Support attached media in `<enclosure>` and `<media:content>` #978

hirnsushi · 2025-03-20T11:13:54Z

hirnsushi
Mar 20, 2025

I'm subsribed to a few feeds (ArsTechnica, any Mastodon profile e.g. Nuclearpasta…) that have preview thumbnails in the article list but don't show these images in article view. The problem seems to be that these images aren't part of <content:encoded> or the <description>, but only show up in <media:content>.

Snippet from the ArsTechnica feed:

<media:content height="648" medium="image" type="image/jpeg" url="https://cdn.arstechnica.net/wp-content/uploads/2025/03/GettyImages-482951216-1152x648.jpg" width="1152">
        <media:thumbnail height="500" url="https://cdn.arstechnica.net/wp-content/uploads/2025/03/GettyImages-482951216-500x500.jpg" width="500"/>
        <media:credit>Getty | Bloomberg</media:credit>
        <media:text>Chickens stand in a barn in Sheffield, Illinois.</media:text>
</media:content>

Snippet from the Mastodon feed:

<media:content url="https://cdn.masto.host/mastodonart/media_attachments/files/114/190/226/759/324/930/original/9943fbbf67c4119c.jpg" type="image/jpeg" fileSize="305425" medium="image">
        <media:rating scheme="urn:simple">nonadult</media:rating>
        <media:description type="plain">Durt can't stop kicking ass. Durt! Durt! Durt!</media:description>
</media:content>

This isn't a huge deal for the ArsTechnica feed, it's only a partial feed anyway and loading full content will often show those article images on top. Full content extraction doesn't work for Mastodon feeds however, following for example comic artists or photographers etc. becomes basically impossible.

I would love an option (or a default) to show those preview images in the article view as well, preferably under the headline and before the main content 🙂

jocmp · 2025-03-20T18:19:18Z

jocmp
Mar 20, 2025
Maintainer

Hey @hirnsushi - thanks for the details here. The XML will be super useful for me to debug. I'll double check, but I don't think there's much media handling for local feeds beyond the custom YouTube parsing. I might be able to write something similar for Mastodon.

5 replies

hirnsushi Mar 21, 2025
Author

I apologize in advance for the infodump 😬

I don't think this should necessarily be specific to Mastodon, apart from ArsTechnica I have so far identified two other regular feeds in my list that exhibit similar issues:

AntennaPod Blog: Special case. Image in <media:content> is only used on https://antennapod.org/blog/ but doesn't show up in an article on the site directly. Not sure how to handle this, I'd show it in article view in Capy Reader regardless.
ArsTechnica: The image is used as a teaser on the site, is part of the header of an article (so arguably a major part of it), but the feed only mentions it in <media:content>.
NewPipe Blog: Yet another different case, the image used to illustrate the articles is in e.g. <enclosure url="https://newpipe.net/img/34c3-press.png" type="image/png" length="46253" />. Not part of <media:content> but rather it's precursor? This is also used in podcast feeds (e.g. Molly White's Citation Needed) so might be worth handling.

No idea how widespread the use of <media:content> is 🤔, it is part of the spec though, so it might show up here and there (and there's probably a bit of an overlap of people using RSS readers who are also subscribed to ArsTechnica)

From my limited understanding (I know nothing about Kotlin, only Bash) for YouTube feeds RichMedia.kt extracts information from <media:foo> and ParsedItem.kt replaces the content with that, if the feed contains media.
That wouldn't work for e.g. Mastodon feeds (or the regular feeds above) without major tweaking, because those contain <media:content> and regular content in <description>/<content:encoded>, here's an example from the Mastodon feed mentioned that shows a full <item> (and illustrates another problem, multiple images):

