#microformats 2019-08-24

2019-08-24 UTC
#
[tantek] Aaronpk can you file that parse poster on video similar to alt on img as a new issue on the mf2 parser with links to this discussion and related issues?
fluffy joined the channel; zakatonDiscord[m left the channel
#
[tantek] [KevinMarks] are we agreed on starting the "rel" repo in the microformats github org?
[jgmac1106], gRegorLove and [vendan] joined the channel
#
[vendan] so, is there a "blessed" implementation that I should be comparing my parser to?
#
aaronpk I think the php and python ones are the most mature? Maybe also the node one
#
KartikPrabhu [vendan]: I wouldn't consider them "blessed" but afaik the php and python ones have had a lot of work and discussion put into them. It would be great to have the test suite updated but someone has to take up that "job"
[fluffy] joined the channel
#
[vendan] I'm willing, I frankly enjoy writing tests/protocol/parser type stuff more then webapps
#
[vendan] I think step one would be to implement a system where I can run a chunk of html against multiple parsers, and then get a diff between all of them.
#
[vendan] then I can just throw the existing test suite against that, and start weeding through parsers disagreeing with each other and with the test suite.
gRegorLove and galaxie joined the channel
#
[vendan] so got a simple "comparison" site set up
#
[vendan] give it a chunk and it parses it with the php and python parsers, and then generates a diff of the json output from both:
#
vendan hrm, slack integration kinda fails when you start doing images
#
aaronpk Oh yea sorry, I still haven't updated that to their new api
#
aaronpk Used to work and they changed how images are handled in their api
#
aaronpk hashtag silo life
#
vendan hah, I do hate slack, but... gotta admit being able to paste images is kinda nice... :(
#
vendan well, the html is still.... extremely primitive, but: https://mf2.andyleap.dev/
#
vendan sad face, the json diff lib I'm using doesn't escape html, but is set up to output html... :(
#
Loqi *sniff*
#
vendan woot, first parsing diff: https://mf2.andyleap.dev/?input=%3Cdiv+class%3D%22h-entry%22%3E%3Cimg+src%3D%22images%2Fphoto.gif%22+%2F%3E%3C%2Fdiv%3E
#
vendan but, that's probably I didn't specify the base url, so that's my bad :(
#
vendan yeah, just a false alarm
#
@t ↩️ The #indieweb has #federated events and RSVPs using #Webmention and #microformats2 h-event. That Homebrew Website Club SF event was a demo thereof: https://tantek.com/2019/233/e1/homebrew-website-club-sf (note RSVPs) Quite good, yet as observed, insufficient for Google Calendar users https://tantek.com/t52B3 (twitter.com/_/status/1165138056661229568)
#
vendan next up is shove the test suite in, make it so it's "expected" is the base the others diff from, and then start exploring
gRegorLove and [prtksxna] joined the channel
#
@metameiert Microformats Analysis: https://php.microformats.io/. (twitter.com/_/status/1165157825229807616)
gRegorLove, [mapkyca], [jgmac1106] and [prtksxna] joined the channel
#
@jgmac1106 ↩️ Spending my Saturday morning redoing the research plan of #questiontheweb: http://jgregorymcverry.com/questionthewebresearchplan.html already cut the chatbots which saddens me, I think we can play a lot with microformats and HTML parsers to provide scaffolds to student writing (https://quickthoughts.jgregorymcverry.com/s/1iAfiV) (twitter.com/_/status/1165256702347927552)
[snarfed] joined the channel
#
[snarfed] good example in the wild of how the mf2 implied u-url rule can be error prone: http://pin13.net/mf2/?url=http%3A%2F%2Fjasonrbriggs.com%2Fmicro%2F2019%2F08%2F12%2F195738.html
#
[snarfed] post is http://jasonrbriggs.com/micro/2019/08/12/195738.html . doesn't actually set u-url to https://fed.brid.gy , it's just the first/only link
#
[snarfed] caused this downstream problem: https://github.com/snarfed/bridgy-fed/issues/56#issuecomment-524556379
#
[snarfed] implied properties, whee
mauz555, [KevinMarks], gRegorLove, [grantcodes] and KartikPrabhu joined the channel; DazaiDiscord[m] left the channel