supported/BasicSupport_Deprecated.java

   1 package be.nikiroo.fanfix.supported;
   2
   3 import java.io.BufferedReader;
   4 import java.io.ByteArrayInputStream;
   5 import java.io.File;
   6 import java.io.IOException;
   7 import java.io.InputStream;
   8 import java.io.InputStreamReader;
   9 import java.net.MalformedURLException;
  10 import java.net.URL;
  11 import java.util.ArrayList;
  12 import java.util.Date;
  13 import java.util.List;
  14 import java.util.Map.Entry;
  15 import java.util.Scanner;
  16
  17 import be.nikiroo.fanfix.Instance;
  18 import be.nikiroo.fanfix.bundles.Config;
  19 import be.nikiroo.fanfix.bundles.StringId;
  20 import be.nikiroo.fanfix.data.Chapter;
  21 import be.nikiroo.fanfix.data.MetaData;
  22 import be.nikiroo.fanfix.data.Paragraph;
  23 import be.nikiroo.fanfix.data.Paragraph.ParagraphType;
  24 import be.nikiroo.fanfix.data.Story;
  25 import be.nikiroo.utils.Image;
  26 import be.nikiroo.utils.Progress;
  27 import be.nikiroo.utils.StringUtils;
  28
  29 /**
  30  * DEPRECATED: use the new Jsoup 'Node' system.
  31  * <p>
  32  * This class is the base class used by the other support classes. It can be
  33  * used outside of this package, and have static method that you can use to get
  34  * access to the correct support class.
  35  * <p>
  36  * It will be used with 'resources' (usually web pages or files).
  37  *
  38  * @author niki
  39  */
  40 @Deprecated
  41 public abstract class BasicSupport_Deprecated extends BasicSupport {
  42         private InputStream in;
  43
  44         // quote chars
  45         private char openQuote = Instance.getInstance().getTrans().getCharacter(StringId.OPEN_SINGLE_QUOTE);
  46         private char closeQuote = Instance.getInstance().getTrans().getCharacter(StringId.CLOSE_SINGLE_QUOTE);
  47         private char openDoubleQuote = Instance.getInstance().getTrans().getCharacter(StringId.OPEN_DOUBLE_QUOTE);
  48         private char closeDoubleQuote = Instance.getInstance().getTrans().getCharacter(StringId.CLOSE_DOUBLE_QUOTE);
  49
  50         // New methods not used in Deprecated mode
  51         @Override
  52         protected String getDesc() throws IOException {
  53                 throw new RuntimeException("should not be used by legacy code");
  54         }
  55
  56         @Override
  57         protected MetaData getMeta() throws IOException {
  58                 throw new RuntimeException("should not be used by legacy code");
  59         }
  60
  61         @Override
  62         protected List<Entry<String, URL>> getChapters(Progress pg)
  63                         throws IOException {
  64                 throw new RuntimeException("should not be used by legacy code");
  65         }
  66
  67         @Override
  68         protected String getChapterContent(URL chapUrl, int number, Progress pg)
  69                         throws IOException {
  70                 throw new RuntimeException("should not be used by legacy code");
  71         }
  72
  73         @Override
  74         public Story process(Progress pg) throws IOException {
  75                 return process(getSource(), pg);
  76         }
  77
  78         //
  79
  80         /**
  81          * Return the {@link MetaData} of this story.
  82          *
  83          * @param source
  84          *            the source of the story
  85          * @param in
  86          *            the input (the main resource)
  87          *
  88          * @return the associated {@link MetaData}, never NULL
  89          *
  90          * @throws IOException
  91          *             in case of I/O error
  92          */
  93         protected abstract MetaData getMeta(URL source, InputStream in)
  94                         throws IOException;
  95
  96         /**
  97          * Return the story description.
  98          *
  99          * @param source
 100          *            the source of the story
 101          * @param in
 102          *            the input (the main resource)
 103          *
 104          * @return the description
 105          *
 106          * @throws IOException
 107          *             in case of I/O error
 108          */
 109         protected abstract String getDesc(URL source, InputStream in)
 110                         throws IOException;
 111
 112         /**
 113          * Return the list of chapters (name and resource).
 114          *
 115          * @param source
 116          *            the source of the story
 117          * @param in
 118          *            the input (the main resource)
 119          * @param pg
 120          *            the optional progress reporter
 121          *
 122          * @return the chapters
 123          *
 124          * @throws IOException
 125          *             in case of I/O error
 126          */
 127         protected abstract List<Entry<String, URL>> getChapters(URL source,
 128                         InputStream in, Progress pg) throws IOException;
 129
 130         /**
 131          * Return the content of the chapter (possibly HTML encoded, if
 132          * {@link BasicSupport_Deprecated#isHtml()} is TRUE).
 133          *
 134          * @param source
 135          *            the source of the story
 136          * @param in
 137          *            the input (the main resource)
 138          * @param number
 139          *            the chapter number
 140          * @param pg
 141          *            the optional progress reporter
 142          *
 143          * @return the content
 144          *
 145          * @throws IOException
 146          *             in case of I/O error
 147          */
 148         protected abstract String getChapterContent(URL source, InputStream in,
 149                         int number, Progress pg) throws IOException;
 150
 151         /**
 152          * Process the given story resource into a partially filled {@link Story}
 153          * object containing the name and metadata, except for the description.
 154          *
 155          * @param url
 156          *            the story resource
 157          *
 158          * @return the {@link Story}
 159          *
 160          * @throws IOException
 161          *             in case of I/O error
 162          */
 163         public Story processMeta(URL url) throws IOException {
 164                 return processMeta(url, true, false, null);
 165         }
 166
 167         /**
 168          * Process the given story resource into a partially filled {@link Story}
 169          * object containing the name and metadata.
