LinkifiedString.java

package org.rascalmpl.uri;

import java.util.ArrayList;
import java.util.List;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class LinkifiedString {
	
	private List<Integer> linkOffsets;
	private List<Integer> linkLengths;
	private List<String> linkTargets;
	private String linkedString; 
	
	
	public String getString() {
		return linkedString;
	}
	
	public boolean containsLinks() {
		return linkOffsets != null && linkOffsets.size() > 0;
	}
	
	public int linkCount() {
		return linkOffsets == null ? 0 : linkOffsets.size();
	}
	
	public int linkOffset(int index) {
		if (linkOffsets == null) {
			throw new IllegalStateException("There are no links"); 
		}
		return linkOffsets.get(index);	
	}
	
	public int linkLength(int index) {
		if (linkLengths == null) {
			throw new IllegalStateException("There are no links"); 
		}
		return linkLengths.get(index);	
	}
	public String linkTarget(int index) {
		if (linkTargets == null) {
			throw new IllegalStateException("There are no links"); 
		}
		return linkTargets.get(index);	
	}
	

	// scan for both \uE007 + [Title](url) and all |loc|'s
	private static final Pattern findLinks = Pattern.compile(
			"(?:\\uE007" // first alternative, Markdown
				+ "\\[([^\\]]*)\\]" // [Title]
				+ "\\(([^\\)]*)\\)" // (link)
			+ ")"
			+ "|"
			+ "(" // or the other alternative, any rascal location 
				+ "\\|[^\\t-\\n\\r\\s\\|]*://[^\\t-\\n\\r\\s\\|]*\\|" // |location|
				+ "(?:\\([^\\)]*\\))?" // (optional offset)
			+ ")");
	
	public LinkifiedString(String input) {
		linkedString = input;
		
		StringBuffer sb = null; 
		Matcher m = findLinks.matcher(input);
		while (m.find()) {
			if (linkOffsets == null) {
				linkOffsets = new ArrayList<Integer>();
				linkLengths = new ArrayList<Integer>();
				linkTargets = new ArrayList<String>();
			}
			if (m.group(3) == null) {
				if (sb == null) {
					// first link found, so we have to recut the string
					sb = new StringBuffer(input.length());
				}
				// markdown link
				String name =  "\u261E " + m.group(1) ;
				String url = m.group(2);
				linkTargets.add(url);
				m.appendReplacement(sb, escapeReplacement(name));
				linkOffsets.add((sb.length() - name.length()) + 1);
				linkLengths.add(name.length());
			}
			else {
				// source location
				String loc = m.group(3);
				linkTargets.add(loc);
				if (sb != null) {
					// we are re-appending
					m.appendReplacement(sb, escapeReplacement(loc));
					linkOffsets.add((sb.length() - loc.length()) + 1);
				}
				else {
					linkOffsets.add(m.start() + 1);
				}
				linkLengths.add(loc.length());
			}
		}
		if (sb != null) {
			// we have a new string 
			m.appendTail(sb);
			linkedString = sb.toString();
		}
	}
	
	private static String escapeReplacement(String replacement) {
		// the $ char should be escaped when we append it as a replacement,
		// since it is used by the regular expression engine to reference a group
		// the \ also needs to be escape, since it is used for escaping in the replacement string
		if (replacement.length() < 20) {
			return replacement.replace("\\", "\\\\").replace("$", "\\$");
		}
		else {
			// okay, "larger" string, lets avoid double loop
			int stringLength = replacement.length();
			StringBuilder sb = new StringBuilder(stringLength);
			for (int currentChar = 0; currentChar < stringLength; currentChar++) {
				char c = replacement.charAt(currentChar);
				if (c == '\\')
					sb.append('\\');
				else if (c == '$' )
					sb.append('\\');
				sb.append(c);
			}
			return sb.toString();
		}
	}

  public String getLinkAt(int offset) {
    if (!containsLinks()) {
      return null;
    }
    for (int i = 0; i < linkOffsets.size() && linkOffsets.get(i) <= offset; i++) {
      if (linkOffsets.get(i) + linkLengths.get(i) >= offset) {
        return linkTargets.get(i);
      }
    }
    return null;
  }
}