<item>
      <guid isPermaLink="true">https://mastodon.art/@nuclearpasta/114184747482894863</guid>
      <link>https://mastodon.art/@nuclearpasta/114184747482894863</link>
      <pubDate>Tue, 18 Mar 2025 18:02:55 +0000</pubDate>
      <description>&lt;p&gt;Chapter 3 was when I started to understand how to draw Joy.&lt;br /&gt;Sweaty Joy appreciation post, I guess.&lt;/p&gt;&lt;p&gt;&lt;a href="https://mastodon.art/tags/art" class="mention hashtag" rel="tag"&gt;#&lt;span&gt;art&lt;/span&gt;&lt;/a&gt; &lt;a href="https://mastodon.art/tags/webcomic" class="mention hashtag" rel="tag"&gt;#&lt;span&gt;webcomic&lt;/span&gt;&lt;/a&gt; &lt;a href="https://mastodon.art/tags/fantasy" class="mention hashtag" rel="tag"&gt;#&lt;span&gt;fantasy&lt;/span&gt;&lt;/a&gt; &lt;a href="https://mastodon.art/tags/comics" class="mention hashtag" rel="tag"&gt;#&lt;span&gt;comics&lt;/span&gt;&lt;/a&gt; &lt;a href="https://mastodon.art/tags/muscle" class="mention hashtag" rel="tag"&gt;#&lt;span&gt;muscle&lt;/span&gt;&lt;/a&gt;&lt;/p&gt;</description>
      <media:content url="https://cdn.masto.host/mastodonart/media_attachments/files/114/184/738/565/086/947/original/d585bbb622a6e8d4.jpg" type="image/jpeg" fileSize="72153" medium="image">
        <media:rating scheme="urn:simple">nonadult</media:rating>
        <media:description type="plain">Joy stands tall</media:description>
      </media:content>
      <media:content url="https://cdn.masto.host/mastodonart/media_attachments/files/114/184/738/597/541/088/original/9beba877ab5e7b9f.jpg" type="image/jpeg" fileSize="283165" medium="image">
        <media:rating scheme="urn:simple">nonadult</media:rating>
        <media:description type="plain">Joy runs from a lizard. She's goofy.</media:description>
      </media:content>
      <media:content url="https://cdn.masto.host/mastodonart/media_attachments/files/114/184/738/563/064/693/original/9ec98ab655e551d5.jpg" type="image/jpeg" fileSize="75142" medium="image">
        <media:rating scheme="urn:simple">nonadult</media:rating>
        <media:description type="plain">Joy ready to fight. </media:description>
      </media:content>
      <media:content url="https://cdn.masto.host/mastodonart/media_attachments/files/114/184/738/657/911/155/original/d7b5e0d5ef32a74d.jpg" type="image/jpeg" fileSize="258831" medium="image">
        <media:rating scheme="urn:simple">nonadult</media:rating>
        <media:description type="plain">detail of promo art for chapter 3. Joy faces the thugs.</media:description>
      </media:content>
      <category>art</category>
      <category>webcomic</category>
      <category>fantasy</category>
      <category>comics</category>
      <category>muscle</category>
</item>

The other issue is that Youtube feeds use <media:description> for the "shownotes" while in the Mastodon example they are used for the image's alt-text, and ArsTechnica doesn't provide alt-text in the feed but sometimes uses <media:text> for image subtitles that could be used for alt-text (from reading the spec that would be the correct place for alt-text anyway).

Having looked a bit closer at this problem I've realized it's far more complex than I initially imagined.
My proposed solution would be to insert media from <media:content>/<enclosure> before itemContent but that would need additional handling for filetype, a gallery or similar for multiple media files (max. 4 for Mastodon, but potentially a lot more), maybe a simple audio player for podcasts.
And then there's the problem with feeds that put the teaser in <media:content> and at the beginning of <content:encoded> which would duplicate it in article view (not that I found any with that behavior yet, but it's possible).

So yeah, I'd understand it if this feature request is out-of-scope 🙈

jocmp Mar 21, 2025
Maintainer

No need to apologize, the more details the better! That's a good point about the combination of media:content in addition to description. As you mentioned, the YouTube case assumes that the existing content can be replaced almost entirely.

Extrapolating, the general issue is that this metadata is usually feed-specific and hard to recombine with the main description or summary tags. My approach for better enclosures has been to write full-content parser. I recently added one for Bluesky. I can write a few more for these feeds if you notice images missing in the full content mode. I'd need to think of a solution for the Mastodon case since that involves sites that follow the same format but exist on separate domains.

hirnsushi Mar 22, 2025
Author

I don't think custom extractors are able to solve the general problem. You could ask the Mastodon WebFinger endpoint if the feed URL contains /user/foo or /@foo (example for the Mastodon profile mentioned above: https://mastodon.art/.well-known/webfinger?resource=acct:[email protected]) to determine if the feed leads to a Mastodon profile, but that would probably only work for (vanilla) Mastodon, there's countless of other ActivityPub software out there that also supports RSS feeds and will deliver content in <media:content> or <enclosure>.

