src/be/nikiroo/fanfix/supported/Text.java

   1 package be.nikiroo.fanfix.supported;
   2
   3 import java.io.File;
   4 import java.io.IOException;
   5 import java.io.InputStream;
   6 import java.net.URISyntaxException;
   7 import java.net.URL;
   8 import java.util.AbstractMap;
   9 import java.util.ArrayList;
  10 import java.util.List;
  11 import java.util.Map.Entry;
  12 import java.util.Scanner;
  13
  14 import org.jsoup.nodes.Document;
  15
  16 import be.nikiroo.fanfix.Instance;
  17 import be.nikiroo.fanfix.bundles.Config;
  18 import be.nikiroo.fanfix.data.MetaData;
  19 import be.nikiroo.utils.Image;
  20 import be.nikiroo.utils.ImageUtils;
  21 import be.nikiroo.utils.Progress;
  22 import be.nikiroo.utils.streams.MarkableFileInputStream;
  23
  24 /**
  25  * Support class for local stories encoded in textual format, with a few rules:
  26  * <ul>
  27  * <li>The title must be on the first line</li>
  28  * <li>The author (preceded by nothing, "by " or "©") must be on the second
  29  * line, possibly with the publication date in parenthesis (i.e., "
  30  * <tt>By Unknown (3rd October 1998)</tt>")</li>
  31  * <li>Chapters must be declared with "<tt>Chapter x</tt>" or "
  32  * <tt>Chapter x: NAME OF THE CHAPTER</tt>", where "<tt>x</tt>" is the chapter
  33  * number</li>
  34  * <li>A description of the story must be given as chapter number 0</li>
  35  * <li>A cover may be present, with the same filename but a PNG, JPEG or JPG
  36  * extension</li>
  37  * </ul>
  38  *
  39  * @author niki
  40  */
  41 class Text extends BasicSupport {
  42         private File sourceFile;
  43         private InputStream in;
  44
  45         protected File getSourceFile() {
  46                 return sourceFile;
  47         }
  48
  49         protected InputStream getInput() {
  50                 if (in != null) {
  51                         try {
  52                                 in.reset();
  53                         } catch (IOException e) {
  54                                 Instance.getInstance().getTraceHandler().error(new IOException("Cannot reset the Text stream", e));
  55                         }
  56
  57                         return in;
  58                 }
  59
  60                 return null;
  61         }
  62
  63         @Override
  64         protected boolean isHtml() {
  65                 return false;
  66         }
  67
  68         @Override
  69         protected Document loadDocument(URL source) throws IOException {
  70                 try {
  71                         sourceFile = new File(source.toURI());
  72                         in = new MarkableFileInputStream(sourceFile);
  73                 } catch (URISyntaxException e) {
  74                         throw new IOException("Cannot load the text document: " + source);
  75                 }
  76
  77                 return null;
  78         }
  79
  80         @Override
  81         protected MetaData getMeta() throws IOException {
  82                 MetaData meta = new MetaData();
  83
  84                 meta.setTitle(getTitle());
  85                 meta.setAuthor(getAuthor());
  86                 meta.setDate(bsHelper.formatDate(getDate()));
  87                 meta.setTags(new ArrayList<String>());
  88                 meta.setSource(getType().getSourceName());
  89                 meta.setUrl(getSourceFile().toURI().toURL().toString());
  90                 meta.setPublisher("");
  91                 meta.setUuid(getSourceFile().toString());
  92                 meta.setLuid("");
  93                 meta.setLang(getLang()); // default is EN
  94                 meta.setSubject(getSourceFile().getParentFile().getName());
  95                 meta.setType(getType().toString());
  96                 meta.setImageDocument(false);
  97                 meta.setCover(getCover(getSourceFile()));
  98
  99                 return meta;
 100         }
 101
 102         private String getLang() {
 103                 @SuppressWarnings("resource") // cannot close, or we loose getInput()!
 104                 Scanner scan = new Scanner(getInput(), "UTF-8");
 105                 scan.useDelimiter("\\n");
 106                 scan.next(); // Title
 107                 scan.next(); // Author (Date)
 108                 String chapter0 = scan.next(); // empty or Chapter 0
 109                 while (chapter0.isEmpty()) {
 110                         chapter0 = scan.next();
 111                 }
 112
 113                 String lang = detectChapter(chapter0, 0);
 114                 if (lang == null) {
 115                         // No description??
