevk>
smola: otherwise you'd switch to no scheme state
12:46
<smola>
annevk: right :)
12:46
<annevk>
smola: and no scheme state would return in failure
12:49
<smola>
annevk: thanks
13:45
<Ms2ger>
<small><font size="+2"><small><b>
14:03
<annevk-cloud>
Ms2ger: nice poem
14:07
<gsnedders>
SimonSapin: Hah. I saw Alex_Gaynor's tweet earlier. Didn't even notice the typo first time. >_>
14:07
<darobin_>
I wonder if anyone ever made an HTML poetry contest
14:08
darobin
in a previous life, placed in a Perl Poetry contest
14:08
darobin
gives annevk-cloud some ROI for Christmas
14:14
<annevk-cloud>
hah
14:37
<Ms2ger>
Robin Berfon, eh
14:49
<darobin>
Ms2ger: indeed, I kept "Berfon" as a nickname for a long time thereafter :)
15:03
<gsnedders>
SimonSapin: https://github.com/html5lib/html5lib-python/pull/133 — can you eyeball that?
15:06
<SimonSapin>
gsnedders: looks good
15:07
<SimonSapin>
I don’t need #129 on PyPI yet
15:08
<gsnedders>
I'm probably gonna try and push a 0.9999 this year, likely with that
15:09
<gsnedders>
SimonSapin: (But yeah, in principle, I can make releases whenever a downstream needs something in. Just say.)
15:09
<SimonSapin>
gsnedders: cool, thanks
15:10
<SimonSapin>
gsnedders: will it ever reach 1 ? :)
15:10
<gsnedders>
SimonSapin: https://github.com/html5lib/html5lib-python/issues?labels=api-breakage&milestone=1&page=1&state=open is what blocks
15:11
<gsnedders>
(Once those are all done, it will reach 1. And then probably just end up going along 1.0.z for a while.)
15:11
<SimonSapin>
I see, makes sense
15:12
<gsnedders>
I think basically nothing else actually /blocks/ 1, though it'd be nice to be up-to-date with the spec by then.
15:14
<gsnedders>
SimonSapin: On PyPI now.
15:17
<SimonSapin>
It’s nice to close bugs by changing install_requires in setup.py :)
16:06
<gsnedders>
SimonSapin: :)
16:41
<GPHemsley>
Ms2ger, gsnedders: Remind me again the situation regarding maintaining a consistent attribute order in anolis/html5lib?
16:42
<gsnedders>
GPHemsley: dunno about anolis
16:43
<GPHemsley>
gsnedders: But there was some setting we'd have to adjust in html5lib to get ordered attributes, right?
16:44
<gsnedders>
GPHemsley: alphabeticalize_attributes on the serializer. Or something with a new almost like that.
16:45
<Ms2ger>
I stuffed that far in the back of my mind :/
16:45
<GPHemsley>
Ms2ger: Has the issue not been affecting anybody? I'd think it'd be making noisy diffs left and right...
16:46
<Ms2ger>
I haven't heard of it
16:47
<GPHemsley>
hmm
16:47
<GPHemsley>
well, I think it'd affect me, so I want to fix it
16:55
<smola>
here again with a URL parsing doubt: what's with chars like ^? it seems that the spec states that it should *not* be percent-encoded when it appears in path; but browsers do.
17:01
<darobin>
Merry Christmas everyone, see you all next year!
17:11
<SimonSapin>
smola: file a bug, as a gift to annevk
17:14
<smola>
SimonSapin: ok, I'll do it. I wanted to ask first since every "bug" I found in the URL Spec ware actually not bugs ;)
17:14
<SimonSapin>
smola: I haven’t checked about this, but maybe annevk-cloud can confirm
17:36
<smola>
annevk: ping again, ^^^, what's with unsafe characters like ^? it seems that the spec states they should not be percent-encoded, but browsers do encode them
17:36
<annevk>
all browsers?
17:37
<smola>
no, I'm not sure :p
17:37
<smola>
I checked Chrome and Firefox
17:37
<smola>
so IE does not encode them?
17:38
<annevk>
might be a bug
17:38
<annevk>
I think I've been fairly conservative with what to escape
17:38
<smola>
ok, let me check other browsers and I'll file the bug with the detailed set of characters
17:38
<annevk>
ta
17:39
<smola>
just wanted to know if I didn't miss something obvious
18:42
<smola>
annevk: bug filed; the behaviour is pretty unconsistent among browsers (and I haven't checked old IE)
18:43
<smola>
Safari 5 has a particularly incosistent behaviour (compared to itself :p)
19:09
<annevk>
smola: thanks
19:09
<annevk>
smola: will likely look at in January somewhere
19:13
<smola>
annevk: great, thanks
20:12
<annevk>
gavin: is the stats thing still kept up to date? the numbers look somewhat similar
20:15
<gavin>
it should be
20:15
<gavin>
oh, but not_gavin isn't here
20:15
<gavin>
I'll look into it
20:16
<gavin_>
oh, it's gavin_ here
20:18
<gavin>
annevk: it all looks like it's working
20:19
<annevk>
gavin: it doesn't have the topic and I believe number of lines has been static for a while
20:19
<annevk>
but I guess I should check again tomorrow to be sure
20:21
<gavin>
annevk: the topic seems to have been last set in 2012, which probably predates this particular stats instance
20:21
<annevk>
ait
20:22
<gavin>
(it only sees changes)