Kenny Root | f8d0f09 | 2010-03-12 08:15:28 -0800 | [diff] [blame] | 1 | /* |
| 2 | * Copyright (C) 2007 The Android Open Source Project |
| 3 | * |
| 4 | * Licensed under the Apache License, Version 2.0 (the "License"); |
| 5 | * you may not use this file except in compliance with the License. |
| 6 | * You may obtain a copy of the License at |
| 7 | * |
| 8 | * http://www.apache.org/licenses/LICENSE-2.0 |
| 9 | * |
| 10 | * Unless required by applicable law or agreed to in writing, software |
| 11 | * distributed under the License is distributed on an "AS IS" BASIS, |
| 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 13 | * See the License for the specific language governing permissions and |
| 14 | * limitations under the License. |
| 15 | */ |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 16 | |
| 17 | package com.google.android.util; |
| 18 | |
Artur Satayev | 2a9f3b8 | 2019-12-10 17:47:55 +0000 | [diff] [blame] | 19 | import android.compat.annotation.UnsupportedAppUsage; |
Artur Satayev | fc46be7 | 2019-11-04 17:50:59 +0000 | [diff] [blame] | 20 | |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 21 | import java.util.ArrayList; |
| 22 | import java.util.HashMap; |
| 23 | import java.util.LinkedHashMap; |
Artur Satayev | fc46be7 | 2019-11-04 17:50:59 +0000 | [diff] [blame] | 24 | import java.util.List; |
| 25 | import java.util.Set; |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 26 | import java.util.regex.Matcher; |
| 27 | import java.util.regex.Pattern; |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 28 | |
| 29 | /** |
| 30 | * |
| 31 | * Logic for parsing a text message typed by the user looking for smileys, |
| 32 | * urls, acronyms,formatting (e.g., '*'s for bold), me commands |
| 33 | * (e.g., "/me is asleep"), and punctuation. |
| 34 | * |
| 35 | * It constructs an array, which breaks the text up into its |
| 36 | * constituent pieces, which we return to the client. |
| 37 | * |
| 38 | */ |
| 39 | public abstract class AbstractMessageParser { |
| 40 | /** |
| 41 | * Interface representing the set of resources needed by a message parser |
| 42 | * |
| 43 | * @author jessan (Jessan Hutchison-Quillian) |
| 44 | */ |
| 45 | public static interface Resources { |
| 46 | |
| 47 | /** Get the known set of URL schemes. */ |
| 48 | public Set<String> getSchemes(); |
| 49 | |
| 50 | /** Get the possible values for the last part of a domain name. |
| 51 | * Values are expected to be reversed in the Trie. |
| 52 | */ |
| 53 | public TrieNode getDomainSuffixes(); |
| 54 | |
| 55 | /** Get the smileys accepted by the parser. */ |
| 56 | public TrieNode getSmileys(); |
| 57 | |
| 58 | /** Get the acronyms accepted by the parser. */ |
| 59 | public TrieNode getAcronyms(); |
| 60 | } |
| 61 | |
| 62 | /** |
| 63 | * Subclasses must define the schemes, domains, smileys and acronyms |
| 64 | * that are necessary for parsing |
| 65 | */ |
| 66 | protected abstract Resources getResources(); |
| 67 | |
| 68 | /** Music note that indicates user is listening to a music track. */ |
| 69 | public static final String musicNote = "\u266B "; |
| 70 | |
| 71 | private String text; |
| 72 | private int nextChar; |
| 73 | private int nextClass; |
| 74 | private ArrayList<Part> parts; |
| 75 | private ArrayList<Token> tokens; |
| 76 | private HashMap<Character,Format> formatStart; |
| 77 | private boolean parseSmilies; |
| 78 | private boolean parseAcronyms; |
| 79 | private boolean parseFormatting; |
| 80 | private boolean parseUrls; |
| 81 | private boolean parseMeText; |
| 82 | private boolean parseMusic; |
| 83 | |
| 84 | /** |
| 85 | * Create a message parser to parse urls, formatting, acronyms, smileys, |
| 86 | * /me text and music |
| 87 | * |
| 88 | * @param text the text to parse |
| 89 | */ |
| 90 | public AbstractMessageParser(String text) { |
| 91 | this(text, true, true, true, true, true, true); |
| 92 | } |
| 93 | |
| 94 | /** |
| 95 | * Create a message parser, specifying the kinds of text to parse |
| 96 | * |
| 97 | * @param text the text to parse |
| 98 | * |
| 99 | */ |
| 100 | public AbstractMessageParser(String text, boolean parseSmilies, |
| 101 | boolean parseAcronyms, boolean parseFormatting, boolean parseUrls, |
| 102 | boolean parseMusic, boolean parseMeText) { |
| 103 | this.text = text; |
| 104 | this.nextChar = 0; |
| 105 | this.nextClass = 10; |
| 106 | this.parts = new ArrayList<Part>(); |
| 107 | this.tokens = new ArrayList<Token>(); |
| 108 | this.formatStart = new HashMap<Character,Format>(); |
| 109 | this.parseSmilies = parseSmilies; |
| 110 | this.parseAcronyms = parseAcronyms; |
| 111 | this.parseFormatting = parseFormatting; |
| 112 | this.parseUrls = parseUrls; |
| 113 | this.parseMusic = parseMusic; |
| 114 | this.parseMeText = parseMeText; |
| 115 | } |
| 116 | |
| 117 | /** Returns the raw text being parsed. */ |
| 118 | public final String getRawText() { return text; } |
| 119 | |
| 120 | /** Return the number of parts. */ |
| 121 | public final int getPartCount() { return parts.size(); } |
| 122 | |
| 123 | /** Return the part at the given index. */ |
| 124 | public final Part getPart(int index) { return parts.get(index); } |
| 125 | |
| 126 | /** Return the list of parts from the parsed text */ |
| 127 | public final List<Part> getParts() { return parts; } |
| 128 | |
| 129 | /** Parses the text string into an internal representation. */ |
| 130 | public void parse() { |
| 131 | // Look for music track (of which there would be only one and it'll be the |
| 132 | // first token) |
| 133 | if (parseMusicTrack()) { |
| 134 | buildParts(null); |
| 135 | return; |
| 136 | } |
| 137 | |
| 138 | // Look for me commands. |
| 139 | String meText = null; |
| 140 | if (parseMeText && text.startsWith("/me") && (text.length() > 3) && |
| 141 | Character.isWhitespace(text.charAt(3))) { |
| 142 | meText = text.substring(0, 4); |
| 143 | text = text.substring(4); |
| 144 | } |
| 145 | |
| 146 | // Break the text into tokens. |
| 147 | boolean wasSmiley = false; |
| 148 | while (nextChar < text.length()) { |
| 149 | if (!isWordBreak(nextChar)) { |
| 150 | if (!wasSmiley || !isSmileyBreak(nextChar)) { |
| 151 | throw new AssertionError("last chunk did not end at word break"); |
| 152 | } |
| 153 | } |
| 154 | |
| 155 | if (parseSmiley()) { |
| 156 | wasSmiley = true; |
| 157 | } else { |
| 158 | wasSmiley = false; |
| 159 | |
| 160 | if (!parseAcronym() && !parseURL() && !parseFormatting()) { |
| 161 | parseText(); |
| 162 | } |
| 163 | } |
| 164 | } |
| 165 | |
| 166 | // Trim the whitespace before and after media components. |
| 167 | for (int i = 0; i < tokens.size(); ++i) { |
| 168 | if (tokens.get(i).isMedia()) { |
| 169 | if ((i > 0) && (tokens.get(i - 1) instanceof Html)) { |
| 170 | ((Html)tokens.get(i - 1)).trimLeadingWhitespace(); |
| 171 | } |
| 172 | if ((i + 1 < tokens.size()) && (tokens.get(i + 1) instanceof Html)) { |
