Html - Parse number with regexp

EvgeniySkiba
November 5, 2024
168 views
1 vote
2 Answers

I have a text block as a paragraph that starts from some number as a string. There could be simple numbers or dotted ones. I have the next paragraph numbers:

1) sovereign control will prevail

1.1. These are the Rules

5.47.1 "Пункт зупинки тролейбуса".

I use the next regexp expression to find the number.

([0-9.,\-\s]+)\b")

Now, I got the new number format
source text

 5.24.1 і 5.24.2 "Зміна напрямку руху на дорозі з розділювальною смугою"

I need to get the number 5.24.1_ 5.24.2

How can regex expression be written to use one all scenario?

Answers

Sometimes, the greater complexity of a code is compensated by a greater ductility and velocity, I know it is not what you are asking for, but…

List<String> getNumbers( String input ) {
   boolean findNumber = true;
   String aux[] = input.split( "" );
   String aux2 = "";
   List<String> out = new ArrayList<>();
   for( int k = 0; k < aux.length; k ++ ) {
      if( findNumber ) {
         if( aux[ k ].matches( "[0-9]" ) ) {
            aux2 += aux[ k ];
            findNumber = false;
         }
      }
      else {
         if( aux[ k ].matches( "[0-9]" )
                 || aux[ k ].matches( "\." ) ) {
            aux2 += aux[ k ];
         }
         else {
            out.add( aux2 );
            aux2 = "";
            findNumber = true;
         }
      }
   }
   return out;
}

I suggest matching any of the following patterns at the start of string only:

alphanumeric characters followed with a ) char
dot-separated digit sequences
dot-separated digit sequences separated with conjunctions (you may then extend this with more conjunctions as need be).

The regex will look like

String regex = "(?U)\G\s*(?:\sі\s+)?(\d+(?:\.\d+)+|\w+\))";

See the regex demo.

Details:

(?U) – Pattern.UNICODE_CHARACTER_CLASS inline flag option (I see you may have Cyrillic letters in the expected matches, so it is required for the w to match them)
G – either start of string or the end of the previous match (so we only allow to match consecutive matches from the start of the string)
s* – zero or more whitespaces
(?:sіs+)? – an optional sequence of a whitespace, і and then one or more whitespaces
(d+(?:.d+)+|w+)) – Group 1: either
- d+(?:.d+)+ – one or more digits and then one or more sequences of a . + one or more digits
- | – or
- w+) – one or more alphanumeric chars and then a ) char.

See the Java testing code online:

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

class Ideone
{
    public static void main (String[] args) throws java.lang.Exception
    {
        String regex = "(?U)\G\s*(?:\sі\s+)?(\d+(?:\.\d+)+|\w+\))";
        Pattern pattern = Pattern.compile(regex);
        List<String> strs = Arrays.asList(
            "а) driver licance",
            "1) sovereign control will prevail",
            "1.1. These are the Rules",
            "5.47.1 "Пункт зупинки тролейбуса".",
            "5.24.1 і 5.24.2 "Зміна напрямку руху на дорозі з розділювальною смугою"");
        for (String str : strs) {
            System.out.println("Input: "" + str + """);
            Matcher matcher = pattern.matcher(str);
            while (matcher.find()){
                System.out.println(matcher.group(1)); 
            } 
            System.out.println("-- END OF STRING PROCESSING --");
        }
    }
}

Output:

Input: "а) driver licance"
а)
-- END OF STRING PROCESSING --
Input: "1) sovereign control will prevail"
1)
-- END OF STRING PROCESSING --
Input: "1.1. These are the Rules"
1.1
-- END OF STRING PROCESSING --
Input: "5.47.1 "Пункт зупинки тролейбуса"."
5.47.1
-- END OF STRING PROCESSING --
Input: "5.24.1 і 5.24.2 "Зміна напрямку руху на дорозі з розділювальною смугою""
5.24.1
5.24.2
-- END OF STRING PROCESSING --

Please signup or login to give your own answer.

Click here to cancel reply.

Html – Parse number with regexp

Answers