#microformats 2019-08-13

2019-08-13 UTC
itsmekntDiscord[, guilbo, [timothy_chamber, [Rose], Tevya, IWSlackGateway, KartikPrabhu, [tonz], [tantek], [grantcodes], [Lewis_Cowles], [KevinMarks], jgmac1106, [pfefferle], [jgmac1106], [xavierroy], nloadholtes, TallTed, jackjamieson, Din_, [snarfed], vesper11, [schmarty], nilocDiscord[m] and sblinnDiscord[m] joined the channel
#
[snarfed] re conversation in #indieweb-dev on whether the category and url properties can be objects instead of strings...
#
[snarfed] i'm no mf2 parsing expert, but based on http://microformats.org/wiki/microformats2-parsing , sounds like u-* properties should always parse to plain string URLs
jackjamieson joined the channel
#
[snarfed] so the interesting question is how something with eg class="h-card u-url" should be parsed
jackjamieson joined the channel
#
[snarfed] sounds like just the plain string URL should go into the url property
#
[snarfed] should the full object also go into children?
jackjamieson joined the channel
#
[snarfed] i would think maybe yes...but this *else* on the parsing page ^ seems to say no: 'else add found elements that are microformats to the "children" array'
#
[snarfed] cc grantcodes
#
aaronpk u- means the "value" is a URL
#
aaronpk if it is a string value, then it'll be a string, but if it's an object, then the "value" of the object is the url
jackjamieson joined the channel
#
aaronpk that's what the "value" property is for
jackjamieson joined the channel
#
[snarfed] huh. so a composite object with a value property would be ok?
#
[snarfed] sgtm, but the last line of http://microformats.org/wiki/microformats2-parsing#parsing_a_u-_property seems to imply that it's always a plain string, right? "return the normalized absolute URL..."
#
aaronpk compare: https://pin13.net/mf2/?id=20190813170957307
#
aaronpk oh one more
#
aaronpk https://pin13.net/mf2/?id=20190813171036701
#
aaronpk look at the "value" of 2 and 3
#
[snarfed] right, i understand. just trying to reconcile that with the parsing spec
#
sknebel [snarfed]: that path only gets entered if it isn't a nested object
#
[snarfed] ok
#
aaronpk imo the parsing spec is not the best source of information for publishers or even consumers
#
[snarfed] oh sure. i was only in there because i was trying to determine whether mf2py was wrong
#
aaronpk gotcha
#
GWG I have commented on that before, aaronpk
im_saeed and [schmarty] joined the channel; truthDiscord[m] left the channel
#
[tantek] The point of a spec is to be precise and unambiguous, not a tutorial
#
[tantek] It should be the best source for precision and minutiae. It’s not for general questions “isn’t x always a y?”
#
GWG Yes
#
GWG But an implementation guide would also help
#
[tantek] GWG feel free to make suggestions on indieweb.org/microformats#Brainstorming for what should go into an implementation guide!
Left_Turn and [snarfed] joined the channel