Last active
August 25, 2016 00:46
-
-
Save facebookegypt/8ceed5db9a9c04697ee4011aae80fa05 to your computer and use it in GitHub Desktop.
Get a spacific link url from a webpahe HTML
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Imports System | |
| Imports System.Net | |
| Imports System.Net.WebClient | |
| Public Class Form1 | |
| Dim tLINK As String = "http://network.bepress.com/explore/?q=&facet=download_type%3A%22PDF%22&facet=discipline%3A%22Law%22" | |
| Dim Inext30 As Integer = 30 | |
| Dim Hlink As String | |
| Public RawHtmlLinks As String | |
| Dim ThisAttrib As String = "fbTooltip" | |
| Dim Link | |
| Public Function StripHtml(htmlContentLines() As String) As String | |
| Dim iStartIndex, iEndIndex As Integer '// used to locate content to extract. | |
| For Each htmlLine As String In htmlContentLines '// loop thru all lines. | |
| If htmlLine.Contains("class=" & ThisAttrib) Then '// locate lines that contain your links. | |
| iStartIndex = htmlLine.IndexOf("http://") '// get start location of link. | |
| iEndIndex = htmlLine.IndexOf("""", iStartIndex) '// get end location of link. | |
| Hlink &= vbCrLf & (htmlLine.Substring(iStartIndex, iEndIndex - iStartIndex)) '// extract link. | |
| End If | |
| Next | |
| Return Hlink | |
| End Function | |
| Sub getdiv() | |
| Dim I As Integer = 0 | |
| Dim divs = WB1.Document.Body.GetElementsByTagName("A") | |
| For Each d As HtmlElement In divs | |
| If InStr(d.GetAttribute("classname").ToString, ThisAttrib, CompareMethod.Text) Then | |
| I += 1 | |
| T2.Text += vbCrLf & I & ("- ") & d.InnerText | |
| RawHtmlLinks += vbCrLf & d.OuterHtml | |
| End If | |
| 'link : http://network.bepress.com/explore/?q=&facet=download_type%3A%22PDF%22&facet=discipline%3A%22Law%22 | |
| Next | |
| T4.Text = StripHtml(RawHtmlLinks.Split(vbCrLf)) | |
| End Sub | |
| Private Sub Form1_Load(sender As System.Object, e As System.EventArgs) Handles MyBase.Load | |
| WB1.Navigate(tLINK) | |
| Do While WB1.ReadyState <> WebBrowserReadyState.Complete | |
| Application.DoEvents() | |
| Loop | |
| T2.Text = "" | |
| getdiv() | |
| End Sub | |
| Private Sub s_Click(sender As System.Object, e As System.EventArgs) Handles s.Click | |
| End Sub | |
| Private Sub Button1_Click(sender As System.Object, e As System.EventArgs) Handles Button1.Click | |
| 'link : http://network.bepress.com/explore/?q=&facet=download_type%3A%22PDF%22&facet=discipline%3A%22Law%22&start=30 | |
| T5.Text &= (T4.Text) & vbCrLf | |
| WB1.Navigate(T1.Text & "&start=" & Inext30) | |
| T2.Text = "" | |
| RawHtmlLinks = "" | |
| Inext30 += 30 | |
| getdiv() | |
| Label4.Text = Inext30 | |
| End Sub | |
| Private Sub WB1_DocumentCompleted(sender As System.Object, e As System.Windows.Forms.WebBrowserDocumentCompletedEventArgs) Handles WB1.DocumentCompleted | |
| 'link : http://network.bepress.com/explore/?q=&facet=download_type%3A%22PDF%22&facet=discipline%3A%22Law%22&start=30 | |
| T5.Text &= (T4.Text) & vbCrLf | |
| WB1.ScriptErrorsSuppressed = True | |
| tLINK = "http://network.bepress.com/explore/?q=&facet=download_type%3A%22PDF%22&facet=discipline%3A%22Law%22&start=" & Inext30 | |
| WB1.Navigate(tLINK) | |
| T2.Text = "" | |
| RawHtmlLinks = "" | |
| Inext30 += 30 | |
| getdiv() | |
| Label4.Text = Inext30 | |
| End Sub | |
| End Class |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment