正则表达式以在Java字符串中查找特定模式

Question

I need to find a particular pattern in a string and extract a substring out of it. Eg of the String:

{'Opening Cost P&L per store','Opening Costs Capital per store','Average Monthly Revenue','GROSS MARGIN %'}] = N:DB('Store Cost', 
    !Country and Region, DB('New Store Plan', !Country and Region, 
    !ID numbers, !Budget version, 'Retailer Type'), ATTRS('New Stores', 
    !New Stores, '}Map_}Link_New Store Plan_3CStore Cost'), DB('New Store Plan', 
    !Country and Region, !ID numbers, !Budget version, 'Size'), 
    DB('New Store Plan', !Country and Region, !ID numbers, !Budget version, 
    'Franchise/Corporate'), 'DATA')

I have to search for:
DB(' only, not other pattern like S:DB(' or DB('} .

Then after finding this pattern I have to take text in list which is available after this pattern and in quotes only eg

DB('Metrics cube-HumanResource',    !country, !Time, !gometric, !Metric Indicators), CONTINUE)​
DB('Metrics cube-InternalProcess',     !country, 'Total of Product', !Time, !gometric, !Metric Indicators), CONTINUE);​

then output will be:

1 - Metrics cube-HumanResource
2 - Metrics cube-InternalProcess​

This is my code. But its not printing anything:

public class StringRegex {
    public static void main(String[] args) {
        String str = "{'Opening Cost P&L per store','Opening Costs Capital per store','Average Monthly Revenue','GROSS MARGIN %'}] = N:DB('Store Cost', \n" +
                "    !Country and Region, DB('New Store Plan', !Country and Region, \n" +
                "    !ID numbers, !Budget version, 'Retailer Type'), ATTRS('New Stores', \n" +
                "    !New Stores, '}Map_}Link_New Store Plan_3CStore Cost'), DB('New Store Plan', \n" +
                "    !Country and Region, !ID numbers, !Budget version, 'Size'), \n" +
                "    DB('New Store Plan', !Country and Region, !ID numbers, !Budget version, \n" +
                "    'Franchise/Corporate'), 'DATA')";
        String[] strArray = str.split(",");
        for(String s : strArray){
            if(s.matches("DB\\('.+'")){
                System.out.println(s);
            }
        }
    }
}

Answer 1

I would go with a Pattern and a Matcher instead of splitting the String . You will get a result easier. Using the following regex :

.*?[^:]DB\((.*?)\).*?

This will capture the content of every DB(.*) not precede with a : . Using lazy quantifier *? will prevent problem like capturing more than just the text until the closing parenthesis.

From Regex101 :

    String regex = ".*?[^:]DB\\((.*?)\\).*?";
    String string = "{'Opening Cost P&L per store','Opening Costs Capital per store','Average Monthly Revenue','GROSS MARGIN %'}] = N:DB('Store Cost', \n"
            + "    !Country and Region, DB('New Store Plan', !Country and Region, \n"
            + "    !ID numbers, !Budget version, 'Retailer Type'), ATTRS('New Stores', \n"
            + "    !New Stores, '}Map_}Link_New Store Plan_3CStore Cost'), DB('New Store Plan', \n"
            + "    !Country and Region, !ID numbers, !Budget version, 'Size'), \n"
            + "    DB('New Store Plan', !Country and Region, !ID numbers, !Budget version, \n" + "    'Franchise/Corporate'), 'DATA')";

    Pattern pattern = Pattern.compile(regex, Pattern.MULTILINE | Pattern.DOTALL);
    Matcher matcher = pattern.matcher(string);

    while (matcher.find()) {
        System.out.println("> " + matcher.group(1));
    }

Results :

> 'New Store Plan', !Country and Region, 
    !ID numbers, !Budget version, 'Retailer Type'  
> 'New Store Plan', 
    !Country and Region, !ID numbers, !Budget version, 'Size'  
> 'New Store Plan', !Country and Region, !ID numbers, !Budget version, 
    'Franchise/Corporate'