| 173 | ((Html)tokens.get(i + 1)).trimTrailingWhitespace(); |
| 174 | } |
| 175 | } |
| 176 | } |
| 177 | |
| 178 | // Remove any empty html tokens. |
| 179 | for (int i = 0; i < tokens.size(); ++i) { |
| 180 | if (tokens.get(i).isHtml() && |
| 181 | (tokens.get(i).toHtml(true).length() == 0)) { |
| 182 | tokens.remove(i); |
| 183 | --i; // visit this index again |
| 184 | } |
| 185 | } |
| 186 | |
| 187 | buildParts(meText); |
| 188 | } |
| 189 | |
| 190 | /** |
| 191 | * Get a the appropriate Token for a given URL |
| 192 | * |
| 193 | * @param text the anchor text |
| 194 | * @param url the url |
| 195 | * |
| 196 | */ |
| 197 | public static Token tokenForUrl(String url, String text) { |
| 198 | if(url == null) { |
| 199 | return null; |
| 200 | } |
| 201 | |
| 202 | //Look for video links |
| 203 | Video video = Video.matchURL(url, text); |
| 204 | if (video != null) { |
| 205 | return video; |
| 206 | } |
| 207 | |
| 208 | // Look for video links. |
| 209 | YouTubeVideo ytVideo = YouTubeVideo.matchURL(url, text); |
| 210 | if (ytVideo != null) { |
| 211 | return ytVideo; |
| 212 | } |
| 213 | |
| 214 | // Look for photo links. |
| 215 | Photo photo = Photo.matchURL(url, text); |
| 216 | if (photo != null) { |
| 217 | return photo; |
| 218 | } |
| 219 | |
| 220 | // Look for photo links. |
| 221 | FlickrPhoto flickrPhoto = FlickrPhoto.matchURL(url, text); |
| 222 | if (flickrPhoto != null) { |
| 223 | return flickrPhoto; |
| 224 | } |
| 225 | |
| 226 | //Not media, so must be a regular URL |
| 227 | return new Link(url, text); |
| 228 | } |
| 229 | |
| 230 | /** |
| 231 | * Builds the parts list. |
| 232 | * |
| 233 | * @param meText any meText parsed from the message |
| 234 | */ |
| 235 | private void buildParts(String meText) { |
| 236 | for (int i = 0; i < tokens.size(); ++i) { |
| 237 | Token token = tokens.get(i); |
| 238 | if (token.isMedia() || (parts.size() == 0) || lastPart().isMedia()) { |
| 239 | parts.add(new Part()); |
| 240 | } |
| 241 | lastPart().add(token); |
| 242 | } |
| 243 | |
| 244 | // The first part inherits the meText of the line. |
| 245 | if (parts.size() > 0) { |
| 246 | parts.get(0).setMeText(meText); |
| 247 | } |
| 248 | } |
| 249 | |
| 250 | /** Returns the last part in the list. */ |
| 251 | private Part lastPart() { return parts.get(parts.size() - 1); } |
| 252 | |
| 253 | /** |
| 254 | * Looks for a music track (\u266B is first character, everything else is |
| 255 | * track info). |
| 256 | */ |
| 257 | private boolean parseMusicTrack() { |
| 258 | |
| 259 | if (parseMusic && text.startsWith(musicNote)) { |
| 260 | addToken(new MusicTrack(text.substring(musicNote.length()))); |
| 261 | nextChar = text.length(); |
| 262 | return true; |
| 263 | } |
| 264 | return false; |
| 265 | } |
| 266 | |
| 267 | /** Consumes all of the text in the next word . */ |
| 268 | private void parseText() { |
| 269 | StringBuilder buf = new StringBuilder(); |
| 270 | int start = nextChar; |
| 271 | do { |
| 272 | char ch = text.charAt(nextChar++); |
| 273 | switch (ch) { |
| 274 | case '<': buf.append("<"); break; |
| 275 | case '>': buf.append(">"); break; |
| 276 | case '&': buf.append("&"); break; |
| 277 | case '"': buf.append("""); break; |
| 278 | case '\'': buf.append("'"); break; |
| 279 | case '\n': buf.append("<br>"); break; |
| 280 | default: buf.append(ch); break; |
| 281 | } |
| 282 | } while (!isWordBreak(nextChar)); |
| 283 | |
| 284 | addToken(new Html(text.substring(start, nextChar), buf.toString())); |
| 285 | } |
| 286 | |
| 287 | /** |
| 288 | * Looks for smileys (e.g., ":)") in the text. The set of known smileys is |
| 289 | * loaded from a file into a trie at server start. |
| 290 | */ |
| 291 | private boolean parseSmiley() { |
| 292 | if(!parseSmilies) { |
| 293 | return false; |
| 294 | } |
| 295 | TrieNode match = longestMatch(getResources().getSmileys(), this, nextChar, |
| 296 | true); |
| 297 | if (match == null) { |
| 298 | return false; |
| 299 | } else { |
| 300 | int previousCharClass = getCharClass(nextChar - 1); |
| 301 | int nextCharClass = getCharClass(nextChar + match.getText().length()); |
| 302 | if ((previousCharClass == 2 || previousCharClass == 3) |
| 303 | && (nextCharClass == 2 || nextCharClass == 3)) { |
| 304 | return false; |
| 305 | } |
| 306 | addToken(new Smiley(match.getText())); |
| 307 | nextChar += match.getText().length(); |
| 308 | return true; |
| 309 | } |
| 310 | } |
| 311 | |
| 312 | /** Looks for acronyms (e.g., "lol") in the text. |
| 313 | */ |
| 314 | private boolean parseAcronym() { |
| 315 | if(!parseAcronyms) { |
| 316 | return false; |
| 317 | } |
| 318 | TrieNode match = longestMatch(getResources().getAcronyms(), this, nextChar); |
| 319 | if (match == null) { |
| 320 | return false; |
| 321 | } else { |
| 322 | addToken(new Acronym(match.getText(), match.getValue())); |
| 323 | nextChar += match.getText().length(); |
| 324 | return true; |
| 325 | } |
| 326 | } |
| 327 | |
| 328 | /** Determines if this is an allowable domain character. */ |
| 329 | private boolean isDomainChar(char c) { |
| 330 | return c == '-' || Character.isLetter(c) || Character.isDigit(c); |
| 331 | } |
| 332 | |
| 333 | /** Determines if the given string is a valid domain. */ |
| 334 | private boolean isValidDomain(String domain) { |
| 335 | // For hostnames, check that it ends with a known domain suffix |
| 336 | if (matches(getResources().getDomainSuffixes(), reverse(domain))) { |
| 337 | return true; |
| 338 | } |
| 339 | return false; |
| 340 | } |
| 341 | |
| 342 | /** |
| 343 | * Looks for a URL in two possible forms: either a proper URL with a known |
| 344 | * scheme or a domain name optionally followed by a path, query, or query. |
| 345 | */ |
| 346 | private boolean parseURL() { |
| 347 | // Make sure this is a valid place to start a URL. |
| 348 | if (!parseUrls || !isURLBreak(nextChar)) { |
| 349 | return false; |
| 350 | } |
| 351 | |
| 352 | int start = nextChar; |
| 353 | |
| 354 | // Search for the first block of letters. |
| 355 | int index = start; |
| 356 | while ((index < text.length()) && isDomainChar(text.charAt(index))) { |
| 357 | index += 1; |
| 358 | } |
| 359 | |
| 360 | String url = ""; |
| 361 | boolean done = false; |
| 362 | |
| 363 | if (index == text.length()) { |
| 364 | return false; |
| 365 | } else if (text.charAt(index) == ':') { |
| 366 | // Make sure this is a known scheme. |
| 367 | String scheme = text.substring(nextChar, index); |
| 368 | if (!getResources().getSchemes().contains(scheme)) { |
| 369 | return false; |
| 370 | } |
| 371 | } else if (text.charAt(index) == '.') { |
| 372 | // Search for the end of the domain name. |
| 373 | while (index < text.length()) { |
| 374 | char ch = text.charAt(index); |
| 375 | if ((ch != '.') && !isDomainChar(ch)) { |
| 376 | break; |
| 377 | } else { |
| 378 | index += 1; |
| 379 | } |
| 380 | } |
| 381 | |