 170          *
 171          * @param url
 172          *            the story resource
 173          * @param close
 174          *            close "this" and "in" when done
 175          * @param getDesc
 176          *            retrieve the description of the story, or not
 177          * @param pg
 178          *            the optional progress reporter
 179          *
 180          * @return the {@link Story}, never NULL
 181          *
 182          * @throws IOException
 183          *             in case of I/O error
 184          */
 185         protected Story processMeta(URL url, boolean close, boolean getDesc,
 186                         Progress pg) throws IOException {
 187                 if (pg == null) {
 188                         pg = new Progress();
 189                 } else {
 190                         pg.setMinMax(0, 100);
 191                 }
 192
 193                 login();
 194                 pg.setProgress(10);
 195
 196                 url = getCanonicalUrl(url);
 197
 198                 setCurrentReferer(url);
 199
 200                 in = openInput(url); // NULL allowed here
 201                 try {
 202                         preprocess(url, getInput());
 203                         pg.setProgress(30);
 204
 205                         Story story = new Story();
 206
 207                         MetaData meta = getMeta(url, getInput());
 208                         meta.setType(getType().toString());
 209                         meta.setSource(getType().getSourceName());
 210                         meta.setPublisher(getType().getSourceName());
 211
 212                         if (meta.getCreationDate() == null
 213                                         || meta.getCreationDate().trim().isEmpty()) {
 214                                 meta.setCreationDate(bsHelper.formatDate(
 215                                                 StringUtils.fromTime(new Date().getTime())));
 216                         }
 217                         story.setMeta(meta);
 218                         pg.put("meta", meta);
 219
 220                         pg.setProgress(50);
 221
 222                         if (meta.getCover() == null) {
 223                                 meta.setCover(getDefaultCover(meta.getSubject()));
 224                         }
 225
 226                         pg.setProgress(60);
 227
 228                         if (getDesc) {
 229                                 String descChapterName = Instance.getInstance().getTrans().getString(StringId.DESCRIPTION);
 230                                 story.getMeta().setResume(makeChapter(url, 0, descChapterName, getDesc(url, getInput()), null));
 231                         }
 232
 233                         pg.setProgress(100);
 234                         return story;
 235                 } finally {
 236                         if (close) {
 237                                 close();
 238
 239                                 if (in != null) {
 240                                         in.close();
 241                                 }
 242                         }
 243                 }
 244         }
 245
 246         /**
 247          * Process the given story resource into a fully filled {@link Story}
 248          * object.
 249          *
 250          * @param url
 251          *            the story resource
 252          * @param pg
 253          *            the optional progress reporter
 254          *
 255          * @return the {@link Story}, never NULL
 256          *
 257          * @throws IOException
 258          *             in case of I/O error
 259          */
 260         protected Story process(URL url, Progress pg) throws IOException {
 261                 if (pg == null) {
 262                         pg = new Progress();
 263                 } else {
 264                         pg.setMinMax(0, 100);
 265                 }
 266
 267                 url = getCanonicalUrl(url);
 268                 pg.setProgress(1);
 269                 try {
 270                         Progress pgMeta = new Progress();
 271                         pg.addProgress(pgMeta, 10);
 272                         Story story = processMeta(url, false, true, pgMeta);
 273                         pg.put("meta", story.getMeta());
 274                         if (!pgMeta.isDone()) {
 275                                 pgMeta.setProgress(pgMeta.getMax()); // 10%
 276                         }
 277
 278                         setCurrentReferer(url);
 279
 280                         Progress pgGetChapters = new Progress();
 281                         pg.addProgress(pgGetChapters, 10);
 282                         story.setChapters(new ArrayList<Chapter>());
 283                         List<Entry<String, URL>> chapters = getChapters(url, getInput(),
 284                                         pgGetChapters);
 285                         if (!pgGetChapters.isDone()) {
 286                                 pgGetChapters.setProgress(pgGetChapters.getMax()); // 20%
 287                         }
 288
 289                         if (chapters != null) {
 290                                 Progress pgChaps = new Progress("Extracting chapters", 0,
 291                                                 chapters.size() * 300);
 292                                 pg.addProgress(pgChaps, 80);
 293
 294                                 long words = 0;
 295                                 int i = 1;
 296                                 for (Entry<String, URL> chap : chapters) {
 297                                         pgChaps.setName("Extracting chapter " + i);
 298                                         InputStream chapIn = null;
 299                                         if (chap.getValue() != null) {
 300                                                 setCurrentReferer(chap.getValue());
 301                                                 chapIn = Instance.getInstance().getCache().open(chap.getValue(), this, false);
 302                                         }
 303                                         pgChaps.setProgress(i * 100);
 304                                         try {
 305                                                 Progress pgGetChapterContent = new Progress();
 306                                                 Progress pgMakeChapter = new Progress();
 307                                                 pgChaps.addProgress(pgGetChapterContent, 100);
 308                                                 pgChaps.addProgress(pgMakeChapter, 100);
 309
 310                                                 String content = getChapterContent(url, chapIn, i,
 311                                                                 pgGetChapterContent);
 312                                                 if (!pgGetChapterContent.isDone()) {
 313                                                         pgGetChapterContent.setProgress(pgGetChapterContent
 314                                                                         .getMax());
 315                                                 }
 316
 317                                                 Chapter cc = makeChapter(url, i, chap.getKey(),
 318                                                                 content, pgMakeChapter);
 319                                                 if (!pgMakeChapter.isDone()) {
 320                                                         pgMakeChapter.setProgress(pgMakeChapter.getMax());
 321                                                 }
 322
 323                                                 words += cc.getWords();
 324                                                 story.getChapters().add(cc);
 325                                         } finally {
 326                                                 if (chapIn != null) {
 327                                                         chapIn.close();
 328                                                 }
 329                                         }
 330
 331                                         i++;
 332                                 }
 333
 334                                 story.getMeta().setWords(words);
 335
 336                                 pgChaps.setName("Extracting chapters");
 337                         } else {
 338                                 pg.setProgress(80);
 339                         }
 340
 341                         // Check for "no chapters" stories
 342                         if (story.getChapters().isEmpty()
 343                                         && story.getMeta().getResume() != null
 344                                         && !story.getMeta().getResume().getParagraphs().isEmpty()) {
 345                                 Chapter resume = story.getMeta().getResume();
 346                                 resume.setName("");
 347                                 resume.setNumber(1);
 348                                 story.getChapters().add(resume);
 349                                 story.getMeta().setWords(resume.getWords());
 350
 351                                 String descChapterName = Instance.getInstance().getTrans()
 352                                                 .getString(StringId.DESCRIPTION);
 353                                 resume = new Chapter(0, descChapterName);
 354                                 story.getMeta().setResume(resume);
 355                         }
 356
 357                         return story;
 358                 } finally {
 359                         close();
 360
 361                         if (in != null) {
 362                                 in.close();
 363                         }
 364                 }
 365         }
 366
 367         /**
 368          * Prepare the support if needed before processing.
