| 1 | package be.nikiroo.utils; |
| 2 | |
| 3 | import java.awt.Image; |
| 4 | import java.awt.geom.AffineTransform; |
| 5 | import java.awt.image.AffineTransformOp; |
| 6 | import java.awt.image.BufferedImage; |
| 7 | import java.io.ByteArrayInputStream; |
| 8 | import java.io.ByteArrayOutputStream; |
| 9 | import java.io.File; |
| 10 | import java.io.IOException; |
| 11 | import java.io.InputStream; |
| 12 | import java.security.MessageDigest; |
| 13 | import java.security.NoSuchAlgorithmException; |
| 14 | import java.text.Normalizer; |
| 15 | import java.text.Normalizer.Form; |
| 16 | import java.text.ParseException; |
| 17 | import java.text.SimpleDateFormat; |
| 18 | import java.util.Base64; |
| 19 | import java.util.Date; |
| 20 | import java.util.regex.Pattern; |
| 21 | |
| 22 | import javax.imageio.ImageIO; |
| 23 | |
| 24 | import org.unbescape.html.HtmlEscape; |
| 25 | import org.unbescape.html.HtmlEscapeLevel; |
| 26 | import org.unbescape.html.HtmlEscapeType; |
| 27 | |
| 28 | /** |
| 29 | * This class offer some utilities based around {@link String}s. |
| 30 | * |
| 31 | * @author niki |
| 32 | */ |
| 33 | public class StringUtils { |
| 34 | /** |
| 35 | * This enum type will decide the alignment of a {@link String} when padding |
| 36 | * is applied or if there is enough horizontal space for it to be aligned. |
| 37 | */ |
| 38 | public enum Alignment { |
| 39 | /** Aligned at left. */ |
| 40 | Beginning, |
| 41 | /** Centered. */ |
| 42 | Center, |
| 43 | /** Aligned at right. */ |
| 44 | End |
| 45 | } |
| 46 | |
| 47 | static private Pattern marks = Pattern |
| 48 | .compile("[\\p{InCombiningDiacriticalMarks}\\p{IsLm}\\p{IsSk}]+"); |
| 49 | |
| 50 | /** |
| 51 | * Fix the size of the given {@link String} either with space-padding or by |
| 52 | * shortening it. |
| 53 | * |
| 54 | * @param text |
| 55 | * the {@link String} to fix |
| 56 | * @param width |
| 57 | * the size of the resulting {@link String} or -1 for a noop |
| 58 | * |
| 59 | * @return the resulting {@link String} of size <i>size</i> |
| 60 | */ |
| 61 | static public String padString(String text, int width) { |
| 62 | return padString(text, width, true, Alignment.Beginning); |
| 63 | } |
| 64 | |
| 65 | /** |
| 66 | * Fix the size of the given {@link String} either with space-padding or by |
| 67 | * optionally shortening it. |
| 68 | * |
| 69 | * @param text |
| 70 | * the {@link String} to fix |
| 71 | * @param width |
| 72 | * the size of the resulting {@link String} if the text fits or |
| 73 | * if cut is TRUE or -1 for a noop |
| 74 | * @param cut |
| 75 | * cut the {@link String} shorter if needed |
| 76 | * @param align |
| 77 | * align the {@link String} in this position if we have enough |
| 78 | * space |
| 79 | * |
| 80 | * @return the resulting {@link String} of size <i>size</i> minimum |
| 81 | */ |
| 82 | static public String padString(String text, int width, boolean cut, |
| 83 | Alignment align) { |
| 84 | |
| 85 | if (width >= 0) { |
| 86 | if (text == null) |
| 87 | text = ""; |
| 88 | |
| 89 | int diff = width - text.length(); |
| 90 | |
| 91 | if (diff < 0) { |
| 92 | if (cut) |
| 93 | text = text.substring(0, width); |
| 94 | } else if (diff > 0) { |
| 95 | if (diff < 2 && align != Alignment.End) |
