| 1 | /* |
|---|---|
| 2 | * Copyright (C) 2007-2010 Júlio Vilmar Gesser. |
| 3 | * Copyright (C) 2011, 2013-2020 The JavaParser Team. |
| 4 | * |
| 5 | * This file is part of JavaParser. |
| 6 | * |
| 7 | * JavaParser can be used either under the terms of |
| 8 | * a) the GNU Lesser General Public License as published by |
| 9 | * the Free Software Foundation, either version 3 of the License, or |
| 10 | * (at your option) any later version. |
| 11 | * b) the terms of the Apache License |
| 12 | * |
| 13 | * You should have received a copy of both licenses in LICENCE.LGPL and |
| 14 | * LICENCE.APACHE. Please refer to those files for details. |
| 15 | * |
| 16 | * JavaParser is distributed in the hope that it will be useful, |
| 17 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 18 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 19 | * GNU Lesser General Public License for more details. |
| 20 | */ |
| 21 | package com.github.javaparser.utils; |
| 22 | |
| 23 | import java.io.IOException; |
| 24 | import java.io.StringWriter; |
| 25 | import java.io.Writer; |
| 26 | import java.util.HashMap; |
| 27 | import java.util.HashSet; |
| 28 | |
| 29 | /** |
| 30 | * Adapted from apache commons-lang3 project. |
| 31 | * <p> |
| 32 | * Unescapes escaped chars in strings. |
| 33 | */ |
| 34 | public final class StringEscapeUtils { |
| 35 | |
| 36 | private StringEscapeUtils() { |
| 37 | } |
| 38 | |
| 39 | /** |
| 40 | * <p>Escapes the characters in a {@code String} using Java String rules.</p> |
| 41 | * <p> |
| 42 | * <p>Deals correctly with quotes and control-chars (tab, backslash, cr, ff, etc.) </p> |
| 43 | * <p> |
| 44 | * <p>So a tab becomes the characters {@code '\\'} and |
| 45 | * {@code 't'}.</p> |
| 46 | * <p> |
| 47 | * <p>The only difference between Java strings and JavaScript strings |
| 48 | * is that in JavaScript, a single quote and forward-slash (/) are escaped.</p> |
| 49 | * <p> |
| 50 | * <p>Example:</p> |
| 51 | * <pre> |
| 52 | * input string: He didn't say, "Stop!" |
| 53 | * output string: He didn't say, \"Stop!\" |
| 54 | * </pre> |
| 55 | * |
| 56 | * @param input String to escape values in, may be null |
| 57 | * @return String with escaped values, {@code null} if null string input |
| 58 | */ |
| 59 | public static String escapeJava(final String input) { |
| 60 | return ESCAPE_JAVA.translate(input); |
| 61 | } |
| 62 | |
| 63 | /** |
| 64 | * <p>Unescapes any Java literals found in the {@code String}. |
| 65 | * For example, it will turn a sequence of {@code '\'} and |
| 66 | * {@code 'n'} into a newline character, unless the {@code '\'} |
| 67 | * is preceded by another {@code '\'}.</p> |
| 68 | * <p> |
| 69 | * This can be replaced by String::translateEscapes in JDK 13 |
| 70 | * |
| 71 | * @param input the {@code String} to unescape, may be null |
| 72 | * @return a new unescaped {@code String}, {@code null} if null string input |
| 73 | */ |
| 74 | public static String unescapeJava(final String input) { |
| 75 | return UNESCAPE_JAVA.translate(input); |
| 76 | } |
| 77 | |
| 78 | public static String unescapeJavaTextBlock(final String input) { |
| 79 | return UNESCAPE_JAVA_TEXT_BLOCK.translate(input); |
| 80 | } |
| 81 | |
| 82 | private static final LookupTranslator JAVA_CTRL_CHARS_UNESCAPE = new LookupTranslator(new String[][]{ |
| 83 | {"\\b", "\b"}, |
