Solar/Swarm
2
1
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

💄 Optimized web articles

Browse Source
This commit is contained in:
LittleSheep
2025-06-26 18:34:51 +08:00
parent 1a137fbb6a
commit f170793928
7 changed files with 93 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
DysonNetwork.Sphere/Connection/WebReader/WebReaderService.cs
View File

@ -2,6 +2,7 @@ using System.Globalization;
using AngleSharp;
using AngleSharp.Dom;
using DysonNetwork.Sphere.Storage;
using HtmlAgilityPack;
namespace DysonNetwork.Sphere.Connection.WebReader;
@ -17,6 +18,30 @@ public class WebReaderService(
private const string LinkPreviewCachePrefix = "scrap:preview:";
private const string LinkPreviewCacheGroup = "scrap:preview";
public async Task<ScrapedArticle> ScrapeArticleAsync(string url, CancellationToken cancellationToken = default)
{
var linkEmbed = await GetLinkPreviewAsync(url, cancellationToken);
var content = await GetArticleContentAsync(url, cancellationToken);
return new ScrapedArticle
{
LinkEmbed = linkEmbed,
Content = content
};
}
private async Task<string?> GetArticleContentAsync(string url, CancellationToken cancellationToken)
{
var httpClient = httpClientFactory.CreateClient("WebReader");
var response = await httpClient.GetAsync(url, cancellationToken);
response.EnsureSuccessStatusCode();
var html = await response.Content.ReadAsStringAsync(cancellationToken);
var doc = new HtmlDocument();
doc.LoadHtml(html);
var articleNode = doc.DocumentNode.SelectSingleNode("//article");
return articleNode?.InnerHtml;
}
/// <summary>
/// Generate a link preview embed from a URL
/// </summary>