#microformats 2018-03-23

2018-03-23 UTC
tantek, [eddie], ben_thatmustbeme, j12t_, [jeremycherfas], chrisaldrich, barpthewire, KartikPrabhu and [asuh] joined the channel
#
Zegnat After a good night’s sleep, I have responded to #32. While I am still of the opinion that accepting non-valid-values (per HTML spec) for attributes, even if they might be correctly defined later in the source document, is dumb, I will set up a browser experiment to find examples in the wild first.
#
Zegnat The issue has a second point though: clarifying what “if any” means for text. Especially since people want to keep empty values around, why is there an exception for text? Or does “if any” mean something else entirely?
#
Zegnat Also: do we agree on dropping leading and trailing whitespace from attribute values, like we do elsewhere in the mf2 spec, at least?
#
Zegnat Please review: https://github.com/microformats/microformats2-parsing/issues/32#issuecomment-375582034
#
Loqi [Zegnat] > I’m not sure it would cause any problems, but it feels a bit weird to me — like the parser is trying to fix a publisher mistake. It was my understanding that the entire `rel-urls` object existed only to assist consumers, it was drafted speci...
nitot_ joined the channel
#
Zegnat !tell aaronpk thank you for the diligent examples! One step closer to better plain text versions in your reader: https://github.com/microformats/microformats2-parsing/issues/15#issuecomment-375595529
#
Loqi Ok, I'll tell them that when I see them next
#
Loqi [Zegnat] It only took 2 months, but I have written a draft specification for handling `textContent` in the microformats context. The algorithm combines: * #17, always remove `<script>` and `<style>` elements, replace `<img>` elements, and strip leading an...
#
Zegnat aaronpk++
#
Loqi aaronpk has 18 karma in this channel (1596 overall)
#
Zegnat !tell gRegorLove thanks for the ++ on the text parsing! The parsing issue has been updated, let discussion commence.
#
Loqi Ok, I'll tell them that when I see them next
[kevinmarks], test and [cb] joined the channel
#
@schussman @kaushalmodi @ChrisAldrich @_dgoldsmith There's an emacs gang on http://micro.blog, too! :) My Hugo theme is HPSTR. It has some plumbing for microformats that gave me a good start, but I've heavily modified it since then as I came to understand them better. (twitter.com/_/status/977189920413335552)
[cb], [kevinmarks], [eddie], KartikPrabhu, [unoabraham], [pfefferle], [cleverdevil], tantek, [manton] and MeanderingCode joined the channel
#
@ChrisAldrich @huby And now the bigger reveal! Since you've liked my tweet, you'll notice that I'm backfeeding reactions and comments from my tweets, so your "like" also lives on my website at the original post which was syndicated to Twitter. http://stream.boffosocko.com/2018/huby-plain-old-semantic-html-with-microformats-in-combination-with (twitter.com/_/status/977218119574917120)
tantek, [snarfed] and [jgmac1106] joined the channel
#
gRegorLove Zegnat: Thanks for catching that unnecessary escaping on the regex. Just pushed a commit: https://github.com/indieweb/php-mf2/pull/161
#
Loqi gRegorLove: Zegnat left you a message 7 hours, 54 minutes ago: thanks for the ++ on the text parsing! The parsing issue has been updated, let discussion commence.
#
Loqi [gRegorLove] #161 Add failing tests and fixes for #158, #160
#
gRegorLove tantek: I'm thinking the rel=tag backcompat for h-review should match h-entry: http://microformats.org/wiki/h-review#Parser_Compatibility v http://microformats.org/wiki/h-entry#Parser_Compatibility
#
Loqi h-review
#
gRegorLove I was going to file an issue but there's no microformats/h-review repo.
#
gRegorLove Same for h-card. It doesn't mention rel-tag in the backcompat section
#
tantek gRegorLove: disagreed. unless you can cite real world hCard and hReview examples that depend on rel-tag
#
tantek we really should constrain any/all backcompat to only what is necessary for real world backcompat. not for theoretical backcompat.
#
gRegorLove Well h-review already has rel-tag in the backcompat section.
#
KartikPrabhu h-review backcompat already has rel-tag just there are not as mnay words as in h-entry
#
KartikPrabhu I agree these words should be added
#
tantek and since h-review is still more in-flux / preliminary, it would be easy to drop it
#
tantek in fact I may just demote it to "proposed"
#
gRegorLove Ok, that would make sense too. So far I've only found 3 still-published hReviews on /hreview-examples-in-wild
#
gRegorLove Many more to review, but they're so old I'll be surprised if many more show up + have rel-tag
#
gRegorLove Actually, here's one with rel-tag: http://jg.typepad.com/ciel/2006/02/daniel_bouluds_.html
#
Loqi [Joan Gelfand] Divine Brunch!
#
tantek ok good enough. that's likely evidence that there are more out there
#
gRegorLove And https://blog.ftwr.co.uk/archives/2006/01/05/the-fox-inn/
#
tantek gRegorLove: go ahead and edit h-review yourself for this since it's not under formal change control yet (because it's still such an early draft in terms of adoption etc.)
