Parse feeds in Python
For example, there is a popular podcast website Buzzsprout that uses Atom for its feeds. Each `item` has no `link` but only an `enclosure`. Here's [an example feed](https://feeds.buzzsprout.com/99850.rss): ```xml ... <item> <itunes:title>With Nick van der Kolk from Love + Radio #17</itunes:title> <title>With Nick van der Kolk from Love + Radio #17</title> <description>...</description> <content:encoded>...</content:encoded> <itunes:author>Andy Clark en Richard den Haring</itunes:author> <enclosure url="https://www.buzzsprout.com/99850/2968933-with-nick-van-der-kolk-from-love-radio-17.mp3" length="20004212" type="audio/mpeg" /> <guid isPermaLink="false">Buzzsprout-2968933</guid> <pubDate>Tue, 10 Mar 2020 02:00:00 -0400</pubDate> <itunes:duration>1665</itunes:duration> <itunes:keywords></itunes:keywords> <itunes:episodeType>bonus</itunes:episodeType> <itunes:explicit>false</itunes:explicit> </item> ... ``` Since there is no link but a **single** enclosure with a URL, this can be used as the definitive `entries[i].link` for this entry. This would parallel how both the `<link>` and enclosures are [mixed together](https://feedparser.readthedocs.io/en/latest/reference-entry-links.html) to produce `entries[i].links`. If there is more than 1 enclosure with a URL, then we can fall back to the current thing since we have no way of knowing which URL would be the primary. But in this case with a single URL, it's clear.
This issue appears to be discussing a feature request or bug report related to the repository. Based on the content, it seems to be still under discussion. The issue was opened by melyux and has received 1 comments.