使用模式从字符串中删除数字和字符

Question

So I have a list of words, like 50,000 of them, and I want to remove certain numbers and letters from them. Specifically, I want to remove anything that has a number from 0-99 followed by either an E or Z, so for example: 4E, 11Z, 11E, 20Z , etc

The words that I want to remove them from look like this:-

• 6S,9,12S-trimethyl-2E,4E,8E,10E-tetradecatetraenoic acid
• 7Z,14Z-eicosadienoic acid
• 13,17,21,25-tetramethyl-5Z-hexacosenoic acid
• CDP-DG(18:1(11Z)/22:6(4Z,7Z,10Z,13Z,16Z,19Z))
• PC(20:4(5Z,8Z,11Z,14Z)/17:2(9Z,12Z))

As you can see the thing I want to remove appears in different ways in the words (as in within a bracket or after a hyphen etc). So far, I've done:

public class EZConfig {

    public static void main(String[] args) throws IOException{

     BufferedReader br = new BufferedReader(new FileReader("C:/Users/colles-a-l-kxc127/Dropbox/PhD/Java/MetabolitesCompiled/src/commonNames"));

        try {

            StringBuilder sb = new StringBuilder();
            String line = br.readLine();

            while (line != null) {

                if(line.contains("[0-99][E|Z]")){

                    System.out.println(line + " TRUE");
                }
                else{
                    System.out.println(line);
                }

                line = br.readLine();
            }

        } finally {
            br.close();
        }
    }
}

Just to see if I can pick up the number/E or Z annotations but I can't seem. I need to basically script something that will remove all those annotations from my list of words. Anyone know what I can do in order to achieve this?

Answer 1

You cannot pass a regular expression to String.contains - or rather, it will be treated as literal.

I would use this draft solution:

// declare as constant somewhere
static final Pattern MY_PATTERN = Pattern.compile("\\d+[EZ]");

Then, instead of your if(line.contains("[0-99][E|Z]")){ statement, you can use:

if (MY_PATTERN.matcher(line).find()) {

On the long run, if you're removing that from your words, you probably want to use:

line = line.replaceAll("\\d+[EZ]", "");

Edit

As newbiedoodle mentions (hadn't noticed), the character class [0-99] will not match a range between 0 and 99 .

If you need to limit your digits to < 100 , you can use \\\\d{1,2} instead of the more generic \\\\d+ .

Notes

To remove [optional] parenthesis surrounding the pattern, an optional hyphen starting it and an optional comma ending it as well, you can use the following idiom: "-?\\\\(?\\\\d+[EZ]\\\\)?,?" .

Note that parenthesis need to be double escaped in this context.