A Simple Web Crawler

This demonstrates a very simple web crawler using the Chilkat Spider component.

Downloads:

MS Windows Visual C/C++ Libraries

Linux/CentOS C/C++ Libraries

MAC OS X C/C++ Libraries

Solaris C/C++ Libraries

C++ Builder Libraries

#include <CkSpider.h>
#include <CkStringArray.h>

void ChilkatSample(void)
    {
    //  The Chilkat Spider component/library is free.
    CkSpider spider;

    CkStringArray seenDomains;
    CkStringArray seedUrls;

    seenDomains.put_Unique(true);
    seedUrls.put_Unique(true);

    seedUrls.Append("http://directory.google.com/Top/Recreation/Outdoors/Hiking/Backpacking/");

    //  Set our outbound URL exclude patterns
    spider.AddAvoidOutboundLinkPattern("*?id=*");
    spider.AddAvoidOutboundLinkPattern("*.mypages.*");
    spider.AddAvoidOutboundLinkPattern("*.personal.*");
    spider.AddAvoidOutboundLinkPattern("*.comcast.*");
    spider.AddAvoidOutboundLinkPattern("*.aol.*");
    spider.AddAvoidOutboundLinkPattern("*~*");

    //  Use a cache so we don't have to re-fetch URLs previously fetched.
    spider.put_CacheDir("c:/spiderCache/");
    spider.put_FetchFromCache(true);
    spider.put_UpdateCache(true);

    while (seedUrls.get_Count() > 0) {

        const char * url;
        url = seedUrls.pop();
        spider.Initialize(url);

        //  Spider 5 URLs of this domain.
        //  but first, save the base domain in seenDomains
        const char * domain;
        domain = spider.getUrlDomain(url);
        seenDomains.Append(spider.getBaseDomain(domain));

        long i;
        bool success;
        for (i = 0; i <= 4; i++) {
            success = spider.CrawlNext();
            if (success != true) {
                break;
            }

            //  Display the URL we just crawled.
            printf("%s\n",spider.lastUrl());

            //  If the last URL was retrieved from cache,
            //  we won't wait.  Otherwise we'll wait 1 second
            //  before fetching the next URL.
            if (spider.get_LastFromCache() != true) {
                spider.SleepMs(1000);
            }

        }

        //  Add the outbound links to seedUrls, except
        //  for the domains we've already seen.
        for (i = 0; i <= spider.get_NumOutboundLinks() - 1; i++) {

            url = spider.getOutboundLink(i);

            domain = spider.getUrlDomain(url);
            const char * baseDomain;
            baseDomain = spider.getBaseDomain(domain);
            if (!seenDomains.Contains(baseDomain)) {
                seedUrls.Append(url);
            }

            //  Don't let our list of seedUrls grow too large.
            if (seedUrls.get_Count() > 1000) {
                break;
            }

        }

    }

    }