Right now, we evidently don't have a way to exclude individual elements inside p-name or e-content or similar properties from contributing to their parsed text value. Image captions are an example: https://snarfed.org/2024-10-30_53932 contains an image with text caption inside e-content, and the caption is purely presentational, so ideally it would be omitted from the parsed content: https://pin13.net/mf2/?url=https://snarfed.org/2024-10-30_53932
@tantek proposes that we expand the mf2 parsing rules to exclude elements with role="presentation" from parsing. Works for me.
Right now, we evidently don't have a way to exclude individual elements inside
p-nameore-contentor similar properties from contributing to their parsed text value. Image captions are an example: https://snarfed.org/2024-10-30_53932 contains an image with text caption insidee-content, and the caption is purely presentational, so ideally it would be omitted from the parsedcontent: https://pin13.net/mf2/?url=https://snarfed.org/2024-10-30_53932@tantek proposes that we expand the mf2 parsing rules to exclude elements with
role="presentation"from parsing. Works for me.https://chat.indieweb.org/microformats/2024-11-01#t1730496257670000