#microformats 2015-05-10

2015-05-10 UTC
KevinMarks__, KevinMarks, adiabatic, KartikPrabhu and tantek joined the channel
#
@anant_ebooks In all outstanding hCalendar issues. (twitter.com/_/status/597224848725823488)
fuzzyhorns, KartikPrabhu, KevinMarks, KevinMarks__, gRegorLove, tantek, KevinMarks_, eschnou, Left_Turn, chiui, adactio, Erkan_Yilmaz and Atamido_ joined the channel
#
@jkphl @kevinmarks Thx for the heads-up! :) #microformats2 parsing is based on @BarnabyWalters' phpmf2, so we'd need to check that one first (twitter.com/_/status/597430422948151297)
Atamido_, Atamido, KartikPrabhu, KevinMarks__ and fuzzyhorns joined the channel
#
@hmans RT @t: “HTML is my API” @aaronpk on @HackerNews’s HTML vs JSON, reliability, and using #microformats2 https://aaronparecki.com/articles/2015/04/26/1/html-is-my-api (ttk.me t4ah1) (twitter.com/_/status/597451249399771137)
KevinMarks_, ChrisUrsich, KevinMarks__, Atamido_, KartikPrabhu and tantek joined the channel
#
kylewm tantek: I read the spec to mean that "e-content h-card" should result in an object with "value": { "value": "...", "html": "..." }
#
kylewm value: "parsed property value per p-*,u-*,dt-*,e-* parsing respectively",
#
tantek hmm
#
tantek so perhaps I have a spec bug as well?
#
tantek or do you think that behavior is desirable?
#
kylewm it's hard to say, we don't really know how people are going to use the "value" property yet
#
tantek that's not true - aaronpk finally figured out that's what he needed to be using (instead of writing hacks) - though I forgot the precise use-case
#
kylewm right, the precise use-case is webmentions, you get "in-reply-to" and expect it to be a list of strings
#
kylewm but it turns out to be a list of objects
#
kylewm and you can just grab the "value" property of the object
#
tantek that would seem to imply that flatter is better
#
kylewm so if you get a "content", expect it to be a list of objects with "html" and "value" and instead get a list of objects with arbitrary properties + value
#
kylewm in the flatter case, you might use the object not even realizing it's an h-card
#
kylewm depends if that is desireable
#
kylewm so i guess that's the use-case i think we need to understand -- when would you want "e-* h-*"
chiui joined the channel
#
tantek that's a good question - I can't think of any off the top of my head
#
tantek hmm - perhaps that webmentions "in-reply-to" use-case is why we decided that "u-in-reply-to h-cite" made more sense? and then the rule about the value of a u-* property with a nested object coming from the u-url of that nested object?
#
tantek really hoping we wrote that down
#
tantek I do remember discussing it and figuring it out - but not sure if IRC or at the microformats dev meetup
#
tantek found it - thanks again to you kylewm for bringing this up: http://microformats.org/wiki/microformats2-parsing-brainstorming#Nested_h-.2A_objects.27_.22value.22_property
#
tantek kylewm: see example expected result here: https://github.com/indieweb/php-mf2/issues/64#issuecomment-100676442
fuzzyhorns joined the channel
#
tantek re: "depends if that is desireable" - I think the answer is yes, it is.
#
tantek thus for an "e-content", you get a list of objects with "html" and "value", and if one of them has nested microformats, then that object also has "type" and "properties" for that nested object.
#
tantek seems like a pretty smooth enhancement for the nested object case, without disturbing those clients that are just looking at objects with "html" and "value"
fuzzyhorns joined the channel
#
tantek edited /microformats2-parsing (+118) "/* parse an element for class microformats */ make it explicit that e-* h-* re-uses the e-* { value:"" } structure (insetad of creating another level of { } structure) - per question/point from kylewm" (view diff)
#
tantek kylewm: made it explicit since your interpretation http://logs.glob.uno/?c=freenode%23microformats&s=10+May+2015&e=10+May+2015#c79315 was possible - but unintended (by the spec)
#
tantek since only one (new) parser supports this at all - this is likely ok
fuzzyhorns joined the channel
#
tantek edited /microformats2 (+51) "/* Python */ another text area unmung" (view diff)
#
kylewm edited /microformats2 (+57) "/* Python */ add another live parser" (view diff)
#
tantek thanks kylewm
Vendan joined the channel