EntityName解析错误- XPathDocument

Question

我正在尝试解析来自不同来源的xml feed，一些新闻feed遵循rss2.0标准，另一些则来自不同的来源，如twitter、facebook (这实际上为我提供了rss2.0)和linkedIn。在我把facebook加入进来之前，一切都运行得很完美。我传递了rss20作为格式，因此它应该遵循与普通rss20相同的标准，并且正好适合我的代码，但是它抛出了错误的An error occurred while parsing EntityName并引用了这一行……

XPathNavigator xpn = new XPathDocument(Server.UrlDecode(XmlHelper.GetXmlFeedUrl(appSettings[x]))).CreateNavigator();

在我的研究中，我发现facebook feed (research)没有标题集。但我不确定这是否适用，或者是否有更好的方法来做这件事。我曾考虑过SyndicationFeed，但由于推特不支持atom或rss，我认为它不会起作用。

这是facebook的网址..。

http://www.facebook.com/feeds/page.php?format=rss20%26id=6198772858

这是我的代码。

protected void Page_Load(object sender, EventArgs e)
{
    XmlFeedItemPath xfip;
    NameValueCollection appSettings = ConfigurationManager.AppSettings;
    List<string> xmlFeeds = appSettings.AllKeys.Where(x => x.StartsWith("XmlFeed")).ToList();
    string currentXmlFeedType;
    if(!string.IsNullOrEmpty(xmlFeedType))
        xmlFeeds.RemoveAll(s => !appSettings[s].Contains(XmlFeedType));

    xmlFeeds.ForEach(x =>
    {   
        currentXmlFeedType = XmlHelper.GetXmlFeedType(appSettings[x]);
        xfip = XmlHelper.GetXmlFeedItemPath(currentXmlFeedType);
        XPathNavigator xpn = new XPathDocument(Server.UrlDecode(XmlHelper.GetXmlFeedUrl(appSettings[x]))).CreateNavigator();
        XmlNamespaceManager xmlnsm = XmlHelper.GetXmlNameSpaceManager(xpn);
        XPathNodeIterator nodes = xpn.Select(xfip.IteratorPath, xmlnsm);
        int i = 0;
        foreach (XPathNavigator node in nodes)
        {
            XmlFeedItems.Add(new XmlFeedItem()
            {
                Title = string.IsNullOrEmpty(xfip.TitlePath) ? xfip.DefaultTitle : node.SelectSingleNode(xfip.TitlePath, xmlnsm).ToString(),
                Link = string.IsNullOrEmpty(xfip.LinkPath) ? null : node.SelectSingleNode(xfip.LinkPath, xmlnsm).ToString(),
                Teaser = string.IsNullOrEmpty(xfip.TeaserPath) ? null : XmlHelper.WrapUrlWithAnchorTags(node.SelectSingleNode(xfip.TeaserPath, xmlnsm).ToString()),
                Source = string.IsNullOrEmpty(xfip.SourcePath) ? null : xpn.SelectSingleNode(xfip.SourcePath, xmlnsm).ToString(),
                SortOrder = i,
                XmlFeedType = currentXmlFeedType 
            });
            i++;
        }
    });

    rptRssFeed.DataSource = XmlFeedItems.OrderBy(x => x.SortOrder).Take(10);
    rptRssFeed.DataBind();
}

Answer 1

我想通了。由于facebook feed没有useragent字符串，我必须更改我的实现以使用HttpWebRequest，并手动为每个xml feed (包括facebook)设置一个useragent……

protected void Page_Load(object sender, EventArgs e)
{
    XmlFeedItemPath xfip;
    NameValueCollection appSettings = ConfigurationManager.AppSettings;
    List<string> xmlFeeds = appSettings.AllKeys.Where(x => x.StartsWith("XmlFeed")).ToList();
    string currentXmlFeedType;
    if(!string.IsNullOrEmpty(xmlFeedType))
        xmlFeeds.RemoveAll(s => !appSettings[s].Contains(XmlFeedType));

    xmlFeeds.ForEach(x =>
    {   
        currentXmlFeedType = XmlHelper.GetXmlFeedType(appSettings[x]);
        xfip = XmlHelper.GetXmlFeedItemPath(currentXmlFeedType);

        var request = (HttpWebRequest)WebRequest.Create(Server.UrlDecode(XmlHelper.GetXmlFeedUrl(appSettings[x])));
        request.Method = "GET";
        request.UserAgent = "Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0)";
        XPathNavigator xpn = new XPathDocument(XmlReader.Create(request.GetResponse().GetResponseStream())).CreateNavigator();
        XmlNamespaceManager xmlnsm = XmlHelper.GetXmlNameSpaceManager(xpn);
        XPathNodeIterator nodes = xpn.Select(xfip.IteratorPath, xmlnsm);

        int i = 0;
        foreach (XPathNavigator node in nodes)
        {
            string publishDate = string.IsNullOrEmpty(xfip.PublishDatePath) ? null : node.SelectSingleNode(xfip.PublishDatePath, xmlnsm).ToString();
            XmlFeedItems.Add(new XmlFeedItem()
            {
                Title = string.IsNullOrEmpty(xfip.TitlePath) ? xfip.DefaultTitle : node.SelectSingleNode(xfip.TitlePath, xmlnsm).ToString(),
                Link = string.IsNullOrEmpty(xfip.LinkPath) ? null : node.SelectSingleNode(xfip.LinkPath, xmlnsm).ToString(),
                Teaser = string.IsNullOrEmpty(xfip.TeaserPath) ? null : XmlHelper.WrapUrlWithAnchorTags(HttpUtility.HtmlDecode(node.SelectSingleNode(xfip.TeaserPath, xmlnsm).ToString())),
                Source = string.IsNullOrEmpty(xfip.SourcePath) ? null : xpn.SelectSingleNode(xfip.SourcePath, xmlnsm).ToString(),
                SortOrder = i,
                XmlFeedType = currentXmlFeedType,
                PublishDate = string.IsNullOrEmpty(publishDate) ? new DateTime() : DateTime.Parse(publishDate.Remove(publishDate.IndexOf(" +")))
            });
            i++;
        }
    });

    rptRssFeed.DataSource = XmlFeedItems.OrderBy(x => x.GetType().GetProperty(sortField)).Take(10);
    rptRssFeed.DataBind();
}