Commit | Line | Data |
---|---|---|
0ffa4754 NR |
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 | private URL currentReferer; // with only one 'r', as in 'HTTP'... | |
44 | ||
45 | // quote chars | |
46 | private char openQuote = Instance.getTrans().getCharacter( | |
47 | StringId.OPEN_SINGLE_QUOTE); | |
48 | private char closeQuote = Instance.getTrans().getCharacter( | |
49 | StringId.CLOSE_SINGLE_QUOTE); | |
50 | private char openDoubleQuote = Instance.getTrans().getCharacter( | |
51 | StringId.OPEN_DOUBLE_QUOTE); | |
52 | private char closeDoubleQuote = Instance.getTrans().getCharacter( | |
53 | StringId.CLOSE_DOUBLE_QUOTE); | |
54 | ||
55 | // New methods not used in Deprecated mode | |
56 | @Override | |
57 | protected String getDesc() throws IOException { | |
58 | throw new RuntimeException("should not be used by legacy code"); | |
59 | } | |
60 | ||
61 | @Override | |
62 | protected MetaData getMeta() throws IOException { | |
63 | throw new RuntimeException("should not be used by legacy code"); | |
64 | } | |
65 | ||
66 | @Override | |
67 | protected List<Entry<String, URL>> getChapters(Progress pg) | |
68 | throws IOException { | |
69 | throw new RuntimeException("should not be used by legacy code"); | |
70 | } | |
71 | ||
72 | @Override | |
73 | protected String getChapterContent(URL chapUrl, int number, Progress pg) | |
74 | throws IOException { | |
75 | throw new RuntimeException("should not be used by legacy code"); | |
76 | } | |
77 | ||
78 | @Override | |
79 | public Story process(Progress pg) throws IOException { | |
80 | return process(getSource(), pg); | |
81 | } | |
82 | ||
83 | // | |
84 | ||
85 | /** | |
86 | * Return the {@link MetaData} of this story. | |
87 | * | |
88 | * @param source | |
89 | * the source of the story | |
90 | * @param in | |
91 | * the input (the main resource) | |
92 | * | |
93 | * @return the associated {@link MetaData}, never NULL | |
94 | * | |
95 | * @throws IOException | |
96 | * in case of I/O error | |
97 | */ | |
98 | protected abstract MetaData getMeta(URL source, InputStream in) | |
99 | throws IOException; | |
100 | ||
101 | /** | |
102 | * Return the story description. | |
103 | * | |
104 | * @param source | |
105 | * the source of the story | |
106 | * @param in | |
107 | * the input (the main resource) | |
108 | * | |
109 | * @return the description | |
110 | * | |
111 | * @throws IOException | |
112 | * in case of I/O error | |
113 | */ | |
114 | protected abstract String getDesc(URL source, InputStream in) | |
115 | throws IOException; | |
116 | ||
117 | /** | |
118 | * Return the list of chapters (name and resource). | |
119 | * | |
120 | * @param source | |
121 | * the source of the story | |
122 | * @param in | |
123 | * the input (the main resource) | |
124 | * @param pg | |
125 | * the optional progress reporter | |
126 | * | |
127 | * @return the chapters | |
128 | * | |
129 | * @throws IOException | |
130 | * in case of I/O error | |
131 | */ | |
132 | protected abstract List<Entry<String, URL>> getChapters(URL source, | |
133 | InputStream in, Progress pg) throws IOException; | |
134 | ||
135 | /** | |
136 | * Return the content of the chapter (possibly HTML encoded, if | |
137 | * {@link BasicSupport_Deprecated#isHtml()} is TRUE). | |
138 | * | |
139 | * @param source | |
140 | * the source of the story | |
141 | * @param in | |
142 | * the input (the main resource) | |
143 | * @param number | |
144 | * the chapter number | |
145 | * @param pg | |
146 | * the optional progress reporter | |
147 | * | |
148 | * @return the content | |
149 | * | |
150 | * @throws IOException | |
151 | * in case of I/O error | |
152 | */ | |
153 | protected abstract String getChapterContent(URL source, InputStream in, | |
154 | int number, Progress pg) throws IOException; | |
155 | ||
156 | /** | |
157 | * Process the given story resource into a partially filled {@link Story} | |
158 | * object containing the name and metadata, except for the description. | |
159 | * | |
160 | * @param url | |
161 | * the story resource | |
162 | * | |
163 | * @return the {@link Story} | |
164 | * | |
165 | * @throws IOException | |
166 | * in case of I/O error | |
167 | */ | |
168 | public Story processMeta(URL url) throws IOException { | |
