#microformats 2022-09-24

2022-09-24 UTC
jacky joined the channel
#
[jgarber] Relatedly, the `<source>` element also supports `srcset` so we may consider its addition to the special parsing rules that currently only apply to `img`.
#
[jgarber] https://developer.mozilla.org/en-US/docs/Web/HTML/Element/source
#
[jgarber] There may be an open issue or two on this in the parsing repo. I recall folks asking about marking up this element for u-photo, u-audio, or u-video.
jacky joined the channel
#
[jgarber] Actually, I misspoke: the `<source>` element’s `srcset` attribute is apparently only permitted within the `<picture>` element.
#
[tantek]4 I think the <source> element is a different kind of complexity, since it impacts more than just img parsing
jacky joined the channel
#
[jgarber] What kind of complexity do you have in mind? (I haven’t been tracking any conversations around `<source>` element parsing, so I’ve no doubt missed some relevant discussions…)
#
[snarfed]1 Parsed data looks great! I don't know how I'd prioritize density vs width descriptors, but probably not a big deal, I can choose arbitrarily
#
[jgarber] [snarfed] Excellent! Thanks for taking a look.
#
[jgarber] > I don’t know how I’d prioritize density vs width descriptors
#
[jgarber] I’m not sure how a consuming application would, either, but I do think there’s higher value in the MF2 parser doing the work of creating the structures shared above (vs. the MF2 parser returning a simple string from `srcset` and requiring every interested consuming app do the work).
ur5us and jacky joined the channel
#
[snarfed]1 Definitely!
gRegorLove_, jacky and gRegorLove__ joined the channel
#
[tantek]4 [jgarber] yes that's good reasoning about putting the parsing in one place, but it's beyond a "nice to have" since relative URLs in the string MUST be resolved by the parser into absolute URLs anyway
angelo, gRegorLove_, jacky, gRegorLove__, gRegor, Awii, barnaby, jacky__ and ur5us joined the channel
#
@RubygemsN micromicro (3.1.0): Extract microformats2-encoded data from HTML documents. https://rubygems.org/gems/micromicro (twitter.com/_/status/1573749664301522944)
#
[jgarber] ☝ There’s `srcset` parsing implementation number one out in the world. 😄
#
capjamesg jgarber++
#
Loqi jgarber has 2 karma in this channel over the last year (4 in all channels)
#
[jgarber] …and live on micromicro.cc: https://micromicro.cc/search?url=https%3A%2F%2Fsixtwothree.org%2Fphotos%2F256
#
capjamesg jgarber is there docs on srcset parsing?
#
capjamesg *are there
#
capjamesg I could try my hand at adding support in mf2py.
#
Loqi micromicro.cc
#
[jgarber] capjamesg I’ve been working off of: https://github.com/microformats/microformats2-parsing/issues/7
#
[jgarber] (in addition to the conversation here over the last few days)
#
Loqi [tantek] #7 Should u-* parsing special case img srcset?
#
[jgarber] I put together a small set of test files here: https://github.com/jgarber623/micromicro/tree/main/spec/support/fixtures/micromicro_test_suite
#
[jgarber] They include many of the examples we discussed here, including some oddball (or outright erroneous) use of the `srcset` attribute.
#
[jgarber] If (or when?) it’s eventually desirable, we could use those tests to update the official microformats/tests repository.
#
gRegor [jgarber]++
#
Loqi [jgarber] has 3 karma in this channel over the last year (5 in all channels)
#
gRegor Once we have 2+ implementations and rough consensus, we can edit the specification and add tests to microformats/tests
#
[tantek]4 [jgarber]++ nice work!
#
Loqi [jgarber] has 4 karma in this channel over the last year (6 in all channels)
#
[tantek]4 [jgarber] you also wrote in that issue #7: "The implementation matches the examples in my comment above along with feedback in the #microformats IRC/Slack channel." which is a good update on progress, however it would be better to fully capture your implementation design in GitHub with the the feedback incorporated, so additional implementers do not have to check microformats IRC logs to figure out what might have changed
#
[tantek]4 even better if you can add a GitHub comment with suggested diffs to the microformats2 parsing spec that would indicate the mf2json output you implemented
#
[tantek]4 also thanks for the tests++
#
Loqi tests has 1 karma in this channel over the last year (3 in all channels)
gRegorLove_ joined the channel
#
capjamesg I'll put together a PR for this into mf2py.
ur5us, roramigator and angelo_ joined the channel