| 382 | // Make sure the domain name has a valid suffix. Since tries look for |
| 383 | // prefix matches, we reverse all the strings to get suffix comparisons. |
| 384 | String domain = text.substring(nextChar, index); |
| 385 | if (!isValidDomain(domain)) { |
| 386 | return false; |
| 387 | } |
| 388 | |
| 389 | // Search for a port. We deal with this specially because a colon can |
| 390 | // also be a punctuation character. |
| 391 | if ((index + 1 < text.length()) && (text.charAt(index) == ':')) { |
| 392 | char ch = text.charAt(index + 1); |
| 393 | if (Character.isDigit(ch)) { |
| 394 | index += 1; |
| 395 | while ((index < text.length()) && |
| 396 | Character.isDigit(text.charAt(index))) { |
| 397 | index += 1; |
| 398 | } |
| 399 | } |
| 400 | } |
| 401 | |
| 402 | // The domain name should be followed by end of line, whitespace, |
| 403 | // punctuation, or a colon, slash, question, or hash character. The |
| 404 | // tricky part here is that some URL characters are also punctuation, so |
| 405 | // we need to distinguish them. Since we looked for ports above, a colon |
| 406 | // is always punctuation here. To distinguish '?' cases, we look at the |
| 407 | // character that follows it. |
| 408 | if (index == text.length()) { |
| 409 | done = true; |
| 410 | } else { |
| 411 | char ch = text.charAt(index); |
| 412 | if (ch == '?') { |
| 413 | // If the next character is whitespace or punctuation (or missing), |
| 414 | // then this question mark looks like punctuation. |
| 415 | if (index + 1 == text.length()) { |
| 416 | done = true; |
| 417 | } else { |
| 418 | char ch2 = text.charAt(index + 1); |
| 419 | if (Character.isWhitespace(ch2) || isPunctuation(ch2)) { |
| 420 | done = true; |
| 421 | } |
| 422 | } |
| 423 | } else if (isPunctuation(ch)) { |
| 424 | done = true; |
| 425 | } else if (Character.isWhitespace(ch)) { |
| 426 | done = true; |
| 427 | } else if ((ch == '/') || (ch == '#')) { |
| 428 | // In this case, the URL is not done. We will search for the end of |
| 429 | // it below. |
| 430 | } else { |
| 431 | return false; |
| 432 | } |
| 433 | } |
| 434 | |
| 435 | // We will assume the user meant HTTP. (One weird case is where they |
| 436 | // type a port of 443. That could mean HTTPS, but they might also want |
| 437 | // HTTP. We'll let them specify if they don't want HTTP.) |
| 438 | url = "http://"; |
| 439 | } else { |
| 440 | return false; |
| 441 | } |
| 442 | |
| 443 | // If the URL is not done, search for the end, which is just before the |
| 444 | // next whitespace character. |
| 445 | if (!done) { |
| 446 | while ((index < text.length()) && |
| 447 | !Character.isWhitespace(text.charAt(index))) { |
| 448 | index += 1; |
| 449 | } |
| 450 | } |
| 451 | |
| 452 | String urlText = text.substring(start, index); |
| 453 | url += urlText; |
| 454 | |
| 455 | // Figure out the appropriate token type. |
| 456 | addURLToken(url, urlText); |
| 457 | |
| 458 | nextChar = index; |
| 459 | return true; |
| 460 | } |
| 461 | |
| 462 | /** |
| 463 | * Adds the appropriate token for the given URL. This might be a simple |
| 464 | * link or it might be a recognized media type. |
| 465 | */ |
| 466 | private void addURLToken(String url, String text) { |
| 467 | addToken(tokenForUrl(url, text)); |
| 468 | } |
| 469 | |
| 470 | /** |
| 471 | * Deal with formatting characters. |
| 472 | * |
| 473 | * Parsing is as follows: |
| 474 | * - Treat all contiguous strings of formatting characters as one block. |
| 475 | * (This method processes one block.) |
| 476 | * - Only a single instance of a particular format character within a block |
| 477 | * is used to determine whether to turn on/off that type of formatting; |
| 478 | * other instances simply print the character itself. |
| 479 | * - If the format is to be turned on, we use the _first_ instance; if it |
| 480 | * is to be turned off, we use the _last_ instance (by appending the |
| 481 | * format.) |
| 482 | * |
| 483 | * Example: |
| 484 | * **string** turns into <b>*string*</b> |
| 485 | */ |
| 486 | private boolean parseFormatting() { |
| 487 | if(!parseFormatting) { |
| 488 | return false; |
| 489 | } |
| 490 | int endChar = nextChar; |
| 491 | while ((endChar < text.length()) && isFormatChar(text.charAt(endChar))) { |
| 492 | endChar += 1; |
| 493 | } |
| 494 | |
| 495 | if ((endChar == nextChar) || !isWordBreak(endChar)) { |
| 496 | return false; |
| 497 | } |
| 498 | |
| 499 | // Keeps track of whether we've seen a character (in map if we've seen it) |
| 500 | // and whether we should append a closing format token (if value in |
| 501 | // map is TRUE). Linked hashmap for consistent ordering. |
| 502 | LinkedHashMap<Character, Boolean> seenCharacters = |
| 503 | new LinkedHashMap<Character, Boolean>(); |
| 504 | |
| 505 | for (int index = nextChar; index < endChar; ++index) { |
| 506 | char ch = text.charAt(index); |
| 507 | Character key = Character.valueOf(ch); |
| 508 | if (seenCharacters.containsKey(key)) { |
| 509 | // Already seen this character, just append an unmatched token, which |
| 510 | // will print plaintext character |
| 511 | addToken(new Format(ch, false)); |
| 512 | } else { |
| 513 | Format start = formatStart.get(key); |
| 514 | if (start != null) { |
| 515 | // Match the start token, and ask an end token to be appended |
| 516 | start.setMatched(true); |
| 517 | formatStart.remove(key); |
| 518 | seenCharacters.put(key, Boolean.TRUE); |
| 519 | } else { |
| 520 | // Append start token |
| 521 | start = new Format(ch, true); |
| 522 | formatStart.put(key, start); |
| 523 | addToken(start); |
| 524 | seenCharacters.put(key, Boolean.FALSE); |
| 525 | } |
| 526 | } |
| 527 | } |
| 528 | |
| 529 | // Append any necessary end tokens |
| 530 | for (Character key : seenCharacters.keySet()) { |
| 531 | if (seenCharacters.get(key) == Boolean.TRUE) { |
| 532 | Format end = new Format(key.charValue(), false); |
| 533 | end.setMatched(true); |
| 534 | addToken(end); |
| 535 | } |
| 536 | } |
| 537 | |
| 538 | nextChar = endChar; |
| 539 | return true; |
| 540 | } |
| 541 | |
| 542 | /** Determines whether the given index could be a possible word break. */ |
| 543 | private boolean isWordBreak(int index) { |
| 544 | return getCharClass(index - 1) != getCharClass(index); |
| 545 | } |
| 546 | |
| 547 | /** Determines whether the given index could be a possible smiley break. */ |
| 548 | private boolean isSmileyBreak(int index) { |
| 549 | if (index > 0 && index < text.length()) { |
| 550 | if (isSmileyBreak(text.charAt(index - 1), text.charAt(index))) { |
| 551 | return true; |
| 552 | } |
| 553 | } |
| 554 | |
| 555 | return false; |
| 556 | } |
| 557 | |
| 558 | /** |
| 559 | * Verifies that the character before the given index is end of line, |
| 560 | * whitespace, or punctuation. |
| 561 | */ |
| 562 | private boolean isURLBreak(int index) { |
| 563 | switch (getCharClass(index - 1)) { |