| 84 | {"\\n", "\n"}, |
| 85 | {"\\t", "\t"}, |
| 86 | {"\\f", "\f"}, |
| 87 | {"\\r", "\r"}}); |
| 88 | |
| 89 | private static final LookupTranslator JAVA_CTRL_CHARS_ESCAPE = new LookupTranslator(new String[][]{ |
| 90 | {"\b", "\\b"}, |
| 91 | {"\n", "\\n"}, |
| 92 | {"\t", "\\t"}, |
| 93 | {"\f", "\\f"}, |
| 94 | {"\r", "\\r"}}); |
| 95 | |
| 96 | private static final CharSequenceTranslator ESCAPE_JAVA = new AggregateTranslator( |
| 97 | new LookupTranslator( |
| 98 | new String[][]{ |
| 99 | {"\"", "\\\""}, |
| 100 | {"\\", "\\\\"}, |
| 101 | }), |
| 102 | JAVA_CTRL_CHARS_ESCAPE); |
| 103 | |
| 104 | private static final CharSequenceTranslator UNESCAPE_JAVA = new AggregateTranslator( |
| 105 | new OctalUnescaper(), |
| 106 | new UnicodeUnescaper(), |
| 107 | JAVA_CTRL_CHARS_UNESCAPE, |
| 108 | new LookupTranslator(new String[][]{ |
| 109 | {"\\\\", "\\"}, |
| 110 | {"\\\"", "\""}, |
| 111 | {"\\'", "'"}, |
| 112 | {"\\", ""}})); |
| 113 | |
| 114 | private static final CharSequenceTranslator UNESCAPE_JAVA_TEXT_BLOCK = new AggregateTranslator( |
| 115 | new OctalUnescaper(), |
| 116 | new UnicodeUnescaper(), |
| 117 | JAVA_CTRL_CHARS_UNESCAPE, |
| 118 | new LookupTranslator(new String[][]{ |
| 119 | {"\\\\", "\\"}, |
| 120 | {"\\\"", "\""}, |
| 121 | {"\\'", "'"}, |
| 122 | {"\\", ""}, |
| 123 | {"\\s", " "}, |
| 124 | {"\\\n", ""}})); |
| 125 | |
| 126 | /** |
| 127 | * Adapted from apache commons-lang3 project. |
| 128 | * <p> |
| 129 | * An API for translating text. |
| 130 | * Its core use is to escape and unescape text. Because escaping and unescaping |
| 131 | * is completely contextual, the API does not present two separate signatures. |
| 132 | * |
| 133 | * @since 3.0 |
| 134 | */ |
| 135 | private static abstract class CharSequenceTranslator { |
| 136 | |
| 137 | /** |
| 138 | * Translate a set of codepoints, represented by an int index into a CharSequence, |
| 139 | * into another set of codepoints. The number of codepoints consumed must be returned, |
| 140 | * and the only IOExceptions thrown must be from interacting with the Writer so that |
| 141 | * the top level API may reliably ignore StringWriter IOExceptions. |
| 142 | * |
| 143 | * @param input CharSequence that is being translated |
| 144 | * @param index int representing the current point of translation |
| 145 | * @param out Writer to translate the text to |
| 146 | * @return int count of codepoints consumed |
| 147 | * @throws IOException if and only if the Writer produces an IOException |
| 148 | */ |
| 149 | protected abstract int translate(CharSequence input, int index, Writer out) throws IOException; |
| 150 | |
| 151 | /** |
| 152 | * Helper for non-Writer usage. |
| 153 | * |
| 154 | * @param input CharSequence to be translated |
| 155 | * @return String output of translation |
| 156 | */ |
| 157 | private String translate(final CharSequence input) { |
| 158 | if (input == null) { |
| 159 | return null; |
| 160 | } |
| 161 | try { |
| 162 | final StringWriter writer = new StringWriter(input.length() * 2); |
| 163 | translate(input, writer); |
| 164 | return writer.toString(); |
| 165 | } catch (final IOException ioe) { |
| 166 | // this should never ever happen while writing to a StringWriter |
| 167 | throw new RuntimeException(ioe); |
| 168 | } |
| 169 | } |
| 170 | |
| 171 | /** |