| 96 | align = Alignment.Beginning; |
| 97 | |
| 98 | switch (align) { |
| 99 | case Beginning: |
| 100 | text = text + new String(new char[diff]).replace('\0', ' '); |
| 101 | break; |
| 102 | case End: |
| 103 | text = new String(new char[diff]).replace('\0', ' ') + text; |
| 104 | break; |
| 105 | case Center: |
| 106 | default: |
| 107 | int pad1 = (diff) / 2; |
| 108 | int pad2 = (diff + 1) / 2; |
| 109 | text = new String(new char[pad1]).replace('\0', ' ') + text |
| 110 | + new String(new char[pad2]).replace('\0', ' '); |
| 111 | break; |
| 112 | } |
| 113 | } |
| 114 | } |
| 115 | |
| 116 | return text; |
| 117 | } |
| 118 | |
| 119 | /** |
| 120 | * Sanitise the given input to make it more Terminal-friendly by removing |
| 121 | * combining characters. |
| 122 | * |
| 123 | * @param input |
| 124 | * the input to sanitise |
| 125 | * @param allowUnicode |
| 126 | * allow Unicode or only allow ASCII Latin characters |
| 127 | * |
| 128 | * @return the sanitised {@link String} |
| 129 | */ |
| 130 | static public String sanitize(String input, boolean allowUnicode) { |
| 131 | return sanitize(input, allowUnicode, !allowUnicode); |
| 132 | } |
| 133 | |
| 134 | /** |
| 135 | * Sanitise the given input to make it more Terminal-friendly by removing |
| 136 | * combining characters. |
| 137 | * |
| 138 | * @param input |
| 139 | * the input to sanitise |
| 140 | * @param allowUnicode |
| 141 | * allow Unicode or only allow ASCII Latin characters |
| 142 | * @param removeAllAccents |
| 143 | * TRUE to replace all accentuated characters by their non |
| 144 | * accentuated counter-parts |
| 145 | * |
| 146 | * @return the sanitised {@link String} |
| 147 | */ |
| 148 | static public String sanitize(String input, boolean allowUnicode, |
| 149 | boolean removeAllAccents) { |
| 150 | |
| 151 | if (removeAllAccents) { |
| 152 | input = Normalizer.normalize(input, Form.NFKD); |
| 153 | input = marks.matcher(input).replaceAll(""); |
| 154 | } |
| 155 | |
| 156 | input = Normalizer.normalize(input, Form.NFKC); |
| 157 | |
| 158 | if (!allowUnicode) { |
| 159 | StringBuilder builder = new StringBuilder(); |
| 160 | for (int index = 0; index < input.length(); index++) { |
| 161 | char car = input.charAt(index); |
| 162 | // displayable chars in ASCII are in the range 32<->255, |
| 163 | // except DEL (127) |
| 164 | if (car >= 32 && car <= 255 && car != 127) { |
| 165 | builder.append(car); |
| 166 | } |
| 167 | } |
| 168 | input = builder.toString(); |
| 169 | } |
| 170 | |
| 171 | return input; |
| 172 | } |
| 173 | |
| 174 | /** |
| 175 | * Convert between time in milliseconds to {@link String} in a "static" way |
| 176 | * (to exchange data over the wire, for instance). |
| 177 | * |
| 178 | * @param time |
| 179 | * the time in milliseconds |
| 180 | * |
| 181 | * @return the time as a {@link String} |
| 182 | */ |
| 183 | static public String fromTime(long time) { |
| 184 | SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss"); |
| 185 | return sdf.format(new Date(time)); |
| 186 | } |
| 187 | |
| 188 | /** |
| 189 | * Convert between time as a {@link String} to milliseconds in a "static" |
| 190 | * way (to exchange data over the wire, for instance). |
| 191 | * |
| 192 | * @param time |
| 193 | * the time as a {@link String} |