 369          *
 370          * @param source
 371          *            the source of the story
 372          * @param in
 373          *            the input (the main resource)
 374          *
 375          * @throws IOException
 376          *             on I/O error
 377          */
 378         @SuppressWarnings("unused")
 379         protected void preprocess(URL source, InputStream in) throws IOException {
 380         }
 381
 382         /**
 383          * Create a {@link Chapter} object from the given information, formatting
 384          * the content as it should be.
 385          *
 386          * @param source
 387          *            the source of the story
 388          * @param number
 389          *            the chapter number
 390          * @param name
 391          *            the chapter name
 392          * @param content
 393          *            the chapter content
 394          * @param pg
 395          *            the optional progress reporter
 396          *
 397          * @return the {@link Chapter}, never NULL
 398          *
 399          * @throws IOException
 400          *             in case of I/O error
 401          */
 402         protected Chapter makeChapter(URL source, int number, String name,
 403                         String content, Progress pg) throws IOException {
 404                 // Chapter name: process it correctly, then remove the possible
 405                 // redundant "Chapter x: " in front of it, or "-" (as in
 406                 // "Chapter 5: - Fun!" after the ": " was automatically added)
 407                 String chapterName = processPara(name).getContent().trim();
 408                 for (String lang : Instance.getInstance().getConfig().getList(Config.CONF_CHAPTER)) {
 409                         String chapterWord = Instance.getInstance().getConfig().getStringX(Config.CONF_CHAPTER, lang);
 410                         if (chapterName.startsWith(chapterWord)) {
 411                                 chapterName = chapterName.substring(chapterWord.length())
 412                                                 .trim();
 413                                 break;
 414                         }
 415                 }
 416
 417                 if (chapterName.startsWith(Integer.toString(number))) {
 418                         chapterName = chapterName.substring(
 419                                         Integer.toString(number).length()).trim();
 420                 }
 421
 422                 while (chapterName.startsWith(":") || chapterName.startsWith("-")) {
 423                         chapterName = chapterName.substring(1).trim();
 424                 }
 425                 //
 426
 427                 Chapter chap = new Chapter(number, chapterName);
 428
 429                 if (content != null) {
 430                         List<Paragraph> paras = makeParagraphs(source, content, pg);
 431                         long words = 0;
 432                         for (Paragraph para : paras) {
 433                                 words += para.getWords();
 434                         }
 435                         chap.setParagraphs(paras);
 436                         chap.setWords(words);
 437                 }
 438
 439                 return chap;
 440
 441         }
 442
 443         /**
 444          * Convert the given content into {@link Paragraph}s.
