RegExp performance for returning contextual search results

Question

Hello,
I have a search results page on which I want to display search term results in context ... the search term plus 15 words on either side.
I've written a function that is working (pasted below). Essentially, it receives as string and then I use regular expressions plus an ArrayList to determine the substing to return. Performance is ok, but I'm wondering if there's a better way to tackle this problem.
Anyone have a suggestion?
Thanks
  public static String returnSnippet(String xmlText, int numberOfWords)
    Pattern myPattern;
    Matcher myMatcher;   
    // Move this out so it's passed in since this is a utility class?
    String[] patternArray = {
      "<chapter-title(.*?)</chapter-title>",
      "<subject-title(.*?)</subject-title>",
      "<resource-metadata>(.*?)</resource-metadata>",
      "<children>(.*?)</children>",
      "<child(.*?)</child>",
      "<a-head(.*?)</a-head>",
      "<b-head(.*?)</b-head>",
      "<lrh>(.*?)</lrh>",
      "<rrh>(.*?)</rrh>",
      "<head(.*?)</head>",
      "<titlegroup(.*?)</titlegroup>",
      "<para role(.*?)</para>",
      "<contributor(.*?)</contributor>",
      "<author(.*?)</author>",
      "<fmname(.*?)</fmname>",
      "<lname(.*?)</lname>",
      "<(.|\n)+?>"
    // Do these two first
    myPattern = Pattern.compile("<hit (.*?)>");
    myMatcher = myPattern.matcher(xmlText);
    xmlText = myMatcher.replaceAll("[hit]");
    myPattern = Pattern.compile("</hit>");
    myMatcher = myPattern.matcher(xmlText);
    xmlText = myMatcher.replaceAll("[/hit]");
    for (int i = 0; i < patternArray.length; i++)
      myPattern = Pattern.compile(patternArray);
myMatcher = myPattern.matcher(xmlText);
xmlText = myMatcher.replaceAll("");
// Add the logic to count words before and after here
// See RegexTestHarness.java in C:\j2sdk1.4.2_11\lib on my machine for notes / test version
myPattern = Pattern.compile("\\[hit\\].*?\\[hit\\]");
myMatcher = myPattern.matcher(xmlText);
if(myMatcher.find()) // Using if captures the first instance only; using while will loop through them all
int hitStart = myMatcher.start();
int hitEnd = myMatcher.end();
myPattern = Pattern.compile("\\s");
myMatcher = myPattern.matcher(xmlText);
ArrayList spaceArray = new ArrayList(100);
while(myMatcher.find())
spaceArray.add(new Integer(myMatcher.start())); // ListArray.add() expects and Object, but int is a primitive type, so create an Integer object
arrayforloop: for(int i=0; i < spaceArray.size(); i++)
When the value of the number (the index of the "space" hit) is greater than or equal to the value of the hitStart
(the index of the "[hit]" start), count backwards and forwards to get the index values for spliting the string.
if( ((Integer)spaceArray.get(i)).intValue() >= hitStart ) // Going from Object -> Integer -> int
int wordIndexStart = ( ( i - numberOfWords ) <= 0 ) ? 0 : i - numberOfWords; // These two get the locations in the array
int wordIndexEnd = ( ( i + numberOfWords ) >= spaceArray.size() ) ? spaceArray.size() : i + numberOfWords;
int substringStart = (wordIndexStart == 0) ? 0 : ((Integer)spaceArray.get(wordIndexStart)).intValue(); // These two get the values of the locations in the array
int substringEnd = ( (Integer)spaceArray.get(wordIndexEnd) ).intValue();
xmlText = xmlText.substring(substringStart, substringEnd);
break arrayforloop;
else
xmlText = "";
// Do these two last
myPattern = Pattern.compile("\\[hit\\]");
myMatcher = myPattern.matcher(xmlText);
xmlText = myMatcher.replaceAll("<span class=\"sr-hit\">");
myPattern = Pattern.compile("\\[hit\\]");
myMatcher = myPattern.matcher(xmlText);
xmlText = myMatcher.replaceAll("</span>");
return xmlText;