| 194 | * |
| 195 | * @return the time in milliseconds |
| 196 | */ |
| 197 | static public long toTime(String display) { |
| 198 | SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss"); |
| 199 | try { |
| 200 | return sdf.parse(display).getTime(); |
| 201 | } catch (ParseException e) { |
| 202 | return -1; |
| 203 | } |
| 204 | } |
| 205 | |
| 206 | /** |
| 207 | * Convert the given {@link Image} object into a Base64 representation of |
| 208 | * the same {@link Image}. object. |
| 209 | * |
| 210 | * @param image |
| 211 | * the {@link Image} object to convert |
| 212 | * |
| 213 | * @return the Base64 representation |
| 214 | * |
| 215 | * @throws IOException |
| 216 | * in case of IO error |
| 217 | */ |
| 218 | static public String fromImage(BufferedImage image) throws IOException { |
| 219 | String imageString = null; |
| 220 | ByteArrayOutputStream out = new ByteArrayOutputStream(); |
| 221 | |
| 222 | ImageIO.write(image, "jpeg", out); |
| 223 | byte[] imageBytes = out.toByteArray(); |
| 224 | |
| 225 | imageString = new String(Base64.getEncoder().encode(imageBytes)); |
| 226 | |
| 227 | out.close(); |
| 228 | |
| 229 | return imageString; |
| 230 | } |
| 231 | |
| 232 | /** |
| 233 | * Convert the given {@link File} image into a Base64 representation of the |
| 234 | * same {@link File}. |
| 235 | * |
| 236 | * @param file |
| 237 | * the {@link File} image to convert |
| 238 | * |
| 239 | * @return the Base64 representation |
| 240 | * |
| 241 | * @throws IOException |
| 242 | * in case of IO error |
| 243 | */ |
| 244 | static public String fromStream(InputStream in) throws IOException { |
| 245 | String fileString = null; |
| 246 | ByteArrayOutputStream out = new ByteArrayOutputStream(); |
| 247 | |
| 248 | byte[] buf = new byte[8192]; |
| 249 | |
| 250 | int c = 0; |
| 251 | while ((c = in.read(buf, 0, buf.length)) > 0) { |
| 252 | out.write(buf, 0, c); |
| 253 | } |
| 254 | out.flush(); |
| 255 | in.close(); |
| 256 | |
| 257 | fileString = new String(Base64.getEncoder().encode(out.toByteArray())); |
| 258 | out.close(); |
| 259 | |
| 260 | return fileString; |
| 261 | } |
| 262 | |
| 263 | /** |
| 264 | * Convert the given Base64 representation of an image into an {@link Image} |
| 265 | * object. |
| 266 | * |
| 267 | * @param b64data |
| 268 | * the {@link Image} in Base64 format |
| 269 | * |
| 270 | * @return the {@link Image} object |
| 271 | * |
| 272 | * @throws IOException |
| 273 | * in case of IO error |
| 274 | */ |
| 275 | static public BufferedImage toImage(String b64data) throws IOException { |
| 276 | ByteArrayInputStream in = new ByteArrayInputStream(Base64.getDecoder() |
| 277 | .decode(b64data)); |
| 278 | return toImage(in); |
| 279 | } |
| 280 | |
| 281 | /** |
| 282 | * Convert the given {@link InputStream} (which must allow calls to |
| 283 | * {@link InputStream#reset()}) into an {@link Image} object. |
| 284 | * |
| 285 | * @param in |
| 286 | * the 'resetable' {@link InputStream} |
| 287 | * |
| 288 | * @return the {@link Image} object |
| 289 | * |
| 290 | * @throws IOException |
| 291 | * in case of IO error |
| 292 | */ |
| 293 | static public BufferedImage toImage(InputStream in) throws IOException { |
| 294 | int orientation; |
| 295 | try { |
