Feature Summary
Content rewrite rules that use Twitter Cards and OpenGraph attribute values
What problem does this feature solve?
Some sites rely so much on JavaScript that basic scraping is not really an option. Some of them do provide useful tags for previews, but Miniflux can't use them because the data is presented as attribute values instead of "text nodes."
Think of Bluesky. An RSS item might point to a post but only show the text content in the description. Requesting the original post is not very useful, as its body won't contain the information you would like to see. If the post is about a picture, I have no way to see it unless I visit the website.
Proposed Solution
Again, considering Bluesky. It provides these in the <head> tag:
og:descriptionog:site_nameog:logotwitter:siteog:typeog:urlog:urlog:imagetwitter:descriptiontwitter:imagetwitter:cardtwitter:label1twitter:value1twitter:label2twitter:value2twitter:label3twitter:value3twitter:label4twitter:value4
I think Miniflux could request the original page and rewrite the content with rules that are able to parse these. So if these are the tags I'm interested in:
<meta property="twitter:description" content="I took this picture of a cat today.">
<meta property="twitter:image" content="https://cdn.foo.bar/baz">
Miniflux could take I took this picture of a cat today. and https://cdn.foo.bar/baz to do so something with them.
Alternatives Considered
No response
Additional Context
No response
Checklist
Feature Summary
Content rewrite rules that use Twitter Cards and OpenGraph attribute values
What problem does this feature solve?
Some sites rely so much on JavaScript that basic scraping is not really an option. Some of them do provide useful tags for previews, but Miniflux can't use them because the data is presented as attribute values instead of "text nodes."
Think of Bluesky. An RSS item might point to a post but only show the text content in the description. Requesting the original post is not very useful, as its body won't contain the information you would like to see. If the post is about a picture, I have no way to see it unless I visit the website.
Proposed Solution
Again, considering Bluesky. It provides these in the
<head>tag:og:descriptionog:site_nameog:logotwitter:siteog:typeog:urlog:urlog:imagetwitter:descriptiontwitter:imagetwitter:cardtwitter:label1twitter:value1twitter:label2twitter:value2twitter:label3twitter:value3twitter:label4twitter:value4I think Miniflux could request the original page and rewrite the content with rules that are able to parse these. So if these are the tags I'm interested in:
Miniflux could take I took this picture of a cat today. and https://cdn.foo.bar/baz to do so something with them.
Alternatives Considered
No response
Additional Context
No response
Checklist