yly
/
NewGaoKaoApi
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

bug fixed

develop
old易 2024-12-11 15:53:56 +08:00
parent fc488ccf1e
commit 8d4b28199a
1 changed files with 14 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
New_College.Tasks/HtmlAgSpider/NationWideNewsAgHelper.cs
View File

@ -11,6 +11,7 @@ using System.Threading.Tasks;
using System.Text.Json; using System.Text.Json;
using New_College.Common.Helper; using New_College.Common.Helper;
using System.Threading; using System.Threading;
using System.Text.RegularExpressions;
namespace New_College.Tasks namespace New_College.Tasks
{ {
public class NationWideNewsAgHelper public class NationWideNewsAgHelper
@ -35,7 +36,7 @@ namespace New_College.Tasks
{ {
resultlist.msg.ForEach(o => resultlist.msg.ForEach(o =>
{ {
Thread.Sleep(100); Thread.Sleep(100);
string newsUrl = $"{itemUrl}{o.uri}"; string newsUrl = $"{itemUrl}{o.uri}";
HtmlDocument doc = webClient.Load(newsUrl); HtmlDocument doc = webClient.Load(newsUrl);
var inntertitle = doc.DocumentNode.SelectSingleNode("//*[@id=\"app\"]/div[2]/h2").InnerText; var inntertitle = doc.DocumentNode.SelectSingleNode("//*[@id=\"app\"]/div[2]/h2").InnerText;
@ -54,19 +55,19 @@ namespace New_College.Tasks
//if (!HtmlHelper.ReplaceHtmlTag(innerhtml).Contains("浏览器")) //if (!HtmlHelper.ReplaceHtmlTag(innerhtml).Contains("浏览器"))
//{ //{
if (innerhtml.Contains(".pdf") || innerhtml.Contains(".doc") || innerhtml.Contains(".docx") || innerhtml.Contains(".xls") || innerhtml.Contains(".xlsx")) if (innerhtml.Contains(".pdf") || innerhtml.Contains(".doc") || innerhtml.Contains(".docx") || innerhtml.Contains(".xls") || innerhtml.Contains(".xlsx"))
{ {
innerhtml = $"{innerhtml}\n若有附件详情，请至本省招生考试院下载附件!!!"; innerhtml = $"{innerhtml}\n若有附件详情，请至本省招生考试院下载附件!!!";
} }
list.Add(new NewsModels() list.Add(new NewsModels()
{ {
title = inntertitle, title = inntertitle,
author = author, author = author,
pubtime = Convert.ToDateTime(createtime), pubtime = Convert.ToDateTime(createtime),
detail = innerhtml.Replace("src=\"", "src=\"https://gaokao.chsi.com.cn") detail = Regex.Replace(innerhtml, @"src=""(?!https?:\/\/)(.*?)""", @"src=""https://gaokao.chsi.com.cn$1""")
}); });
// } // }
}); });
} }
} }