1 /* 2 * Copyright (c) 1999, 2017, Oracle and/or its affiliates. All rights reserved. 3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 4 * 5 * This code is free software; you can redistribute it and/or modify it 6 * under the terms of the GNU General Public License version 2 only, as 7 * published by the Free Software Foundation. Oracle designates this 8 * particular file as subject to the "Classpath" exception as provided 9 * by Oracle in the LICENSE file that accompanied this code. 10 * 11 * This code is distributed in the hope that it will be useful, but WITHOUT 12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 14 * version 2 for more details (a copy is included in the LICENSE file that 15 * accompanied this code). 16 * 17 * You should have received a copy of the GNU General Public License version 18 * 2 along with this work; if not, write to the Free Software Foundation, 19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 20 * 21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 22 * or visit www.oracle.com if you need additional information or have any 23 * questions. 24 */ 25 26 package com.sun.tools.javac.util; 27 28 /** Utility class for static conversion methods between numbers 29 * and strings in various formats. 30 * 31 * <p>Note regarding UTF-8. 32 * The JVMS defines its own version of the UTF-8 format so that it 33 * contains no zero bytes (modified UTF-8). This is not actually the same 34 * as Charset.forName("UTF-8"). 35 * 36 * <p> 37 * See also: 38 * <ul> 39 * <li><a href="http://docs.oracle.com/javase/specs/jvms/se7/html/jvms-4.html#jvms-4.4.7"> 40 * JVMS 4.4.7 </a></li> 41 * <li><a href="http://docs.oracle.com/javase/7/docs/api/java/io/DataInput.html#modified-utf-8"> 42 java.io.DataInput: Modified UTF-8 </a></li> 43 <li><a href="https://en.wikipedia.org/wiki/UTF-8#Modified_UTF-8"> 44 Modified UTF-8 (wikipedia) </a></li> 45 * </ul> 46 * 47 * The methods here support modified UTF-8. 48 * 49 * <p><b>This is NOT part of any supported API. 50 * If you write code that depends on this, you do so at your own risk. 51 * This code and its internal interfaces are subject to change or 52 * deletion without notice.</b> 53 */ 54 public class Convert { 55 56 /** Convert string to integer. 57 */ string2int(String s, int radix)58 public static int string2int(String s, int radix) 59 throws NumberFormatException { 60 if (radix == 10) { 61 return Integer.parseInt(s, radix); 62 } else { 63 char[] cs = s.toCharArray(); 64 int limit = Integer.MAX_VALUE / (radix/2); 65 int n = 0; 66 for (char c : cs) { 67 int d = Character.digit(c, radix); 68 if (n < 0 || 69 n > limit || 70 n * radix > Integer.MAX_VALUE - d) 71 throw new NumberFormatException(); 72 n = n * radix + d; 73 } 74 return n; 75 } 76 } 77 78 /** Convert string to long integer. 79 */ string2long(String s, int radix)80 public static long string2long(String s, int radix) 81 throws NumberFormatException { 82 if (radix == 10) { 83 return Long.parseLong(s, radix); 84 } else { 85 char[] cs = s.toCharArray(); 86 long limit = Long.MAX_VALUE / (radix/2); 87 long n = 0; 88 for (char c : cs) { 89 int d = Character.digit(c, radix); 90 if (n < 0 || 91 n > limit || 92 n * radix > Long.MAX_VALUE - d) 93 throw new NumberFormatException(); 94 n = n * radix + d; 95 } 96 return n; 97 } 98 } 99 100 /* Conversion routines between names, strings, and byte arrays in Utf8 format 101 */ 102 103 /** Convert `len' bytes from utf8 to characters. 104 * Parameters are as in System.arraycopy 105 * Return first index in `dst' past the last copied char. 106 * @param src The array holding the bytes to convert. 107 * @param sindex The start index from which bytes are converted. 108 * @param dst The array holding the converted characters.. 109 * @param dindex The start index from which converted characters 110 * are written. 111 * @param len The maximum number of bytes to convert. 