 445          *
 446          * @param source
 447          *            the source URL of the story
 448          * @param content
 449          *            the textual content
 450          * @param pg
 451          *            the optional progress reporter
 452          *
 453          * @return the {@link Paragraph}s (can be empty, but never NULL)
 454          *
 455          * @throws IOException
 456          *             in case of I/O error
 457          */
 458         protected List<Paragraph> makeParagraphs(URL source, String content,
 459                         Progress pg) throws IOException {
 460                 if (pg == null) {
 461                         pg = new Progress();
 462                 }
 463
 464                 if (isHtml()) {
 465                         // Special <HR> processing:
 466                         content = content.replaceAll("(<hr [^>]*>)|(<hr/>)|(<hr>)",
 467                                         "<br/>* * *<br/>");
 468                 }
 469
 470                 List<Paragraph> paras = new ArrayList<Paragraph>();
 471                 if (content != null && !content.trim().isEmpty()) {
 472                         if (isHtml()) {
 473                                 String[] tab = content.split("(<p>|</p>|<br>|<br/>)");
 474                                 pg.setMinMax(0, tab.length);
 475                                 int i = 1;
 476                                 for (String line : tab) {
 477                                         if (line.startsWith("[") && line.endsWith("]")) {
 478                                                 pg.setName("Extracting image " + i);
 479                                         }
 480                                         paras.add(makeParagraph(source, line.trim()));
 481                                         pg.setProgress(i++);
 482                                 }
 483                                 pg.setName(null);
 484                         } else {
 485                                 List<String> lines = new ArrayList<String>();
 486                                 BufferedReader buff = null;
 487                                 try {
 488                                         buff = new BufferedReader(
 489                                                         new InputStreamReader(new ByteArrayInputStream(
 490                                                                         content.getBytes("UTF-8")), "UTF-8"));
 491                                         for (String line = buff.readLine(); line != null; line = buff
 492                                                         .readLine()) {
 493                                                 lines.add(line.trim());
 494                                         }
 495                                 } finally {
 496                                         if (buff != null) {
 497                                                 buff.close();
 498                                         }
 499                                 }
 500
 501                                 pg.setMinMax(0, lines.size());
 502                                 int i = 0;
 503                                 for (String line : lines) {
 504                                         if (line.startsWith("[") && line.endsWith("]")) {
 505                                                 pg.setName("Extracting image " + i);
 506                                         }
 507                                         paras.add(makeParagraph(source, line));
 508                                         pg.setProgress(i++);
 509                                 }
 510                                 pg.setName(null);
 511                         }
 512
 513                         // Check quotes for "bad" format
 514                         List<Paragraph> newParas = new ArrayList<Paragraph>();
 515                         for (Paragraph para : paras) {
 516                                 newParas.addAll(requotify(para));
 517                         }
 518                         paras = newParas;
 519
 520                         // Remove double blanks/brks
 521                         fixBlanksBreaks(paras);
 522                 }
 523
 524                 return paras;
 525         }
 526
 527         /**
 528          * Convert the given line into a single {@link Paragraph}.
 529          *
 530          * @param source
 531          *            the source URL of the story
 532          * @param line
 533          *            the textual content of the paragraph
 534          *
 535          * @return the {@link Paragraph}, never NULL
 536          */
 537         private Paragraph makeParagraph(URL source, String line) {
 538                 Image image = null;
 539                 if (line.startsWith("[") && line.endsWith("]")) {
 540                         image = getImage(this, source, line.substring(1, line.length() - 1)
 541                                         .trim());
 542                 }
 543
 544                 if (image != null) {
 545                         return new Paragraph(image);
 546                 }
 547
 548                 return processPara(line);
 549         }
 550
 551         /**
 552          * Fix the {@link ParagraphType#BLANK}s and {@link ParagraphType#BREAK}s of
 553          * those {@link Paragraph}s.
 554          * <p>
 555          * The resulting list will not contain a starting or trailing blank/break
 556          * nor 2 blanks or breaks following each other.
 557          *
 558          * @param paras
 559          *            the list of {@link Paragraph}s to fix
 560          */
 561         protected void fixBlanksBreaks(List<Paragraph> paras) {
 562                 boolean space = false;
 563                 boolean brk = true;
 564                 for (int i = 0; i < paras.size(); i++) {
 565                         Paragraph para = paras.get(i);
 566                         boolean thisSpace = para.getType() == ParagraphType.BLANK;
 567                         boolean thisBrk = para.getType() == ParagraphType.BREAK;
 568
 569                         if (i > 0 && space && thisBrk) {
 570                                 paras.remove(i - 1);
 571                                 i--;
 572                         } else if ((space || brk) && (thisSpace || thisBrk)) {
 573                                 paras.remove(i);
 574                                 i--;
 575                         }
 576
 577                         space = thisSpace;
 578                         brk = thisBrk;
 579                 }
 580
 581                 // Remove blank/brk at start
 582                 if (paras.size() > 0
 583                                 && (paras.get(0).getType() == ParagraphType.BLANK || paras.get(
 584                                                 0).getType() == ParagraphType.BREAK)) {
 585                         paras.remove(0);
 586                 }
 587
 588                 // Remove blank/brk at end
 589                 int last = paras.size() - 1;
 590                 if (paras.size() > 0
 591                                 && (paras.get(last).getType() == ParagraphType.BLANK || paras
 592                                                 .get(last).getType() == ParagraphType.BREAK)) {
 593                         paras.remove(last);
 594                 }
 595         }
 596
 597         /**
 598          * Get the default cover related to this subject (see <tt>.info</tt> files).
 599          *
 600          * @param subject
 601          *            the subject
 602          *
 603          * @return the cover if any, or NULL
 604          */
 605         static Image getDefaultCover(String subject) {
 606                 if (subject != null && !subject.isEmpty() && Instance.getInstance().getCoverDir() != null) {
 607                         try {
 608                                 File fileCover = new File(Instance.getInstance().getCoverDir(), subject);
 609                                 return getImage(null, fileCover.toURI().toURL(), subject);
 610                         } catch (MalformedURLException e) {
 611                         }
 612                 }
 613
 614                 return null;
 615         }
 616
 617         /**
 618          * Return the list of supported image extensions.
 619          *
 620          * @param emptyAllowed
 621          *            TRUE to allow an empty extension on first place, which can be
 622          *            used when you may already have an extension in your input but
 623          *            are not sure about it
 624          *
 625          * @return the extensions
 626          */
 627         static String[] getImageExt(boolean emptyAllowed) {
 628                 if (emptyAllowed) {
 629                         return new String[] { "", ".png", ".jpg", ".jpeg", ".gif", ".bmp" };
 630                 }
 631
 632                 return new String[] { ".png", ".jpg", ".jpeg", ".gif", ".bmp" };
 633         }
 634
 635         /**
 636          * Check if the given resource can be a local image or a remote image, then
 637          * refresh the cache with it if it is.