169 | return processMeta(url, true, false, null); | |
170 | } | |
171 | ||
172 | /** | |
173 | * Process the given story resource into a partially filled {@link Story} | |
174 | * object containing the name and metadata. | |
175 | * | |
176 | * @param url | |
177 | * the story resource | |
178 | * @param close | |
179 | * close "this" and "in" when done | |
180 | * @param getDesc | |
181 | * retrieve the description of the story, or not | |
182 | * @param pg | |
183 | * the optional progress reporter | |
184 | * | |
185 | * @return the {@link Story}, never NULL | |
186 | * | |
187 | * @throws IOException | |
188 | * in case of I/O error | |
189 | */ | |
190 | protected Story processMeta(URL url, boolean close, boolean getDesc, | |
191 | Progress pg) throws IOException { | |
192 | if (pg == null) { | |
193 | pg = new Progress(); | |
194 | } else { | |
195 | pg.setMinMax(0, 100); | |
196 | } | |
197 | ||
198 | login(); | |
199 | pg.setProgress(10); | |
200 | ||
201 | url = getCanonicalUrl(url); | |
202 | ||
203 | setCurrentReferer(url); | |
204 | ||
205 | in = openInput(url); // NULL allowed here | |
206 | try { | |
207 | preprocess(url, getInput()); | |
208 | pg.setProgress(30); | |
209 | ||
210 | Story story = new Story(); | |
211 | MetaData meta = getMeta(url, getInput()); | |
212 | if (meta.getCreationDate() == null | |
213 | || meta.getCreationDate().isEmpty()) { | |
214 | meta.setCreationDate(StringUtils.fromTime(new Date().getTime())); | |
215 | } | |
216 | story.setMeta(meta); | |
217 | ||
218 | pg.setProgress(50); | |
219 | ||
220 | if (meta.getCover() == null) { | |
221 | meta.setCover(getDefaultCover(meta.getSubject())); | |
222 | } | |
223 | ||
224 | pg.setProgress(60); | |
225 | ||
226 | if (getDesc) { | |
227 | String descChapterName = Instance.getTrans().getString( | |
228 | StringId.DESCRIPTION); | |
229 | story.getMeta().setResume( | |
230 | makeChapter(url, 0, descChapterName, | |
231 | getDesc(url, getInput()), null)); | |
232 | } | |
233 | ||
234 | pg.setProgress(100); | |
235 | return story; | |
236 | } finally { | |
237 | if (close) { | |
238 | close(); | |
239 | ||
240 | if (in != null) { | |
241 | in.close(); | |
242 | } | |
243 | } | |
244 | } | |
245 | } | |
246 | ||
247 | /** | |
248 | * Process the given story resource into a fully filled {@link Story} | |
249 | * object. | |
250 | * | |
251 | * @param url | |
252 | * the story resource | |
253 | * @param pg | |
254 | * the optional progress reporter | |
255 | * | |
256 | * @return the {@link Story}, never NULL | |
257 | * | |
258 | * @throws IOException | |
259 | * in case of I/O error | |
260 | */ | |
261 | protected Story process(URL url, Progress pg) throws IOException { | |
262 | if (pg == null) { | |
263 | pg = new Progress(); | |
264 | } else { | |
265 | pg.setMinMax(0, 100); | |
266 | } | |
267 | ||
268 | url = getCanonicalUrl(url); | |
269 | pg.setProgress(1); | |
270 | try { | |
271 | Progress pgMeta = new Progress(); | |
272 | pg.addProgress(pgMeta, 10); | |
273 | Story story = processMeta(url, false, true, pgMeta); | |
274 | if (!pgMeta.isDone()) { | |
275 | pgMeta.setProgress(pgMeta.getMax()); // 10% | |
276 | } | |
277 | ||
278 | pg.setName("Retrieving " + story.getMeta().getTitle()); | |
279 | ||
280 | setCurrentReferer(url); | |
281 | ||
282 | Progress pgGetChapters = new Progress(); | |
283 | pg.addProgress(pgGetChapters, 10); | |
284 | story.setChapters(new ArrayList<Chapter>()); | |
285 | List<Entry<String, URL>> chapters = getChapters(url, getInput(), | |
286 | pgGetChapters); | |
287 | if (!pgGetChapters.isDone()) { | |
288 | pgGetChapters.setProgress(pgGetChapters.getMax()); // 20% | |
289 | } | |
290 | ||
291 | if (chapters != null) { | |
292 | Progress pgChaps = new Progress("Extracting chapters", 0, | |
293 | chapters.size() * 300); | |
294 | pg.addProgress(pgChaps, 80); | |
295 | ||
296 | long words = 0; | |
297 | int i = 1; | |
298 | for (Entry<String, URL> chap : chapters) { | |
299 | pgChaps.setName("Extracting chapter " + i); | |
300 | InputStream chapIn = null; | |
301 | if (chap.getValue() != null) { | |
302 | setCurrentReferer(chap.getValue()); | |
303 | chapIn = Instance.getCache().open(chap.getValue(), | |
304 | this, false); | |
305 | } | |
306 | pgChaps.setProgress(i * 100); | |
307 | try { | |
