Tell me more ×
Stack Overflow is a question and answer site for professional and enthusiast programmers. It's 100% free, no registration required.
up vote 2 down vote favorite
1

Code:

static short state = 0;
static int td_number = 0;

public static void main(String[] args) {

final Pattern p = Pattern.compile("^[\\s]*?\\d+\\.\\d+[\\s]*?");
final short TD_ENTRY = 0;
final short NO_ENTRY = 1;

HTMLEditorKit.ParserCallback callback = new HTMLEditorKit.ParserCallback() {
    public void handleText(char[] data, int pos) {
    switch (state) {
    case NO_ENTRY:
        break;
    case TD_ENTRY: {
        // We are in the right table column
        // Create string from char array
        String s = new String(data);
        Matcher m = p.matcher(s);
        boolean b = m.matches();
        // Check if data information has correct format (0.0)
        if (b) {

        }
    }
        break;
    default:
        break;
    }
    state = NO_ENTRY;
    }

    public void handleStartTag(HTML.Tag tag, MutableAttributeSet set, int pos) {
    if (tag == HTML.Tag.TD) {
      //[...]
    }
    }
};

Reader reader = new StringReader(html);
try {
    new ParserDelegator().parse(reader, callback, false);
} catch (IOException e) {
}
}

I am trying to parse HTML with Regular Expressions. The program reads the content of td tags within an html table. The content in the table cell should fit a special pattern defined in Pattern p.

The main problem is now that the regex pattern does not match for cell content like this " 0.1". But if I define the String s manually with the value (" 0.1") in the code the pattern matches.

Furthermore if I copy the content of char[] data in debug mode and define s with this copied content the pattern does also not fit although it looks the same like the manually defined value from above.

Is it possible to find out which whitespace characters are really read?

It seems that the whitespace is not always a whitespace and therefore does not match with regex class [\s]. Is this possible?

EDIT:

Thanks for answers. It was really a whitespace character (\xA0) which was not recognized by \s regex class.

For all of you which downvote (really frustrating) my question simply missunderstood me. Maybe the problem was really the sentence "I want to parse HTML with regex" but in fact I simply have content from a HTML table cell with unknown whitespace characters ;-).

I think I had got the same problems with a library like jsoup.

share|improve this question
7  
This is your main problem: Parsing Html The Cthulhu Way‌​. Instead of RegEx, use a library like jsoup – Luiggi Mendoza Jan 14 at 15:27
1  
You are breaking one of the 10 commandments of programming. DO NOT use regular expressions to parse HTML. Read this: codinghorror.com/blog/2009/11/parsing-html-the-cthulhu-way.html – Eric Leschinski Jan 14 at 15:31
@jlordo yes, but the trim() function does not remove the whitespaces for whatever reason. – sk2212 Jan 14 at 15:33
"I am trying to parse HTML with Regular Expressions." Whenever you find yourself saying that, stop what you're doing and step away from the keyboard. Then, learn how to use an XML parser. Then--only then--should you continue. – Jack Maney Jan 14 at 15:35
1  
HTML often contain non-breaking spaces (code-point 160) which are not matched by \s or by the definition of space used by String.trim(). \s matches [ \t\n\x0B\f\r] only and trim treats only code-points < 0x20 as space. – Mike Samuel Jan 14 at 15:37
show 1 more comment

2 Answers

up vote 1 down vote accepted

In Java regexes, the non-breaking space character (NBSP, U+00A0) is traditionally not treated as whitespace for the purpose of matching \s. If that's what's causing your problem, you just need to add it to your existing whitespace class:

"^[\\s\\xA0]*\\d+\\.\\d+[\\s\\xA0]*$"

There are other Unicode whitespace characters that aren't matched by \s, but none of them are anywhere as common as the NBSP.

Alternatively, if you're running Java 7+ you can specify UNICODE_CHARACTER_CLASS mode and go on using \s.

share|improve this answer
up vote 0 down vote

Your code snippet is too long, but as far as I understand you just need pattern to match something like 0.0, 10.52 etc, i.e. floating point numbers? Use pattern \\d+\\.\\d+.

\d+ means 1..n digits \. means dot. A single dot . in regex means "any character"

Here is the usage example:

String str = "123.456";
Pattern p = Pattern.compile("\\d+\\.\\d+");
Matcher m = p.matcher(str);
if (m.matches()) {
    // do something.
}

BTW, pay attention that matches() matches full line. If you want to match part of line use find() instead. I personally always use find() and use start and end line markers ^ and $ into regex itself when needed.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.