 116                         lang = detectChapter(chapter0, 1);
 117                 }
 118
 119                 if (lang == null) {
 120                         lang = "en";
 121                 } else {
 122                         lang = lang.toLowerCase();
 123                 }
 124
 125                 return lang;
 126         }
 127
 128         private String getTitle() {
 129                 @SuppressWarnings("resource") // cannot close, or we loose getInput()!
 130                 Scanner scan = new Scanner(getInput(), "UTF-8");
 131                 scan.useDelimiter("\\n");
 132                 return scan.next();
 133         }
 134
 135         private String getAuthor() {
 136                 @SuppressWarnings("resource") // cannot close, or we loose getInput()!
 137                 Scanner scan = new Scanner(getInput(), "UTF-8");
 138                 scan.useDelimiter("\\n");
 139                 scan.next();
 140                 String authorDate = scan.next();
 141
 142                 String author = authorDate;
 143                 int pos = authorDate.indexOf('(');
 144                 if (pos >= 0) {
 145                         author = authorDate.substring(0, pos);
 146                 }
 147
 148                 return bsHelper.fixAuthor(author);
 149         }
 150
 151         private String getDate() {
 152                 @SuppressWarnings("resource") // cannot close, or we loose getInput()!
 153                 Scanner scan = new Scanner(getInput(), "UTF-8");
 154                 scan.useDelimiter("\\n");
 155                 scan.next();
 156                 String authorDate = scan.next();
 157
 158                 String date = "";
 159                 int pos = authorDate.indexOf('(');
 160                 if (pos >= 0) {
 161                         date = authorDate.substring(pos + 1).trim();
 162                         pos = date.lastIndexOf(')');
 163                         if (pos >= 0) {
 164                                 date = date.substring(0, pos).trim();
 165                         }
 166                 }
 167
 168                 return date;
 169         }
 170
 171         @Override
 172         protected String getDesc() throws IOException {
 173                 return getChapterContent(null, 0, null).trim();
 174         }
 175
 176         protected Image getCover(File sourceFile) {
 177                 String path = sourceFile.getName();
 178
 179                 for (String ext : new String[] { ".txt", ".text", ".story" }) {
 180                         if (path.endsWith(ext)) {
 181                                 path = path.substring(0, path.length() - ext.length());
 182                         }
 183                 }
 184
 185                 Image cover = bsImages.getImage(this, sourceFile.getParentFile(), path);
 186                 if (cover != null) {
 187                         try {
 188                                 File tmp = Instance.getInstance().getTempFiles().createTempFile("test_cover_image");
 189                                 ImageUtils.getInstance().saveAsImage(cover, tmp, "png");
 190                                 tmp.delete();
 191                         } catch (IOException e) {
 192                                 cover = null;
 193                         }
 194                 }
 195
 196                 return cover;
 197         }
 198
 199         @Override
 200         protected List<Entry<String, URL>> getChapters(Progress pg)
 201                         throws IOException {
 202                 List<Entry<String, URL>> chaps = new ArrayList<Entry<String, URL>>();
 203                 @SuppressWarnings("resource") // cannot close, or we loose getInput()!
 204                 Scanner scan = new Scanner(getInput(), "UTF-8");
 205                 scan.useDelimiter("\\n");
 206                 String line = "first is not empty";
 207                 while (scan.hasNext()) {
 208                         boolean prevLineEmpty = line.trim().isEmpty();
 209                         line = scan.next();
 210                         if (prevLineEmpty && detectChapter(line, chaps.size() + 1) != null) {
 211                                 String chapName = Integer.toString(chaps.size() + 1);
 212                                 int pos = line.indexOf(':');
 213                                 if (pos >= 0 && pos + 1 < line.length()) {
 214                                         chapName = line.substring(pos + 1).trim();
 215                                 }
 216
 217                                 chaps.add(new AbstractMap.SimpleEntry<String, URL>(//
 218                                                 chapName, //
 219                                                 getSourceFile().toURI().toURL()));
 220                         }
 221                 }
 222
 223                 return chaps;
 224         }
 225
 226         @Override
 227         protected String getChapterContent(URL source, int number, Progress pg)
 228                         throws IOException {
 229                 StringBuilder builder = new StringBuilder();
 230                 @SuppressWarnings("resource") // cannot close, or we loose getInput()!