308 | Progress pgGetChapterContent = new Progress(); | |
309 | Progress pgMakeChapter = new Progress(); | |
310 | pgChaps.addProgress(pgGetChapterContent, 100); | |
311 | pgChaps.addProgress(pgMakeChapter, 100); | |
312 | ||
313 | String content = getChapterContent(url, chapIn, i, | |
314 | pgGetChapterContent); | |
315 | if (!pgGetChapterContent.isDone()) { | |
316 | pgGetChapterContent.setProgress(pgGetChapterContent | |
317 | .getMax()); | |
318 | } | |
319 | ||
320 | Chapter cc = makeChapter(url, i, chap.getKey(), | |
321 | content, pgMakeChapter); | |
322 | if (!pgMakeChapter.isDone()) { | |
323 | pgMakeChapter.setProgress(pgMakeChapter.getMax()); | |
324 | } | |
325 | ||
326 | words += cc.getWords(); | |
327 | story.getChapters().add(cc); | |
328 | story.getMeta().setWords(words); | |
329 | } finally { | |
330 | if (chapIn != null) { | |
331 | chapIn.close(); | |
332 | } | |
333 | } | |
334 | ||
335 | i++; | |
336 | } | |
337 | ||
338 | pgChaps.setName("Extracting chapters"); | |
339 | } else { | |
340 | pg.setProgress(80); | |
341 | } | |
342 | ||
343 | return story; | |
344 | ||
345 | } finally { | |
346 | close(); | |
347 | ||
348 | if (in != null) { | |
349 | in.close(); | |
350 | } | |
351 | } | |
352 | } | |
353 | ||
354 | /** | |
355 | * Prepare the support if needed before processing. | |
356 | * | |
357 | * @param source | |
358 | * the source of the story | |
359 | * @param in | |
360 | * the input (the main resource) | |
361 | * | |
362 | * @throws IOException | |
363 | * on I/O error | |
364 | */ | |
365 | @SuppressWarnings("unused") | |
366 | protected void preprocess(URL source, InputStream in) throws IOException { | |
367 | } | |
368 | ||
369 | /** | |
370 | * Create a {@link Chapter} object from the given information, formatting | |
371 | * the content as it should be. | |
372 | * | |
373 | * @param source | |
374 | * the source of the story | |
375 | * @param number | |
376 | * the chapter number | |
377 | * @param name | |
378 | * the chapter name | |
379 | * @param content | |
380 | * the chapter content | |
381 | * @param pg | |
382 | * the optional progress reporter | |
383 | * | |
384 | * @return the {@link Chapter} | |
385 | * | |
386 | * @throws IOException | |
387 | * in case of I/O error | |
388 | */ | |
389 | protected Chapter makeChapter(URL source, int number, String name, | |
390 | String content, Progress pg) throws IOException { | |
391 | // Chapter name: process it correctly, then remove the possible | |
392 | // redundant "Chapter x: " in front of it, or "-" (as in | |
393 | // "Chapter 5: - Fun!" after the ": " was automatically added) | |
394 | String chapterName = processPara(name).getContent().trim(); | |
395 | for (String lang : Instance.getConfig().getString(Config.CHAPTER) | |
396 | .split(",")) { | |
397 | String chapterWord = Instance.getConfig().getStringX( | |
398 | Config.CHAPTER, lang); | |
399 | if (chapterName.startsWith(chapterWord)) { | |
400 | chapterName = chapterName.substring(chapterWord.length()) | |
401 | .trim(); | |
402 | break; | |
403 | } | |
404 | } | |
405 | ||
406 | if (chapterName.startsWith(Integer.toString(number))) { | |
407 | chapterName = chapterName.substring( | |
408 | Integer.toString(number).length()).trim(); | |
409 | } | |
410 | ||
411 | while (chapterName.startsWith(":") || chapterName.startsWith("-")) { | |
412 | chapterName = chapterName.substring(1).trim(); | |
413 | } | |
414 | // | |
415 | ||
416 | Chapter chap = new Chapter(number, chapterName); | |
417 | ||
418 | if (content != null) { | |
419 | List<Paragraph> paras = makeParagraphs(source, content, pg); | |
420 | long words = 0; | |
421 | for (Paragraph para : paras) { | |
422 | words += para.getWords(); | |
423 | } | |
424 | chap.setParagraphs(paras); | |
425 | chap.setWords(words); | |
426 | } | |
427 | ||
428 | return chap; | |
429 | ||
430 | } | |
431 | ||
432 | /** | |
433 | * Convert the given content into {@link Paragraph}s. | |
434 | * | |
435 | * @param source | |
436 | * the source URL of the story | |
437 | * @param content | |
438 | * the textual content | |
439 | * @param pg | |
440 | * the optional progress reporter | |
441 | * | |
442 | * @return the {@link Paragraph}s | |
443 | * | |
444 | * @throws IOException | |
445 | * in case of I/O error | |
446 | */ | |
447 | protected List<Paragraph> makeParagraphs(URL source, String content, | |
448 | Progress pg) throws IOException { | |
449 | if (pg == null) { | |
450 | pg = new Progress(); | |
451 | } | |
452 | ||
453 | if (isHtml()) { | |
454 | // Special <HR> processing: | |
455 | content = content.replaceAll("(<hr [^>]*>)|(<hr/>)|(<hr>)", | |
456 | "<br/>* * *<br/>"); | |
457 | } | |
458 | ||
459 | List<Paragraph> paras = new ArrayList<Paragraph>(); | |
460 | ||
461 | if (content != null && !content.trim().isEmpty()) { | |
462 | if (isHtml()) { | |
463 | String[] tab = content.split("(<p>|</p>|<br>|<br/>)"); | |
464 | pg.setMinMax(0, tab.length); | |
465 | int i = 1; | |
466 | for (String line : tab) { | |
467 | if (line.startsWith("[") && line.endsWith("]")) { | |
468 | pg.setName("Extracting image " + i); | |
469 | } | |
470 | paras.add(makeParagraph(source, line.trim())); | |
471 | pg.setProgress(i++); | |
472 | } | |
473 | pg.setName(null); | |
474 | } else { | |
475 | List<String> lines = new ArrayList<String>(); | |
476 | BufferedReader buff = null; | |
477 | try { | |
478 | buff = new BufferedReader( | |
479 | new InputStreamReader(new ByteArrayInputStream( | |
480 | content.getBytes("UTF-8")), "UTF-8")); | |
481 | for (String line = buff.readLine(); line != null; line = buff | |
482 | .readLine()) { | |
483 | lines.add(line.trim()); | |
484 | } | |
485 | } finally { | |
486 | if (buff != null) { | |
487 | buff.close(); | |
488 | } | |
489 | } | |
490 | ||
491 | pg.setMinMax(0, lines.size()); | |
492 | int i = 0; | |
493 | for (String line : lines) { | |
494 | if (line.startsWith("[") && line.endsWith("]")) { | |
495 | pg.setName("Extracting image " + i); | |
496 | } | |
497 | paras.add(makeParagraph(source, line)); | |
498 | pg.setProgress(i++); | |
499 | } | |
500 | pg.setName(null); | |
501 | } | |
502 | ||
503 | // Check quotes for "bad" format | |
504 | List<Paragraph> newParas = new ArrayList<Paragraph>(); | |
505 | for (Paragraph para : paras) { | |
506 | newParas.addAll(requotify(para)); | |
507 | } | |
508 | paras = newParas; | |
509 | ||
510 | // Remove double blanks/brks | |
511 | fixBlanksBreaks(paras); | |
512 | } | |
513 | ||
514 | return paras; | |
515 | } | |
516 | ||
517 | /** | |
518 | * Convert the given line into a single {@link Paragraph}. | |
519 | * | |
520 | * @param source | |
521 | * the source URL of the story | |
522 | * @param line | |
523 | * the textual content of the paragraph | |
524 | * | |
525 | * @return the {@link Paragraph} | |
526 | */ | |
527 | private Paragraph makeParagraph(URL source, String line) { | |
528 | Image image = null; | |
529 | if (line.startsWith("[") && line.endsWith("]")) { | |
530 | image = getImage(this, source, line.substring(1, line.length() - 1) | |
531 | .trim()); | |
532 | } | |
533 | ||
534 | if (image != null) { | |
535 | return new Paragraph(image); | |
536 | } | |
537 | ||
538 | return processPara(line); | |
539 | } | |
540 | ||
541 | /** | |
542 | * Fix the {@link ParagraphType#BLANK}s and {@link ParagraphType#BREAK}s of | |
543 | * those {@link Paragraph}s. | |
544 | * <p> | |
545 | * The resulting list will not contain a starting or trailing blank/break | |
546 | * nor 2 blanks or breaks following each other. | |
547 | * | |
548 | * @param paras | |
549 | * the list of {@link Paragraph}s to fix | |
550 | */ | |
551 | protected void fixBlanksBreaks(List<Paragraph> paras) { | |
552 | boolean space = false; | |
553 | boolean brk = true; | |
554 | for (int i = 0; i < paras.size(); i++) { | |
555 | Paragraph para = paras.get(i); | |
556 | boolean thisSpace = para.getType() == ParagraphType.BLANK; | |
557 | boolean thisBrk = para.getType() == ParagraphType.BREAK; | |
558 | ||
559 | if (i > 0 && space && thisBrk) { | |
560 | paras.remove(i - 1); | |
561 | i--; | |
562 | } else if ((space || brk) && (thisSpace || thisBrk)) { | |
563 | paras.remove(i); | |
564 | i--; | |
565 | } | |
566 | ||
567 | space = thisSpace; | |
568 | brk = thisBrk; | |
569 | } | |
570 | ||
571 | // Remove blank/brk at start | |
572 | if (paras.size() > 0 | |
573 | && (paras.get(0).getType() == ParagraphType.BLANK || paras.get( | |
574 | 0).getType() == ParagraphType.BREAK)) { | |
575 | paras.remove(0); | |
576 | } | |
577 | ||
578 | // Remove blank/brk at end | |
