bitcoinmagazine.nl feed issues

[AbelLykens]

https://bitcoinmagazine.nl/sitemap/news.xml

does not work for me. W3C thinks it is OK: https://validator.w3.org/feed/check.cgi?url=https%3A%2F%2Fbitcoinmagazine.nl%2Fsitemap%2Fnews.xml

Error I get:
malformed xml string. parsing error : DOMDocument::loadXML(): StartTag: invalid element name in Entity, line: 1 (2)

[SMillerDev]

You should disable auto-discovery when adding it because the self reference is wrong.

[AbelLykens]

That works. But why? What does/should auto discover do?

[Grotax]

It's a feature of the feed library that we use "feed-io".
It can search on websites for RSS links and returns them.

It was implemented in news to improve the usability for inexperienced users. Well and some websites do have an RSS feed but do not advertise it anywhere, instead of searching the source code myself I can just give the link to news.

The problem is that some feeds have a problem with the self-reference.

A feed is an XML document and it should include a link to itself, so that someone with the file knows where it came from, to look for updates and so on.

And some feeds put a link to their main website, which is not the idea of this reference.
Or they put a relative URL /my/cool/RSS, which is also not allowed by the standard.

And that leads to news not finding a valid URL.
The code could certainly be improved.

[SMillerDev]

Auto-discover sees what the current page references as the feed and imports that. For example it could read CNN.com and figure out how to get the rss feed.

It breaks however when an rss feed says it's in a different location than it actually is.

[AbelLykens]

Thanks both!