 638          *
 639          * @param source
 640          *            the story source
 641          * @param line
 642          *            the resource to check
 643          *
 644          * @return the image if found, or NULL
 645          *
 646          */
 647         static Image getImage(BasicSupport_Deprecated support, URL source,
 648                         String line) {
 649                 URL url = getImageUrl(support, source, line);
 650                 if (url != null) {
 651                         if ("file".equals(url.getProtocol())) {
 652                                 if (new File(url.getPath()).isDirectory()) {
 653                                         return null;
 654                                 }
 655                         }
 656                         InputStream in = null;
 657                         try {
 658                                 in = Instance.getInstance().getCache().open(url, getSupport(url), true);
 659                                 Image img = new Image(in);
 660                                 if (img.getSize() == 0) {
 661                                         img.close();
 662                                         throw new IOException(
 663                                                         "Empty image not accepted");
 664                                 }
 665                                 return img;
 666                         } catch (IOException e) {
 667                         } finally {
 668                                 if (in != null) {
 669                                         try {
 670                                                 in.close();
 671                                         } catch (IOException e) {
 672                                         }
 673                                 }
 674                         }
 675                 }
 676
 677                 return null;
 678         }
 679
 680         /**
 681          * Check if the given resource can be a local image or a remote image, then
 682          * refresh the cache with it if it is.
 683          *
 684          * @param source
 685          *            the story source
 686          * @param line
 687          *            the resource to check
 688          *
 689          * @return the image URL if found, or NULL
 690          *
 691          */
 692         static URL getImageUrl(BasicSupport_Deprecated support, URL source,
 693                         String line) {
 694                 URL url = null;
 695
 696                 if (line != null) {
 697                         // try for files
 698                         if (source != null) {
 699                                 try {
 700                                         String relPath = null;
 701                                         String absPath = null;
 702                                         try {
 703                                                 String path = new File(source.getFile()).getParent();
 704                                                 relPath = new File(new File(path), line.trim())
 705                                                                 .getAbsolutePath();
 706                                         } catch (Exception e) {
 707                                                 // Cannot be converted to path (one possibility to take
 708                                                 // into account: absolute path on Windows)
 709                                         }
 710                                         try {
 711                                                 absPath = new File(line.trim()).getAbsolutePath();
 712                                         } catch (Exception e) {
 713                                                 // Cannot be converted to path (at all)
 714                                         }
 715
 716                                         for (String ext : getImageExt(true)) {
 717                                                 File absFile = new File(absPath + ext);
 718                                                 File relFile = new File(relPath + ext);
 719                                                 if (absPath != null && absFile.exists()
 720                                                                 && absFile.isFile()) {
 721                                                         url = absFile.toURI().toURL();
 722                                                 } else if (relPath != null && relFile.exists()
 723                                                                 && relFile.isFile()) {
 724                                                         url = relFile.toURI().toURL();
 725                                                 }
 726                                         }
 727                                 } catch (Exception e) {
 728                                         // Should not happen since we control the correct arguments
 729                                 }
 730                         }
 731
 732                         if (url == null) {
 733                                 // try for URLs
 734                                 try {
 735                                         for (String ext : getImageExt(true)) {
 736                                                 if (Instance.getInstance().getCache().check(new URL(line + ext), true)) {
 737                                                         url = new URL(line + ext);
 738                                                         break;
 739                                                 }
 740                                         }
 741
 742                                         // try out of cache
 743                                         if (url == null) {
 744                                                 for (String ext : getImageExt(true)) {
 745                                                         try {
 746                                                                 url = new URL(line + ext);
 747                                                                 Instance.getInstance().getCache().refresh(url, support, true);
 748                                                                 break;
 749                                                         } catch (IOException e) {
 750                                                                 // no image with this ext
 751                                                                 url = null;
 752                                                         }
 753                                                 }
 754                                         }
 755                                 } catch (MalformedURLException e) {
 756                                         // Not an url
 757                                 }
 758                         }
 759
 760                         // refresh the cached file
 761                         if (url != null) {
 762                                 try {
 763                                         Instance.getInstance().getCache().refresh(url, support, true);
 764                                 } catch (IOException e) {
 765                                         // woops, broken image
 766                                         url = null;
 767                                 }
 768                         }
 769                 }
 770
 771                 return url;
 772         }
 773
 774         /**
 775          * Open the input file that will be used through the support.
 776          * <p>
 777          * Can return NULL, in which case you are supposed to work without an
 778          * {@link InputStream}.
 779          *
 780          * @param source
 781          *            the source {@link URL}
 782          *
 783          * @return the {@link InputStream}
 784          *
 785          * @throws IOException
 786          *             in case of I/O error
 787          */
 788         protected InputStream openInput(URL source) throws IOException {
 789                 return Instance.getInstance().getCache().open(source, this, false);
 790         }
 791
 792         /**
 793          * Reset then return {@link BasicSupport_Deprecated#in}.
 794          *
 795          * @return {@link BasicSupport_Deprecated#in}
 796          */
 797         protected InputStream getInput() {
 798                 return reset(in);
 799         }
 800
 801         /**
 802          * Check quotes for bad format (i.e., quotes with normal paragraphs inside)
 803          * and requotify them (i.e., separate them into QUOTE paragraphs and other
 804          * paragraphs (quotes or not)).
