如何从Feed链接对象数组中获取图片、PDF类型的目标URL

Instead of relying on fixed array indices (which breaks if the feed structure changes), we can iterate through all links in each entry and filter them based on their MIME type (or relation attribute, if needed). Here's how to rewrite your code:

Modified Code #

URL url = new URL("https://catalog.example.com/publicdomain/browse/en/recent.atom");
XmlReader xmlReader = null;
try {
    xmlReader = new XmlReader(url);
    SyndFeed feeder = new SyndFeedInput().build(xmlReader);
    
    // Use generic iterator for type safety
    for (Iterator<SyndEntry> entryIterator = feeder.getEntries().iterator(); entryIterator.hasNext();) {
        SyndEntry syndEntry = entryIterator.next();
        System.out.println("Author: " + syndEntry.getAuthor());
        
        // Iterate through all links in the entry
        for (SyndLink link : syndEntry.getLinks()) {
            String linkType = link.getType();
            
            // Filter links by MIME type: images and PDFs
            if (linkType != null && (linkType.startsWith("image/") || linkType.equals("application/pdf"))) {
                System.out.println("\nRelevant Link Found:");
                System.out.println("  MIME Type: " + linkType);
                System.out.println("  URL: " + link.getHref());
                
                // Optional: Check relation for thumbnail-specific images
                String linkRel = link.getRel();
                if (linkRel != null) {
                    System.out.println("  Relation: " + linkRel);
                }
            }
        }
        System.out.println("------------------------");
    }
} finally {
    if (xmlReader != null) xmlReader.close();
}

Key Changes Explained #

• No fixed indices: We loop directly through syndEntry.getLinks() instead of converting to an array and hardcoding positions. This makes the code resilient to feed structure changes.
• MIME type filtering: We check if the link's type starts with image/ (covers all image formats like JPEG, PNG) or exactly matches application/pdf (for PDFs). This is the most reliable way to identify content types.
• Optional relation check: If you specifically need thumbnails (like your original str[3] example), you can add an extra condition for the rel attribute (e.g., linkRel.equals("http://opds-example.org/image/thumbnail")).
• Type safety: Using Iterator<SyndEntry> avoids unnecessary casting and makes the code cleaner.

Example Output #

For your original thumbnail link, the output would look like:

Author: John Doe

Relevant Link Found:
  MIME Type: image/jpeg
  URL: https://covers.example.net/book/6709.jpg
  Relation: http://opds-example.org/image/thumbnail
------------------------

内容的提问来源于stack exchange，提问作者user2341596