A few examples:

GoToSocial example feed snippet

<item>
  <title>Hiya, we&#39;ve just done bugfix release 0.17.1 for #GoToSocial.&#xA;&#xA;https://github.com/superseriousbusiness/gotosocial/releases/tag...</title>
  <link>https://gts.superseriousbusiness.org/@gotosocial/statuses/01JAQJ25W45BNPZHX44GC2YVQ6</link>
  <description>@[email protected] posted 1 attachment: &#34;Hiya, we&#39;ve just done bugfix release 0.17.1 for #GoToSocial.&#xA;&#xA;https://github.com/superseriousbusiness/gotosocial/releases/tag/v0.17.1&#xA;&#xA;Fixes a bug in the streaming API that may have been causing...</description>
  <content:encoded><![CDATA[<p>Hiya, we've just done bugfix release 0.17.1 for <a href="https://gts.superseriousbusiness.org/tags/gotosocial" class="mention hashtag" rel="tag nofollow noreferrer noopener" target="_blank">#<span>GoToSocial</span></a>.</p><p><a href="https://github.com/superseriousbusiness/gotosocial/releases/tag/v0.17.1" rel="nofollow noreferrer noopener" target="_blank">https://github.com/superseriousbusiness/gotosocial/releases/tag/v0.17.1</a></p><p>Fixes a bug in the streaming API that may have been causing problems for some clients, and adds alt-text for default avatar + header images.</p><p>Updating from 0.17.0 is very simple and quick, but if you're updating from 0.16.0 or below then please read the migration notes carefully and be aware of the big database migrations from 0.17.0.</p><p>Enjoy!</p>]]></content:encoded>
  <author>@[email protected]</author>
  <enclosure url="https://gts.superseriousbusiness.org/fileserver/016VP4S3BP3QSJBACEQ6J1VFDX/attachment/original/01JAQJ11AV3AE9CSXF0PFNYV01.png" length="12845739" type="image/png"></enclosure>
  <guid isPermaLink="true">https://gts.superseriousbusiness.org/@gotosocial/statuses/01JAQJ25W45BNPZHX44GC2YVQ6</guid>
  <pubDate>Mon, 21 Oct 2024 14:50:34 +0200</pubDate>
  <source>https://gts.superseriousbusiness.org/@gotosocial/feed.rss</source>
</item>

