#microformats 2018-08-14

2018-08-14 UTC
strugee and [jgarber] joined the channel
#
[jgarber]
Good morning/afternoon/evening!
#
[jgarber]
I’ve got a couple of PRs open on the microformats/tests repo that should be fairly straightforward. Anyone have some time to review (and if all looks good merge)?
#
Loqi
[jgarber623] #92 Update self-closing p-* properties test results
#
Loqi
[jgarber623] #91 Update license file
hober, globbot and reidab joined the channel
#
@nagahitoyuki
#SearchConsole の検索での見え方の #構造化データ で、#Microformats の #マークアップ はコンテンツが表示されてないと検出されないと分かった。ブログの記事/追加ページの下の方にupdated:更新日を付けて表示して #エラー を回避するようにした。#デザイン も納得した。https://www.nagahitoyuki.com/
(twitter.com/_/status/1029172001456369665)
[jgmac1106] joined the channel
#
Loqi
jgarber has 1 karma in this channel over the last year (7 in all channels)
#
ben_thatmustbeme
for #92 it agrees with PHP, Ruby, and Python parsers https://ben.thatmustbe.me/mf2tests/
#
[jgarber]
[ben_thatmustbeme] Thanks! 🙂
#
ben_thatmustbeme
i'm seeing that some of the issues with the md5s are just a matter of sorting better
#
ben_thatmustbeme
i tried to get it to sort, but it didn't get everything
#
[jgarber]
Oh, hello. That’s a handy little web page.
#
[jgarber]
The Ruby parser’s got a way to go yet.
#
KartikPrabhu
just to note: not all the tests are upto spec
#
ben_thatmustbeme
Correct, another reason I wanted this page, if the parsers all agree on something different than the test suite
#
ben_thatmustbeme
I have a branch with the implied names fix. I just wanted to fix the tests first, but I think it might make sense to disable those tests for a time
#
gRegorLove
Are there mf2 tests with implied name issues? I thought I got all those.
#
gRegorLove
#92 looks good. I can merge.
#
[jgarber]
Thanks, gRegorLove!
#
ben_thatmustbeme
Implied url at least
#
gRegorLove
Oh, haven't checked on that yet, just implied name.
#
[jgarber]
Oh, while we’re here… I believe this issue can be closed out, too: https://github.com/microformats/tests/issues/80
#
Loqi
[willnorris] #80 h-event date tests use vcp formatting when it shouldn't
#
ben_thatmustbeme
i had been working on a list
#
ben_thatmustbeme
but then i hit the geo stuff with mf1 and got side-tracked
#
gRegorLove
I'm working on adding some tests where name should *not* be implied currently.
#
gRegorLove
I can include some implied photo/url updates too if you have the list handy
elimisteve joined the channel
#
ben_thatmustbeme
all of the /includes/ stuff is just not implemented it seems
#
ben_thatmustbeme
all 3 parsers just ignore it
#
ben_thatmustbeme
page is updated by the way
#
KartikPrabhu
ben_thatmustbeme: what inludes stuff?
#
ben_thatmustbeme
class="include"
#
KartikPrabhu
hmm never seen that in the spec
#
KartikPrabhu
hmmm but there is no backcompat for that
#
KartikPrabhu
so it won't be parsed by mf2 parsers
#
ben_thatmustbeme
i believe it was something you had to parse through a seperate piece, i'm trying to find exactly where
#
ben_thatmustbeme
looks like tantek started brainstorming back in april to replace it with itemref
#
ben_thatmustbeme
either way, that will still be super annoying in my implementation
#
KartikPrabhu
huh! I don't see it anywhere in the mf2 parsing or backcompat so I wouldn't worry about implementing it
#
gRegorLove
The April edits mostly appear to be cleanup of /microformats2 into other pages. https://chat.indieweb.org/microformats/2018-04-03
#
gRegorLove
Yeah, I'm not going to worry about include-pattern for now.
#
ben_thatmustbeme
nor am i, thats what tantek had said when i was re-writing the parser
#
gRegorLove
Hm, so should we remove them or flag them as for mf1 parsers only somehow?
#
aaronpk
wtf is include pattern? I've never heard of that before!