579 | int last = paras.size() - 1; | |
580 | if (paras.size() > 0 | |
581 | && (paras.get(last).getType() == ParagraphType.BLANK || paras | |
582 | .get(last).getType() == ParagraphType.BREAK)) { | |
583 | paras.remove(last); | |
584 | } | |
585 | } | |
586 | ||
587 | /** | |
588 | * Get the default cover related to this subject (see <tt>.info</tt> files). | |
589 | * | |
590 | * @param subject | |
591 | * the subject | |
592 | * | |
593 | * @return the cover if any, or NULL | |
594 | */ | |
595 | static Image getDefaultCover(String subject) { | |
596 | if (subject != null && !subject.isEmpty() | |
597 | && Instance.getCoverDir() != null) { | |
598 | try { | |
599 | File fileCover = new File(Instance.getCoverDir(), subject); | |
600 | return getImage(null, fileCover.toURI().toURL(), subject); | |
601 | } catch (MalformedURLException e) { | |
602 | } | |
603 | } | |
604 | ||
605 | return null; | |
606 | } | |
607 | ||
608 | /** | |
609 | * Return the list of supported image extensions. | |
610 | * | |
611 | * @param emptyAllowed | |
612 | * TRUE to allow an empty extension on first place, which can be | |
613 | * used when you may already have an extension in your input but | |
614 | * are not sure about it | |
615 | * | |
616 | * @return the extensions | |
617 | */ | |
618 | static String[] getImageExt(boolean emptyAllowed) { | |
619 | if (emptyAllowed) { | |
620 | return new String[] { "", ".png", ".jpg", ".jpeg", ".gif", ".bmp" }; | |
621 | } | |
622 | ||
623 | return new String[] { ".png", ".jpg", ".jpeg", ".gif", ".bmp" }; | |
624 | } | |
625 | ||
626 | /** | |
627 | * Check if the given resource can be a local image or a remote image, then | |
628 | * refresh the cache with it if it is. | |
629 | * | |
630 | * @param source | |
631 | * the story source | |
632 | * @param line | |
633 | * the resource to check | |
634 | * | |
635 | * @return the image if found, or NULL | |
636 | * | |
637 | */ | |
638 | static Image getImage(BasicSupport_Deprecated support, URL source, | |
639 | String line) { | |
640 | URL url = getImageUrl(support, source, line); | |
641 | if (url != null) { | |
642 | if ("file".equals(url.getProtocol())) { | |
643 | if (new File(url.getPath()).isDirectory()) { | |
644 | return null; | |
645 | } | |
646 | } | |
647 | InputStream in = null; | |
648 | try { | |
649 | in = Instance.getCache().open(url, getSupport(url), true); | |
650 | return new Image(in); | |
651 | } catch (IOException e) { | |
652 | } finally { | |
653 | if (in != null) { | |
654 | try { | |
655 | in.close(); | |
656 | } catch (IOException e) { | |
657 | } | |
658 | } | |
659 | } | |
660 | } | |
661 | ||
662 | return null; | |
663 | } | |
664 | ||
665 | /** | |
666 | * Check if the given resource can be a local image or a remote image, then | |
667 | * refresh the cache with it if it is. | |
668 | * | |
669 | * @param source | |
670 | * the story source | |
671 | * @param line | |
672 | * the resource to check | |
673 | * | |
674 | * @return the image URL if found, or NULL | |
675 | * | |
676 | */ | |
677 | static URL getImageUrl(BasicSupport_Deprecated support, URL source, | |
678 | String line) { | |
679 | URL url = null; | |
680 | ||
681 | if (line != null) { | |
682 | // try for files | |
683 | if (source != null) { | |
684 | try { | |
685 | ||
686 | String relPath = null; | |
687 | String absPath = null; | |
688 | try { | |
689 | String path = new File(source.getFile()).getParent(); | |
690 | relPath = new File(new File(path), line.trim()) | |
691 | .getAbsolutePath(); | |
692 | } catch (Exception e) { | |
693 | // Cannot be converted to path (one possibility to take | |
694 | // into account: absolute path on Windows) | |
695 | } | |
696 | try { | |
697 | absPath = new File(line.trim()).getAbsolutePath(); | |
698 | } catch (Exception e) { | |
699 | // Cannot be converted to path (at all) | |
700 | } | |
701 | ||
702 | for (String ext : getImageExt(true)) { | |
703 | File absFile = new File(absPath + ext); | |
704 | File relFile = new File(relPath + ext); | |
705 | if (absPath != null && absFile.exists() | |
706 | && absFile.isFile()) { | |
707 | url = absFile.toURI().toURL(); | |
708 | } else if (relPath != null && relFile.exists() | |
709 | && relFile.isFile()) { | |
710 | url = relFile.toURI().toURL(); | |
711 | } | |
712 | } | |
