#microformats 2018-08-01

2018-08-01 UTC
dougbeal|mb1, KevinMarks, [jgmac1106], Syfer, jgmac1106_, timfi, [miklb], ssbr28 and spirit_pact joined the channel
#
KartikPrabhu SPAM
[cjwillcock] and sparr15 joined the channel
#
aaronpk adminhelp can someone make Loqi an op?
Guest29805, bmos, theWhisper_11 and DataComputist joined the channel
#
KartikPrabhu aaronpk: can Loqi kick people here too?
#
aaronpk If someone makes Loqi an op
#
KartikPrabhu KevinMarks: ^
[tantek], Arokh18, [ryan339] and nOgAnOo joined the channel
#
aaronpk [tantek] if you make Loqi an admin here he'll kick those boys
#
aaronpk Bots haha damn autocorrect
#
[tantek] Idk if I can do that from Slack. Let me get to a desk
#
aaronpk Yea prob not
#
aaronpk it's /msg ChanServ OP #microformats Loqi
ozy25, moondoggy13 and tantek joined the channel
#
tantek aaronpk, you and gRegorLove should be ops automatically since you're both admins http://microformats.org/wiki/admins
#
Loqi microformats admins
#
aaronpk Automatically?
#
tantek I mean should as in I need to set the flag to make that so
#
aaronpk Ah
strugee4 joined the channel
#
tantek aaronpk: done. you and gRegorLove should auto-op when you enter
#
tantek aaronpk this should give you power to op Loqi as needed for him to do his thing
#
aaronpk woo
[matpacker], __idiot__, Yoda16, [pfefferle], trisk29, badpixel25, Sagan17, philipashlock, [chrisaldrich], apetresc9, Jamesofur16, [snarfed], [metbril], alkyl6, brykr14, cholcombe6, GeekDude1, RichiH1, CalimeroTeknik29, Dominian23, indy, Nightmare27, DLange19, PityDaFool2, linuxmodder12, kaniini_, kambiz13, ben_zen29, Nazca28 and Kaja____ joined the channel
#
@jgmac1106 ↩️ Wow that is pretty cool. Microformats2 are different type of metadata than RDFa in that mf2 use existing classes rather than properties: https://indieweb.org/RDFa but they can exist together. I checked out the “add metadata menu” Is there a way to add… http://jgregorymcverry.com/5382-2/ (twitter.com/_/status/1024589292969119744)
timeless17, Xlbrag_, lebster8, his, atomicthumbs9, some_weirdo28, Freejack19, irinix15, indy, OPK6, Guest42469, enyc17, brackets2, eth223 and jackjamieson joined the channel
#
@00d4n1 ↩️ tbh micropub is pretty messy to retrofit onto an existing non-microformats2 application? it's pretty much just mf2 dropped into a request body, and so there's a bunch of two-way conversions needed if your native data format is not mf2 idk, something to keep in mind (twitter.com/_/status/1024657876114010112)
[pfefferle] joined the channel
#
@aaronpk ↩️ that'd be awesome. I just wrote an ActivityStreams-to-Microformats JSON converter which wasn't that hard. It shouldn't be that hard to go the other way around. (twitter.com/_/status/1024659636207550467)
testiiiiiiing, Kirito, [snarfed], rasengan6, dan3wik, clarjon125 and tantek joined the channel
#
aaronpk is reviewing php-mf2 PRs
#
aaronpk just merged https://github.com/indieweb/php-mf2/pull/170
#
Loqi [Zegnat] #170 Move resolve step last in u-* parsing
#
aaronpk and https://github.com/indieweb/php-mf2/pull/175
#
Loqi [Zegnat] #175 Do not mutate DOMDocument
#
Zegnat Woop!
#
aaronpk Zegnat: does https://github.com/indieweb/php-mf2/pull/177 look right to you?
#
Loqi [gRegorLove] #177 Fix XPaths used in implied photo parsing
#
Zegnat I haven’t gotten to php-mf2 in my after-vacation-catch-up yet ...
#
Zegnat I’ll have a quick loo
#
Zegnat s/loo/look/
NvpkD1y7Ez joined the channel
#
Zegnat aaronpk, that PR looks right to me, and if it doesn’t break old tests and manages these new tests, I would say it is a definitive improvement.
gRegorLove joined the channel
#
Zegnat I see little harm in merging.
#
aaronpk sounds good
#
Zegnat This test driven development has upsides this way ;)
#
aaronpk yep
ablackack8, [snarfed], tsp, nug700 and NvpkD1y7Ez joined the channel; KartikPrabhu left the channel
#
aaronpk I think I am going to release this as 0.4.4
#
aaronpk the changes in 0.4.4-alpha i've been using for a while now
#
gRegorLove I think there's still discussions about the new textContent algorithm. Should we at least get the img src update in before 0.4.4?
#
aaronpk hm this is a lot still left in the milestone https://github.com/indieweb/php-mf2/milestone/2
#
gRegorLove I just added several today; those don't have to be for 0.4.4
#
aaronpk ah k
#
gRegorLove Most are pretty simple, but I'm ok with pushing them back.
#
gRegorLove This is the img src issue I was thinking we'd want to get in though: https://github.com/indieweb/php-mf2/issues/180
#
Loqi [gRegorLove] #180 Fix <img> handling in implied p-name
#
gRegorLove And the discussions about whitespace in pre: https://github.com/microformats/microformats2-parsing/issues/15
#
Loqi [Zegnat] #15 What should mf2 textContent parsing result in? User expectation vs. DOM specification.