PeerTube example feed snippet

<item>
    <title><![CDATA[Chorus Frog Serenades You]]></title>
    <link>https://lone.earth/w/wjp4jxYRMFY219timSDdft</link>
    <guid>https://lone.earth/w/wjp4jxYRMFY219timSDdft</guid>
    <pubDate>Mon, 10 Mar 2025 23:49:51 GMT</pubDate>
    <description><![CDATA[a chorus frog faces the camera and repeatedly inflates their throat to croak. in the background many other frogs can be heard]]></description>
    <content:encoded><![CDATA[<p>a chorus frog faces the camera and repeatedly inflates their throat to croak. in the background many other frogs can be heard</p>
    ]]></content:encoded>
    <dc:creator>nature</dc:creator>
    <category>Animals</category>
    <enclosure length="52358470" type="video/mp4" url="https://lone.earth/download/videos/generate/f583c1d7-121b-451d-97ee-39118f96881f?videoFileIds=560172"/>
    <media:community>
        <media:statistics views="19"/>
    </media:community>
    <media:embed url="https://lone.earth/videos/embed/wjp4jxYRMFY219timSDdft"/>
    <media:player url="https://lone.earth/w/wjp4jxYRMFY219timSDdft"/>
    <media:group>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/1d6bef5b-091f-456f-8943-79839e3a3663-2160.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/998fc455-f4cd-411a-b28f-48ac7e0c404d-720.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/e6594a36-ff09-4145-be0b-8a9d463afb60-480.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/e95c0111-5e14-45e0-ae97-2e5fbd7c3b51-144.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/fb6e0729-312b-45cf-8039-7eefa4394dc4-0.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/0ca5aca1-bac5-4750-87a7-30f643c5c761-2160-hls.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/07a2ef0b-b55f-4213-b8bc-73f89b836e77-720-hls.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/0f5f049b-116d-48af-99f6-117eb7093ec3-480-hls.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/f99d7fb2-d469-42b8-a9e0-4b7ba606ecda-144-hls.torrent" isDefault="false"/>
        <media:peerLink type="application/x-bittorrent" href="https://lone.earth/lazy-static/torrents/feeaa11b-27b7-4568-8952-35d95368502d-0-hls.torrent" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="2160" fileSize="52358470" url="https://lone.earth/static/web-videos/0b977812-d647-47e5-849d-d6dfaab17c03-2160.mp4" framerate="30" duration="27" isDefault="true"/>
        <media:content type="video/mp4" medium="video" height="720" fileSize="7858804" url="https://lone.earth/static/web-videos/4188d30c-80a1-4625-8e54-c357b8ecc989-720.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="480" fileSize="4383670" url="https://lone.earth/static/web-videos/e4f5e226-fd25-4386-9463-aa53e16d58f0-480.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="144" fileSize="1396815" url="https://lone.earth/static/web-videos/e18ceef5-339b-4479-8e45-bb2bce31f871-144.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="audio/mp4" medium="video" height="0" fileSize="873508" url="https://lone.earth/static/web-videos/33125596-958f-4195-99ad-9512637a1fa0-0.mp4" framerate="0" duration="27" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="2160" fileSize="52347348" url="https://lone.earth/static/streaming-playlists/hls/f583c1d7-121b-451d-97ee-39118f96881f/b7b65acd-1552-4255-86f3-1a99cb0f728b-2160-fragmented.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="720" fileSize="7848582" url="https://lone.earth/static/streaming-playlists/hls/f583c1d7-121b-451d-97ee-39118f96881f/45e67516-e014-4bb6-b9a5-248f37f4e368-720-fragmented.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="480" fileSize="4373744" url="https://lone.earth/static/streaming-playlists/hls/f583c1d7-121b-451d-97ee-39118f96881f/039ccc6a-1ce2-48e6-8514-26169125c241-480-fragmented.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="video/mp4" medium="video" height="144" fileSize="1385857" url="https://lone.earth/static/streaming-playlists/hls/f583c1d7-121b-451d-97ee-39118f96881f/c16466ea-6374-4dc0-9fd1-646a19736c15-144-fragmented.mp4" framerate="30" duration="27" isDefault="false"/>
        <media:content type="audio/mp4" medium="video" height="0" fileSize="874574" url="https://lone.earth/static/streaming-playlists/hls/f583c1d7-121b-451d-97ee-39118f96881f/44617fb3-4ae7-4164-9e3a-affd76855557-0-fragmented.mp4" framerate="0" duration="27" isDefault="false"/>
    </media:group>
    <media:thumbnail url="https://lone.earth/lazy-static/previews/3feeed83-cfac-4950-b2a7-abe687a94e9f.jpg"/>
    <media:thumbnail url="https://lone.earth/lazy-static/thumbnails/119b7306-e36c-4d40-91ab-1fa7aa372d3c.jpg"/>
    <media:rating>nonadult</media:rating>
    <media:title type="plain">Chorus Frog Serenades You</media:title>
    <media:description type="plain">a chorus frog faces the camera and repeatedly inflates their throat to croak. in the background many other frogs can be heard</media:description>
</item>

Pixelfed example feed snippet (notable because it actually has duplicate images in media:content and the main content, and it's an atom feed)

<entry>
  <id>https://pixelfed.art/p/abonbon/808378417108635329</id>
  <title>Happy Spring Equinox! 🌸

