How to get a RegEx to extract only uppercase from string

[Piet Souris]

et cetera, et cetera.

@Mike,

can you supply us with some sentences that formed the motivation of your opening post, and tell in each case what should be the outcome?

[Darryl Burke]

Stephan van Hulst wrote:

Darryl Burke wrote:What's wrong with sentence.replaceAll("(?:\\b)[a-z]*(?:\\b)", "")?  And if the double spaces left behind are an issue, that could be chained to .replace("  ", " ").

That regular expression does not take into account that the sentence could consist of characters other than a-z, A-Z and spaces.

Ah, I see.

Stephan van Hulst wrote:Also, the second replace would have to be performed multiple times until there are no more changes.

Um, no.

String#replace(CharSequence target, CharSequence replacement) wrote:Replaces each substring of this string that matches the literal target sequence with the specified literal replacement sequence.

[Stephan van Hulst]

Yes, but if you have three spaces in a row, it will replace the first two with a single space, but the two spaces that are left over will not be replaced, because the matcher has moved past the first one (which was a replacement).

[Stephan van Hulst]

Here is a short program that illustrates what I mean:

final class Test {
  public static void main(String... args) {
    String result = "UPPERCASE some lowercase words WORDS".replaceAll("(?:\\b)[a-z]*(?:\\b)", "").replace("  ", " ");
    System.out.println(result);
  }
}

Output:UPPERCASE  WORDS

[Darryl Burke]

Thank you Stephan, I see where I went wrong.

Still assuming that only words with all-lowercase (i.e. [a-z]) are to be removed, how about this?

public class UppercaseWords {
  public static void main(String[] args) {
    String sentence = "UPPERCASE some lowercase words WORDS";
    System.out.println(sentence.replaceAll("(?:\\b)[a-z]*(?:\\b)", "").replaceAll(" (?= )", ""));
  }
}

[Carey Brown]

This is more straightforward.

public class UppercaseWords {
  public static void main(String[] args) {
    String sentence = "UPPERCASE some lowercase words WORDS";
    System.out.println(sentence.replaceAll("(?:\\b)[a-z]*(?:\\b)", "").replaceAll("\\s+", " "));
  }
}

[Stephan van Hulst]

Carey, that suffers from the same problem. If the replacement results in multiple spaces in a row, they won't be recursively replaced.

[Carey Brown]

Stephan van Hulst wrote:Carey, that suffers from the same problem. If the replacement results in multiple spaces in a row, they won't be recursively replaced.

Look again. Doesn't need to be recursive.

Try

System.out.println( "A B  C   D    E     F".replaceAll( "\\s+", " " ) );

[Carey Brown]

replaceAll( "\\s+", " " )

The "\\s" means any white space character. The "+" says one or more. So you have a replacement pattern of one or more adjacent spaces to be replaced by a single space.

[Campbell Ritchie]

I haven't been following this th‍read carefully; once it go onto the second page it started to get long‑winded. But a long time ago somebody showed how to extract upper case letters from the String with a Stream. So why do you insist on wanting a regex?

[Darryl Burke]

Regexes are fun!

(So are Streams...)

[Stephan van Hulst]

Carey Brown wrote:

replaceAll( "\\s+", " " )

The "\\s" means any white space character. The "+" says one or more. So you have a replacement pattern of one or more adjacent spaces to be replaced by a single space.

Yes, but it won't replace the replacements that result in multiple spaces in a row.
Nevermind, it was a long weekend