#microformats 2026-02-03

2026-02-03 UTC
gRegorLove_, gRegorLove__, gRegor, barnaby and [lazcorp] joined the channel
#
[lazcorp] I've been looking at http://openmentions.com and was wondering what microformat you'd use to tag/link from your blogpost to the relevant openmentions topic page. Would it be a link with a u-syndication class?
ehmry joined the channel
#
jkingweb I'm a bit frustrated now. A sentence in the VCP specification I thought meant one thing I realize now could mean something else entirely, but I don't have sufficient knowledge of the context in which it was written to say which makes more sense.
#
jkingweb "If by parsing the "value" element(s), at least a specific date has been found, then the "value" is overall valid, and the parser assembles the overall datetime value by..."
#
jkingweb I presume that "overall valid" mean that you should use the concatenation rules it goes on to describe, but what happens if it's not valid? I had assumed that VCP processing fails and the Mf2 algorithm continues as if there were no value classes, but it occurred to me last night that this was written before Mf2, so would you be expected to instead treat it as if it would a p- and use the normal concatenation rules in the general
#
[tantek] Yes jkingweb that was very likely written before mf2, and there may be some gaps or possibly inconsistencies that need fixing
#
jkingweb Actually now that I test the latter assumption, implementations don't do that, either. There's no consistency at all in the face of failure of date parsing. Most do completely nonsensical things.
#
[tantek] Can you drop a link to the specific subsection with that quoted text and I'll take a look in a bit
#
jkingweb https://microformats.org/wiki/value-class-pattern#Date_and_time_parsing
#
jkingweb Last paragraph of the section, just before the expected output formats. There's nothing wrong with the text itself, I don't think; it just leaves unsaid what to do if the assumption of validity doesn't hold.
#
jkingweb MicroMicro, the Go parser, and my parser assume VCP fails and continue with Mf2 processing. The Elixir parser crashes. The Rust parser proceeds as if non-datetime VCP rules apply, unless there's a valid time, which it then concatenates with an empty string (this is clearly wrong). The other six I have at hand are various kinds of weird.
[schmarty], saccharin, pfych, [pfefferle], barnaby, gRegor, [jeremycherfas], [artlung], [lazcorp], [snarfed], btrem, [Mike_Mai], Kupietz and [aciccarello] joined the channel