| 564 | case 2: |
| 565 | case 3: |
| 566 | case 4: |
| 567 | return false; |
| 568 | |
| 569 | case 0: |
| 570 | case 1: |
| 571 | default: |
| 572 | return true; |
| 573 | } |
| 574 | } |
| 575 | |
| 576 | /** Returns the class for the character at the given index. */ |
| 577 | private int getCharClass(int index) { |
| 578 | if ((index < 0) || (text.length() <= index)) { |
| 579 | return 0; |
| 580 | } |
| 581 | |
| 582 | char ch = text.charAt(index); |
| 583 | if (Character.isWhitespace(ch)) { |
| 584 | return 1; |
| 585 | } else if (Character.isLetter(ch)) { |
| 586 | return 2; |
| 587 | } else if (Character.isDigit(ch)) { |
| 588 | return 3; |
| 589 | } else if (isPunctuation(ch)) { |
| 590 | // For punctuation, we return a unique value every time so that they are |
| 591 | // always different from any other character. Punctuation should always |
| 592 | // be considered a possible word break. |
| 593 | return ++nextClass; |
| 594 | } else { |
| 595 | return 4; |
| 596 | } |
| 597 | } |
| 598 | |
| 599 | /** |
| 600 | * Returns true if <code>c1</code> could be the last character of |
| 601 | * a smiley and <code>c2</code> could be the first character of |
| 602 | * a different smiley, if {@link #isWordBreak} would not already |
| 603 | * recognize that this is possible. |
| 604 | */ |
| 605 | private static boolean isSmileyBreak(char c1, char c2) { |
| 606 | switch (c1) { |
| 607 | /* |
| 608 | * These characters can end smileys, but don't normally end words. |
| 609 | */ |
| 610 | case '$': case '&': case '*': case '+': case '-': |
| 611 | case '/': case '<': case '=': case '>': case '@': |
| 612 | case '[': case '\\': case ']': case '^': case '|': |
| 613 | case '}': case '~': |
| 614 | switch (c2) { |
| 615 | /* |
| 616 | * These characters can begin smileys, but don't normally |
| 617 | * begin words. |
| 618 | */ |
| 619 | case '#': case '$': case '%': case '*': case '/': |
| 620 | case '<': case '=': case '>': case '@': case '[': |
| 621 | case '\\': case '^': case '~': |
| 622 | return true; |
| 623 | } |
| 624 | } |
| 625 | |
| 626 | return false; |
| 627 | } |
| 628 | |
| 629 | /** Determines whether the given character is punctuation. */ |
| 630 | private static boolean isPunctuation(char ch) { |
| 631 | switch (ch) { |
| 632 | case '.': case ',': case '"': case ':': case ';': |
| 633 | case '?': case '!': case '(': case ')': |
| 634 | return true; |
| 635 | |
| 636 | default: |
| 637 | return false; |
| 638 | } |
| 639 | } |
| 640 | |
| 641 | /** |
| 642 | * Determines whether the given character is the beginning or end of a |
| 643 | * section with special formatting. |
| 644 | */ |
| 645 | private static boolean isFormatChar(char ch) { |
| 646 | switch (ch) { |
| 647 | case '*': case '_': case '^': |
| 648 | return true; |
| 649 | |
| 650 | default: |
| 651 | return false; |
| 652 | } |
| 653 | } |
| 654 | |
| 655 | /** Represents a unit of parsed output. */ |
| 656 | public static abstract class Token { |
Artur Satayev | fc46be7 | 2019-11-04 17:50:59 +0000 | [diff] [blame] | 657 | @UnsupportedAppUsage(implicitMember = |
| 658 | "values()[Lcom/google/android/util/AbstractMessageParser$Token$Type;") |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 659 | public enum Type { |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 660 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 661 | HTML ("html"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 662 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 663 | FORMAT ("format"), // subtype of HTML |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 664 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 665 | LINK ("l"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 666 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 667 | SMILEY ("e"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 668 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 669 | ACRONYM ("a"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 670 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 671 | MUSIC ("m"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 672 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 673 | GOOGLE_VIDEO ("v"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 674 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 675 | YOUTUBE_VIDEO ("yt"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 676 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 677 | PHOTO ("p"), |
Andrei Onea | eecddd5 | 2019-03-27 10:32:55 +0000 | [diff] [blame] | 678 | @UnsupportedAppUsage |
The Android Open Source Project | 9066cfe | 2009-03-03 19:31:44 -0800 | [diff] [blame] | 679 | FLICKR ("f"); |
| 680 | |
| 681 | //stringreps for HTML and FORMAT don't really matter |
| 682 | //because they don't define getInfo(), which is where it is used |
| 683 | //For the other types, code depends on their stringreps |
| 684 | private String stringRep; |
| 685 | |
| 686 | Type(String stringRep) { |
| 687 | this.stringRep = stringRep; |
| 688 | } |
| 689 | |
| 690 | /** {@inheritDoc} */ |
| 691 | public String toString() { |
| 692 | return this.stringRep; |
| 693 | } |
| 694 | } |
| 695 | |
| 696 | protected Type type; |
| 697 | protected String text; |
| 698 | |
| 699 | protected Token(Type type, String text) { |
| 700 | this.type = type; |
| 701 | this.text = text; |
| 702 | } |
| 703 | |
| 704 | /** Returns the type of the token. */ |
| 705 | public Type getType() { return type; } |
| 706 | |
| 707 | /** |
| 708 | * Get the relevant information about a token |
| 709 | * |
| 710 | * @return a list of strings representing the token, not null |
| 711 | * The first item is always a string representation of the type |
| 712 | */ |
| 713 | public List<String> getInfo() { |
| 714 | List<String> info = new ArrayList<String>(); |
| 715 | info.add(getType().toString()); |
| 716 | return info; |
| 717 | } |
| 718 | |
| 719 | /** Returns the raw text of the token. */ |
| 720 | public String getRawText() { return text; } |
| 721 | |
| 722 | public boolean isMedia() { return false; } |
| 723 | public abstract boolean isHtml(); |
| 724 | public boolean isArray() { return !isHtml(); } |
| 725 | |
| 726 | public String toHtml(boolean caps) { throw new AssertionError("not html"); } |
| 727 | |
| 728 | // The token can change the caps of the text after that point. |
| 729 | public boolean controlCaps() { return false; } |
| 730 | public boolean setCaps() { return false; } |
| 731 | } |
| 732 | |
| 733 | /** Represents a simple string of html text. */ |
| 734 | public static class Html extends Token { |
| 735 | private String html; |
| 736 | |
| 737 | public Html(String text, String html) { |
| 738 | super(Type.HTML, text); |
| 739 | this.html = html; |
| 740 | } |
| 741 | |
| 742 | public boolean isHtml() { return true; } |
| 743 | public String toHtml(boolean caps) { |
