net.vidageek.crawler.component.DefaultLinkNormalizer Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of crawler Show documentation

Simple and extensible crawler.

The newest version!

package net.vidageek.crawler.component;

/**
 * @author jonasabreu
 */
public class DefaultLinkNormalizer implements LinkNormalizer {

	private final String beginUrl;

	public DefaultLinkNormalizer(final String beginUrl) {
		if ((beginUrl == null) || (beginUrl.trim().length() == 0)) {
			throw new IllegalArgumentException("beginUrl cannot be null or empty");
		}
		this.beginUrl = beginUrl;
	}

	public String normalize(String url) {

		url = url.replaceAll("&", "&");

		return UrlUtils.resolveUrl(beginUrl, url);
	}

}