 231                 Scanner scan = new Scanner(getInput(), "UTF-8");
 232                 scan.useDelimiter("\\n");
 233                 scan.next(); // title
 234                 scan.next(); // author
 235                 scan.next(); // date or empty
 236                 Boolean inChap = null;
 237                 String line = "";
 238                 while (scan.hasNext()) {
 239                         if (number == 0 && !line.trim().isEmpty()) {
 240                                 // We found pre-chapter content, we are checking for
 241                                 // Chapter 0 (fake chapter) --> keep the content
 242                                 if (inChap == null)
 243                                         inChap = true;
 244                         }
 245                         line = scan.next();
 246                         if ((inChap == null || !inChap) && detectChapter(line, number) != null) {
 247                                 inChap = true;
 248                         } else if (detectChapter(line, number + 1) != null) {
 249                                 break;
 250                         } else if (inChap != null && inChap) {
 251                                 builder.append(line);
 252                                 builder.append("\n");
 253                         }
 254                 }
 255
 256                 return builder.toString();
 257         }
 258
 259         @Override
 260         protected void close() {
 261                 InputStream in = getInput();
 262                 if (in != null) {
 263                         try {
 264                                 in.close();
 265                         } catch (IOException e) {
 266                                 Instance.getInstance().getTraceHandler()
 267                                                 .error(new IOException("Cannot close the text source file input", e));
 268                         }
 269                 }
 270
 271                 super.close();
 272         }
 273
 274         @Override
 275         protected boolean supports(URL url) {
 276                 return supports(url, false);
 277         }
 278
 279         /**
 280          * Check if we supports this {@link URL}, that is, if the info file can be
 281          * found OR not found.
 282          * <p>
 283          * It must also be a file, not another kind of URL.
 284          *
 285          * @param url
 286          *            the {@link URL} to check
 287          * @param info
 288          *            TRUE to require the info file, FALSE to forbid the info file
 289          *
 290          * @return TRUE if it is supported
 291          */
 292         protected boolean supports(URL url, boolean info) {
 293                 if (!"file".equals(url.getProtocol())) {
 294                         return false;
 295                 }
 296
 297                 boolean infoPresent = false;
 298                 File file;
 299                 try {
 300                         file = new File(url.toURI());
 301                         file = assureNoTxt(file);
 302                         file = new File(file.getPath() + ".info");
 303                 } catch (URISyntaxException e) {
 304                         Instance.getInstance().getTraceHandler().error(e);
 305                         file = null;
 306                 }
 307
 308                 infoPresent = (file != null && file.exists());
 309
 310                 return infoPresent == info;
 311         }
 312
 313         /**
 314          * Remove the ".txt" (or ".text") extension if it is present.
 315          *
 316          * @param file
 317          *            the file to process
 318          *
 319          * @return the same file or a copy of it without the ".txt" extension if it
 320          *         was present
 321          */
 322         protected File assureNoTxt(File file) {
 323                 for (String ext : new String[] { ".txt", ".text" }) {
 324                         if (file.getName().endsWith(ext)) {
 325                                 file = new File(file.getPath().substring(0,
 326                                                 file.getPath().length() - ext.length()));
 327                         }
 328                 }
 329
 330                 return file;
 331         }
 332
 333         /**
 334          * Check if the given line looks like the given starting chapter in a
 335          * supported language, and return the language if it does (or NULL if not).
 336          *
 337          * @param line
 338          *            the line to check
 339          * @param number
 340          *            the specific chapter number to check for
 341          *
 342          * @return the language or NULL
 343          */
 344         static private String detectChapter(String line, int number) {
 345                 line = line.toUpperCase();
 346                 for (String lang : Instance.getInstance().getConfig().getList(Config.CONF_CHAPTER)) {
 347                         String chapter = Instance.getInstance().getConfig().getStringX(Config.CONF_CHAPTER, lang);
 348                         if (chapter != null && !chapter.isEmpty()) {
 349                                 chapter = chapter.toUpperCase() + " ";
 350                                 if (line.startsWith(chapter)) {
 351                                         // We want "[CHAPTER] [number]: [name]", with ": [name]"
 352                                         // optional
 353                                         String test = line.substring(chapter.length()).trim();
 354
 355                                         String possibleNum = test.trim();
 356                                         if (possibleNum.indexOf(':') > 0) {
 357                                                 possibleNum = possibleNum.substring(0,
 358                                                                 possibleNum.indexOf(':')).trim();
 359                                         }
 360
 361                                         if (test.startsWith(Integer.toString(number))) {
 362                                                 test = test
 363                                                                 .substring(Integer.toString(number).length())
 364                                                                 .trim();
 365                                                 if (test.isEmpty() || test.startsWith(":")) {
 366                                                         return lang;
 367                                                 }
 368                                         }
 369                                 }
 370                         }
 371                 }
 372
 373                 return null;
 374         }
 375 }