Help with java regex

Question

Hey, I've been struggling with this regex and I'm out of ideas. I have this types of strings (not all of them are here, but only this 2 types) and I have to extract the part between the th tags.

<th class="tip" title='manje'>manje</th>
<th class="tip" title='ne d.'>ne d.</th>
<th class="tip" title='manje'>manje</th>
<th class="tip" title='točno'>točno</th>
<th class="tip" title='više'>više</th>
<th class="tip" title='m./t.'>m./t.</th>
<th class="tip" title='v./t.'>v./t.</th>
<th class="tip">daje</th>
<th class="tip">X2</th>
<th class="tip">12</th>

I've tried some combinations bu I only get the value if there is no that attribute "title" in th tag.

This pattern only extracts the content if there is no "title" attribute in th tag:

Pattern pattern = Pattern.compile("<th class=\"tip\"[\\s*|[.]{0,20}]>(.*?)\\s*</th>");

This one also:

Pattern patternType = Pattern.compile("<th class=\"tip\"[\\s*|[.]{0,20}]>(.*?)\\s*</th>");

Any suggestions? Tnx

Careful with the "[]", it defines a single character to be matched. I'm not sure what it is actually trying to match for "[.]". — toto2
– toto2, Commented Jun 4, 2011 at 12:33
It would probably be more appropriate to use the dom and sax utilities anyway since you are reading XML. Those are part of the Java SE base packages. — toto2
– toto2, Commented Jun 4, 2011 at 12:40

Roland Illig · Accepted Answer · 2011-06-04 11:57:20Z

5

Regular expressions are not suitable in all cases. Use Jsoup instead:

package so6235727;

import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;

public class PrintContent {

  private static final String html = //
      "<th class=\"tip\" title='manje'>manje</th>\r\n" + //
      "<th class=\"tip\" title='ne d.'>ne d.</th>\r\n" + //
      "<th class=\"tip\" title='manje'>manje</th>\r\n" + //
      "<th class=\"tip\" title='točno'>točno</th>\r\n" + //
      "<th class=\"tip\" title='više'>više</th>\r\n" + //
      "<th class=\"tip\" title='m./t.'>m./t.</th>\r\n" + //
      "<th class=\"tip\" title='v./t.'>v./t.</th>\r\n" + //
      "<th class=\"tip\">daje</th>\r\n" + //
      "<th class=\"tip\">X2</th>\r\n" + //
      "<th class=\"tip\">12</th>\r\n";

  public static void main(String[] args) {
    Document jsoup = Jsoup.parse(html);
    Elements headings = jsoup.select("th.tip");
    for (Element element : headings) {
      System.out.println(element.text());
    }
  }
}

See how easy this is?

answered Jun 4, 2011 at 11:57

Roland Illig

41.9k12 gold badges92 silver badges127 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

vale4674 Over a year ago

Yes, but I have a really big HTML page that I have to parse and regex where my first hint. But this tool looks really great. I'll try it.

Yuri Stuken · Accepted Answer · 2011-06-04 11:22:26Z

0

Try this one:

Pattern pattern = Pattern.compile("<th class=\"tip\"[^>]*>(.*)</th>");

answered Jun 4, 2011 at 11:22

Yuri Stuken

13.8k1 gold badge30 silver badges23 bronze badges

1 Comment

vale4674 Over a year ago

Woohoo :D this one workes for me. Tnx. Good idea with matching until ">".

Billy Moon · Accepted Answer · 2011-06-04 11:20:07Z

0

Try this:

Pattern pattern = Pattern.compile("<th[^>]*>(.*?)\\s*</th>");

answered Jun 4, 2011 at 11:20

Billy Moon

58.9k27 gold badges148 silver badges245 bronze badges

1 Comment

vale4674 Over a year ago

Tried it but it doesn't work for me. stuken.yuri's regex works. tnx anyway

Hovercraft Full Of Eels · Accepted Answer · 2011-06-04 12:38:08Z

0

What the heck, one more Pattern answer attempt, this one with look ahead and look behind:

Pattern pattern = Pattern.compile("(?<=<th .{0,100}>).*(?=</th>)");

EDIT 1
Regarding I tried it and it doesn't work in any case: perhaps your harness is different from mine:

import java.util.Scanner;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class Foo1 {
   private static final String FOO_TXT = "Foo1.txt";

   public static void main(String[] args) {
      Pattern pattern = Pattern.compile("(?<=<th .{0,100}>).*(?=</th>)");

      Scanner scan = new Scanner(Foo1.class.getResourceAsStream(FOO_TXT));
      while (scan.hasNextLine()) {
         String line = scan.nextLine();
         System.out.println("Line: " + line);
         Matcher match = pattern.matcher(line);
         if (match.find()) {
            System.out.println("Match: " + match.group());
         } else {
            System.out.println("No match found");
         }
      }
   }
}

This assumes that the text file is named Foo1.txt and that it is located with the class files.

edited Jun 4, 2011 at 12:38

answered Jun 4, 2011 at 11:31

Hovercraft Full Of Eels

286k25 gold badges268 silver badges391 bronze badges

4 Comments

toto2 Over a year ago

Weird... your pattern really does not work on my machine and the other ones do. (I removed my comment saying it does not work. My harness is different, but I don't see what difference it might do on the matches...)

toto2 Over a year ago

If I change your lookahead stuff ("(?<=") with a simpler non-matching group tag ("(?:") it does work.

Hovercraft Full Of Eels Over a year ago

@toto: perhaps if you edited your original post and showed us your test harness, preferably as a small compilable and runnable program, I could better see why my regex fails for you.

toto2 Over a year ago

I put my code as an answer (unfortunately). I have to run now, but I'll check tonight to see if you have any comment. I'm running Java 7 (beta) on Windows 7.

toto2 · Accepted Answer · 2011-06-06 14:21:33Z

I'm including my test code because it seems I have positive/negative matches when others have negative/positive matches.

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

public class Regex {

public static void test(String patternString) {
    System.out.println("Test with pattern: " + patternString);
    Pattern pattern = Pattern.compile(patternString);
    String[] testStrings = {"<th class=\"tip\" title='manje'>manje</th>", "<th class=\"tip\">daje</th>"};
    for (String testString : testStrings) {
        System.out.println("> Test on " + testString);
        Matcher matcher = pattern.matcher(testString);
        if (matcher.matches()) {
            System.out.println(">> number of matches in group = " + matcher.groupCount());
            for (int i = 1; i <= matcher.groupCount(); i++) {
                System.out.println(">>group " + i + " is " + matcher.group(i));
            }
        } else {
            System.out.println(">> no match");
        }
    }
    System.out.println("");
}

/**
 * @param args the command line arguments
 */
public static void main(String[] args) {
    test("<th class=\"tip\"[\\s*|[.]{0,20}]>(.*?)\\s*</th>"); // op
    test("<th[^>]*>(.*?)\\s*</th>"); // Billy Moon
    test("<th class=\"tip\"[^>]*>(.*)</th>"); // stuken.yuri
    test("(?<=<th .{0,100}>).*(?=</th>)"); // Hovercraft full of Eels
    test("(?:<th .{0,100}>).*(?:</th>)");
}
}

My output is that I get a match for Billy Moon and stuken.yuri, but no match for the OP or Hovercraft. I would be curious to see if others get the same. I am using Java 7 beta with Windows 7.

Collectives™ on Stack Overflow

Help with java regex

5 Answers 5

1 Comment

1 Comment

1 Comment

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

5 Answers 5

1 Comment

1 Comment

1 Comment

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related