#
tantek " rel=tag backcompat for h-review should match h-entry"
#
tantek citing those examples
#
gRegorLove Will do
#
tantek in edit summary
[kim_landwehr] joined the channel
#
gregorlove edited /h-review (+308) "/* Parser Compatibility */ rel=tag backcompat for h-review should match h-entry. Examples in wild: http://jg.typepad.com/ciel/2006/02/daniel_bouluds_.html and http://blog.ftwr.co.uk/archives/2006/01/" (view diff)
[jeremycherfas] and [manton] joined the channel
#
Zegnat gRegorLove, is there a roadmap for next php-mf2 release?
#
gRegorLove No. I expect aaronpk will want to release 0.4.2 with the vendor prefix fix asap, which I'm also in favor of.
#
gRegorLove A roadmap for future releases would be nice.
#
gRegorLove Your unit tests PR reminded me I haven't tried the test suite in quite a while. I was focused on adding rel-urls and some other things to make it pass more of those, or get closer to it at least.
#
Zegnat Can always set milestones on issues, to give an idea of in what release you’d want to land it
#
gRegorLove So I'm definitely interested in getting it to pass all of mf2/tests
#
Zegnat I am looking into both the accuracy of how I run it, and preparing a possible PR to mf2/tests for some things. Might get to that this weekend.
#
Zegnat IF you want to coordinate, let me know
#
gRegorLove Haven't commented on that PR yet but was wondering if it should go into master, since they won't pass CI
#
Zegnat Probably hold of for now, especially if we are ramping up to a 0.4.2 release
#
Zegnat I am currently split between putting my textContent algo into PHP, and looking more into tests.
#
gRegorLove Haven't done milestones in github, I'll check into that
#
gRegorLove Yeah, let's definitely coordinate.
#
gRegorLove I can probably get a PR for https://github.com/indieweb/php-mf2/issues/157 this weekend, so that could probably go in 0.4.2 as well.
#
Loqi [gRegorLove] #157 parser compatibility: hentry/hreview rel=tag
[eddie] and [kevinmarks] joined the channel
#
aaronpk I'm in favor of more quick point releases for bug fixes in general
#
Loqi aaronpk: KartikPrabhu left you a message 22 hours, 37 minutes ago: snarfed: do you guys consume the rel-urls from the parsed mf2?
#
Loqi aaronpk: Zegnat left you a message 22 hours, 19 minutes ago: Sneakpeak of text content extraction that (should) match all of your whitespace examples: https://wiki.zegnat.net/media/textparsing.html
#
Loqi aaronpk: Zegnat left you a message 10 hours, 54 minutes ago: thank you for the diligent examples! One step closer to better plain text versions in your reader: https://github.com/microformats/microformats2-parsing/issues/15#issuecomment-375595529
#
aaronpk I do use rel-URLs somewhere
#
Zegnat Have you had any problems finding certain values, aaronpk? There were a few minor things I fixed in the PHP parser, but I don’t think anything major.
tantek joined the channel
#
aaronpk I haven't noticed
#
masterseo edited /get-started (+32) (view diff)
#
aaronpk Master SEO 🤔
#
zegnat edited /get-started (-32) "Undo revision 66732 by [[Special:Contributions/Masterseo|Masterseo]] ([[User talk:Masterseo|Talk]]) - spam" (view diff)
#
Zegnat Have they caught on to the disfunctioning captcha, or is this just the general spam, hmm
#
tantek general spam
#
tantek edited /Special:Log/block () "blocked [[User:Masterseo]] with an expiry time of infinite (account creation disabled): Spamming links to external sites: noob move for a "Master"" (view diff)
#
Zegnat Maybe the accounts that made the previous few edits should be blocked to, if you have a moment, tantek?
#
Zegnat I reverted them pretty quickly, but I obviously can’t block
#
tantek ok will check Recent Changes
toomim joined the channel
#
Zegnat thanks!
#
tantek Zegnat note you can always say "adminhelp: ..." in the channel which any admins online at the time should see, e.g. Phae can help us block too
#
Zegnat Ooh, interesting! I never knew that! What enables that?
#
tantek those of us that are admins putting the word "adminhelp" in the keyword highlight list of our IRC clients ;)
gRegorLove joined the channel
#
Zegnat Aah, hahaha
#
Loqi nice
#
Zegnat I thought maybe ChanServ sending notices through the server or something fancy :P
#
tantek nevermind the hacks behind the curtain ;)
#
Zegnat hacks-behind-the-curtain++
#
Loqi hacks-behind-the-curtain has 1 karma
#
Zegnat Or maybe it is a case of manual until it hurts, really
#
tantek Zegnat: re: blocking other spammers - pretty sure I got them yesterday - can you verify? http://microformats.org/wiki/Special:RecentChanges
#
Loqi microformats community
#
Zegnat Oh, I must have missed those messages in chat! All is good then, tantek 👍
[kevinmarks] and [tantek] joined the channel