fork download
copy 
  1. using System.Collections.Generic;
  2. using System.IO;
  3. using System.Linq;
  4. using System.Text;
  5. using System.Xml;
  6. using System.Xml.Linq;
  7. // System.Web
  8. using System.Net;
  9. // SGMLReader
  10. // http://d...content-available-to-author-only...h.com/SgmlReader
  11. using Sgml;
  12.  
  13. namespace Ranking
  14. {
  15.     class Program
  16.     {
  17.         static void Main(string[] args)
  18.         {
  19.             string rankingUrl = "http://w...content-available-to-author-only...o.jp/ranking";
  20.  
  21.             var doc = LoadHtml(rankingUrl);
  22.  
  23.             var ranking = from x in doc.Descendants("div")
  24.                           let a = x.Attribute("class")
  25.                           where a != null && a.Value == "ranking_box"
  26.                           // <a href="watch/sm10000000"... → sm10000000
  27.                           select x.Descendants("a").First().Attribute("href").Value.Substring(6);
  28.  
  29.             /* ranking.xml
  30.              * <ranking>
  31.              *   <video id="sm*">
  32.              *     <tag>*</tag>...
  33.              *   </video>...
  34.              * </ranking>
  35.              */
  36.             var cache = new XDocument(
  37.                 new XElement("ranking",
  38.                 from id in ranking
  39.                 select new XElement("video",
  40.                     new XAttribute("id", id),
  41.                     from tag in GetTags(id)
  42.                     select new XElement("tag") { Value = tag })));
  43.  
  44.             cache.Save("ranking.xml");
  45.         }
  46.  
  47.         static IEnumerable<string> GetTags(string videoId)
  48.         {
  49.             string getthumbinfoUrl = "http://e...content-available-to-author-only...o.jp/api/getthumbinfo/";
  50.  
  51.             var doc = XDocument.Load(getthumbinfoUrl + videoId);
  52.             return from x in doc.Descendants("tag")
  53.                    select x.Value;
  54.         }
  55.  
  56.         static XDocument LoadHtml(string url)
  57.         {
  58.             using (var stream = new WebClient().OpenRead(url))
  59.             using (var reader = new StreamReader(stream, Encoding.UTF8))
  60.             {
  61.                 return ParseHtml(reader);
  62.             }
  63.         }
  64.  
  65.         static XDocument ParseHtml(TextReader reader)
  66.         {
  67.             var sgmlReader = new SgmlReader()
  68.             {
  69.                 DocType = "HTML",
  70.                 WhitespaceHandling = WhitespaceHandling.All,
  71.                 CaseFolding = CaseFolding.ToLower,
  72.                 InputStream = reader,
  73.             };
  74.             return XDocument.Load(sgmlReader);
  75.         }
  76.     }
  77. }
  78.  
Not running #stdin #stdout 0s 0KB
comments (?)
stdin
copy 
Standard input is empty
stdout
copy 
Standard output is empty
https://ideone.com/h5ZOU
language:
C# (gmcs 5.20.1)
created:
11 years ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!