713 | } catch (Exception e) { | |
714 | // Should not happen since we control the correct arguments | |
715 | } | |
716 | } | |
717 | ||
718 | if (url == null) { | |
719 | // try for URLs | |
720 | try { | |
721 | for (String ext : getImageExt(true)) { | |
722 | if (Instance.getCache() | |
723 | .check(new URL(line + ext), true)) { | |
724 | url = new URL(line + ext); | |
725 | break; | |
726 | } | |
727 | } | |
728 | ||
729 | // try out of cache | |
730 | if (url == null) { | |
731 | for (String ext : getImageExt(true)) { | |
732 | try { | |
733 | url = new URL(line + ext); | |
734 | Instance.getCache().refresh(url, support, true); | |
735 | break; | |
736 | } catch (IOException e) { | |
737 | // no image with this ext | |
738 | url = null; | |
739 | } | |
740 | } | |
741 | } | |
742 | } catch (MalformedURLException e) { | |
743 | // Not an url | |
744 | } | |
745 | } | |
746 | ||
747 | // refresh the cached file | |
748 | if (url != null) { | |
749 | try { | |
750 | Instance.getCache().refresh(url, support, true); | |
751 | } catch (IOException e) { | |
752 | // woops, broken image | |
753 | url = null; | |
754 | } | |
755 | } | |
756 | } | |
757 | ||
758 | return url; | |
759 | } | |
760 | ||
761 | /** | |
762 | * Open the input file that will be used through the support. | |
763 | * <p> | |
764 | * Can return NULL, in which case you are supposed to work without an | |
765 | * {@link InputStream}. | |
766 | * | |
767 | * @param source | |
768 | * the source {@link URL} | |
769 | * | |
770 | * @return the {@link InputStream} | |
771 | * | |
772 | * @throws IOException | |
773 | * in case of I/O error | |
774 | */ | |
775 | protected InputStream openInput(URL source) throws IOException { | |
776 | return Instance.getCache().open(source, this, false); | |
777 | } | |
778 | ||
779 | /** | |
780 | * Reset then return {@link BasicSupport_Deprecated#in}. | |
781 | * | |
782 | * @return {@link BasicSupport_Deprecated#in} | |
783 | */ | |
784 | protected InputStream getInput() { | |
785 | return reset(in); | |
786 | } | |
787 | ||
788 | /** | |
789 | * Check quotes for bad format (i.e., quotes with normal paragraphs inside) | |
790 | * and requotify them (i.e., separate them into QUOTE paragraphs and other | |
791 | * paragraphs (quotes or not)). | |
792 | * | |
793 | * @param para | |
794 | * the paragraph to requotify (not necessarily a quote) | |
795 | * | |
796 | * @return the correctly (or so we hope) quotified paragraphs | |
797 | */ | |
798 | protected List<Paragraph> requotify(Paragraph para) { | |
799 | List<Paragraph> newParas = new ArrayList<Paragraph>(); | |
800 | ||
801 | if (para.getType() == ParagraphType.QUOTE | |
802 | && para.getContent().length() > 2) { | |
803 | String line = para.getContent(); | |
804 | boolean singleQ = line.startsWith("" + openQuote); | |
805 | boolean doubleQ = line.startsWith("" + openDoubleQuote); | |
806 | ||
807 | // Do not try when more than one quote at a time | |
808 | // (some stories are not easily readable if we do) | |
809 | if (singleQ | |
810 | && line.indexOf(closeQuote, 1) < line | |
811 | .lastIndexOf(closeQuote)) { | |
812 | newParas.add(para); | |
813 | return newParas; | |
814 | } | |
815 | if (doubleQ | |
816 | && line.indexOf(closeDoubleQuote, 1) < line | |
817 | .lastIndexOf(closeDoubleQuote)) { | |
818 | newParas.add(para); | |
819 | return newParas; | |
820 | } | |
821 | // | |
822 | ||
823 | if (!singleQ && !doubleQ) { | |
824 | line = openDoubleQuote + line + closeDoubleQuote; | |
825 | newParas.add(new Paragraph(ParagraphType.QUOTE, line, para | |
826 | .getWords())); | |
827 | } else { | |
828 | char open = singleQ ? openQuote : openDoubleQuote; | |
829 | char close = singleQ ? closeQuote : closeDoubleQuote; | |
830 | ||
831 | int posDot = -1; | |
832 | boolean inQuote = false; | |
833 | int i = 0; | |
834 | for (char car : line.toCharArray()) { | |
835 | if (car == open) { | |
836 | inQuote = true; | |
837 | } else if (car == close) { | |
838 | inQuote = false; | |
839 | } else if (car == '.' && !inQuote) { | |
840 | posDot = i; | |
841 | break; | |
842 | } | |
843 | i++; | |
844 | } | |
845 | ||
846 | if (posDot >= 0) { | |
847 | String rest = line.substring(posDot + 1).trim(); | |
848 | line = line.substring(0, posDot + 1).trim(); | |