| 296 | orientation = getExifTransorm(in); |
| 297 | } catch (Exception e) { |
| 298 | // no EXIF transform, ok |
| 299 | orientation = -1; |
| 300 | } |
| 301 | |
| 302 | in.reset(); |
| 303 | BufferedImage image = ImageIO.read(in); |
| 304 | |
| 305 | if (image == null) { |
| 306 | throw new IOException("Failed to convert input to image"); |
| 307 | } |
| 308 | |
| 309 | // Note: this code has been found on internet; |
| 310 | // thank you anonymous coder. |
| 311 | int width = image.getWidth(); |
| 312 | int height = image.getHeight(); |
| 313 | AffineTransform affineTransform = new AffineTransform(); |
| 314 | |
| 315 | switch (orientation) { |
| 316 | case 1: |
| 317 | break; |
| 318 | case 2: // Flip X |
| 319 | affineTransform.scale(-1.0, 1.0); |
| 320 | affineTransform.translate(-width, 0); |
| 321 | break; |
| 322 | case 3: // PI rotation |
| 323 | affineTransform.translate(width, height); |
| 324 | affineTransform.rotate(Math.PI); |
| 325 | break; |
| 326 | case 4: // Flip Y |
| 327 | affineTransform.scale(1.0, -1.0); |
| 328 | affineTransform.translate(0, -height); |
| 329 | break; |
| 330 | case 5: // - PI/2 and Flip X |
| 331 | affineTransform.rotate(-Math.PI / 2); |
| 332 | affineTransform.scale(-1.0, 1.0); |
| 333 | break; |
| 334 | case 6: // -PI/2 and -width |
| 335 | affineTransform.translate(height, 0); |
| 336 | affineTransform.rotate(Math.PI / 2); |
| 337 | break; |
| 338 | case 7: // PI/2 and Flip |
| 339 | affineTransform.scale(-1.0, 1.0); |
| 340 | affineTransform.translate(-height, 0); |
| 341 | affineTransform.translate(0, width); |
| 342 | affineTransform.rotate(3 * Math.PI / 2); |
| 343 | break; |
| 344 | case 8: // PI / 2 |
| 345 | affineTransform.translate(0, width); |
| 346 | affineTransform.rotate(3 * Math.PI / 2); |
| 347 | break; |
| 348 | default: |
| 349 | affineTransform = null; |
| 350 | break; |
| 351 | } |
| 352 | |
| 353 | if (affineTransform != null) { |
| 354 | AffineTransformOp affineTransformOp = new AffineTransformOp( |
| 355 | affineTransform, AffineTransformOp.TYPE_BILINEAR); |
| 356 | |
| 357 | BufferedImage transformedImage = new BufferedImage(height, width, |
| 358 | image.getType()); |
| 359 | transformedImage = affineTransformOp |
| 360 | .filter(image, transformedImage); |
| 361 | |
| 362 | image = transformedImage; |
| 363 | } |
| 364 | // |
| 365 | |
| 366 | return image; |
| 367 | } |
| 368 | |
| 369 | /** |
| 370 | * Return a hash of the given {@link String}. |
| 371 | * |
| 372 | * @param input |
| 373 | * the input data |
| 374 | * |
| 375 | * @return the hash |
| 376 | */ |
| 377 | static public String getHash(String input) { |
| 378 | try { |
| 379 | MessageDigest md = MessageDigest.getInstance("MD5"); |
| 380 | md.update(input.getBytes()); |
| 381 | byte byteData[] = md.digest(); |
| 382 | |
| 383 | StringBuffer hexString = new StringBuffer(); |
| 384 | for (int i = 0; i < byteData.length; i++) { |
| 385 | String hex = Integer.toHexString(0xff & byteData[i]); |
| 386 | if (hex.length() == 1) |
| 387 | hexString.append('0'); |
| 388 | hexString.append(hex); |
| 389 | } |
| 390 | |
| 391 | return hexString.toString(); |
| 392 | } catch (NoSuchAlgorithmException e) { |
| 393 | return input; |
| 394 | } |
| 395 | } |
| 396 | |
| 397 | /** |
