Newsbot.Collector/Newsbot.Collector.Services/HtmlParser/HeadParserClient.cs

201 lines
5.6 KiB
C#
Raw Normal View History

using HtmlAgilityPack;
using Newsbot.Collector.Domain.Exceptions;
namespace Newsbot.Collector.Services.HtmlParser;
public class HeadParserClient
{
private const string XPathHeadMetaTag = "//head/meta";
private const string XPathBodyMetaTag = "//body/meta";
private const string XPathLinkTag = "//head/link";
private readonly string _htmlContent;
public HeadParserClient(string htmlContent, bool useBrowser = false)
{
_htmlContent = htmlContent;
Data = new HeadParserModel();
}
public HeadParserModel Data { get; set; }
public void Parse()
{
Data.Title = GetMetaTitle();
Data.Description = GetMetaDescription();
Data.Image = GetMetaImage();
Data.Url = GetMetaUrl();
Data.PageType = GetMetaPageType();
Data.ColorTheme = GetMetaColorTheme();
Data.FeedUri = GetSiteFeed();
Data.YoutubeChannelID = GetYouTubeChannelId();
}
private List<HtmlNode> CollectMetaTags()
{
var htmlDoc = new HtmlDocument();
htmlDoc.LoadHtml(_htmlContent);
var allTags = new List<HtmlNode>();
var headerTags = htmlDoc.DocumentNode.SelectNodes(XPathHeadMetaTag).ToList();
allTags.AddRange(headerTags);
try
{
var bodyTags = htmlDoc.DocumentNode.SelectNodes(XPathBodyMetaTag).ToList();
allTags.AddRange(bodyTags);
}
catch
{
// no tags found in the body and that's ok.
// we check the body thanks to Youtube.
}
return allTags;
}
private List<HtmlNode> CollectLinkTags()
{
var htmlDoc = new HtmlDocument();
htmlDoc.LoadHtml(_htmlContent);
var links = htmlDoc.DocumentNode.SelectNodes(XPathLinkTag).ToList();
return links;
}
private string GetTagValue(string Tag, List<HtmlNode> html)
{
foreach (var meta in html)
{
if (meta.Attributes.Count == 0) continue;
;
//Console.WriteLine($"Name={meta.Attributes[0].Name} & Value={meta.Attributes[0].Value}");
if (meta.Attributes[0].Value.Contains(Tag) == false) continue;
return meta.Attributes[1].Value;
}
return "";
}
private string FindFirstResult(string[] tags, List<HtmlNode> htmlTags)
{
foreach (var tag in tags)
{
var res = GetTagValue(tag, htmlTags);
if (res == "") continue;
return res;
}
return "";
}
public string GetMetaTitle()
{
var htmlTags = CollectMetaTags();
string[] tags = { "twitter:title", "og:title", "title" };
return FindFirstResult(tags, htmlTags);
}
public string GetMetaDescription()
{
var htmlTags = CollectMetaTags();
string[] tags = { "twitter:description", "og:description", "description" };
return FindFirstResult(tags, htmlTags);
}
public string GetMetaImage()
{
var htmlTags = CollectMetaTags();
string[] tags = { "twitter:image", "og:image", "image" };
return FindFirstResult(tags, htmlTags);
}
public string GetMetaUrl()
{
var htmlTags = CollectMetaTags();
string[] tags = { "twitter:url", "og:url", "url" };
return FindFirstResult(tags, htmlTags);
}
public string GetMetaPageType()
{
var htmlTags = CollectMetaTags();
string[] tags = { "og:type", "type" };
return FindFirstResult(tags, htmlTags);
}
public string GetMetaColorTheme()
{
var htmlTags = CollectMetaTags();
string[] tags = { "theme-color" };
return FindFirstResult(tags, htmlTags);
}
public string GetYouTubeChannelId()
{
var htmlTags = CollectMetaTags();
string[] tags = { "og:url", "channelId" };
var results = FindFirstResult(tags, htmlTags);
var id = results.Replace("https://www.youtube.com/channel/", "");
return id;
}
/// <summary>
/// This will parse the headers looking for known keys that will contain a RSS feed link.
/// If the feed is not found, this will throw an exception (MissingHeaderValueException).
/// </summary>
/// <returns></returns>
public string GetSiteFeed()
{
var htmlTags = CollectLinkTags();
var tags = new[] { "alternate" };
try
{
var attr = FindFirstAttribute(tags, htmlTags);
foreach (var item in attr)
{
if (item.Name != "href") continue;
var uri = item.Value;
if (uri.StartsWith("//")) uri = uri.Replace("//", "https://");
return uri;
}
return "";
}
catch
{
// not found
return "";
}
}
private HtmlAttributeCollection FindFirstAttribute(string[] tags, List<HtmlNode> htmlTags)
{
foreach (var tag in tags)
try
{
var res = GetValidAttribute(tag, htmlTags);
return res;
}
catch (MissingHeaderValueException)
{
// Nothing was found in the given tag but we will keep looking till we finish all the entries.
}
throw new MissingHeaderValueException("Unable to find the requested value");
}
private HtmlAttributeCollection GetValidAttribute(string Tag, List<HtmlNode> html)
{
foreach (var meta in html)
{
if (meta.Attributes[0].Value.Contains(Tag) == false) continue;
return meta.Attributes;
}
throw new MissingHeaderValueException("Site does not expose requested tag.");
}
}