  #PixelArt #Spring #PixelFedArt #FediArt #MastoArt #DigitalArt</title>
  <updated>2025-03-20T16:47:13.000Z</updated>
  <author>
      <name>abonbon</name>
      <uri>https://pixelfed.art/abonbon</uri>
  </author>
  <content type="html">
    <![CDATA[
      <div class="media-gallery">
        <img class="media-item" src="https://media.pixelfed.art/public/m/_v2/789397385881629934/549719332-a3f277/NkAPurvfkam8/xypu1HAtxJc9tIKG7vbqe2R3bZxCF27fAjB7vyMN.png" alt="pixel art portrait of a lily of the valley fairy, she has pink hair, and holding a stalk of lily of the valley">
      </div>
      <p style="padding:10px;">Happy Spring Equinox! 🌸<br />
      <br />
      <a href="https://pixelfed.art/discover/tags/PixelArt?src=hash" title="#PixelArt" class="u-url hashtag" rel="external nofollow noopener">#PixelArt</a> <a href="https://pixelfed.art/discover/tags/Spring?src=hash" title="#Spring" class="u-url hashtag" rel="external nofollow noopener">#Spring</a> <a href="https://pixelfed.art/discover/tags/PixelFedArt?src=hash" title="#PixelFedArt" class="u-url hashtag" rel="external nofollow noopener">#PixelFedArt</a> <a href="https://pixelfed.art/discover/tags/FediArt?src=hash" title="#FediArt" class="u-url hashtag" rel="external nofollow noopener">#FediArt</a> <a href="https://pixelfed.art/discover/tags/MastoArt?src=hash" title="#MastoArt" class="u-url hashtag" rel="external nofollow noopener">#MastoArt</a> <a href="https://pixelfed.art/discover/tags/DigitalArt?src=hash" title="#DigitalArt" class="u-url hashtag" rel="external nofollow noopener">#DigitalArt</a></p>
    ]]>
  </content>
  <link rel="alternate" href="https://pixelfed.art/p/abonbon/808378417108635329" />
  <summary type="html">Happy Spring Equinox! 🌸&lt;br /&gt;
&lt;br /&gt;
&lt;a href=&quot;https://pixelfed.art/discover/tags/PixelArt?src=hash&quot; title=&quot;#PixelArt&quot; class=&quot;u-url hashtag&quot; rel=&quot;external nofollow noopener&quot;&gt;#PixelArt&lt;/a&gt; &lt;a href=&quot;https://pixelfed.art/discover/tags/Spring?src=hash&quot; title=&quot;#Spring&quot; class=&quot;u-url hashtag&quot; rel=&quot;external nofollow noopener&quot;&gt;#Spring&lt;/a&gt; &lt;a href=&quot;https://pixelfed.art/discover/tags/PixelFedArt?src=hash&quot; title=&quot;#PixelFedArt&quot; class=&quot;u-url hashtag&quot; rel=&quot;external nofollow noopener&quot;&gt;#PixelFedArt&lt;/a&gt; &lt;a href=&quot;https://pixelfed.art/discover/tags/FediArt?src=hash&quot; title=&quot;#FediArt&quot; class=&quot;u-url hashtag&quot; rel=&quot;external nofollow noopener&quot;&gt;#FediArt&lt;/a&gt; &lt;a href=&quot;https://pixelfed.art/discover/tags/MastoArt?src=hash&quot; title=&quot;#MastoArt&quot; class=&quot;u-url hashtag&quot; rel=&quot;external nofollow noopener&quot;&gt;#MastoArt&lt;/a&gt; &lt;a href=&quot;https://pixelfed.art/discover/tags/DigitalArt?src=hash&quot; title=&quot;#DigitalArt&quot; class=&quot;u-url hashtag&quot; rel=&quot;external nofollow noopener&quot;&gt;#DigitalArt&lt;/a&gt;</summary>
  <media:content
    url="https://media.pixelfed.art/public/m/_v2/789397385881629934/549719332-a3f277/NkAPurvfkam8/xypu1HAtxJc9tIKG7vbqe2R3bZxCF27fAjB7vyMN.png"
    type="image/png"
    medium="image"
  />
</entry>

YouTube can be separated from the other use cases since the feed specifies <feed xmlns:yt="http://www.youtube.com/xml/schemas/2015" […] in the beginning, the other feeds behave fairly consistently.
Generally there's media in <enclosure> and/or <media:content> that is used as (in the example feeds):

Header image: The image illustrates and is at the top of an article e.g. ArsTechnica, AntennaPod Blog, NewPipe Blog.
Main content: The media is the main content, additional text below or next to it on the original site. Instagram-like for Pixelfed, YouTube-like for Peertube, and the usual way podcast websites are displayed with an audio player at the top.
Attached content: Usually how social media sites like BlueSky, Mastodon, etc. display content, text with media below.

Showing the media between title and text content could potentially work for all of them, typical Twitter-style social media posts reference the media included so having it on top of the text would also make sense. Multiple images would have to be rendered as a gallery, for the PeerTube example the main video is marked with isDefault="true" and the rest isn't needed.
Metadata doesn't have to be too complicated, <media:text> for the media's alt-text since that's the spec, fall back to <media:description> if that isn't available.
No need to change anything about how text content is parsed, the feeds behave like regular feeds in that regard.
To detect duplicates setImageFromContent could be expanded to get a list of included media in the main content and discard additional media if the URL matches.