| 398 | * Return the EXIF transformation flag of this image if any. |
| 399 | * |
| 400 | * <p> |
| 401 | * Note: this code has been found on internet; thank you anonymous coder. |
| 402 | * </p> |
| 403 | * |
| 404 | * @param in |
| 405 | * the data {@link InputStream} |
| 406 | * |
| 407 | * @return the transformation flag if any |
| 408 | * |
| 409 | * @throws IOException |
| 410 | * in case of IO error |
| 411 | */ |
| 412 | static private int getExifTransorm(InputStream in) throws IOException { |
| 413 | int[] exif_data = new int[100]; |
| 414 | int set_flag = 0; |
| 415 | int is_motorola = 0; |
| 416 | |
| 417 | /* Read File head, check for JPEG SOI + Exif APP1 */ |
| 418 | for (int i = 0; i < 4; i++) |
| 419 | exif_data[i] = in.read(); |
| 420 | |
| 421 | if (exif_data[0] != 0xFF || exif_data[1] != 0xD8 |
| 422 | || exif_data[2] != 0xFF || exif_data[3] != 0xE1) |
| 423 | return -2; |
| 424 | |
| 425 | /* Get the marker parameter length count */ |
| 426 | int length = (in.read() << 8 | in.read()); |
| 427 | |
| 428 | /* Length includes itself, so must be at least 2 */ |
| 429 | /* Following Exif data length must be at least 6 */ |
| 430 | if (length < 8) |
| 431 | return -1; |
| 432 | length -= 8; |
| 433 | /* Read Exif head, check for "Exif" */ |
| 434 | for (int i = 0; i < 6; i++) |
| 435 | exif_data[i] = in.read(); |
| 436 | |
| 437 | if (exif_data[0] != 0x45 || exif_data[1] != 0x78 |
| 438 | || exif_data[2] != 0x69 || exif_data[3] != 0x66 |
| 439 | || exif_data[4] != 0 || exif_data[5] != 0) |
| 440 | return -1; |
| 441 | |
| 442 | /* Read Exif body */ |
| 443 | length = length > exif_data.length ? exif_data.length : length; |
| 444 | for (int i = 0; i < length; i++) |
| 445 | exif_data[i] = in.read(); |
| 446 | |
| 447 | if (length < 12) |
| 448 | return -1; /* Length of an IFD entry */ |
| 449 | |
| 450 | /* Discover byte order */ |
| 451 | if (exif_data[0] == 0x49 && exif_data[1] == 0x49) |
| 452 | is_motorola = 0; |
| 453 | else if (exif_data[0] == 0x4D && exif_data[1] == 0x4D) |
| 454 | is_motorola = 1; |
| 455 | else |
| 456 | return -1; |
| 457 | |
| 458 | /* Check Tag Mark */ |
| 459 | if (is_motorola == 1) { |
| 460 | if (exif_data[2] != 0) |
| 461 | return -1; |
| 462 | if (exif_data[3] != 0x2A) |
| 463 | return -1; |
| 464 | } else { |
| 465 | if (exif_data[3] != 0) |
| 466 | return -1; |
| 467 | if (exif_data[2] != 0x2A) |
| 468 | return -1; |
| 469 | } |
| 470 | |
| 471 | /* Get first IFD offset (offset to IFD0) */ |
| 472 | int offset; |
| 473 | if (is_motorola == 1) { |
| 474 | if (exif_data[4] != 0) |
| 475 | return -1; |
| 476 | if (exif_data[5] != 0) |
| 477 | return -1; |
| 478 | offset = exif_data[6]; |
| 479 | offset <<= 8; |
| 480 | offset += exif_data[7]; |
| 481 | } else { |
| 482 | if (exif_data[7] != 0) |
| 483 | return -1; |
| 484 | if (exif_data[6] != 0) |
| 485 | return -1; |
| 486 | offset = exif_data[5]; |
| 487 | offset <<= 8; |
| 488 | offset += exif_data[4]; |
| 489 | } |
| 490 | if (offset > length - 2) |
| 491 | return -1; /* check end of data segment */ |
| 492 | |
| 493 | /* Get the number of directory entries contained in this IFD */ |
| 494 | int number_of_tags; |
| 495 | if (is_motorola == 1) { |
| 496 | number_of_tags = exif_data[offset]; |
| 497 | number_of_tags <<= 8; |