 805          *
 806          * @param para
 807          *            the paragraph to requotify (not necessarily a quote)
 808          *
 809          * @return the correctly (or so we hope) quotified paragraphs
 810          */
 811         protected List<Paragraph> requotify(Paragraph para) {
 812                 List<Paragraph> newParas = new ArrayList<Paragraph>();
 813
 814                 if (para.getType() == ParagraphType.QUOTE
 815                                 && para.getContent().length() > 2) {
 816                         String line = para.getContent();
 817                         boolean singleQ = line.startsWith("" + openQuote);
 818                         boolean doubleQ = line.startsWith("" + openDoubleQuote);
 819
 820                         // Do not try when more than one quote at a time
 821                         // (some stories are not easily readable if we do)
 822                         if (singleQ
 823                                         && line.indexOf(closeQuote, 1) < line
 824                                                         .lastIndexOf(closeQuote)) {
 825                                 newParas.add(para);
 826                                 return newParas;
 827                         }
 828                         if (doubleQ
 829                                         && line.indexOf(closeDoubleQuote, 1) < line
 830                                                         .lastIndexOf(closeDoubleQuote)) {
 831                                 newParas.add(para);
 832                                 return newParas;
 833                         }
 834                         //
 835
 836                         if (!singleQ && !doubleQ) {
 837                                 line = openDoubleQuote + line + closeDoubleQuote;
 838                                 newParas.add(new Paragraph(ParagraphType.QUOTE, line, para
 839                                                 .getWords()));
 840                         } else {
 841                                 char open = singleQ ? openQuote : openDoubleQuote;
 842                                 char close = singleQ ? closeQuote : closeDoubleQuote;
 843
 844                                 int posDot = -1;
 845                                 boolean inQuote = false;
 846                                 int i = 0;
 847                                 for (char car : line.toCharArray()) {
 848                                         if (car == open) {
 849                                                 inQuote = true;
 850                                         } else if (car == close) {
 851                                                 inQuote = false;
 852                                         } else if (car == '.' && !inQuote) {
 853                                                 posDot = i;
 854                                                 break;
 855                                         }
 856                                         i++;
 857                                 }
 858
 859                                 if (posDot >= 0) {
 860                                         String rest = line.substring(posDot + 1).trim();
 861                                         line = line.substring(0, posDot + 1).trim();
 862                                         long words = 1;
 863                                         for (char car : line.toCharArray()) {
 864                                                 if (car == ' ') {
 865                                                         words++;
 866                                                 }
 867                                         }
 868                                         newParas.add(new Paragraph(ParagraphType.QUOTE, line, words));
 869                                         if (!rest.isEmpty()) {
 870                                                 newParas.addAll(requotify(processPara(rest)));
 871                                         }
 872                                 } else {
 873                                         newParas.add(para);
 874                                 }
 875                         }
 876                 } else {
 877                         newParas.add(para);
 878                 }
 879
 880                 return newParas;
 881         }
 882
 883         /**
 884          * Process a {@link Paragraph} from a raw line of text.
 885          * <p>
 886          * Will also fix quotes and HTML encoding if needed.
 887          *
 888          * @param line
 889          *            the raw line
 890          *
 891          * @return the processed {@link Paragraph}, never NULL
 892          */
 893         protected Paragraph processPara(String line) {
 894                 line = ifUnhtml(line).trim();
 895
 896                 boolean space = true;
 897                 boolean brk = true;
 898                 boolean quote = false;
 899                 boolean tentativeCloseQuote = false;
 900                 char prev = '\0';
 901                 int dashCount = 0;
 902                 long words = 1;
 903
 904                 StringBuilder builder = new StringBuilder();
 905                 for (char car : line.toCharArray()) {
 906                         if (car != '-') {
 907                                 if (dashCount > 0) {
 908                                         // dash, ndash and mdash: - – —
 909                                         // currently: always use mdash
 910                                         builder.append(dashCount == 1 ? '-' : '—');
 911                                 }
 912                                 dashCount = 0;
 913                         }
 914
 915                         if (tentativeCloseQuote) {
 916                                 tentativeCloseQuote = false;
 917                                 if (Character.isLetterOrDigit(car)) {
 918                                         builder.append("'");
 919                                 } else {
 920                                         // handle double-single quotes as double quotes
 921                                         if (prev == car) {
 922                                                 builder.append(closeDoubleQuote);
 923                                                 continue;
 924                                         }
 925
 926                                         builder.append(closeQuote);
 927                                 }
 928                         }
 929
 930                         switch (car) {
 931                         case ' ': // note: unbreakable space
 932                         case ' ':
 933                         case '\t':
 934                         case '\n': // just in case
 935                         case '\r': // just in case
 936                                 if (builder.length() > 0
 937                                                 && builder.charAt(builder.length() - 1) != ' ') {
 938                                         words++;
 939                                 }
 940                                 builder.append(' ');
 941                                 break;
 942
 943                         case '\'':
 944                                 if (space || (brk && quote)) {
 945                                         quote = true;
 946                                         // handle double-single quotes as double quotes
 947                                         if (prev == car) {
 948                                                 builder.deleteCharAt(builder.length() - 1);
 949                                                 builder.append(openDoubleQuote);
 950                                         } else {
 951                                                 builder.append(openQuote);
 952                                         }
 953                                 } else if (prev == ' ' || prev == car) {
 954                                         // handle double-single quotes as double quotes
 955                                         if (prev == car) {
 956                                                 builder.deleteCharAt(builder.length() - 1);
 957                                                 builder.append(openDoubleQuote);
 958                                         } else {
 959                                                 builder.append(openQuote);
 960                                         }
 961                                 } else {
 962                                         // it is a quote ("I'm off") or a 'quote' ("This
 963                                         // 'good' restaurant"...)
