1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000 1.2 +++ b/src/share/jaxws_classes/com/sun/xml/internal/bind/WhiteSpaceProcessor.java Wed Apr 27 01:27:09 2016 +0800 1.3 @@ -0,0 +1,198 @@ 1.4 +/* 1.5 + * Copyright (c) 1997, 2011, Oracle and/or its affiliates. All rights reserved. 1.6 + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 1.7 + * 1.8 + * This code is free software; you can redistribute it and/or modify it 1.9 + * under the terms of the GNU General Public License version 2 only, as 1.10 + * published by the Free Software Foundation. Oracle designates this 1.11 + * particular file as subject to the "Classpath" exception as provided 1.12 + * by Oracle in the LICENSE file that accompanied this code. 1.13 + * 1.14 + * This code is distributed in the hope that it will be useful, but WITHOUT 1.15 + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 1.16 + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 1.17 + * version 2 for more details (a copy is included in the LICENSE file that 1.18 + * accompanied this code). 1.19 + * 1.20 + * You should have received a copy of the GNU General Public License version 1.21 + * 2 along with this work; if not, write to the Free Software Foundation, 1.22 + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 1.23 + * 1.24 + * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 1.25 + * or visit www.oracle.com if you need additional information or have any 1.26 + * questions. 1.27 + */ 1.28 + 1.29 +package com.sun.xml.internal.bind; 1.30 + 1.31 +/** 1.32 + * Processes white space normalization. 1.33 + * 1.34 + * @since 1.0 1.35 + */ 1.36 +public abstract class WhiteSpaceProcessor { 1.37 + 1.38 +// benchmarking (see test/src/ReplaceTest.java in the CVS Attic) 1.39 +// showed that this code is slower than the current code. 1.40 +// 1.41 +// public static String replace(String text) { 1.42 +// final int len = text.length(); 1.43 +// StringBuffer result = new StringBuffer(len); 1.44 +// 1.45 +// for (int i = 0; i < len; i++) { 1.46 +// char ch = text.charAt(i); 1.47 +// if (isWhiteSpace(ch)) 1.48 +// result.append(' '); 1.49 +// else 1.50 +// result.append(ch); 1.51 +// } 1.52 +// 1.53 +// return result.toString(); 1.54 +// } 1.55 + 1.56 + public static String replace(String text) { 1.57 + return replace( (CharSequence)text ).toString(); 1.58 + } 1.59 + 1.60 + /** 1.61 + * @since 2.0 1.62 + */ 1.63 + public static CharSequence replace(CharSequence text) { 1.64 + int i=text.length()-1; 1.65 + 1.66 + // look for the first whitespace char. 1.67 + while( i>=0 && !isWhiteSpaceExceptSpace(text.charAt(i)) ) 1.68 + i--; 1.69 + 1.70 + if( i<0 ) 1.71 + // no such whitespace. replace(text)==text. 1.72 + return text; 1.73 + 1.74 + // we now know that we need to modify the text. 1.75 + // allocate a char array to do it. 1.76 + StringBuilder buf = new StringBuilder(text); 1.77 + 1.78 + buf.setCharAt(i--,' '); 1.79 + for( ; i>=0; i-- ) 1.80 + if( isWhiteSpaceExceptSpace(buf.charAt(i))) 1.81 + buf.setCharAt(i,' '); 1.82 + 1.83 + return new String(buf); 1.84 + } 1.85 + 1.86 + /** 1.87 + * Equivalent of {@link String#trim()}. 1.88 + * @since 2.0 1.89 + */ 1.90 + public static CharSequence trim(CharSequence text) { 1.91 + int len = text.length(); 1.92 + int start = 0; 1.93 + 1.94 + while( start<len && isWhiteSpace(text.charAt(start)) ) 1.95 + start++; 1.96 + 1.97 + int end = len-1; 1.98 + 1.99 + while( end>start && isWhiteSpace(text.charAt(end)) ) 1.100 + end--; 1.101 + 1.102 + if(start==0 && end==len-1) 1.103 + return text; // no change 1.104 + else 1.105 + return text.subSequence(start,end+1); 1.106 + } 1.107 + 1.108 + public static String collapse(String text) { 1.109 + return collapse( (CharSequence)text ).toString(); 1.110 + } 1.111 + 1.112 + /** 1.113 + * This is usually the biggest processing bottleneck. 1.114 + * 1.115 + * @since 2.0 1.116 + */ 1.117 + public static CharSequence collapse(CharSequence text) { 1.118 + int len = text.length(); 1.119 + 1.120 + // most of the texts are already in the collapsed form. 1.121 + // so look for the first whitespace in the hope that we will 1.122 + // never see it. 1.123 + int s=0; 1.124 + while(s<len) { 1.125 + if(isWhiteSpace(text.charAt(s))) 1.126 + break; 1.127 + s++; 1.128 + } 1.129 + if(s==len) 1.130 + // the input happens to be already collapsed. 1.131 + return text; 1.132 + 1.133 + // we now know that the input contains spaces. 1.134 + // let's sit down and do the collapsing normally. 1.135 + 1.136 + StringBuilder result = new StringBuilder(len /*allocate enough size to avoid re-allocation*/ ); 1.137 + 1.138 + if(s!=0) { 1.139 + for( int i=0; i<s; i++ ) 1.140 + result.append(text.charAt(i)); 1.141 + result.append(' '); 1.142 + } 1.143 + 1.144 + boolean inStripMode = true; 1.145 + for (int i = s+1; i < len; i++) { 1.146 + char ch = text.charAt(i); 1.147 + boolean b = isWhiteSpace(ch); 1.148 + if (inStripMode && b) 1.149 + continue; // skip this character 1.150 + 1.151 + inStripMode = b; 1.152 + if (inStripMode) 1.153 + result.append(' '); 1.154 + else 1.155 + result.append(ch); 1.156 + } 1.157 + 1.158 + // remove trailing whitespaces 1.159 + len = result.length(); 1.160 + if (len > 0 && result.charAt(len - 1) == ' ') 1.161 + result.setLength(len - 1); 1.162 + // whitespaces are already collapsed, 1.163 + // so all we have to do is to remove the last one character 1.164 + // if it's a whitespace. 1.165 + 1.166 + return result; 1.167 + } 1.168 + 1.169 + /** 1.170 + * Returns true if the specified string is all whitespace. 1.171 + */ 1.172 + public static boolean isWhiteSpace(CharSequence s) { 1.173 + for( int i=s.length()-1; i>=0; i-- ) 1.174 + if(!isWhiteSpace(s.charAt(i))) 1.175 + return false; 1.176 + return true; 1.177 + } 1.178 + 1.179 + /** returns true if the specified char is a white space character. */ 1.180 + public static boolean isWhiteSpace(char ch) { 1.181 + // most of the characters are non-control characters. 1.182 + // so check that first to quickly return false for most of the cases. 1.183 + if( ch>0x20 ) return false; 1.184 + 1.185 + // other than we have to do four comparisons. 1.186 + return ch == 0x9 || ch == 0xA || ch == 0xD || ch == 0x20; 1.187 + } 1.188 + 1.189 + /** 1.190 + * Returns true if the specified char is a white space character 1.191 + * but not 0x20. 1.192 + */ 1.193 + protected static boolean isWhiteSpaceExceptSpace(char ch) { 1.194 + // most of the characters are non-control characters. 1.195 + // so check that first to quickly return false for most of the cases. 1.196 + if( ch>=0x20 ) return false; 1.197 + 1.198 + // other than we have to do four comparisons. 1.199 + return ch == 0x9 || ch == 0xA || ch == 0xD; 1.200 + } 1.201 +}