| 172 | * Translate an input onto a Writer. This is intentionally final as its algorithm is |
| 173 | * tightly coupled with the abstract method of this class. |
| 174 | * |
| 175 | * @param input CharSequence that is being translated |
| 176 | * @param out Writer to translate the text to |
| 177 | * @throws IOException if and only if the Writer produces an IOException |
| 178 | */ |
| 179 | private void translate(final CharSequence input, final Writer out) throws IOException { |
| 180 | if (out == null) { |
| 181 | throw new IllegalArgumentException("The Writer must not be null"); |
| 182 | } |
| 183 | if (input == null) { |
| 184 | return; |
| 185 | } |
| 186 | int pos = 0; |
| 187 | final int len = input.length(); |
| 188 | while (pos < len) { |
| 189 | final int consumed = translate(input, pos, out); |
| 190 | if (consumed == 0) { |
| 191 | // inlined implementation of Character.toChars(Character.codePointAt(input, pos)) |
| 192 | // avoids allocating temp char arrays and duplicate checks |
| 193 | char c1 = input.charAt(pos); |
| 194 | out.write(c1); |
| 195 | pos++; |
| 196 | if (Character.isHighSurrogate(c1) && pos < len) { |
| 197 | char c2 = input.charAt(pos); |
| 198 | if (Character.isLowSurrogate(c2)) { |
| 199 | out.write(c2); |
| 200 | pos++; |
| 201 | } |
| 202 | } |
| 203 | continue; |
| 204 | } |
| 205 | // contract with translators is that they have to understand codepoints |
| 206 | // and they just took care of a surrogate pair |
| 207 | for (int pt = 0; pt < consumed; pt++) { |
| 208 | pos += Character.charCount(Character.codePointAt(input, pos)); |
| 209 | } |
| 210 | } |
| 211 | } |
| 212 | } |
| 213 | |
| 214 | /** |
| 215 | * Adapted from apache commons-lang3 project. |
| 216 | * <p> |
| 217 | * Translates a value using a lookup table. |
| 218 | * |
| 219 | * @since 3.0 |
| 220 | */ |
| 221 | private static class LookupTranslator extends CharSequenceTranslator { |
| 222 | |
| 223 | private final HashMap<String, String> lookupMap; |
| 224 | private final HashSet<Character> prefixSet; |
| 225 | private final int shortest; |
| 226 | private final int longest; |
| 227 | |
| 228 | /** |
| 229 | * Define the lookup table to be used in translation |
| 230 | * <p> |
| 231 | * Note that, as of Lang 3.1, the key to the lookup table is converted to a |
| 232 | * java.lang.String. This is because we need the key to support hashCode and |
| 233 | * equals(Object), allowing it to be the key for a HashMap. See LANG-882. |
| 234 | * |
| 235 | * @param lookup CharSequence[][] table of size [*][2] |
| 236 | */ |
| 237 | private LookupTranslator(final CharSequence[]... lookup) { |
| 238 | lookupMap = new HashMap<>(); |
| 239 | prefixSet = new HashSet<>(); |
| 240 | int _shortest = Integer.MAX_VALUE; |
| 241 | int _longest = 0; |
| 242 | if (lookup != null) { |
| 243 | for (final CharSequence[] seq : lookup) { |
| 244 | this.lookupMap.put(seq[0].toString(), seq[1].toString()); |
| 245 | this.prefixSet.add(seq[0].charAt(0)); |
| 246 | final int sz = seq[0].length(); |
| 247 | if (sz < _shortest) { |
| 248 | _shortest = sz; |
| 249 | } |
| 250 | if (sz > _longest) { |
| 251 | _longest = sz; |
| 252 | } |
| 253 | } |
| 254 | } |
| 255 | shortest = _shortest; |
| 256 | longest = _longest; |
| 257 | } |
| 258 | |
| 259 | /** |
| 260 | * {@inheritDoc} |
| 261 | */ |
| 262 | @Override |
