How to find images in Feed items

[arminreiter]

How to find images in Feed items

[VDWWD]

Hi Armin,

What i have done is the following:

XDocument xdoc = XDocument.Parse(reader.OriginalDocument);
XNamespace xns = xdoc.Root.GetDefaultNamespace();

string Image = "";
BaseFeedItem bfi = item.SpecificItem;

if (bfi.Element.Descendants().Any(x => x.Name.LocalName == xns + "content"))
{
    Image = bfi.Element.Descendants().First(x => x.Name.LocalName == xns + "content").Attribute("url").Value;
}
else if (bfi.Element.Descendants().Any(x => x.Name.LocalName == "enclosure"))
{
    Image = bfi.Element.Descendants().First(x => x.Name.LocalName == "enclosure").Attribute("url").Value;
}

The parsing to XDocument is necessary to get the XNamespace that is needed to find media:content.

And thanks for the great Library!
Regards,
Erwin

[arminreiter]

Hello @tustc79,

the reason for that is, that there is a RSS 2.0 standard and there are many extensions.

The RSS 2.0 standard says, that the image is stored in the enclosure tag.
But as already mentioned, there are extensions (tags). E.g. the itunes feeds have some extensions to store additional information like the ItunesOwner. The media tag that is used in the second RSS feed is such an extension. Most of the RSS 2.0 feeds use the enclosure tag, but there are also some that use one of the available exensions.
For such cases, each item has a property Element which contains all tags of the item. You can use that to read the image, but you have to select it on your own.

So to solve it, there are different ways:
1.) use the Element property and check it on your own
2.) extend the FeedReader so that it supports the media namespace and implement the image selection mode outside
3.) extend the FeedReader so that it supports the media namespace and extend it by the image selection (but please be aware that there are many feeds and those must still be supported).

I'll leave this issue open as I'll probably find some time in the future to find a solution for it and implement it.

best regards,
Armin

[VDWWD]

Another addition I made is the following snippet. In one aspnet application ReadAsync(string url) was working fine, while in another it resulted in a deadlock? (site never stopped loading).

        /// <summary>
        /// reads a feed from an url, not in async to escape possible deadlock. the url must be a feed. Use ParseFeedUrlsFromHtml to
        /// parse the feeds from a url which is not a feed.
        /// </summary>
        /// <param name="url">the url to a feed</param>
        /// <returns>parsed feed</returns>
        public static Feed ReadNotAsync(string url)
        {
            string feedContent = string.Empty;

            using (HttpClient client = new HttpClient())
            {
                HttpResponseMessage response = client.GetAsync(new Uri(url)).Result;

                if (response.IsSuccessStatusCode)
                {
                    feedContent = response.Content.ReadAsStringAsync().Result;
                }
            }

            return FeedParser.GetFeed(feedContent);
        }

[ceemafour]

I've raised a PR for this issue https://github.com/codehollow/FeedReader/pull/12.
Please review

[arminreiter]

thanks to @ceemafour for fixing this issue. Updated NuGet package 1.1.1.1 contains the fixes.

[aliusman]

Hi Armin,

What i have done is the following:

XDocument xdoc = XDocument.Parse(reader.OriginalDocument);
XNamespace xns = xdoc.Root.GetDefaultNamespace();

string Image = "";
BaseFeedItem bfi = item.SpecificItem;

if (bfi.Element.Descendants().Any(x => x.Name.LocalName == xns + "content"))
{
    Image = bfi.Element.Descendants().First(x => x.Name.LocalName == xns + "content").Attribute("url").Value;
}
else if (bfi.Element.Descendants().Any(x => x.Name.LocalName == "enclosure"))
{
    Image = bfi.Element.Descendants().First(x => x.Name.LocalName == "enclosure").Attribute("url").Value;
}

The parsing to XDocument is necessary to get the XNamespace that is needed to find media:content.

And thanks for the great Library!
Regards,
Erwin

Hi, Can you please explain how you made this work? Cheers