| 744 | return caps ? html.toUpperCase() : html; |
| 745 | } |
| 746 | /** |
| 747 | * Not supported. Info should not be needed for this type |
| 748 | */ |
| 749 | public List<String> getInfo() { |
| 750 | throw new UnsupportedOperationException(); |
| 751 | } |
| 752 | |
| 753 | public void trimLeadingWhitespace() { |
| 754 | text = trimLeadingWhitespace(text); |
| 755 | html = trimLeadingWhitespace(html); |
| 756 | } |
| 757 | |
| 758 | public void trimTrailingWhitespace() { |
| 759 | text = trimTrailingWhitespace(text); |
| 760 | html = trimTrailingWhitespace(html); |
| 761 | } |
| 762 | |
| 763 | private static String trimLeadingWhitespace(String text) { |
| 764 | int index = 0; |
| 765 | while ((index < text.length()) && |
| 766 | Character.isWhitespace(text.charAt(index))) { |
| 767 | ++index; |
| 768 | } |
| 769 | return text.substring(index); |
| 770 | } |
| 771 | |
| 772 | public static String trimTrailingWhitespace(String text) { |
| 773 | int index = text.length(); |
| 774 | while ((index > 0) && Character.isWhitespace(text.charAt(index - 1))) { |
| 775 | --index; |
| 776 | } |
| 777 | return text.substring(0, index); |
| 778 | } |
| 779 | } |
| 780 | |
| 781 | /** Represents a music track token at the beginning. */ |
| 782 | public static class MusicTrack extends Token { |
| 783 | private String track; |
| 784 | |
| 785 | public MusicTrack(String track) { |
| 786 | super(Type.MUSIC, track); |
| 787 | this.track = track; |
| 788 | } |
| 789 | |
| 790 | public String getTrack() { return track; } |
| 791 | |
| 792 | public boolean isHtml() { return false; } |
| 793 | |
| 794 | public List<String> getInfo() { |
| 795 | List<String> info = super.getInfo(); |
| 796 | info.add(getTrack()); |
| 797 | return info; |
| 798 | } |
| 799 | } |
| 800 | |
| 801 | /** Represents a link that was found in the input. */ |
| 802 | public static class Link extends Token { |
| 803 | private String url; |
| 804 | |
| 805 | public Link(String url, String text) { |
| 806 | super(Type.LINK, text); |
| 807 | this.url = url; |
| 808 | } |
| 809 | |
| 810 | public String getURL() { return url; } |
| 811 | |
| 812 | public boolean isHtml() { return false; } |
| 813 | |
| 814 | public List<String> getInfo() { |
| 815 | List<String> info = super.getInfo(); |
| 816 | info.add(getURL()); |
| 817 | info.add(getRawText()); |
| 818 | return info; |
| 819 | } |
| 820 | } |
| 821 | |
| 822 | /** Represents a link to a Google Video. */ |
| 823 | public static class Video extends Token { |
| 824 | /** Pattern for a video URL. */ |
| 825 | private static final Pattern URL_PATTERN = Pattern.compile( |
| 826 | "(?i)http://video\\.google\\.[a-z0-9]+(?:\\.[a-z0-9]+)?/videoplay\\?" |
| 827 | + ".*?\\bdocid=(-?\\d+).*"); |
| 828 | |
| 829 | private String docid; |
| 830 | |
| 831 | public Video(String docid, String text) { |
| 832 | super(Type.GOOGLE_VIDEO, text); |
| 833 | this.docid = docid; |
| 834 | } |
| 835 | |
| 836 | public String getDocID() { return docid; } |
| 837 | |
| 838 | public boolean isHtml() { return false; } |
| 839 | public boolean isMedia() { return true; } |
| 840 | |
| 841 | /** Returns a Video object if the given url is to a video. */ |
| 842 | public static Video matchURL(String url, String text) { |
| 843 | Matcher m = URL_PATTERN.matcher(url); |
| 844 | if (m.matches()) { |
| 845 | return new Video(m.group(1), text); |
| 846 | } else { |
| 847 | return null; |
| 848 | } |
| 849 | } |
| 850 | |
| 851 | public List<String> getInfo() { |
| 852 | List<String> info = super.getInfo(); |
| 853 | info.add(getRssUrl(docid)); |
| 854 | info.add(getURL(docid)); |
| 855 | return info; |
| 856 | } |
| 857 | |
| 858 | /** Returns the URL for the RSS description of the given video. */ |
| 859 | public static String getRssUrl(String docid) { |
| 860 | return "http://video.google.com/videofeed" |
| 861 | + "?type=docid&output=rss&sourceid=gtalk&docid=" + docid; |
| 862 | } |
| 863 | |
| 864 | /** (For testing purposes:) Returns a video URL with the given parts. */ |
| 865 | public static String getURL(String docid) { |
| 866 | return getURL(docid, null); |
| 867 | } |
| 868 | |
| 869 | /** (For testing purposes:) Returns a video URL with the given parts. */ |
| 870 | public static String getURL(String docid, String extraParams) { |
| 871 | if (extraParams == null) { |
| 872 | extraParams = ""; |
| 873 | } else if (extraParams.length() > 0) { |
| 874 | extraParams += "&"; |
| 875 | } |
| 876 | return "http://video.google.com/videoplay?" + extraParams |
| 877 | + "docid=" + docid; |
| 878 | } |
| 879 | } |
| 880 | |
| 881 | /** Represents a link to a YouTube video. */ |
| 882 | public static class YouTubeVideo extends Token { |
| 883 | /** Pattern for a video URL. */ |
| 884 | private static final Pattern URL_PATTERN = Pattern.compile( |
| 885 | "(?i)http://(?:[a-z0-9]+\\.)?youtube\\.[a-z0-9]+(?:\\.[a-z0-9]+)?/watch\\?" |
| 886 | + ".*\\bv=([-_a-zA-Z0-9=]+).*"); |
| 887 | |
| 888 | private String docid; |
| 889 | |
| 890 | public YouTubeVideo(String docid, String text) { |
| 891 | super(Type.YOUTUBE_VIDEO, text); |
| 892 | this.docid = docid; |
| 893 | } |
| 894 | |
| 895 | public String getDocID() { return docid; } |
| 896 | |
| 897 | public boolean isHtml() { return false; } |
| 898 | public boolean isMedia() { return true; } |
| 899 | |
| 900 | /** Returns a Video object if the given url is to a video. */ |
| 901 | public static YouTubeVideo matchURL(String url, String text) { |
| 902 | Matcher m = URL_PATTERN.matcher(url); |
| 903 | if (m.matches()) { |
| 904 | return new YouTubeVideo(m.group(1), text); |
| 905 | } else { |
| 906 | return null; |
| 907 | } |
| 908 | } |
| 909 | |
| 910 | public List<String> getInfo() { |
| 911 | List<String> info = super.getInfo(); |
| 912 | info.add(getRssUrl(docid)); |
| 913 | info.add(getURL(docid)); |
| 914 | return info; |
| 915 | } |
| 916 | |
| 917 | /** Returns the URL for the RSS description of the given video. */ |
| 918 | public static String getRssUrl(String docid) { |
| 919 | return "http://youtube.com/watch?v=" + docid; |
| 920 | } |
| 921 | |
| 922 | /** (For testing purposes:) Returns a video URL with the given parts. */ |
| 923 | public static String getURL(String docid) { |
| 924 | return getURL(docid, null); |
| 925 | } |
| 926 | |
| 927 | /** (For testing purposes:) Returns a video URL with the given parts. */ |
| 928 | public static String getURL(String docid, String extraParams) { |
| 929 | if (extraParams == null) { |
| 930 | extraParams = ""; |
| 931 | } else if (extraParams.length() > 0) { |
| 932 | extraParams += "&"; |
| 933 | } |
| 934 | return "http://youtube.com/watch?" + extraParams + "v=" + docid; |
| 935 | } |
| 936 | |
| 937 | /** (For testing purposes:) Returns a video URL with the given parts. |
| 938 | * @param http If true, includes http:// |
| 939 | * @param prefix If non-null/non-blank, adds to URL before youtube.com. |