112 */ utf2chars(byte[] src, int sindex, char[] dst, int dindex, int len)113 public static int utf2chars(byte[] src, int sindex, 114 char[] dst, int dindex, 115 int len) { 116 int i = sindex; 117 int j = dindex; 118 int limit = sindex + len; 119 while (i < limit) { 120 int b = src[i++] & 0xFF; 121 if (b >= 0xE0) { 122 b = (b & 0x0F) << 12; 123 b = b | (src[i++] & 0x3F) << 6; 124 b = b | (src[i++] & 0x3F); 125 } else if (b >= 0xC0) { 126 b = (b & 0x1F) << 6; 127 b = b | (src[i++] & 0x3F); 128 } 129 dst[j++] = (char)b; 130 } 131 return j; 132 } 133 134 /** Return bytes in Utf8 representation as an array of characters. 135 * @param src The array holding the bytes. 136 * @param sindex The start index from which bytes are converted. 137 * @param len The maximum number of bytes to convert. 138 */ utf2chars(byte[] src, int sindex, int len)139 public static char[] utf2chars(byte[] src, int sindex, int len) { 140 char[] dst = new char[len]; 141 int len1 = utf2chars(src, sindex, dst, 0, len); 142 char[] result = new char[len1]; 143 System.arraycopy(dst, 0, result, 0, len1); 144 return result; 145 } 146 147 /** Return all bytes of a given array in Utf8 representation 148 * as an array of characters. 149 * @param src The array holding the bytes. 150 */ utf2chars(byte[] src)151 public static char[] utf2chars(byte[] src) { 152 return utf2chars(src, 0, src.length); 153 } 154 155 /** Return bytes in Utf8 representation as a string. 156 * @param src The array holding the bytes. 157 * @param sindex The start index from which bytes are converted. 158 * @param len The maximum number of bytes to convert. 159 */ utf2string(byte[] src, int sindex, int len)160 public static String utf2string(byte[] src, int sindex, int len) { 161 char dst[] = new char[len]; 162 int len1 = utf2chars(src, sindex, dst, 0, len); 163 return new String(dst, 0, len1); 164 } 165 166 /** Return all bytes of a given array in Utf8 representation 167 * as a string. 168 * @param src The array holding the bytes. 169 */ utf2string(byte[] src)170 public static String utf2string(byte[] src) { 171 return utf2string(src, 0, src.length); 172 } 173 174 /** Copy characters in source array to bytes in target array, 175 * converting them to Utf8 representation. 176 * The target array must be large enough to hold the result. 177 * returns first index in `dst' past the last copied byte. 178 * @param src The array holding the characters to convert. 179 * @param sindex The start index from which characters are converted. 180 * @param dst The array holding the converted characters.. 181 * @param dindex The start index from which converted bytes 182 * are written. 183 * @param len The maximum number of characters to convert. 184 */ chars2utf(char[] src, int sindex, byte[] dst, int dindex, int len)185 public static int chars2utf(char[] src, int sindex, 186 byte[] dst, int dindex, 187 int len) { 188 int j = dindex; 189 int limit = sindex + len; 190 for (int i = sindex; i < limit; i++) { 191 char ch = src[i]; 192 if (1 <= ch && ch <= 0x7F) { 193 dst[j++] = (byte)ch; 194 } else if (ch <= 0x7FF) { 195 dst[j++] = (byte)(0xC0 | (ch >> 6)); 196 dst[j++] = (byte)(0x80 | (ch & 0x3F)); 197 } else { 198 dst[j++] = (byte)(0xE0 | (ch >> 12)); 199 dst[j++] = (byte)(0x80 | ((ch >> 6) & 0x3F)); 200 dst[j++] = (byte)(0x80 | (ch & 0x3F)); 201 } 202 } 203 return j; 204 } 205 206 /** Return characters as an array of bytes in Utf8 representation. 207 * @param src The array holding the characters. 208 * @param sindex The start index from which characters are converted. 209 * @param len The maximum number of characters to convert. 210 */ chars2utf(char[] src, int sindex, int len)211 public static byte[] chars2utf(char[] src, int sindex, int len) { 212 byte[] dst = new byte[len * 3]; 213 int len1 = chars2utf(src, sindex, dst, 0, len); 214 byte[] result = new byte[len1]; 215 System.arraycopy(dst, 0, result, 0, len1); 216 return result; 217 } 218 219 /** Return all characters in given array as an array of bytes 220 * in Utf8 representation. 221 * @param src The array holding the characters. 222 */ chars2utf(char[] src)223 public static byte[] chars2utf(char[] src) { 224 return chars2utf(src, 0, src.length); 225 } 226 227 /** Return string as an array of bytes in in Utf8 representation. 