-1

I have followed the solution below this method , but when it comes to extracting string between the codes, there are errors in the regex pattern or pattern cannot match the section of string

such as

This is a test [url] http://www.google.com.hk [/url] and [img] http://www.abc.com/test.png [/img]

how to get data between quotes in java?

Community
  • 1
  • 1
Jeff Bootsholz
  • 2,971
  • 15
  • 70
  • 141
  • Could you show us your pattern, how are you using it and what problems you get? – Pshemo Nov 26 '13 at 15:28
  • Have a look at [this question and its answers](http://stackoverflow.com/questions/3064453/regex-bbcode-to-html?rq=1)! – tilpner Nov 26 '13 at 15:32

2 Answers2

2

You did not state if you have nested patterns, so here is an example to start you off.

You need to double escape \\ here since \ is also an escape character for strings. 

String s = "This is a test [url] http://www.google.com.hk [/url]\n"
         + " and [img] http://www.abc.com/test.png [/img]";

Pattern p = Pattern.compile("\\[[^\\]]*\\]([^\\]]*)\\[[^\\]]*\\]");
Matcher m = p.matcher(s);
while (m.find()) {
  System.out.println(m.group(1).trim());
}

See working demo

Regular expression:

\[               '['
[^\]]*           any character except: '\]' (0 or more times)
 \]              ']'
(                group and capture to \1:
 [^\]]*          any character except: '\]' (0 or more times)
)                end of \1
\[               '['
 [^\]]*          any character except: '\]' (0 or more times)
\]               ']'

If you want to be specific on img or url code tags, you could use the following.

Pattern p = Pattern.compile("(?i)\\[(?:url|img)\\]([^\\]]*)\\[\\/(?:url|img)\\]");
hwnd
  • 69,796
  • 4
  • 95
  • 132
1

Here's a regex that assumes that the names of your tags ("bbcodes") consist only of word characters. It also checks that the name of the closing tag matches the name of the opening tag. (This is done by referring to the name of the opening tag by the backrefence \1.)

\[(\w+)\](.+?)\[/\1\]
   ^^^    ^^^
    1      2

Backreferences:

  1. The name of the opening tag. (e.g., url)
  2. A non-greedy match of all characters between the opening and closing tags.

Here's a demonstration. (Also see the live demo.)

import java.util.*;
import java.util.regex.*;

class RegexTester
{
    public static void main (String[] args)
    {
        String s =
              "This is a test [url] http://www.google.com.hk [/url]\n"
            + " and [img] http://www.abc.com/test.png [/img]";

        Pattern p = Pattern.compile("\\[(\\w+)\\](.+?)\\[/\\1\\]");

        Matcher m = p.matcher(s);
        while (m.find()) {
            System.out.println("Match=[" + m.group(2).trim() + "]");
        }
    }
}

Expected Result

Match=[http://www.google.com.hk]
Match=[http://www.abc.com/test.png]
DavidRR
  • 18,291
  • 25
  • 109
  • 191
  • how about if handling the following case ? [quote][quote] This is a test [url] http://www.google.com.hk [/url] and [img] http://www.abc.com/test.png [/img] [/quote][/quote] – Jeff Bootsholz Nov 27 '13 at 01:32
  • @Raju - Using regular expressions to parse **arbitrarily nested tags** is generally not considered the best approach. (Although some enhanced regex implementations make this possible.) See [Can regular expressions be used to match nested patterns?](http://stackoverflow.com/q/133601/1497596) and [Using regular expressions to parse HTML: why not?](http://stackoverflow.com/q/590747/1497596). – DavidRR Nov 28 '13 at 00:54
  • I really don't understand how the theory mentioned can be applied to he implementation of getting the values of nested tags – Jeff Bootsholz Nov 28 '13 at 08:09
  • @Raju - I found the first part of this article helpful: [PHP: Recursive Patterns](http://php.net/manual/en/regexp.reference.recursive.php). In particular, note the mention of the special item `(?R)`. Also see the topic **Recursive patterns** at [Wikipedia: Perl Compatible Regular Expressions](https://en.wikipedia.org/wiki/Perl_Compatible_Regular_Expressions). – DavidRR Nov 30 '13 at 01:37