| 263 | protected int translate(final CharSequence input, final int index, final Writer out) throws IOException { |
| 264 | // check if translation exists for the input at position index |
| 265 | if (prefixSet.contains(input.charAt(index))) { |
| 266 | int max = longest; |
| 267 | if (index + longest > input.length()) { |
| 268 | max = input.length() - index; |
| 269 | } |
| 270 | // implement greedy algorithm by trying maximum match first |
| 271 | for (int i = max; i >= shortest; i--) { |
| 272 | final CharSequence subSeq = input.subSequence(index, index + i); |
| 273 | final String result = lookupMap.get(subSeq.toString()); |
| 274 | |
| 275 | if (result != null) { |
| 276 | out.write(result); |
| 277 | return i; |
| 278 | } |
| 279 | } |
| 280 | } |
| 281 | return 0; |
| 282 | } |
| 283 | } |
| 284 | |
| 285 | /** |
| 286 | * Adapted from apache commons-lang3 project. |
| 287 | * <p> |
| 288 | * Executes a sequence of translators one after the other. Execution ends whenever |
| 289 | * the first translator consumes codepoints from the input. |
| 290 | * |
| 291 | * @since 3.0 |
| 292 | */ |
| 293 | private static class AggregateTranslator extends CharSequenceTranslator { |
| 294 | |
| 295 | private final CharSequenceTranslator[] translators; |
| 296 | |
| 297 | /** |
| 298 | * Specify the translators to be used at creation time. |
| 299 | * |
| 300 | * @param translators CharSequenceTranslator array to aggregate |
| 301 | */ |
| 302 | private AggregateTranslator(final CharSequenceTranslator... translators) { |
| 303 | this.translators = translators == null ? null : translators.clone(); |
| 304 | } |
| 305 | |
| 306 | /** |
| 307 | * The first translator to consume codepoints from the input is the 'winner'. |
| 308 | * Execution stops with the number of consumed codepoints being returned. |
| 309 | * {@inheritDoc} |
| 310 | */ |
| 311 | @Override |
| 312 | protected int translate(final CharSequence input, final int index, final Writer out) throws IOException { |
| 313 | for (final CharSequenceTranslator translator : translators) { |
| 314 | final int consumed = translator.translate(input, index, out); |
| 315 | if (consumed != 0) { |
| 316 | return consumed; |
| 317 | } |
| 318 | } |
| 319 | return 0; |
| 320 | } |
| 321 | |
| 322 | } |
| 323 | |
| 324 | /** |
| 325 | * Adapted from apache commons-lang3 project. |
| 326 | * <p> |
| 327 | * Translate escaped octal Strings back to their octal values. |
| 328 | * <p> |
| 329 | * For example, "\45" should go back to being the specific value (a %). |
| 330 | * <p> |
| 331 | * Note that this currently only supports the viable range of octal for Java; namely |
| 332 | * 1 to 377. This is because parsing Java is the main use case. |
| 333 | * |
| 334 | * @since 3.0 |
| 335 | */ |
| 336 | private static class OctalUnescaper extends CharSequenceTranslator { |
| 337 | |
| 338 | /** |
| 339 | * {@inheritDoc} |
| 340 | */ |
| 341 | @Override |
| 342 | protected int translate(final CharSequence input, final int index, final Writer out) throws IOException { |
| 343 | final int remaining = input.length() - index - 1; // how many characters left, ignoring the first \ |
| 344 | final StringBuilder builder = new StringBuilder(); |
| 345 | if (input.charAt(index) == '\\' && remaining > 0 && isOctalDigit(input.charAt(index + 1))) { |