#
KartikPrabhu
welcome to the club! :P
#
Loqi
nice
#
gRegorLove
You broke the chain Loqi
#
aaronpk
hahahaha
#
gRegorLove
The microformats wiki is the gift that keeps giving
#
aaronpk
Come on Loqi
#
ben_thatmustbeme
gives loqi a broken chain
#
Loqi
throws the broken chain
#
ben_thatmustbeme
you keep that
#
gRegorLove
Huh, looks like the node parser follows the includes and itemref
#
ben_thatmustbeme
h-resume/affiliation has the implied u-url issue was i mentioned
#
gRegorLove
ok, updated
#
gRegorLove
updated h-adr/geourl
#
ben_thatmustbeme
microformats-v2/h-entry/impliedvalue-nested looks to be the implied name issue
#
ben_thatmustbeme
microformats-v1/hnews/all.json was the geo one that was thowing me off
#
gRegorLove
h-entry/impliedvalue has explicit p-name though?
#
gRegorLove
Do you have latest master? that one was updated July 22
#
ben_thatmustbeme
ah, i might not
#
ben_thatmustbeme
weird though, i just pulled from composer and it didn't have that
#
ben_thatmustbeme
which is supposed to pull off of master
#
KartikPrabhu
is there any backcompat defined fr hnews?
[chrisaldrich] joined the channel
#
gRegorLove
heh, "source-org" in that hnews/all.json too. wut?
#
gRegorLove
oh, helps if I look at the right spec
#
ben_thatmustbeme
microformats-mixed/h-resume/mixedroots another implied name
#
ben_thatmustbeme
microformats-v2/h-recipe/all has an implied url
#
gRegorLove
updated h-resume/mixedroots
#
gRegorLove
h-recipe/all also updated July 22
#
ben_thatmustbeme
i literally just did a pull before i ran that test to see it wrong
#
gRegorLove
Which name is wrong?
#
ben_thatmustbeme
not name, url
#
gRegorLove
oops, my bad
#
ben_thatmustbeme
flickr url i believe
#
gRegorLove
ok, updated
#
ben_thatmustbeme
microformats-v2/h-adr geourl i'm seeing another implied url
#
gRegorLove
Finished reviewing v2 tests, I think they're all good for implied url now
#
gRegorLove
Pushed those updates to https://github.com/gRegorLove/tests. Stepping away for a bit.
#
Loqi
[gRegorLove] tests: Microformats test suite
Omnious joined the channel
#
ben_thatmustbeme
huh, okay, question for you
#
ben_thatmustbeme
if no explicit "name" property, and no other p-* or e-* properties, and no nested microformats,
#
ben_thatmustbeme
if its a u-url h-cite ...
#
ben_thatmustbeme
its a property, not just a random nested item
#
ben_thatmustbeme
this is what it throwing me off
#
ben_thatmustbeme
nested can mean anywhere correct?
#
ben_thatmustbeme
So much for me getting to bed early
[jgmac1106], strugee and [chrisaldrich] joined the channel
#
gRegorLove
Right, nested includes on properties as well as child microformats, like h-feed > h-entry
#
gRegorLove
Since implied name was intended for much simpler cases like minimal h-cards. Once nested microformats come in, it makes for a messy, mostly useless implied name.
Humvee9, TehNut, [pfefferle], circle, MobileMatt, adactio, barpthewire, [kevinmarks], [jgmac1106], [metbril], badpixel5, phuzion28 and tantek__ joined the channel
#
aaronpk
edited /rel-payment (+203) "/* Implementations */ transistor.fm"
(view diff)
[jgmac1106] joined the channel
#
tantek
edited /h-resume (+215) "/* Examples in the Wild */ archive link for old example"
(view diff)
Guest51933 joined the channel
#
aaronpk
whoa did the lang parsing make it in to the production php parser or did i forget that i enabled that flag on pin13?
jackjamieson, Shnaw, TallTed, [iambismark], d9b4bef92, [schmarty], [keithjgrant], [chrisaldrich] and Theking^25 joined the channel
#
gRegorLove
I noticed that. I thought you'd enabled it, aaronpk
#
gRegorLove
It's on php.microformats.io though, so maybe it did make it to production
#
gRegorLove
or is that a redirect to pin13?
#
gRegorLove
master branch still has the flag set to false
#
aaronpk
no the one on microformats.io is on heroku
#
aaronpk
ah the flag is enabled
#
aaronpk
both use that repo
[jgmac1106] joined the channel
#
gRegorLove
Maybe we could add it as a checkbox toggle, labeled experimental
#
aaronpk
that'd be nice
tantek, tantek__, burnout_, [eddie], [cleverdevil], Turbo-Folker, [tantek], Humvee26, [metbril], [iambismark], [chrisaldrich], [grantcodes], WSPR20, rory10, adactio_, [pfefferle], eNbass, MikeoftheEast, [jgmac1106], strugee, Nietzsche15, Yatekii26, barpthewire, hober, [kevinmarks], Ragnor27, stephen872, KartikPrabhu and milkii joined the channel