849 | long words = 1; | |
850 | for (char car : line.toCharArray()) { | |
851 | if (car == ' ') { | |
852 | words++; | |
853 | } | |
854 | } | |
855 | newParas.add(new Paragraph(ParagraphType.QUOTE, line, words)); | |
856 | if (!rest.isEmpty()) { | |
857 | newParas.addAll(requotify(processPara(rest))); | |
858 | } | |
859 | } else { | |
860 | newParas.add(para); | |
861 | } | |
862 | } | |
863 | } else { | |
864 | newParas.add(para); | |
865 | } | |
866 | ||
867 | return newParas; | |
868 | } | |
869 | ||
870 | /** | |
871 | * Process a {@link Paragraph} from a raw line of text. | |
872 | * <p> | |
873 | * Will also fix quotes and HTML encoding if needed. | |
874 | * | |
875 | * @param line | |
876 | * the raw line | |
877 | * | |
878 | * @return the processed {@link Paragraph} | |
879 | */ | |
880 | protected Paragraph processPara(String line) { | |
881 | line = ifUnhtml(line).trim(); | |
882 | ||
883 | boolean space = true; | |
884 | boolean brk = true; | |
885 | boolean quote = false; | |
886 | boolean tentativeCloseQuote = false; | |
887 | char prev = '\0'; | |
888 | int dashCount = 0; | |
889 | long words = 1; | |
890 | ||
891 | StringBuilder builder = new StringBuilder(); | |
892 | for (char car : line.toCharArray()) { | |
893 | if (car != '-') { | |
894 | if (dashCount > 0) { | |
895 | // dash, ndash and mdash: - – — | |
896 | // currently: always use mdash | |
897 | builder.append(dashCount == 1 ? '-' : '—'); | |
898 | } | |
899 | dashCount = 0; | |
900 | } | |
901 | ||
902 | if (tentativeCloseQuote) { | |
903 | tentativeCloseQuote = false; | |
904 | if (Character.isLetterOrDigit(car)) { | |
905 | builder.append("'"); | |
906 | } else { | |
907 | // handle double-single quotes as double quotes | |
908 | if (prev == car) { | |
909 | builder.append(closeDoubleQuote); | |
910 | continue; | |
911 | } | |
912 | ||
913 | builder.append(closeQuote); | |
914 | } | |
915 | } | |
916 | ||
917 | switch (car) { | |
918 | case ' ': // note: unbreakable space | |
919 | case ' ': | |
920 | case '\t': | |
921 | case '\n': // just in case | |
922 | case '\r': // just in case | |
923 | if (builder.length() > 0 | |
924 | && builder.charAt(builder.length() - 1) != ' ') { | |
925 | words++; | |
926 | } | |
927 | builder.append(' '); | |
928 | break; | |
929 | ||
930 | case '\'': | |
931 | if (space || (brk && quote)) { | |
932 | quote = true; | |
933 | // handle double-single quotes as double quotes | |
934 | if (prev == car) { | |
935 | builder.deleteCharAt(builder.length() - 1); | |
936 | builder.append(openDoubleQuote); | |
937 | } else { | |
938 | builder.append(openQuote); | |
939 | } | |
940 | } else if (prev == ' ' || prev == car) { | |
941 | // handle double-single quotes as double quotes | |
942 | if (prev == car) { | |
943 | builder.deleteCharAt(builder.length() - 1); | |
944 | builder.append(openDoubleQuote); | |
945 | } else { | |
946 | builder.append(openQuote); | |
947 | } | |
948 | } else { | |
949 | // it is a quote ("I'm off") or a 'quote' ("This | |
950 | // 'good' restaurant"...) | |
951 | tentativeCloseQuote = true; | |
952 | } | |
953 | break; | |
954 | ||
955 | case '"': | |
956 | if (space || (brk && quote)) { | |
957 | quote = true; | |
958 | builder.append(openDoubleQuote); | |
959 | } else if (prev == ' ') { | |
960 | builder.append(openDoubleQuote); | |
961 | } else { | |
962 | builder.append(closeDoubleQuote); | |
963 | } | |
964 | break; | |
965 | ||
966 | case '-': | |
967 | if (space) { | |
968 | quote = true; | |
969 | } else { | |
970 | dashCount++; | |
971 | } | |
972 | space = false; | |
973 | break; | |
974 | ||
975 | case '*': | |
976 | case '~': | |
977 | case '/': | |
978 | case '\\': | |
979 | case '<': | |
980 | case '>': | |
981 | case '=': | |
982 | case '+': | |
983 | case '_': | |
984 | case '–': | |
985 | case '—': | |
986 | space = false; | |
987 | builder.append(car); | |
988 | break; | |
989 | ||
990 | case '‘': | |
991 | case '`': | |
992 | case '‹': | |
993 | case '﹁': | |
994 | case '〈': | |
995 | case '「': | |
996 | if (space || (brk && quote)) { | |
997 | quote = true; | |
998 | builder.append(openQuote); | |
999 | } else { | |
1000 | // handle double-single quotes as double quotes | |
1001 | if (prev == car) { | |
1002 | builder.deleteCharAt(builder.length() - 1); | |