| 346 | final int next = index + 1; |
| 347 | final int next2 = index + 2; |
| 348 | final int next3 = index + 3; |
| 349 | |
| 350 | // we know this is good as we checked it in the if block above |
| 351 | builder.append(input.charAt(next)); |
| 352 | |
| 353 | if (remaining > 1 && isOctalDigit(input.charAt(next2))) { |
| 354 | builder.append(input.charAt(next2)); |
| 355 | if (remaining > 2 && isZeroToThree(input.charAt(next)) && isOctalDigit(input.charAt(next3))) { |
| 356 | builder.append(input.charAt(next3)); |
| 357 | } |
| 358 | } |
| 359 | |
| 360 | out.write(Integer.parseInt(builder.toString(), 8)); |
| 361 | return 1 + builder.length(); |
| 362 | } |
| 363 | return 0; |
| 364 | } |
| 365 | |
| 366 | /** |
| 367 | * Checks if the given char is an octal digit. Octal digits are the character representations of the digits 0 to |
| 368 | * 7. |
| 369 | * |
| 370 | * @param ch the char to check |
| 371 | * @return true if the given char is the character representation of one of the digits from 0 to 7 |
| 372 | */ |
| 373 | private boolean isOctalDigit(final char ch) { |
| 374 | return ch >= '0' && ch <= '7'; |
| 375 | } |
| 376 | |
| 377 | /** |
| 378 | * Checks if the given char is the character representation of one of the digit from 0 to 3. |
| 379 | * |
| 380 | * @param ch the char to check |
| 381 | * @return true if the given char is the character representation of one of the digits from 0 to 3 |
| 382 | */ |
| 383 | private boolean isZeroToThree(final char ch) { |
| 384 | return ch >= '0' && ch <= '3'; |
| 385 | } |
| 386 | } |
| 387 | |
| 388 | /** |
| 389 | * Adapted from apache commons-lang3 project. |
| 390 | * <p> |
| 391 | * Translates escaped Unicode values of the form \\u+\d\d\d\d back to |
| 392 | * Unicode. It supports multiple 'u' characters and will work with or |
| 393 | * without the +. |
| 394 | * |
| 395 | * @since 3.0 |
| 396 | */ |
| 397 | private static class UnicodeUnescaper extends CharSequenceTranslator { |
| 398 | |
| 399 | /** |
| 400 | * {@inheritDoc} |
| 401 | */ |
| 402 | @Override |
| 403 | protected int translate(final CharSequence input, final int index, final Writer out) throws IOException { |
| 404 | if (input.charAt(index) == '\\' && index + 1 < input.length() && input.charAt(index + 1) == 'u') { |
| 405 | // consume optional additional 'u' chars |
| 406 | int i = 2; |
| 407 | while (index + i < input.length() && input.charAt(index + i) == 'u') { |
| 408 | i++; |
| 409 | } |
| 410 | |
| 411 | if (index + i < input.length() && input.charAt(index + i) == '+') { |
| 412 | i++; |
| 413 | } |
| 414 | |
| 415 | if (index + i + 4 <= input.length()) { |
| 416 | // Get 4 hex digits |
| 417 | final CharSequence unicode = input.subSequence(index + i, index + i + 4); |
| 418 | |
| 419 | try { |
| 420 | final int value = Integer.parseInt(unicode.toString(), 16); |
| 421 | out.write((char) value); |
| 422 | } catch (final NumberFormatException nfe) { |
| 423 | throw new IllegalArgumentException("Unable to parse unicode value: " + unicode, nfe); |
| 424 | } |
| 425 | return i + 4; |
| 426 | } |
| 427 | throw new IllegalArgumentException("Less than 4 hex digits in unicode value: '" + input.subSequence(index, input.length()) |
| 428 | + "' due to end of CharSequence"); |
| 429 | } |
| 430 | return 0; |
| 431 | } |
| 432 | } |
| 433 | |
| 434 | } |
| 435 |
Members