Crawling the Web

If the Chilkat Spider component only crawls a single site, how do you crawl the Web? The answer is simple: as you crawl a site, the spider collects outbound links and makes them accessible to you. You may then instantiate an instance of the Spider object for each site, and crawl it. The task of keeping track of what sites you've already crawled is left to you (for now). This example retrieves the home page of http://www.joelonsoftware.com/ and displays the outbound links.

Download Chilkat Spider ActiveX

uses
    Windows, Messages, SysUtils, Variants, Classes, Graphics, Controls, Forms,
    Dialogs, StdCtrls,
    SPIDERXLib_TLB,
    SPIDERXLib_TLB,
    OleCtrls;

...

procedure TForm1.Button1Click(Sender: TObject);
var
spider: TSpider;
success: Integer;
i: Integer;

begin
//  The Chilkat Spider component/library is free.
spider := TSpider.Create(Self);

//  The Initialize method may be called with just the domain name,
//  such as "www.joelonsoftware.com" or a full URL.  If you pass only
//  the domain name, you must add URLs to the unspidered list by calling
//  AddUnspidered.  Otherwise, the URL you pass to Initialize is the 1st
//  URL in the unspidered list.
spider.Initialize('www.joelonsoftware.com');

spider.AddUnspidered('http://www.joelonsoftware.com/');

success := spider.CrawlNext();

for i := 0 to spider.NumOutboundLinks - 1 do
  begin
    Memo1.Lines.Add(spider.GetOutboundLink(i));
  end;

end;