 964                                         tentativeCloseQuote = true;
 965                                 }
 966                                 break;
 967
 968                         case '"':
 969                                 if (space || (brk && quote)) {
 970                                         quote = true;
 971                                         builder.append(openDoubleQuote);
 972                                 } else if (prev == ' ') {
 973                                         builder.append(openDoubleQuote);
 974                                 } else {
 975                                         builder.append(closeDoubleQuote);
 976                                 }
 977                                 break;
 978
 979                         case '-':
 980                                 if (space) {
 981                                         quote = true;
 982                                 } else {
 983                                         dashCount++;
 984                                 }
 985                                 space = false;
 986                                 break;
 987
 988                         case '*':
 989                         case '~':
 990                         case '/':
 991                         case '\\':
 992                         case '<':
 993                         case '>':
 994                         case '=':
 995                         case '+':
 996                         case '_':
 997                         case '–':
 998                         case '—':
 999                                 space = false;
1000                                 builder.append(car);
1001                                 break;
1002
1003                         case '‘':
1004                         case '`':
1005                         case '‹':
1006                         case '﹁':
1007                         case '〈':
1008                         case '「':
1009                                 if (space || (brk && quote)) {
1010                                         quote = true;
1011                                         builder.append(openQuote);
1012                                 } else {
1013                                         // handle double-single quotes as double quotes
1014                                         if (prev == car) {
1015                                                 builder.deleteCharAt(builder.length() - 1);
1016                                                 builder.append(openDoubleQuote);
1017                                         } else {
1018                                                 builder.append(openQuote);
1019                                         }
1020                                 }
1021                                 space = false;
1022                                 brk = false;
1023                                 break;
1024
1025                         case '’':
1026                         case '›':
1027                         case '﹂':
1028                         case '〉':
1029                         case '」':
1030                                 space = false;
1031                                 brk = false;
1032                                 // handle double-single quotes as double quotes
1033                                 if (prev == car) {
1034                                         builder.deleteCharAt(builder.length() - 1);
1035                                         builder.append(closeDoubleQuote);
1036                                 } else {
1037                                         builder.append(closeQuote);
1038                                 }
1039                                 break;
1040
1041                         case '«':
1042                         case '“':
1043                         case '﹃':
1044                         case '《':
1045                         case '『':
1046                                 if (space || (brk && quote)) {
1047                                         quote = true;
1048                                         builder.append(openDoubleQuote);
1049                                 } else {
1050                                         builder.append(openDoubleQuote);
1051                                 }
1052                                 space = false;
1053                                 brk = false;
1054                                 break;
1055
1056                         case '»':
1057                         case '”':
1058                         case '﹄':
1059                         case '》':
1060                         case '』':
1061                                 space = false;
1062                                 brk = false;
1063                                 builder.append(closeDoubleQuote);
1064                                 break;
1065
1066                         default:
1067                                 space = false;
1068                                 brk = false;
1069                                 builder.append(car);
1070                                 break;
1071                         }
1072
1073                         prev = car;
1074                 }
1075
1076                 if (tentativeCloseQuote) {
1077                         tentativeCloseQuote = false;
1078                         builder.append(closeQuote);
1079                 }
1080
1081                 line = builder.toString().trim();
1082
1083                 ParagraphType type = ParagraphType.NORMAL;
1084                 if (space) {
1085                         type = ParagraphType.BLANK;
1086                 } else if (brk) {
1087                         type = ParagraphType.BREAK;
1088                 } else if (quote) {
1089                         type = ParagraphType.QUOTE;
1090                 }
1091
1092                 return new Paragraph(type, line, words);
1093         }
1094
1095         /**
1096          * Remove the HTML from the input <b>if</b>
1097          * {@link BasicSupport_Deprecated#isHtml()} is true.
1098          *
1099          * @param input
1100          *            the input
1101          *
1102          * @return the no html version if needed
1103          */
1104         private String ifUnhtml(String input) {
1105                 if (isHtml() && input != null) {
1106                         return StringUtils.unhtml(input);
1107                 }
1108
1109                 return input;
1110         }
1111
1112         /**
1113          * Reset the given {@link InputStream} and return it.
1114          *
1115          * @param in
1116          *            the {@link InputStream} to reset
1117          *
1118          * @return the same {@link InputStream} after reset
1119          */
1120         static protected InputStream reset(InputStream in) {
1121                 try {
1122                         if (in != null) {
1123                                 in.reset();
1124                         }
1125                 } catch (IOException e) {
1126                 }
1127
1128                 return in;
1129         }
1130
1131         /**
1132          * Return the first line from the given input which correspond to the given
1133          * selectors.
1134          *
1135          * @param in
1136          *            the input
1137          * @param needle
1138          *            a string that must be found inside the target line (also
1139          *            supports "^" at start to say "only if it starts with" the
1140          *            needle)
1141          * @param relativeLine
1142          *            the line to return based upon the target line position (-1 =
1143          *            the line before, 0 = the target line...)
1144          *
1145          * @return the line, or NULL if not found
1146          */
1147         static protected String getLine(InputStream in, String needle,
1148                         int relativeLine) {
1149                 return getLine(in, needle, relativeLine, true);
1150         }
1151
1152         /**
1153          * Return a line from the given input which correspond to the given
1154          * selectors.
1155          *
1156          * @param in
1157          *            the input
1158          * @param needle
1159          *            a string that must be found inside the target line (also
1160          *            supports "^" at start to say "only if it starts with" the
1161          *            needle)
1162          * @param relativeLine
1163          *            the line to return based upon the target line position (-1 =
1164          *            the line before, 0 = the target line...)
