#microformats 2015-05-17

2015-05-17 UTC
KevinMarks and adiabatic joined the channel
#
adiabatic not that schema.org things and microformats things are mutually exclusive, but…why would I prefer to mark things up with microformats instead of schema.org things?
#
tantek adiabatic: because it's less work, and microformats are far minimal and well thought out
#
tantek as well based on existing standards, instead of making up whatever Google wants to make up (which they have a history of doing and abandoning)
KartikPrabhu joined the channel
#
adiabatic ah, ok.
KartikPrabhu and fuzzyhorns joined the channel
#
tantek adiabatic: some of this (Google's various efforts, that they've abandoned one after another) is documented in this blog post, along with others: http://microformats.org/2012/06/25/microformats-org-at-7
KartikPrabhu, KevinMarks__ and KevinMarks joined the channel
#
kylewm edited /microformats2 (+0) "/* Parsers */ alpha sort" (view diff)
fuzzyhorns and KartikPrabhu joined the channel
#
KartikPrabhu would be nice to have a universal test-suite for microformats so that we can add data about how far along each parser is to the wiki
#
Vendan I was working over one, but I've run into multiple issues with it
#
Vendan including stuff that wasn't being tested, but should be
#
KartikPrabhu kylewm: why does your hosted parser sort the dictionary in a different way that mine/KevinMarks' ?
#
kylewm don't know
#
kylewm python 3?
#
KartikPrabhu Vendan: that^ is also a potential problem. How does one compare the final output when even 2 runs of mf2py yield differently sorted results
#
Vendan was basing it off of https://github.com/tobiastom/tests, but I'm not sure about certain things
#
KartikPrabhu kylewm: possible
#
kylewm dict's .equals should handle that
#
Vendan KartikPrabhu, in golang, I can do a deepequal that doesn't care about ordering of maps like that
#
Vendan only issue is it gets too harsh, so if you add features, that counts as a fail
#
KartikPrabhu kylewm: but presumably a universal mf parser test would not use a dict yes?
#
Vendan i.e. rel-urls support would currently cause fails
#
Vendan not sure that that's an issue though
#
KartikPrabhu Vendan: that is fine. because it does fail the rel-urls test
#
KartikPrabhu maybe there should be a way to say "wrong parsing" vs "not supported feature"
#
kylewm KartikPrabhu: Im not sure I understand the problem. Any language that compares maps should handle arbitrary ordering
#
KartikPrabhu kylewm: so you think it should be possible to just have a text or json file for final output and python can read both the HTML source and the expected output?
#
KartikPrabhu and compare them? I guess that is what the tests in mf2py are currently doing?
#
Vendan that's what the above tests are set up for
#
KartikPrabhu oh ok! Vendan then it is just that those tests are out of date
#
KartikPrabhu ?
#
Vendan yeah
#
KartikPrabhu I see
#
Vendan and getting more so as time goes on, unfortunately
#
KartikPrabhu tbh I am a programming n00b by most accounts so I don't really know how these things work :P
#
Vendan see: https://travis-ci.org/andyleap/microformats
#
Vendan that's my parser running against those tests, I've lost about 4 recently due to changes in the spec
#
KartikPrabhu Vendan: maybe you could update the tests as a fork of https://github.com/tantek/tests and send PRs. I am sure tantek would appreciate it
#
KartikPrabhu I would try but given my limited experience with testing code I am sure I'll mess something up
#
kylewm glennjones is the author of the test suite
#
kylewm tobiastom just processed them to split them into json and html files
#
kylewm (that's not necessarily in reference to anything, just a general fyi)
#
KartikPrabhu yeah i guessed as much :)
#
Vendan there's some of the test fixes so far
#
Vendan I'm honestly developing the parser as I need it for my indieweb stuff at this point, unless someone else wants to use it
KartikPrabhu joined the channel
#
Vendan oh, you were gone
#
Vendan there's some of the test fixes so far
#
Vendan I'm honestly developing the parser as I need it for my indieweb stuff at this point, unless someone else wants to use it
#
Vendan pushed some of my test fixes up at https://github.com/andyleap/tests
#
KartikPrabhu Vendan: I re-read the logs :) my connection is a bit spotty
#
Vendan ah, ok
#
KartikPrabhu Vendan++ for updating tests.
#
Loqi Vendan has 12 karma
#
Vendan that's one reason I use znc on my vps to connect
KartikPrabhu, elf-pavlik, tantek, Left_Turn, chiui, KevinMarks__, KevinMarks, eschnou, fuzzyhorns and elf-pavlik_ joined the channel
#
@bendiken RT @graphityhq: http://Schema.org 2.0 vocabulary public release: http://blog.schema.org/2015/05/schema.html #SEO #RDF #LinkedData #microformats (twitter.com/_/status/600032931042623488)
fuzzyhorns and KartikPrabhu joined the channel
#
@ustoll RT @graphityhq: http://Schema.org 2.0 vocabulary public release: http://blog.schema.org/2015/05/schema.html #SEO #RDF #LinkedData #microformats (twitter.com/_/status/600052152338374657)
KartikPrabhu, KevinMarks__, fuzzyhorns, KevinMarks, gRegorLove and Atamido joined the channel