FileDocCategorySizeDatePackage
WhiteSpaceProcessor.javaAPI DocJava SE 6 API5023Tue Jun 10 00:27:04 BST 2008javax.xml.bind

WhiteSpaceProcessor

public abstract class WhiteSpaceProcessor extends Object
Processes white space normalization.
since
1.0

Fields Summary
Constructors Summary
Methods Summary
public static java.lang.Stringcollapse(java.lang.String text)

        return collapse( (CharSequence)text ).toString();
    
public static java.lang.CharSequencecollapse(java.lang.CharSequence text)
This is usually the biggest processing bottleneck.

since
2.0

        int len = text.length();

        // most of the texts are already in the collapsed form.
        // so look for the first whitespace in the hope that we will
        // never see it.
        int s=0;
        while(s<len) {
            if(isWhiteSpace(text.charAt(s)))
                break;
            s++;
        }
        if(s==len)
            // the input happens to be already collapsed.
            return text;

        // we now know that the input contains spaces.
        // let's sit down and do the collapsing normally.

        StringBuilder result = new StringBuilder(len /*allocate enough size to avoid re-allocation*/ );

        if(s!=0) {
            for( int i=0; i<s; i++ )
                result.append(text.charAt(i));
            result.append(' ");
        }

        boolean inStripMode = true;
        for (int i = s+1; i < len; i++) {
            char ch = text.charAt(i);
            boolean b = isWhiteSpace(ch);
            if (inStripMode && b)
                continue; // skip this character

            inStripMode = b;
            if (inStripMode)
                result.append(' ");
            else
                result.append(ch);
        }

        // remove trailing whitespaces
        len = result.length();
        if (len > 0 && result.charAt(len - 1) == ' ")
            result.setLength(len - 1);
        // whitespaces are already collapsed,
        // so all we have to do is to remove the last one character
        // if it's a whitespace.

        return result;
    
public static final booleanisWhiteSpace(java.lang.CharSequence s)
Returns true if the specified string is all whitespace.

        for( int i=s.length()-1; i>=0; i-- )
            if(!isWhiteSpace(s.charAt(i)))
                return false;
        return true;
    
public static final booleanisWhiteSpace(char ch)
returns true if the specified char is a white space character.

        // most of the characters are non-control characters.
        // so check that first to quickly return false for most of the cases.
        if( ch>0x20 )   return false;

        // other than we have to do four comparisons.
        return ch == 0x9 || ch == 0xA || ch == 0xD || ch == 0x20;
    
protected static final booleanisWhiteSpaceExceptSpace(char ch)
Returns true if the specified char is a white space character but not 0x20.

        // most of the characters are non-control characters.
        // so check that first to quickly return false for most of the cases.
        if( ch>=0x20 )   return false;

        // other than we have to do four comparisons.
        return ch == 0x9 || ch == 0xA || ch == 0xD;
    
public static java.lang.Stringreplace(java.lang.String text)

        return replace( (CharSequence)text ).toString();
    
public static java.lang.CharSequencereplace(java.lang.CharSequence text)

since
2.0

        int i=text.length()-1;

        // look for the first whitespace char.
        while( i>=0 && !isWhiteSpaceExceptSpace(text.charAt(i)) )
            i--;

        if( i<0 )
            // no such whitespace. replace(text)==text.
            return text;

        // we now know that we need to modify the text.
        // allocate a char array to do it.
        StringBuilder buf = new StringBuilder(text);

        buf.setCharAt(i--,' ");
        for( ; i>=0; i-- )
            if( isWhiteSpaceExceptSpace(buf.charAt(i)))
                buf.setCharAt(i,' ");

        return new String(buf);
    
public static java.lang.CharSequencetrim(java.lang.CharSequence text)
Equivalent of {@link String#trim()}.

since
2.0

        int len = text.length();
        int start = 0;

        while( start<len && isWhiteSpace(text.charAt(start)) )
            start++;

        int end = len-1;

        while( end>start && isWhiteSpace(text.charAt(end)) )
            end--;

        if(start==0 && end==len-1)
            return text;    // no change
        else
            return text.subSequence(start,end+1);