用Java正则表达式拆分

Question

I have a string like:

Snt:It was the most widespread day of environmental action in the planet's history
====================
-----------
Snt:Five years ago, I was working for just over minimum wage
====================
-----------

and I want to split the string with

====================
-----------

and ofcourse remove Snt: from the first of sentences. what is the best way?

I used this regular expression, but it didnt work!

String[] content1 =content.split("\\n\\====================\\n\\-----------\\n");

Thanks in advance.

Answer 1

What about

Pattern p = Pattern.compile("^Snt:(.*)$", Pattern.MULTILINE);
Matcher m = p.matcher(str);

while (m.find()) {
    String sentence = m.group(1);
}

Rather than hacking around with split and doing extra parsing, this just looks for lines beginning with "Snt," then captures whatever follows.

Answer 2

Because of the way the data is structured, I would reverse the concept from a split, to be a matcher instead., This allows you to mathc the Snt nicely as well:

private static final String VAL = "Snt:It was the most widespread day of environmental action in the planet's history\n"
        + "====================\n"
        + "-----------\n"
        + "Snt:Five years ago, I was working for just over minimum wage\n"
        + "====================\n"
        + "-----------";

public static void main(String[] args) {
    List<String> phrases = new ArrayList<String>();
    Matcher mat = Pattern.compile("Snt:(.+?)\n={20}\n-{11}\\s*").matcher(VAL);
    while (mat.find()) {
        phrases.add(mat.group(1));
    }

    System.out.printf("Value: %s%n", phrases); 
}

I use the regex: "Snt:(.+?)\\n={20}\\n-{11}\\\\s*"

This assumes that the first word in the file is the Snt: , and then it groups the next phrase, until the delimiter. It will consume any trailing whitespace, making the expression ready for the next record.

The upside of this process is that the match matches a single record, instead of having an expression that matches part of the end of one record, an perhaps the beginning of the next.

Answer 3

Because there is no newline exist at the last, it won't match the last == , -- lines. You need to add end of the line anchor $ at the last as an alternative to \\n in your regex.

String s = "Snt:It was the most widespread day of environmental action in the planet's history\n" +
"====================\n" +
"-----------\n" +
"Snt:Five years ago, I was working for just over minimum wage\n" +
"====================\n" +
"-----------";
String m = s.replaceAll("(?m)^Snt:", "");
String[] tok = m.split("\\n\\====================\\n\\-----------(?:\\n|$)");
System.out.println(Arrays.toString(tok));

Output:

[It was the most widespread day of environmental action in the planet's history, Five years ago, I was working for just over minimum wage]

Answer 4

Matcher m = Pattern.compile("([^=\\-]+)([=\\-]+[\\t\\n\\s]*)+").matcher(str);   
while (m.find()) {
    String match = m.group(1);
    System.out.println(match);
}