If I understand RssFeedHandler.kt correctly…

RssKeyword.Item.Enclosure.value -> {
                    val type = node.attr(RssKeyword.Item.Type.value)
                    when {
                        type.contains("image") -> {
                            channelFactory.articleBuilder.image(node.attr(RssKeyword.Url.value))
                        }

                        type.contains("audio") -> {
                            channelFactory.articleBuilder.audioIfNull(node.attr(RssKeyword.Url.value))
                        }

                        type.contains("video") -> {
                            channelFactory.articleBuilder.videoIfNull(node.attr(RssKeyword.Url.value))
                        }

                        type.isNullOrBlank() -> {
                            channelFactory.articleBuilder.image(node.text())
                        }
                    }
                }

…there's already code to handle <enclosure> but it's not actually used anywhere at the moment, right? 🤔
Could that be used as a starting point to handle <media:content> as well?

hirnsushi Mar 26, 2025
Author

I decided to take a step back and look at how this problem is handled elsewhere because reinventing the wheel is probably not a good idea in this case.. Maybe FreshRSS is a good example since that is supported as a backend.
So here is my limited understanding (I also don't know any PHP) how FreshRSS (and the simplepie library used) deals with that:

Item.php parses <enclosure> and <media:content> into separate attributes. I find it pretty compelling that <media:content> just gets lumped in with <enclosure> and everything just ends up in enclosures:

$this->data['enclosures'][] = $this->registry->create(Enclosure::class, [$url, $type, $length, null, $bitrate, $captions, $categories, $channels, $copyrights, $credits, $description, $duration, $expression, $framerate, $hashes, $height, $keywords, $lang, $medium, $player, $ratings, $restrictions, $samplingrate, $thumbnails, $title, $width]);

Entry.php then constructs HTML code from that and combines it with the regular content (I can't tell if it gets added before or after 🤔)

The parsing is pretty involved so maybe that should happen upstream in RSS-Parser? That whole problem feels very much out of scope for a simple RSS reader 🙈

An interim step could be to support one of the enclosures that are already parsed (maybe audio for people asking for podcast support, no fancy player, just HTML5 <audio> tags if that's possible, similar to how YouTube is handled) to see if it's possible to neatly integrate that in the UI.
If that works it could be expanded to the other types which would mean that if <media:content> gets parsed at some point it could be mapped to enclosure, similar to what FreshRSS is doing.

jocmp Mar 27, 2025
Maintainer

FreshRSS is a good example. I see what they're doing by appending the enclosures to the end. Here's a zoomed out view of Ars Technica for instance. The two images at the end are media:content.

<media:content height="648" medium="image" type="image/jpeg" url="https://cdn.arstechnica.net/wp-content/uploads/2023/07/cygnusng18-1152x648.jpg" width="1152">
  <media:thumbnail height="500" url="https://cdn.arstechnica.net/wp-content/uploads/2023/07/cygnusng18-500x500.jpg" width="500"/>
  <media:credit>NASA</media:credit>
  <media:text>Northrop Grumman's Cygnus spacecraft is pictured at the International Space Station. 
  </media:text>
</media:content>

It's possible to extract the content image alongside the post just like the YouTube example. Here's something I hacked together. I'll need a little more time to make the code shippable:

Before	After

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Support attached media in `<enclosure>` and `<media:content>` #978

{{title}}

Replies: 1 comment 5 replies

{{title}}

{{title}}

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

{{title}}

Select a reply

Support attached media in <enclosure> and <media:content> #978

hirnsushi Mar 20, 2025

Replies: 1 comment · 5 replies

jocmp Mar 20, 2025 Maintainer

hirnsushi Mar 21, 2025 Author

jocmp Mar 21, 2025 Maintainer

hirnsushi Mar 22, 2025 Author

hirnsushi Mar 26, 2025 Author

jocmp Mar 27, 2025 Maintainer

Support attached media in `<enclosure>` and `<media:content>` #978

hirnsushi
Mar 20, 2025

Replies: 1 comment 5 replies

jocmp
Mar 20, 2025
Maintainer

hirnsushi Mar 21, 2025
Author

jocmp Mar 21, 2025
Maintainer

hirnsushi Mar 22, 2025
Author

hirnsushi Mar 26, 2025
Author

jocmp Mar 27, 2025
Maintainer