1165          * @param first
1166          *            takes the first result (as opposed to the last one, which will
1167          *            also always spend the input)
1168          *
1169          * @return the line, or NULL if not found
1170          */
1171         static protected String getLine(InputStream in, String needle,
1172                         int relativeLine, boolean first) {
1173                 String rep = null;
1174
1175                 reset(in);
1176
1177                 List<String> lines = new ArrayList<String>();
1178                 @SuppressWarnings("resource")
1179                 Scanner scan = new Scanner(in, "UTF-8");
1180                 int index = -1;
1181                 scan.useDelimiter("\\n");
1182                 while (scan.hasNext()) {
1183                         lines.add(scan.next());
1184
1185                         if (index == -1) {
1186                                 if (needle.startsWith("^")) {
1187                                         if (lines.get(lines.size() - 1).startsWith(
1188                                                         needle.substring(1))) {
1189                                                 index = lines.size() - 1;
1190                                         }
1191
1192                                 } else {
1193                                         if (lines.get(lines.size() - 1).contains(needle)) {
1194                                                 index = lines.size() - 1;
1195                                         }
1196                                 }
1197                         }
1198
1199                         if (index >= 0 && index + relativeLine < lines.size()) {
1200                                 rep = lines.get(index + relativeLine);
1201                                 if (first) {
1202                                         break;
1203                                 }
1204                         }
1205                 }
1206
1207                 return rep;
1208         }
1209
1210         /**
1211          * Return the text between the key and the endKey (and optional subKey can
1212          * be passed, in this case we will look for the key first, then take the
1213          * text between the subKey and the endKey).
1214          * <p>
1215          * Will only match the first line with the given key if more than one are
1216          * possible. Which also means that if the subKey or endKey is not found on
1217          * that line, NULL will be returned.
1218          *
1219          * @param in
1220          *            the input
1221          * @param key
1222          *            the key to match (also supports "^" at start to say
1223          *            "only if it starts with" the key)
1224          * @param subKey
1225          *            the sub key or NULL if none
1226          * @param endKey
1227          *            the end key or NULL for "up to the end"
1228          * @return the text or NULL if not found
1229          */
1230         static protected String getKeyLine(InputStream in, String key,
1231                         String subKey, String endKey) {
1232                 return getKeyText(getLine(in, key, 0), key, subKey, endKey);
1233         }
1234
1235         /**
1236          * Return the text between the key and the endKey (and optional subKey can
1237          * be passed, in this case we will look for the key first, then take the
1238          * text between the subKey and the endKey).
1239          *
1240          * @param in
1241          *            the input
1242          * @param key
1243          *            the key to match (also supports "^" at start to say
1244          *            "only if it starts with" the key)
1245          * @param subKey
1246          *            the sub key or NULL if none
1247          * @param endKey
1248          *            the end key or NULL for "up to the end"
1249          * @return the text or NULL if not found
1250          */
1251         static protected String getKeyText(String in, String key, String subKey,
1252                         String endKey) {
1253                 String result = null;
1254
1255                 String line = in;
1256                 if (line != null && line.contains(key)) {
1257                         line = line.substring(line.indexOf(key) + key.length());
1258                         if (subKey == null || subKey.isEmpty() || line.contains(subKey)) {
1259                                 if (subKey != null) {
1260                                         line = line.substring(line.indexOf(subKey)
1261                                                         + subKey.length());
1262                                 }
1263                                 if (endKey == null || line.contains(endKey)) {
1264                                         if (endKey != null) {
1265                                                 line = line.substring(0, line.indexOf(endKey));
1266                                                 result = line;
1267                                         }
1268                                 }
1269                         }
1270                 }
1271
1272                 return result;
1273         }
1274
1275         /**
1276          * Return the text between the key and the endKey (optional subKeys can be
1277          * passed, in this case we will look for the subKeys first, then take the
1278          * text between the key and the endKey).
1279          *
1280          * @param in
1281          *            the input
1282          * @param key
1283          *            the key to match
1284          * @param endKey
1285          *            the end key or NULL for "up to the end"
1286          * @param afters
1287          *            the sub-keys to find before checking for key/endKey
1288          *
1289          * @return the text or NULL if not found
1290          */
1291         static protected String getKeyTextAfter(String in, String key,
1292                         String endKey, String... afters) {
1293
1294                 if (in != null && !in.isEmpty()) {
1295                         int pos = indexOfAfter(in, 0, afters);
1296                         if (pos < 0) {
1297                                 return null;
1298                         }
1299
1300                         in = in.substring(pos);
1301                 }
1302
1303                 return getKeyText(in, key, null, endKey);
1304         }
1305
1306         /**
1307          * Return the first index after all the given "afters" have been found in
1308          * the {@link String}, or -1 if it was not possible.
1309          *
1310          * @param in
1311          *            the input
1312          * @param startAt
1313          *            start at this position in the string
1314          * @param afters
1315          *            the sub-keys to find before checking for key/endKey
1316          *
1317          * @return the text or NULL if not found
1318          */
1319         static protected int indexOfAfter(String in, int startAt, String... afters) {
1320                 int pos = -1;
1321                 if (in != null && !in.isEmpty()) {
1322                         pos = startAt;
1323                         if (afters != null) {
1324                                 for (int i = 0; pos >= 0 && i < afters.length; i++) {
1325                                         String subKey = afters[i];
1326                                         if (!subKey.isEmpty()) {
1327                                                 pos = in.indexOf(subKey, pos);
1328                                                 if (pos >= 0) {
1329                                                         pos += subKey.length();
1330                                                 }
1331                                         }
1332                                 }
1333                         }
1334                 }
1335
1336                 return pos;
1337         }
1338 }