| 940 | * (e.g., prefix="br." --> "br.youtube.com") |
| 941 | */ |
| 942 | public static String getPrefixedURL(boolean http, String prefix, |
| 943 | String docid, String extraParams) { |
| 944 | String protocol = ""; |
| 945 | |
| 946 | if (http) { |
| 947 | protocol = "http://"; |
| 948 | } |
| 949 | |
| 950 | if (prefix == null) { |
| 951 | prefix = ""; |
| 952 | } |
| 953 | |
| 954 | if (extraParams == null) { |
| 955 | extraParams = ""; |
| 956 | } else if (extraParams.length() > 0) { |
| 957 | extraParams += "&"; |
| 958 | } |
| 959 | |
| 960 | return protocol + prefix + "youtube.com/watch?" + extraParams + "v=" + |
| 961 | docid; |
| 962 | } |
| 963 | } |
| 964 | |
| 965 | /** Represents a link to a Picasa photo or album. */ |
| 966 | public static class Photo extends Token { |
| 967 | /** Pattern for an album or photo URL. */ |
| 968 | // TODO (katyarogers) searchbrowse includes search lists and tags, |
| 969 | // it follows a different pattern than albums - would be nice to add later |
| 970 | private static final Pattern URL_PATTERN = Pattern.compile( |
| 971 | "http://picasaweb.google.com/([^/?#&]+)/+((?!searchbrowse)[^/?#&]+)(?:/|/photo)?(?:\\?[^#]*)?(?:#(.*))?"); |
| 972 | |
| 973 | private String user; |
| 974 | private String album; |
| 975 | private String photo; // null for albums |
| 976 | |
| 977 | public Photo(String user, String album, String photo, String text) { |
| 978 | super(Type.PHOTO, text); |
| 979 | this.user = user; |
| 980 | this.album = album; |
| 981 | this.photo = photo; |
| 982 | } |
| 983 | |
| 984 | public String getUser() { return user; } |
| 985 | public String getAlbum() { return album; } |
| 986 | public String getPhoto() { return photo; } |
| 987 | |
| 988 | public boolean isHtml() { return false; } |
| 989 | public boolean isMedia() { return true; } |
| 990 | |
| 991 | /** Returns a Photo object if the given url is to a photo or album. */ |
| 992 | public static Photo matchURL(String url, String text) { |
| 993 | Matcher m = URL_PATTERN.matcher(url); |
| 994 | if (m.matches()) { |
| 995 | return new Photo(m.group(1), m.group(2), m.group(3), text); |
| 996 | } else { |
| 997 | return null; |
| 998 | } |
| 999 | } |
| 1000 | |
| 1001 | public List<String> getInfo() { |
| 1002 | List<String> info = super.getInfo(); |
| 1003 | info.add(getRssUrl(getUser())); |
| 1004 | info.add(getAlbumURL(getUser(), getAlbum())); |
| 1005 | if (getPhoto() != null) { |
| 1006 | info.add(getPhotoURL(getUser(), getAlbum(), getPhoto())); |
| 1007 | } else { |
| 1008 | info.add((String)null); |
| 1009 | } |
| 1010 | return info; |
| 1011 | } |
| 1012 | |
| 1013 | /** Returns the URL for the RSS description of the user's albums. */ |
| 1014 | public static String getRssUrl(String user) { |
| 1015 | return "http://picasaweb.google.com/data/feed/api/user/" + user + |
| 1016 | "?category=album&alt=rss"; |
| 1017 | } |
| 1018 | |
| 1019 | /** Returns the URL for an album. */ |
| 1020 | public static String getAlbumURL(String user, String album) { |
| 1021 | return "http://picasaweb.google.com/" + user + "/" + album; |
| 1022 | } |
| 1023 | |
| 1024 | /** Returns the URL for a particular photo. */ |
| 1025 | public static String getPhotoURL(String user, String album, String photo) { |
| 1026 | return "http://picasaweb.google.com/" + user + "/" + album + "/photo#" |
| 1027 | + photo; |
| 1028 | } |
| 1029 | } |
| 1030 | |
| 1031 | /** Represents a link to a Flickr photo or album. */ |
| 1032 | public static class FlickrPhoto extends Token { |
| 1033 | /** Pattern for a user album or photo URL. */ |
| 1034 | private static final Pattern URL_PATTERN = Pattern.compile( |
| 1035 | "http://(?:www.)?flickr.com/photos/([^/?#&]+)/?([^/?#&]+)?/?.*"); |
| 1036 | private static final Pattern GROUPING_PATTERN = Pattern.compile( |
| 1037 | "http://(?:www.)?flickr.com/photos/([^/?#&]+)/(tags|sets)/" + |
| 1038 | "([^/?#&]+)/?"); |
| 1039 | |
| 1040 | private static final String SETS = "sets"; |
| 1041 | private static final String TAGS = "tags"; |
| 1042 | |
| 1043 | private String user; |
| 1044 | private String photo; // null for user album |
| 1045 | private String grouping; // either "tags" or "sets" |
| 1046 | private String groupingId; // sets or tags identifier |
| 1047 | |
| 1048 | public FlickrPhoto(String user, String photo, String grouping, |
| 1049 | String groupingId, String text) { |
| 1050 | super(Type.FLICKR, text); |
| 1051 | |
| 1052 | /* System wide tags look like the URL to a Flickr user. */ |
| 1053 | if (!TAGS.equals(user)) { |
| 1054 | this.user = user; |
| 1055 | // Don't consider slide show URL a photo |
| 1056 | this.photo = (!"show".equals(photo) ? photo : null); |
| 1057 | this.grouping = grouping; |
| 1058 | this.groupingId = groupingId; |
| 1059 | } else { |
| 1060 | this.user = null; |
| 1061 | this.photo = null; |
| 1062 | this.grouping = TAGS; |
| 1063 | this.groupingId = photo; |
| 1064 | } |
| 1065 | } |
| 1066 | |
| 1067 | public String getUser() { return user; } |
| 1068 | public String getPhoto() { return photo; } |
| 1069 | public String getGrouping() { return grouping; } |
| 1070 | public String getGroupingId() { return groupingId; } |
| 1071 | |
| 1072 | public boolean isHtml() { return false; } |
| 1073 | public boolean isMedia() { return true; } |
| 1074 | |
| 1075 | /** |
| 1076 | * Returns a FlickrPhoto object if the given url is to a photo or Flickr |
| 1077 | * user. |
| 1078 | */ |
| 1079 | public static FlickrPhoto matchURL(String url, String text) { |
| 1080 | Matcher m = GROUPING_PATTERN.matcher(url); |
| 1081 | if (m.matches()) { |
| 1082 | return new FlickrPhoto(m.group(1), null, m.group(2), m.group(3), text); |
| 1083 | } |
| 1084 | |
| 1085 | m = URL_PATTERN.matcher(url); |
| 1086 | if (m.matches()) { |
| 1087 | return new FlickrPhoto(m.group(1), m.group(2), null, null, text); |
| 1088 | } else { |
| 1089 | return null; |
| 1090 | } |
| 1091 | } |
| 1092 | |
| 1093 | public List<String> getInfo() { |
| 1094 | List<String> info = super.getInfo(); |
| 1095 | info.add(getUrl()); |
| 1096 | info.add(getUser() != null ? getUser() : ""); |
| 1097 | info.add(getPhoto() != null ? getPhoto() : ""); |
| 1098 | info.add(getGrouping() != null ? getGrouping() : ""); |
| 1099 | info.add(getGroupingId() != null ? getGroupingId() : ""); |
| 1100 | return info; |
| 1101 | } |
| 1102 | |
| 1103 | public String getUrl() { |
| 1104 | if (SETS.equals(grouping)) { |
| 1105 | return getUserSetsURL(user, groupingId); |
| 1106 | } else if (TAGS.equals(grouping)) { |
| 1107 | if (user != null) { |
| 1108 | return getUserTagsURL(user, groupingId); |
| 1109 | } else { |
| 1110 | return getTagsURL(groupingId); |
| 1111 | } |
| 1112 | } else if (photo != null) { |
| 1113 | return getPhotoURL(user, photo); |
| 1114 | } else { |
| 1115 | return getUserURL(user); |
| 1116 | } |
| 1117 | } |
| 1118 | |
| 1119 | /** Returns the URL for the RSS description. */ |
