#microformats 2025-08-27

2025-08-27 UTC
jimmyjames, gotra, [tantek], [pfefferle], [Sophia_wood], [lazcorp], gRegor, failxontour, srxl, trwnh, srxl_, The_Decryptor, barnaby, [KevinMarks]1, [Sophia_wood]1, [artlung]1, [artlung], [KevinMarks], konsti, duanin2 and [snarfed] joined the channel
#
[snarfed] interesting mf2 failure on https://hidde.blog/component-conformance/ , they put u-url on the post title <span>
#
[snarfed] ...so parsing it results in "https://hidde.blog/component-conformance/Can components conform to WCAG" as the url property: https://pin13.net/mf2/?url=https://hidde.blog/component-conformance/
#
gRegor Ah interesting. Looks like "else get the textContent... return the normalized absolute URL of the gotten value following the containing document's language's rules for resolving relative URLs" https://microformats.org/wiki/microformats2-parsing#parsing_a_u-_property
#
gRegor Discussion in https://github.com/microformats/microformats2-parsing/issues/9
#
Loqi [preview] [sknebel] #9 "return the normalized absolute URL" for invalid URLs?
#
[snarfed] yeah I get how it got there, by adding the base URL
#
[snarfed] the more interesting point is that even with those kinds of transforms, afaik mf2 parsing doesn't really aim to verify parsed data against any formats or other constraints, so consumers always need to check themselves
[morganm], barnaby and [aciccarello] joined the channel
#
[tantek] [snarfed] did you try the indiewebify h-entry validator? What does it say (or fail to point out) ?
#
[snarfed] doesn't complain, but I guess I'm not too surprised, it's not really a comprehensive validator. https://indiewebify.me/validate-h-entry/?url=https://hidde.blog/component-conformance/
#
[tantek] Worth filing an issue on Indiewebify to improve its h-entry validator to catch that type of error