228 */ string2utf(String s)229 public static byte[] string2utf(String s) { 230 return chars2utf(s.toCharArray()); 231 } 232 233 /** 234 * Escapes each character in a string that has an escape sequence or 235 * is non-printable ASCII. Leaves non-ASCII characters alone. 236 */ quote(String s)237 public static String quote(String s) { 238 StringBuilder buf = new StringBuilder(); 239 for (int i = 0; i < s.length(); i++) { 240 buf.append(quote(s.charAt(i))); 241 } 242 return buf.toString(); 243 } 244 245 /** 246 * Escapes a character if it has an escape sequence or is 247 * non-printable ASCII. Leaves non-ASCII characters alone. 248 */ quote(char ch)249 public static String quote(char ch) { 250 switch (ch) { 251 case '\b': return "\\b"; 252 case '\f': return "\\f"; 253 case '\n': return "\\n"; 254 case '\r': return "\\r"; 255 case '\t': return "\\t"; 256 case '\'': return "\\'"; 257 case '\"': return "\\\""; 258 case '\\': return "\\\\"; 259 default: 260 return (isPrintableAscii(ch)) 261 ? String.valueOf(ch) 262 : String.format("\\u%04x", (int) ch); 263 } 264 } 265 266 /** 267 * Is a character printable ASCII? 268 */ isPrintableAscii(char ch)269 private static boolean isPrintableAscii(char ch) { 270 return ch >= ' ' && ch <= '~'; 271 } 272 273 /** Escape all unicode characters in string. 274 */ escapeUnicode(String s)275 public static String escapeUnicode(String s) { 276 int len = s.length(); 277 int i = 0; 278 while (i < len) { 279 char ch = s.charAt(i); 280 if (ch > 255) { 281 StringBuilder buf = new StringBuilder(); 282 buf.append(s.substring(0, i)); 283 while (i < len) { 284 ch = s.charAt(i); 285 if (ch > 255) { 286 buf.append("\\u"); 287 buf.append(Character.forDigit((ch >> 12) % 16, 16)); 288 buf.append(Character.forDigit((ch >> 8) % 16, 16)); 289 buf.append(Character.forDigit((ch >> 4) % 16, 16)); 290 buf.append(Character.forDigit((ch ) % 16, 16)); 291 } else { 292 buf.append(ch); 293 } 294 i++; 295 } 296 s = buf.toString(); 297 } else { 298 i++; 299 } 300 } 301 return s; 302 } 303 304 /* Conversion routines for qualified name splitting 305 */ 306 /** Return the last part of a qualified name. 307 * @param name the qualified name 308 * @return the last part of the qualified name 309 */ shortName(Name name)310 public static Name shortName(Name name) { 311 int start = name.lastIndexOf((byte)'.') + 1; 312 int end = name.getByteLength(); 313 if (start == 0 && end == name.length()) { 314 return name; 315 } 316 return name.subName( 317 name.lastIndexOf((byte)'.') + 1, name.getByteLength()); 318 } 319 320 /** Return the last part of a qualified name from its string representation 321 * @param name the string representation of the qualified name 322 * @return the last part of the qualified name 323 */ shortName(String name)324 public static String shortName(String name) { 325 return name.substring(name.lastIndexOf('.') + 1); 326 } 327 328 /** Return the package name of a class name, excluding the trailing '.', 329 * "" if not existent. 330 */ packagePart(Name classname)331 public static Name packagePart(Name classname) { 332 return classname.subName(0, classname.lastIndexOf((byte)'.')); 333 } 334 packagePart(String classname)335 public static String packagePart(String classname) { 336 int lastDot = classname.lastIndexOf('.'); 337 return (lastDot < 0 ? "" : classname.substring(0, lastDot)); 338 } 339 enclosingCandidates(Name name)340 public static List<Name> enclosingCandidates(Name name) { 341 List<Name> names = List.nil(); 342 int index; 343 while ((index = name.lastIndexOf((byte)'$')) > 0) { 344 name = name.subName(0, index); 345 names = names.prepend(name); 346 } 347 return names; 348 } 349 classCandidates(Name name)350 public static List<Name> classCandidates(Name name) { 351 List<Name> names = List.nil(); 352 String nameStr = name.toString(); 353 int index = -1; 354 while ((index = nameStr.indexOf('.', index + 1)) > 0) { 355 String pack = nameStr.substring(0, index + 1); 356 String clz = nameStr.substring(index + 1).replace('.', '$'); 357 names = names.prepend(name.table.names.fromString(pack + clz)); 358 } 359 return names.reverse(); 360 } 361 } 362