| 1120 | public static String getRssUrl(String user) { |
| 1121 | return null; |
| 1122 | } |
| 1123 | |
| 1124 | /** Returns the URL for a particular tag. */ |
| 1125 | public static String getTagsURL(String tag) { |
| 1126 | return "http://flickr.com/photos/tags/" + tag; |
| 1127 | } |
| 1128 | |
| 1129 | /** Returns the URL to the user's Flickr homepage. */ |
| 1130 | public static String getUserURL(String user) { |
| 1131 | return "http://flickr.com/photos/" + user; |
| 1132 | } |
| 1133 | |
| 1134 | /** Returns the URL for a particular photo. */ |
| 1135 | public static String getPhotoURL(String user, String photo) { |
| 1136 | return "http://flickr.com/photos/" + user + "/" + photo; |
| 1137 | } |
| 1138 | |
| 1139 | /** Returns the URL for a user tag photo set. */ |
| 1140 | public static String getUserTagsURL(String user, String tagId) { |
| 1141 | return "http://flickr.com/photos/" + user + "/tags/" + tagId; |
| 1142 | } |
| 1143 | |
| 1144 | /** Returns the URL for user set. */ |
| 1145 | public static String getUserSetsURL(String user, String setId) { |
| 1146 | return "http://flickr.com/photos/" + user + "/sets/" + setId; |
| 1147 | } |
| 1148 | } |
| 1149 | |
| 1150 | /** Represents a smiley that was found in the input. */ |
| 1151 | public static class Smiley extends Token { |
| 1152 | // TODO: Pass the SWF URL down to the client. |
| 1153 | |
| 1154 | public Smiley(String text) { |
| 1155 | super(Type.SMILEY, text); |
| 1156 | } |
| 1157 | |
| 1158 | public boolean isHtml() { return false; } |
| 1159 | |
| 1160 | public List<String> getInfo() { |
| 1161 | List<String> info = super.getInfo(); |
| 1162 | info.add(getRawText()); |
| 1163 | return info; |
| 1164 | } |
| 1165 | } |
| 1166 | |
| 1167 | /** Represents an acronym that was found in the input. */ |
| 1168 | public static class Acronym extends Token { |
| 1169 | private String value; |
| 1170 | // TODO: SWF |
| 1171 | |
| 1172 | public Acronym(String text, String value) { |
| 1173 | super(Type.ACRONYM, text); |
| 1174 | this.value = value; |
| 1175 | } |
| 1176 | |
| 1177 | public String getValue() { return value; } |
| 1178 | |
| 1179 | public boolean isHtml() { return false; } |
| 1180 | |
| 1181 | public List<String> getInfo() { |
| 1182 | List<String> info = super.getInfo(); |
| 1183 | info.add(getRawText()); |
| 1184 | info.add(getValue()); |
| 1185 | return info; |
| 1186 | } |
| 1187 | } |
| 1188 | |
| 1189 | /** Represents a character that changes formatting. */ |
| 1190 | public static class Format extends Token { |
| 1191 | private char ch; |
| 1192 | private boolean start; |
| 1193 | private boolean matched; |
| 1194 | |
| 1195 | public Format(char ch, boolean start) { |
| 1196 | super(Type.FORMAT, String.valueOf(ch)); |
| 1197 | this.ch = ch; |
| 1198 | this.start = start; |
| 1199 | } |
| 1200 | |
| 1201 | public void setMatched(boolean matched) { this.matched = matched; } |
| 1202 | |
| 1203 | public boolean isHtml() { return true; } |
| 1204 | |
| 1205 | public String toHtml(boolean caps) { |
| 1206 | // This character only implies special formatting if it was matched. |
| 1207 | // Otherwise, it was just a plain old character. |
| 1208 | if (matched) { |
| 1209 | return start ? getFormatStart(ch) : getFormatEnd(ch); |
| 1210 | } else { |
| 1211 | // We have to make sure we escape HTML characters as usual. |
| 1212 | return (ch == '"') ? """ : String.valueOf(ch); |
| 1213 | } |
| 1214 | } |
| 1215 | |
| 1216 | /** |
| 1217 | * Not supported. Info should not be needed for this type |
| 1218 | */ |
| 1219 | public List<String> getInfo() { |
| 1220 | throw new UnsupportedOperationException(); |
| 1221 | } |
| 1222 | |
| 1223 | public boolean controlCaps() { return (ch == '^'); } |
| 1224 | public boolean setCaps() { return start; } |
| 1225 | |
| 1226 | private String getFormatStart(char ch) { |
| 1227 | switch (ch) { |
| 1228 | case '*': return "<b>"; |
| 1229 | case '_': return "<i>"; |
| 1230 | case '^': return "<b><font color=\"#005FFF\">"; // TODO: all caps |
| 1231 | case '"': return "<font color=\"#999999\">\u201c"; |
| 1232 | default: throw new AssertionError("unknown format '" + ch + "'"); |
| 1233 | } |
| 1234 | } |
| 1235 | |
| 1236 | private String getFormatEnd(char ch) { |
| 1237 | switch (ch) { |
| 1238 | case '*': return "</b>"; |
| 1239 | case '_': return "</i>"; |
| 1240 | case '^': return "</font></b>"; // TODO: all caps |
| 1241 | case '"': return "\u201d</font>"; |
| 1242 | default: throw new AssertionError("unknown format '" + ch + "'"); |
| 1243 | } |
| 1244 | } |
| 1245 | } |
| 1246 | |
| 1247 | /** Adds the given token to the parsed output. */ |
| 1248 | private void addToken(Token token) { |
| 1249 | tokens.add(token); |
| 1250 | } |
| 1251 | |
| 1252 | /** Converts the entire message into a single HTML display string. */ |
| 1253 | public String toHtml() { |
| 1254 | StringBuilder html = new StringBuilder(); |
| 1255 | |
| 1256 | for (Part part : parts) { |
| 1257 | boolean caps = false; |
| 1258 | |
| 1259 | html.append("<p>"); |
| 1260 | for (Token token : part.getTokens()) { |
| 1261 | if (token.isHtml()) { |
| 1262 | html.append(token.toHtml(caps)); |
| 1263 | } else { |
| 1264 | switch (token.getType()) { |
| 1265 | case LINK: |
| 1266 | html.append("<a href=\""); |
| 1267 | html.append(((Link)token).getURL()); |
| 1268 | html.append("\">"); |
| 1269 | html.append(token.getRawText()); |
| 1270 | html.append("</a>"); |
| 1271 | break; |
| 1272 | |
| 1273 | case SMILEY: |
| 1274 | // TODO: link to an appropriate image |
| 1275 | html.append(token.getRawText()); |
| 1276 | break; |
| 1277 | |
| 1278 | case ACRONYM: |
| 1279 | html.append(token.getRawText()); |
| 1280 | break; |
| 1281 | |
| 1282 | case MUSIC: |
| 1283 | // TODO: include a music glyph |
| 1284 | html.append(((MusicTrack)token).getTrack()); |
| 1285 | break; |
| 1286 | |
| 1287 | case GOOGLE_VIDEO: |
| 1288 | // TODO: include a Google Video icon |
| 1289 | html.append("<a href=\""); |
| 1290 | html.append(((Video)token).getURL(((Video)token).getDocID())); |
| 1291 | html.append("\">"); |
| 1292 | html.append(token.getRawText()); |
| 1293 | html.append("</a>"); |
| 1294 | break; |
| 1295 | |
| 1296 | case YOUTUBE_VIDEO: |
| 1297 | // TODO: include a YouTube icon |
| 1298 | html.append("<a href=\""); |
| 1299 | html.append(((YouTubeVideo)token).getURL( |
| 1300 | ((YouTubeVideo)token).getDocID())); |
| 1301 | html.append("\">"); |
| 1302 | html.append(token.getRawText()); |
| 1303 | html.append("</a>"); |
| 1304 | break; |
| 1305 | |
| 1306 | case PHOTO: { |
| 1307 | // TODO: include a Picasa Web icon |
| 1308 | html.append("<a href=\""); |
| 1309 | html.append(Photo.getAlbumURL( |
| 1310 | ((Photo)token).getUser(), ((Photo)token).getAlbum())); |
| 1311 | html.append("\">"); |
| 1312 | html.append(token.getRawText()); |
| 1313 | html.append("</a>"); |
| 1314 | break; |
| 1315 | } |
| 1316 | |
