fix #584

[antoniosanct]

Hi, everyone:

I sent a new patch according to #584 because new 2.x master branch has come! Maybe it wouldn't be an issue. Please review and make your comments.

Regards,
Antonio.

[PatrickGotthard]

Hi @antoniosanct,

I think the comment element was intentionally not set on SyndFeed. When converting

<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
    <channel>
        <title>Channel Title</title>
        <link>https://example.com</link>
        <description>Channel Description</description>
        <item>
            <title>Item Title</title>
            <link>https://example.com/item</link>
            <description>Item Description</description>
            <comments>https://example.com/comments</comments>
        </item>
    </channel>
</rss>

to Atom 1.0, the following XML will be produced:

<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>Channel Title</title>
  <link rel="alternate" href="https://example.com" />
  <subtitle>Channel Description</subtitle>
  <entry>
    <title>Item Title</title>
    <link rel="alternate" href="https://example.com/item" />
    <link rel="comments" type="text/html" href="https://example.com/comments" />
    <author>
      <name />
    </author>
    <id>https://example.com/item</id>
    <summary type="html">Item Description</summary>
  </entry>
</feed>

According to RFC 4287 (The Atom Syndication Format) only IANA-registered link relations are allowed and there's simply no matching type for "comments".

Regards,
Patrick

[PatrickGotthard]

Hi @antoniosanct,

any update about this? Seems that we produce invalid Atom feeds when merging this...

Regards,
Patrick

[antoniosanct]

Hi, @PatrickGotthard

Hhmmm that's interesting... Then the test was bad implemented by this reason? If you think that the test is correct in his actual form, then close PR without merging it.

However, I only modified RSS 0.94 to pass the test. Does this behaviour affect the rest of converters? For example, TestSyndFeedAtom10.java @ rome use illegal rel values in atom_1.0.xml. If Rome controls which rel you can render, these tests (testEntry0, testEntry1) would be invalid by definition.

Regards,
Antonio.