private static string GenerateFeedItemKey(SimpleFeedlyFeedItem item)
{
if (string.IsNullOrWhiteSpace(item.Id))
{
if (string.IsNullOrWhiteSpace(item.Link))
{
return(null);
}
else
{
return(item.Link);
}
}
else
{
return(item.Id);
}
}
/// <summary>
/// GetFeedsFromChannel
/// </summary>
/// <param name="feedUrl">feed Url</param>
/// <param name="crawlerEngine">crawler engine</param>
/// <param name="isRest">
/// Normally we call this method two times, first times with 'default' channel's crawler engine, and the last times for the rest crawler engine
/// isRest = false: FIRST TIMES
/// isRest = true: LAST TIMES
/// </param>
/// <param name="engine"></param>
/// <param name="error"></param>
/// <returns></returns>
public static SimpleFeedlyFeed GetFeedsFromChannel(string feedUrl, RssCrawlerEngine?defaultCrawlerEngine, bool isRest, out RssCrawlerEngine engine, out Exception error)
{
error = null;
engine = RssCrawlerEngine.CodeHollowFeedReader;
SimpleFeedlyFeed result = new SimpleFeedlyFeed();
var items = new List <SimpleFeedlyFeedItem>();
var status = false;
foreach (RssCrawlerEngine engineLoop in (RssCrawlerEngine[])Enum.GetValues(typeof(RssCrawlerEngine)))
{
if (status)
{
break;
}
var canRun = false;
if (defaultCrawlerEngine == null)
{
canRun = true;
}
else
{
if (!isRest) // first time
{
canRun = engineLoop == defaultCrawlerEngine;
}
else
{
canRun = engineLoop != defaultCrawlerEngine;
}
}
if (!canRun)
{
continue;
}
if (engineLoop == RssCrawlerEngine.CodeHollowFeedReader && canRun)
{
if (!status)
{
try
{
var feed = CodeHollow.FeedReader.FeedReader.ReadAsync(feedUrl).GetAwaiter().GetResult();
foreach (var item in feed.Items)
{
var feedItem = new SimpleFeedlyFeedItem
{
Id = item.Id,
Title = string.IsNullOrWhiteSpace(item.Title) ? item.Link : item.Title,
Link = item.Link,
Description = item.Description,
PublishingDate = item.PublishingDate ?? DateTime.Now,
Author = item.Author,
Content = item.Content
};
items.Add(feedItem);
}
engine = RssCrawlerEngine.CodeHollowFeedReader;
status = true;
}
catch (Exception ex)
{
error = ex;
}
}
}
if (engineLoop == RssCrawlerEngine.SyndicationFeed && canRun)
{
if (!status)
{
try
{
XmlReaderSettings settings = new XmlReaderSettings();
settings.DtdProcessing = DtdProcessing.Parse;
using (var reader = XmlReader.Create(feedUrl, settings))
{
var feed = System.ServiceModel.Syndication.SyndicationFeed.Load(reader);
reader.Close();
foreach (System.ServiceModel.Syndication.SyndicationItem item in feed.Items)
{
var feedItem = new SimpleFeedlyFeedItem();
var link = item.Links.FirstOrDefault()?.Uri.ToString();
link = string.IsNullOrWhiteSpace(link) ? item.Id : link;
feedItem.Id = item.Id;
feedItem.Title = string.IsNullOrWhiteSpace(item.Title?.Text) ? link : item.Title.Text;
feedItem.Link = link;
feedItem.Description = item.Summary?.Text;
feedItem.PublishingDate = item.PublishDate.UtcDateTime;
feedItem.Author = item.Authors.FirstOrDefault()?.Name ?? string.Empty;
feedItem.Content = item.Content?.ToString();
items.Add(feedItem);
}
}
engine = RssCrawlerEngine.SyndicationFeed;
status = true;
}
catch (Exception ex)
{
error = ex;
}
}
}
if (engineLoop == RssCrawlerEngine.ParseRssByXml && canRun)
{
if (!status)
{
try
{
var xmlString = string.Empty;
using (WebClient client = new WebClient())
{
var htmlData = client.DownloadData(feedUrl);
xmlString = System.Text.Encoding.UTF8.GetString(htmlData);
// ReplaceHexadecimalSymbols
string r = "[\x00-\x08\x0B\x0C\x0E-\x1F\x26]";
xmlString = Regex.Replace(xmlString, r, "", RegexOptions.Compiled);
}
XmlDocument rssXmlDoc = new XmlDocument();
rssXmlDoc.LoadXml(xmlString);
// Parse the Items in the RSS file
XmlNodeList rssNodes = rssXmlDoc.SelectNodes("rss/channel/item");
// Iterate through the items in the RSS file
foreach (XmlNode rssNode in rssNodes)
{
var feedItem = new SimpleFeedlyFeedItem();
XmlNode rssSubNode = rssNode.SelectSingleNode("link");
feedItem.Link = rssSubNode != null ? rssSubNode.InnerText : null;
rssSubNode = rssNode.SelectSingleNode("title");
feedItem.Title = rssSubNode != null ? rssSubNode.InnerText : null;
feedItem.Title = string.IsNullOrWhiteSpace(feedItem.Title) ? feedItem.Link : feedItem.Title;
rssSubNode = rssNode.SelectSingleNode("description");
feedItem.Description = rssSubNode != null ? rssSubNode.InnerText : null;
rssSubNode = rssNode.SelectSingleNode("pubDate");
DateTime pubDate = DateTime.Now;
if (rssSubNode != null)
{
if (DateTime.TryParse(rssSubNode.InnerText, out DateTime tmpDate))
{
pubDate = tmpDate;
}
}
feedItem.PublishingDate = pubDate;
if (!string.IsNullOrWhiteSpace(feedItem.Link))
{
items.Add(feedItem);
}
}
engine = RssCrawlerEngine.ParseRssByXml;
status = true;
}
catch (Exception ex)
{
error = ex;
}
}
}
}
// isRest == false => if it's the first time, we maybe need to call 2nd times
// status == false => we maybe need to call 2nd times if current engines did not return anything
// defaultCrawlerEngine = null will process rss with all engines, therefor we don't need to call 2nd times
if (isRest == false && !status && defaultCrawlerEngine != null)
{
return(GetFeedsFromChannel(feedUrl, defaultCrawlerEngine, true, out engine, out error));
}
if (!status)
{
return(null);
}
else
{
error = null;
result.Items = items;
return(result);
}
}