| 1317 | case FLICKR: |
| 1318 | // TODO: include a Flickr icon |
| 1319 | Photo p = (Photo) token; |
| 1320 | html.append("<a href=\""); |
| 1321 | html.append(((FlickrPhoto)token).getUrl()); |
| 1322 | html.append("\">"); |
| 1323 | html.append(token.getRawText()); |
| 1324 | html.append("</a>"); |
| 1325 | break; |
| 1326 | |
| 1327 | default: |
| 1328 | throw new AssertionError("unknown token type: " + token.getType()); |
| 1329 | } |
| 1330 | } |
| 1331 | |
| 1332 | if (token.controlCaps()) { |
| 1333 | caps = token.setCaps(); |
| 1334 | } |
| 1335 | } |
| 1336 | html.append("</p>\n"); |
| 1337 | } |
| 1338 | |
| 1339 | return html.toString(); |
| 1340 | } |
| 1341 | |
| 1342 | /** Returns the reverse of the given string. */ |
| 1343 | protected static String reverse(String str) { |
| 1344 | StringBuilder buf = new StringBuilder(); |
| 1345 | for (int i = str.length() - 1; i >= 0; --i) { |
| 1346 | buf.append(str.charAt(i)); |
| 1347 | } |
| 1348 | return buf.toString(); |
| 1349 | } |
| 1350 | |
| 1351 | public static class TrieNode { |
| 1352 | private final HashMap<Character,TrieNode> children = |
| 1353 | new HashMap<Character,TrieNode>(); |
| 1354 | private String text; |
| 1355 | private String value; |
| 1356 | |
| 1357 | public TrieNode() { this(""); } |
| 1358 | public TrieNode(String text) { |
| 1359 | this.text = text; |
| 1360 | } |
| 1361 | |
| 1362 | public final boolean exists() { return value != null; } |
| 1363 | public final String getText() { return text; } |
| 1364 | public final String getValue() { return value; } |
| 1365 | public void setValue(String value) { this.value = value; } |
| 1366 | |
| 1367 | public TrieNode getChild(char ch) { |
| 1368 | return children.get(Character.valueOf(ch)); |
| 1369 | } |
| 1370 | |
| 1371 | public TrieNode getOrCreateChild(char ch) { |
| 1372 | Character key = Character.valueOf(ch); |
| 1373 | TrieNode node = children.get(key); |
| 1374 | if (node == null) { |
| 1375 | node = new TrieNode(text + String.valueOf(ch)); |
| 1376 | children.put(key, node); |
| 1377 | } |
| 1378 | return node; |
| 1379 | } |
| 1380 | |
| 1381 | /** Adds the given string into the trie. */ |
| 1382 | public static void addToTrie(TrieNode root, String str, String value) { |
| 1383 | int index = 0; |
| 1384 | while (index < str.length()) { |
| 1385 | root = root.getOrCreateChild(str.charAt(index++)); |
| 1386 | } |
| 1387 | root.setValue(value); |
| 1388 | } |
| 1389 | } |
| 1390 | |
| 1391 | |
| 1392 | |
| 1393 | /** Determines whether the given string is in the given trie. */ |
| 1394 | private static boolean matches(TrieNode root, String str) { |
| 1395 | int index = 0; |
| 1396 | while (index < str.length()) { |
| 1397 | root = root.getChild(str.charAt(index++)); |
| 1398 | if (root == null) { |
| 1399 | break; |
| 1400 | } else if (root.exists()) { |
| 1401 | return true; |
| 1402 | } |
| 1403 | } |
| 1404 | return false; |
| 1405 | } |
| 1406 | |
| 1407 | /** |
| 1408 | * Returns the longest substring of the given string, starting at the given |
| 1409 | * index, that exists in the trie. |
| 1410 | */ |
| 1411 | private static TrieNode longestMatch( |
| 1412 | TrieNode root, AbstractMessageParser p, int start) { |
| 1413 | return longestMatch(root, p, start, false); |
| 1414 | } |
| 1415 | |
| 1416 | /** |
| 1417 | * Returns the longest substring of the given string, starting at the given |
| 1418 | * index, that exists in the trie, with a special tokenizing case for |
| 1419 | * smileys if specified. |
| 1420 | */ |
| 1421 | private static TrieNode longestMatch( |
| 1422 | TrieNode root, AbstractMessageParser p, int start, boolean smiley) { |
| 1423 | int index = start; |
| 1424 | TrieNode bestMatch = null; |
| 1425 | while (index < p.getRawText().length()) { |
| 1426 | root = root.getChild(p.getRawText().charAt(index++)); |
| 1427 | if (root == null) { |
| 1428 | break; |
| 1429 | } else if (root.exists()) { |
| 1430 | if (p.isWordBreak(index)) { |
| 1431 | bestMatch = root; |
| 1432 | } else if (smiley && p.isSmileyBreak(index)) { |
| 1433 | bestMatch = root; |
| 1434 | } |
| 1435 | } |
| 1436 | } |
| 1437 | return bestMatch; |
| 1438 | } |
| 1439 | |
| 1440 | |
| 1441 | /** Represents set of tokens that are delivered as a single message. */ |
| 1442 | public static class Part { |
| 1443 | private String meText; |
| 1444 | private ArrayList<Token> tokens; |
| 1445 | |
| 1446 | public Part() { |
| 1447 | this.tokens = new ArrayList<Token>(); |
| 1448 | } |
| 1449 | |
| 1450 | public String getType(boolean isSend) { |
| 1451 | return (isSend ? "s" : "r") + getPartType(); |
| 1452 | } |
| 1453 | |
| 1454 | private String getPartType() { |
| 1455 | if (isMedia()) { |
| 1456 | return "d"; |
| 1457 | } else if (meText != null) { |
| 1458 | return "m"; |
| 1459 | } else { |
| 1460 | return ""; |
| 1461 | } |
| 1462 | } |
| 1463 | |
| 1464 | public boolean isMedia() { |
| 1465 | return (tokens.size() == 1) && tokens.get(0).isMedia(); |
| 1466 | } |
| 1467 | /** |
| 1468 | * Convenience method for getting the Token of a Part that represents |
| 1469 | * a media Token. Parts of this kind will always only have a single Token |
| 1470 | * |
| 1471 | * @return if this.isMedia(), |
| 1472 | * returns the Token representing the media contained in this Part, |
| 1473 | * otherwise returns null; |
| 1474 | */ |
| 1475 | public Token getMediaToken() { |
| 1476 | if(isMedia()) { |
| 1477 | return tokens.get(0); |
| 1478 | } |
| 1479 | return null; |
| 1480 | } |
| 1481 | |
| 1482 | /** Adds the given token to this part. */ |
| 1483 | public void add(Token token) { |
| 1484 | if (isMedia()) { |
| 1485 | throw new AssertionError("media "); |
| 1486 | } |
| 1487 | tokens.add(token); |
| 1488 | } |
| 1489 | |
| 1490 | public void setMeText(String meText) { |
| 1491 | this.meText = meText; |
| 1492 | } |
| 1493 | |
| 1494 | /** Returns the original text of this part. */ |
| 1495 | public String getRawText() { |
| 1496 | StringBuilder buf = new StringBuilder(); |
| 1497 | if (meText != null) { |
| 1498 | buf.append(meText); |
| 1499 | } |
| 1500 | for (int i = 0; i < tokens.size(); ++i) { |
| 1501 | buf.append(tokens.get(i).getRawText()); |
| 1502 | } |
| 1503 | return buf.toString(); |
| 1504 | } |
| 1505 | |
| 1506 | /** Returns the tokens in this part. */ |
| 1507 | public ArrayList<Token> getTokens() { return tokens; } |
| 1508 | |
| 1509 | /** Adds the tokens into the given builder as an array. */ |
| 1510 | // public void toArray(JSArrayBuilder array) { |
| 1511 | // if (isMedia()) { |
| 1512 | // // For media, we send its array (i.e., we don't wrap this in another |
| 1513 | // // array as we do for non-media parts). |
| 1514 | // tokens.get(0).toArray(array); |
| 1515 | // } else { |
| 1516 | // array.beginArray(); |
| 1517 | // addToArray(array); |
| 1518 | // array.endArray(); |
| 1519 | // } |
| 1520 | // } |
| 1521 | } |
| 1522 | } |