First word in string java

None of these answers appears to define what the OP might mean by a «word». As others have already said, a «word boundary» may be a comma, and certainly can’t be counted on to be a space, or even «white space» (i.e. also tabs, newlines, etc.)

At the simplest, I’d say the word has to consist of any Unicode letters, and any digits. Even this may not be right: a String may not qualify as a word if it contains numbers, or starts with a number. Furthermore, what about hyphens, or apostrophes, of which there are presumably several variants in the whole of Unicode? All sorts of discussions of this kind and many others will apply not just to English but to all other languages, including non-human language, scientific notation, etc. It’s a big topic.

But a start might be this (NB written in Groovy):

String givenString = "one two9 thr0ee four"
// String givenString = "oňňÜÐæne;:tŵo9===tĥr0eè? four!"
// String givenString = "mouse"
// String givenString = "&&^^^%"

String[] substrings = givenString.split( '[^\p{L}^\d]+' )

println "substrings |$substrings|"

println "first word |${substrings[0]}|"

This works OK for the first, second and third givenStrings. For «&&^^^%» it says that the first «word» is a zero-length string, and the second is «^^^». Actually a leading zero-length token is String.split‘s way of saying «your given String starts not with a token but a delimiter».

NB in regex p{L} means «any Unicode letter». The parameter of String.split is of course what defines the «delimiter pattern»… i.e. a clump of characters which separates tokens.

NB2 Performance issues are irrelevant for a discussion like this, and almost certainly for all contexts.

NB3 My first port of call was Apache Commons’ StringUtils package. They are likely to have the most effective and best engineered solutions for this sort of thing. But nothing jumped out… https://commons.apache.org/proper/commons-lang/apidocs/org/apache/commons/lang3/StringUtils.html … although something of use may be lurking there.

I have to be able to input any two words as a string

The zero, one, infinity design rule says there is no such thing as two. Lets design it to work with any number of words.

String words = "One two many lots"; // This will be our input

and then invoke and display the first word returned from the method,

So we need a method that takes a String and returns a String.

// Method that returns the first word
public static String firstWord(String input) {
    return input.split(" ")[0]; // Create array of words and return the 0th word
}

static lets us call it from main without needing to create instances of anything. public lets us call it from another class if we want.

.split(" ") creates an array of Strings delimited at every space.
[0] indexes into that array and gives the first word since arrays in java are zero indexed (they start counting at 0).

and the method has to be a for loop method

Ah crap, then we have to do it the hard way.

// Method that returns the first word
public static String firstWord(String input) {
    String result = "";  // Return empty string if no space found

    for(int i = 0; i < input.length(); i++)
    {
        if(input.charAt(i) == ' ')
        {
            result = input.substring(0, i);
            break; // because we're done
        }
    }

    return result; 
}

I kind of know how to use substring, and I know how to return the first word by just using .substring(0,x) x being how long the first word is.

There it is, using those methods you mentioned and the for loop. What more could you want?

But how can I make it so that no matter what phrase I use for the string, it will always return the first word?

Man you’re picky OK fine:

// Method that returns the first word
public static String firstWord(String input) {
    String result = input;  // if no space found later, input is the first word

    for(int i = 0; i < input.length(); i++)
    {
        if(input.charAt(i) == ' ')
        {
            result = input.substring(0, i);
            break;
        }
    }

    return result; 
}

Put it all together it looks like this:

public class FirstWord {

    public static void main(String[] args) throws Exception
    {
        String words = "One two many lots"; // This will be our input
        System.out.println(firstWord(words)); 
    }

    // Method that returns the first word
    public static String firstWord(String input) {

        for(int i = 0; i < input.length(); i++)
        {
            if(input.charAt(i) == ' ')
            {
                return input.substring(0, i);
            }
        }

        return input; 
    }    
}

And it prints this:

One

Hey wait, you changed the firstWord method there.

Yeah I did. This style avoids the need for a result string. Multiple returns are frowned on by old programmers that never got used to garbage collected languages or using finally. They want one place to clean up their resources but this is java so we don’t care. Which style you should use depends on your instructor.

And please explain what you do, because this is my first year in a CS class. Thank you!

What do I do? I post awesome!

Hope it helps.

11 ответов

String mystring = "the quick brown fox";
String arr[] = mystring.split(" ", 2);

String firstWord = arr[0];   //the
String theRest = arr[1];     //quick brown fox

String input = "hello world, this is a line of text";

int i = input.indexOf(' ');
String word = input.substring(0, i);
String rest = input.substring(i);

text.substring(0, text.indexOf(' ')); 

  private String getFirstWord(String text) {
    int index = text.indexOf(' ');
    if (index > -1) { // Check if there is more than one word.
      return text.substring(0, index); // Extract first word.
    } else {
      return text; // Text is the first word itself.
    }
  }

str.contains(" ") ? str.split(" ")[0] : str

    String s = "Hello World, I'm the rest.";
    String[] result = s.split(" ", 2);
    String first = result[0];
    String rest = result[1];
    System.out.println("First: " + first);
    System.out.println("Rest: " + rest);

    // prints =>
    // First: Hello
    // Rest: World, I'm the rest.

import org.apache.commons.lang3.StringUtils;

...
StringUtils.substringBefore("Grigory Kislin", " ")

String test = "123_456";
String value = test.split("_")[0];
System.out.println(value);

123

final String str = "This is a long sentence";
final String[] arr = str.split(" ", 2);
System.out.println(Arrays.toString(arr));

String anotherPalindrome = "Niagara. O roar again!"; 
String roar = anotherPalindrome.substring(11, 15); 

None of these answers appears to define what the OP might mean by a «word». As others have already said, a «word boundary» may be a comma, and certainly can’t be counted on to be a space, or even «white space» (i.e. also tabs, newlines, etc.)

At the simplest, I’d say the word has to consist of any Unicode letters, and any digits. Even this may not be right: a String may not qualify as a word if it contains numbers, or starts with a number. Furthermore, what about hyphens, or apostrophes, of which there are presumably several variants in the whole of Unicode? All sorts of discussions of this kind and many others will apply not just to English but to all other languages, including non-human language, scientific notation, etc. It’s a big topic.

But a start might be this (NB written in Groovy):

String givenString = "one two9 thr0ee four"
// String givenString = "oňňÜÐæne;:tŵo9===tĥr0eè? four!"
// String givenString = "mouse"
// String givenString = "&&^^^%"

String[] substrings = givenString.split( '[^\p{L}^\d]+' )

println "substrings |$substrings|"

println "first word |${substrings[0]}|"

This works OK for the first, second and third givenStrings. For «&&^^^%» it says that the first «word» is a zero-length string, and the second is «^^^». Actually a leading zero-length token is String.split‘s way of saying «your given String starts not with a token but a delimiter».

NB in regex p{L} means «any Unicode letter». The parameter of String.split is of course what defines the «delimiter pattern»… i.e. a clump of characters which separates tokens.

NB2 Performance issues are irrelevant for a discussion like this, and almost certainly for all contexts.

NB3 My first port of call was Apache Commons’ StringUtils package. They are likely to have the most effective and best engineered solutions for this sort of thing. But nothing jumped out… https://commons.apache.org/proper/commons-lang/apidocs/org/apache/commons/lang3/StringUtils.html … although something of use may be lurking there.