| 498 | number_of_tags += exif_data[offset + 1]; |
| 499 | } else { |
| 500 | number_of_tags = exif_data[offset + 1]; |
| 501 | number_of_tags <<= 8; |
| 502 | number_of_tags += exif_data[offset]; |
| 503 | } |
| 504 | if (number_of_tags == 0) |
| 505 | return -1; |
| 506 | offset += 2; |
| 507 | |
| 508 | /* Search for Orientation Tag in IFD0 */ |
| 509 | for (;;) { |
| 510 | if (offset > length - 12) |
| 511 | return -1; /* check end of data segment */ |
| 512 | /* Get Tag number */ |
| 513 | int tagnum; |
| 514 | if (is_motorola == 1) { |
| 515 | tagnum = exif_data[offset]; |
| 516 | tagnum <<= 8; |
| 517 | tagnum += exif_data[offset + 1]; |
| 518 | } else { |
| 519 | tagnum = exif_data[offset + 1]; |
| 520 | tagnum <<= 8; |
| 521 | tagnum += exif_data[offset]; |
| 522 | } |
| 523 | if (tagnum == 0x0112) |
| 524 | break; /* found Orientation Tag */ |
| 525 | if (--number_of_tags == 0) |
| 526 | return -1; |
| 527 | offset += 12; |
| 528 | } |
| 529 | |
| 530 | /* Get the Orientation value */ |
| 531 | if (is_motorola == 1) { |
| 532 | if (exif_data[offset + 8] != 0) |
| 533 | return -1; |
| 534 | set_flag = exif_data[offset + 9]; |
| 535 | } else { |
| 536 | if (exif_data[offset + 9] != 0) |
| 537 | return -1; |
| 538 | set_flag = exif_data[offset + 8]; |
| 539 | } |
| 540 | if (set_flag > 8) |
| 541 | return -1; |
| 542 | |
| 543 | return set_flag; |
| 544 | } |
| 545 | |
| 546 | /** |
| 547 | * Remove the HTML content from the given input, and un-html-ize the rest. |
| 548 | * |
| 549 | * @param html |
| 550 | * the HTML-encoded content |
| 551 | * |
| 552 | * @return the HTML-free equivalent content |
| 553 | */ |
| 554 | public static String unhtml(String html) { |
| 555 | StringBuilder builder = new StringBuilder(); |
| 556 | |
| 557 | int inTag = 0; |
| 558 | for (char car : html.toCharArray()) { |
| 559 | if (car == '<') { |
| 560 | inTag++; |
| 561 | } else if (car == '>') { |
| 562 | inTag--; |
| 563 | } else if (inTag <= 0) { |
| 564 | builder.append(car); |
| 565 | } |
| 566 | } |
| 567 | |
| 568 | return HtmlEscape.unescapeHtml(builder.toString()); |
| 569 | } |
| 570 | |
| 571 | /** |
| 572 | * Escape the given {@link String} so it can be used in XML, as content. |
| 573 | * |
| 574 | * @param input |
| 575 | * the input {@link String} |
| 576 | * |
| 577 | * @return the escaped {@link String} |
| 578 | */ |
| 579 | public static String xmlEscape(String input) { |
| 580 | if (input == null) { |
| 581 | return ""; |
| 582 | } |
| 583 | |
| 584 | return HtmlEscape.escapeHtml(input, |
| 585 | HtmlEscapeType.HTML4_NAMED_REFERENCES_DEFAULT_TO_HEXA, |
| 586 | HtmlEscapeLevel.LEVEL_1_ONLY_MARKUP_SIGNIFICANT); |
| 587 | } |
| 588 | |
| 589 | /** |
| 590 | * Escape the given {@link String} so it can be used in XML, as text content |
| 591 | * inside double-quotes. |
| 592 | * |
| 593 | * @param input |
| 594 | * the input {@link String} |
| 595 | * |
| 596 | * @return the escaped {@link String} |
| 597 | */ |
| 598 | public static String xmlEscapeQuote(String input) { |
| 599 | if (input == null) { |
| 600 | return ""; |
| 601 | } |
| 602 | |
| 603 | return HtmlEscape.escapeHtml(input, |
| 604 | HtmlEscapeType.HTML4_NAMED_REFERENCES_DEFAULT_TO_HEXA, |
| 605 | HtmlEscapeLevel.LEVEL_1_ONLY_MARKUP_SIGNIFICANT); |
| 606 | } |
| 607 | } |