How to Code Web Crawlers in C++

How to Code Web Crawlers in C++

Check out this program.

#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);

    //  You will need to change the start URL to something else...
    seedUrls.Append("http://something.whateverYouWant.com/");

    //  Set outbound URL exclude patterns
    //  URLs matching any of these patterns will not be added to the
    //  collection of outbound links.
    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("%sn",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;
            }

        }

    }

    }

How to Code Web Crawlers in C++

I guess you are looking for a sample web spider written in C++. You can easily get very simple codes for web crawlers on the web but to make it work, you should also know how to code and compile with C++ because not all samples will work. You can encounter a sample code that has errors that needs correction and when this happens your knowledge in C++ will be handy.

You can try Whalebot. It is an open-source web spider which is designed to be fast, simple, and memory efficient. It was designed as a targeted spider but you can also use it as a common. Visit Whalebot 0.02 to download. It features start/stop/resume fetching sessions; and simple configuration from text files and command lines.

Another sample web crawler is the Gist web crawler. It uses boost_regex and boost_algorithm. I can’t post the entire code because of its length. If you want to see it and download, visit Gist Simplest Web Crawler in C++.