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

bug fixed

develop
old易 2024-12-11 15:44:25 +08:00
parent bb0799c9e5
commit fc488ccf1e
2 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
New_College.Tasks/HostedService/JobTimedSpiderService.cs
View File

@ -79,7 +79,7 @@ namespace New_College.Tasks
var list = anghelper.HtmlCreatePageData(p.RegionCode); var list = anghelper.HtmlCreatePageData(p.RegionCode);
list.ForEach(async c => list.ForEach(async c =>
{ {
// c.title = c.title.Replace($"{p.SimpleName}：", ""); c.title = c.title.Replace($"{p.SimpleName}：", "");
var newsinfo = await newsInfoServices.Query(e => e.Title.Trim() == c.title); var newsinfo = await newsInfoServices.Query(e => e.Title.Trim() == c.title);
if (!newsinfo.Any() && c.pubtime.Year > DateTime.Now.Year - 1) if (!newsinfo.Any() && c.pubtime.Year > DateTime.Now.Year - 1)
{ {

8
New_College.Tasks/HtmlAgSpider/NationWideNewsAgHelper.cs
View File

@ -35,7 +35,7 @@ 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;
@ -51,8 +51,8 @@ namespace New_College.Tasks
} }
} }
var innerhtml = doc.DocumentNode.SelectSingleNode("//*[@id=\"article_dnull\"]").InnerHtml; var innerhtml = doc.DocumentNode.SelectSingleNode("//*[@id=\"article_dnull\"]").InnerHtml;
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"))
{ {
@ -66,7 +66,7 @@ namespace New_College.Tasks
pubtime = Convert.ToDateTime(createtime), pubtime = Convert.ToDateTime(createtime),
detail = innerhtml.Replace("src=\"", "src=\"https://gaokao.chsi.com.cn") detail = innerhtml.Replace("src=\"", "src=\"https://gaokao.chsi.com.cn")
}); });
} // }
}); });
} }
} }