#
gRegorLove Concerned about results like this in production parser: https://github.com/microformats/microformats2-parsing/issues/15#issuecomment-407632168
#
Loqi [gRegorLove] In [#microformats](https://chat.indieweb.org/microformats/2018-07-25#t1532481944885400) today there was some discussion about `<pre>` and whitespace. The results of this algorithm don't match with @aaronpk's [whitespace-test #11](https://pin13.net/mf...
#
gRegorLove (just updated that comment with the HTML inline)
#
aaronpk hmm
snarfed and jackjami_ joined the channel
#
gRegorLove Not sure. Maybe it's an edge case not many will run into, so it's ok?
#
gRegorLove ..for the short-term
krysjonaz joined the channel
#
aaronpk any objection to me turning on the more aggressive anti-spam measures here?
#
snarfed +1
#
aaronpk basically Loqi tries to +v people who aren't spammers, and only +v people can talk in the channel
#
snarfed so sknebel, do you think our next task for mf2py is to collect unit test(s) for https://github.com/microformats/mf2py/issues/108 ?
#
Loqi [kartikprabhu] #108 deepcopy error on <use> inside <svg>
#
snarfed to confirm whether the revert PR works well, in addition to your updated #114? and then we decide between the two?
#
sknebel snarfed: an option. I'd vastly prefer the revert, been thinking about how to ensure it works
#
sknebel one part obviously is manual review, should be completable but doesn't cover changes, for checking if it's safe the best I can think of is serializing the source DOM before and after, and checking if the resulting strings are the same
#
sknebel I have to spend some time with the code to make sure, but I think we really only touch it in backcompat
echarlie joined the channel
#
sknebel and figure out what operations are safe
#
sknebel we add children, that's safe I'm fairly sure
#
sknebel and we replace classes, have to check how that can be done properly
#
sknebel and of course, if we only modify in backcompat, we can use the code from #114 to only copy there
#
sknebel and not always copy
#
sknebel (which also means we avoid the soup-copy, which reparses completely
#
sknebel right now the revert PR does modify during backcompat in a way that's partially visible
#
aaronpk channel is now in red alert mode
snarfed and RussellB2812 joined the channel
#
snarfed sknebel: sounds good. thank you!
#
snarfed i expect this is the only remaining thing we need to settle before releasing 1.1.2
#
snarfed i'll probably defer to you since your head is deeper in it, but i'm happy to take all the administrative work, cut the release, docs, anything else!
#
sknebel great!
#
sknebel I should probably dump my mindstate about this topic somewhere as am I doing it
#
Loqi yea
#
aaronpk gRegorLove: that <pre> whitespace issue isn't filed as a php parser issue?
#
Loqi Welcome, aaronpk. Since you weren't yet recognized as a real person, your message wasn't sent to the channel. I'm repeating it below for you.
#
Loqi aaronpk: gRegorLove: that <pre> whitespace issue isn't filed as a php parser issue?
#
aaronpk thanks Loqi
#
gRegorLove haha
#
Loqi you're welcome, aaronpk
#
gRegorLove I don't think so. Zegnat's algorithm will need to be updated with whatever the resolution is. I think snarfed was using that to add the same support to mf2py.
#
gRegorLove So I think discussion has been focused more around the mf2-parsing issue
#
aaronpk ah
#
Zegnat Last I heard Kartik was already running a slightly newer version that didn’t even get fully documented yet.
#
Zegnat I’ll be trying to find some time this week to get the algorithm on the wiki, updated, and ready for more discussion
#
aaronpk hm, well seeing as that bug is currently in 0.4.4-alpha and i've been using that for a few months now, i'm inclined t o just ship it
#
sknebel yeah, mf2py used it as a basis, but e.g. does handle <pre> already, which I don't think was in Zegnats yet
#
gRegorLove I'm ok with that if there's no objections.
#
gRegorLove sknebel, released version or just in master?
#
sknebel released
snarfed joined the channel
#
aaronpk gRegorLove: can I move those 4 issues to 0.4.5?
#
gRegorLove Sure
#
gRegorLove I think Github does automatically
#
gRegorLove One of those was originally 0.4.3 :)
#
aaronpk automatically how?
#
gRegorLove Not sure, based on the release tag?
#
gRegorLove Prob best to do it manually
KartikPrabhu, snarfed, ozymandias26, xMopxShell and radiofree2 joined the channel
#
aaronpk alright time to cut the release
jackjamieson joined the channel
#
snarfed aaronpk++
#
Loqi aaronpk has 21 karma in this channel (1701 overall)
#
gRegorLove woo!
#
Loqi Welcome, gRegorLove. Since you weren't yet recognized as a real person, your message wasn't sent to the channel. I'm repeating it below for you.
#
Loqi gRegorLove: woo!
#
snarfed "real person" wow harsh
#
gRegorLove I want to be a real boy :/
#
gRegorLove Maybe "verified person"
#
aaronpk hm, another idea
#
aaronpk we'll see if that makes more sense next time
#
Zegnat (gRegorLove, Loqi might be on to your cyborg status.)
#
gRegorLove hehe
#
aaronpk https://github.com/indieweb/php-mf2/releases/tag/v0.4.4
#
aaronpk 🎉
ckeltz10, moonlight10, jackjamieson, Smeef6, spot26, Sheraf and KartikPrabhu joined the channel
#
gRegorLove aaronpk++
#
Loqi aaronpk has 22 karma in this channel (1702 overall)
Ovius, circle, jak25, KartikPrabhu, disi29, [dougbeal], [jgmac1106], Ragnor20, snarfed, Guest43996, boredguy, xfil, puff and bairdmich joined the channel; snarfed left the channel