Answer 1

Wow. Let me ask you if this is what you meant (code below).
It's shaved about 200 milliseconds off the resopnse time. I went from 600-800 milliseconds down to 400-600 to process a page of results. That's a fantastic improvement.
  private static Pattern regex_chaptertitle = Pattern.compile("<chapter-title(.*?)</chapter-title>");
  private static Pattern regex_subjecttitle = Pattern.compile("<subject-title(.*?)</subject-title>");
  private static Pattern regex_resourcemetadata = Pattern.compile("<resource-metadata>(.*?)</resource-metadata>");
  private static Pattern regex_children = Pattern.compile("<children>(.*?)</children>");
  private static Pattern regex_child = Pattern.compile("<child(.*?)</child>");
  private static Pattern regex_ahead = Pattern.compile("<a-head(.*?)</a-head>");
  private static Pattern regex_bhead = Pattern.compile("<b-head(.*?)</b-head>");
  private static Pattern regex_lrh = Pattern.compile("<lrh>(.*?)</lrh>");
  private static Pattern regex_rrh = Pattern.compile("<rrh>(.*?)</rrh>");
  private static Pattern regex_head = Pattern.compile("<head(.*?)</head>");
  private static Pattern regex_titlegroup = Pattern.compile("<titlegroup(.*?)</titlegroup>");
  private static Pattern regex_pararole = Pattern.compile("<para role(.*?)</para>");
  private static Pattern regex_contributor = Pattern.compile("<contributor(.*?)</contributor>");
  private static Pattern regex_author = Pattern.compile("<author(.*?)</author>");
  private static Pattern regex_fname = Pattern.compile("<fmname(.*?)</fmname>");
  private static Pattern regex_lname = Pattern.compile("<lname(.*?)</lname>");
  private static Pattern regex_alltags = Pattern.compile("<(.|\n)+?>");
  public static String returnSnippet(String xmlText, int numberOfWords)
    int wordIndexStart;
    int wordIndexEnd;
    int substringStart;
    int substringEnd;
    Pattern myPattern;
    Matcher myMatcher;       
    // Do these two first
    myPattern = Pattern.compile("<ixiahit (.*?)>");
    myMatcher = myPattern.matcher(xmlText);
    xmlText = myMatcher.replaceAll("[ixiahit]");
    myPattern = Pattern.compile("</ixiahit>");
    myMatcher = myPattern.matcher(xmlText);
    xmlText = myMatcher.replaceAll("[/ixiahit]");
    myMatcher = regex_chaptertitle.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_subjecttitle.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_resourcemetadata.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_children.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_child.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_ahead.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_bhead.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_lrh.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_rrh.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_head.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_titlegroup.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_pararole.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_contributor.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_author.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_fname.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_lname.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    myMatcher = regex_alltags.matcher(xmlText);
    xmlText = myMatcher.replaceAll("");
    // Add the logic to count words before and after here
    // See RegexTestHarness.java in C:\j2sdk1.4.2_11\lib on my machine for notes / test version
    myPattern = Pattern.compile("\\[ixiahit\\].*?\\[/ixiahit\\]");
    myMatcher = myPattern.matcher(xmlText);
    if(myMatcher.find()) // Using if captures the first instance only; using while will loop through them all
      int hitStart = myMatcher.start();
      int hitEnd = myMatcher.end();
      myPattern = Pattern.compile("\\s");
      myMatcher = myPattern.matcher(xmlText);
      ArrayList spaceArray = new ArrayList(100);
      while(myMatcher.find())
        spaceArray.add(new Integer(myMatcher.start())); // ListArray.add() expects and Object, but int is a primitive type, so create an Integer object       
      arrayforloop: for(int i=0; i < spaceArray.size(); i++)
        When the value of the number (the index of the "space" hit) is greater than or equal to the value of the hitStart
        (the index of the "[ixiahit]" start), count backwards and forwards to get the index values for spliting the string.
        if( ((Integer)spaceArray.get(i)).intValue() >= hitStart ) // Going from Object -> Integer -> int
          wordIndexStart = ( ( i - numberOfWords ) <= 0 ) ? 0 : i - numberOfWords; // These two get the locations in the array
          wordIndexEnd = ( ( i + numberOfWords ) >= spaceArray.size() ) ? spaceArray.size() : i + numberOfWords;
          substringStart = (wordIndexStart == 0) ? 0 : ((Integer)spaceArray.get(wordIndexStart)).intValue(); // These two get the values of the locations in the array
          substringEnd = ( (Integer)spaceArray.get(wordIndexEnd) ).intValue();
          xmlText = xmlText.substring(substringStart, substringEnd);         
          break arrayforloop;
    else
      xmlText = "";
    // Do these two last
    myPattern = Pattern.compile("\\[ixiahit\\]");
    myMatcher = myPattern.matcher(xmlText);
    xmlText = myMatcher.replaceAll("<span class=\"sr-hit\">");
    myPattern = Pattern.compile("\\[/ixiahit\\]");
    myMatcher = myPattern.matcher(xmlText);
    xmlText = myMatcher.replaceAll("</span>");
    return xmlText;
  }