1003 | builder.append(openDoubleQuote); | |
1004 | } else { | |
1005 | builder.append(openQuote); | |
1006 | } | |
1007 | } | |
1008 | space = false; | |
1009 | brk = false; | |
1010 | break; | |
1011 | ||
1012 | case '’': | |
1013 | case '›': | |
1014 | case '﹂': | |
1015 | case '〉': | |
1016 | case '」': | |
1017 | space = false; | |
1018 | brk = false; | |
1019 | // handle double-single quotes as double quotes | |
1020 | if (prev == car) { | |
1021 | builder.deleteCharAt(builder.length() - 1); | |
1022 | builder.append(closeDoubleQuote); | |
1023 | } else { | |
1024 | builder.append(closeQuote); | |
1025 | } | |
1026 | break; | |
1027 | ||
1028 | case '«': | |
1029 | case '“': | |
1030 | case '﹃': | |
1031 | case '《': | |
1032 | case '『': | |
1033 | if (space || (brk && quote)) { | |
1034 | quote = true; | |
1035 | builder.append(openDoubleQuote); | |
1036 | } else { | |
1037 | builder.append(openDoubleQuote); | |
1038 | } | |
1039 | space = false; | |
1040 | brk = false; | |
1041 | break; | |
1042 | ||
1043 | case '»': | |
1044 | case '”': | |
1045 | case '﹄': | |
1046 | case '》': | |
1047 | case '』': | |
1048 | space = false; | |
1049 | brk = false; | |
1050 | builder.append(closeDoubleQuote); | |
1051 | break; | |
1052 | ||
1053 | default: | |
1054 | space = false; | |
1055 | brk = false; | |
1056 | builder.append(car); | |
1057 | break; | |
1058 | } | |
1059 | ||
1060 | prev = car; | |
1061 | } | |
1062 | ||
1063 | if (tentativeCloseQuote) { | |
1064 | tentativeCloseQuote = false; | |
1065 | builder.append(closeQuote); | |
1066 | } | |
1067 | ||
1068 | line = builder.toString().trim(); | |
1069 | ||
1070 | ParagraphType type = ParagraphType.NORMAL; | |
1071 | if (space) { | |
1072 | type = ParagraphType.BLANK; | |
1073 | } else if (brk) { | |
1074 | type = ParagraphType.BREAK; | |
1075 | } else if (quote) { | |
1076 | type = ParagraphType.QUOTE; | |
1077 | } | |
1078 | ||
1079 | return new Paragraph(type, line, words); | |
1080 | } | |
1081 | ||
1082 | /** | |
1083 | * Remove the HTML from the input <b>if</b> | |
1084 | * {@link BasicSupport_Deprecated#isHtml()} is true. | |
1085 | * | |
1086 | * @param input | |
1087 | * the input | |
1088 | * | |
1089 | * @return the no html version if needed | |
1090 | */ | |
1091 | private String ifUnhtml(String input) { | |
1092 | if (isHtml() && input != null) { | |
1093 | return StringUtils.unhtml(input); | |
1094 | } | |
1095 | ||
1096 | return input; | |
1097 | } | |
1098 | ||
1099 | /** | |
1100 | * Reset the given {@link InputStream} and return it. | |
1101 | * | |
1102 | * @param in | |
1103 | * the {@link InputStream} to reset | |
1104 | * | |
1105 | * @return the same {@link InputStream} after reset | |
1106 | */ | |
1107 | static protected InputStream reset(InputStream in) { | |
1108 | try { | |
1109 | if (in != null) { | |
1110 | in.reset(); | |
1111 | } | |
1112 | } catch (IOException e) { | |
1113 | } | |
1114 | ||
1115 | return in; | |
1116 | } | |
1117 | ||
1118 | /** | |
1119 | * Return the first line from the given input which correspond to the given | |
1120 | * selectors. | |
1121 | * | |
1122 | * @param in | |
1123 | * the input | |
1124 | * @param needle | |
1125 | * a string that must be found inside the target line (also | |
1126 | * supports "^" at start to say "only if it starts with" the | |
1127 | * needle) | |
1128 | * @param relativeLine | |
1129 | * the line to return based upon the target line position (-1 = | |
1130 | * the line before, 0 = the target line...) | |
1131 | * | |
1132 | * @return the line | |
1133 | */ | |
1134 | static protected String getLine(InputStream in, String needle, | |
1135 | int relativeLine) { | |
1136 | return getLine(in, needle, relativeLine, true); | |
1137 | } | |
1138 | ||
1139 | /** | |
1140 | * Return a line from the given input which correspond to the given | |
1141 | * selectors. | |
1142 | * | |
1143 | * @param in | |
1144 | * the input | |
1145 | * @param needle | |
1146 | * a string that must be found inside the target line (also | |
1147 | * supports "^" at start to say "only if it starts with" the | |
1148 | * needle) | |
1149 | * @param relativeLine | |
1150 | * the line to return based upon the target line position (-1 = | |
1151 | * the line before, 0 = the target line...) | |
1152 | * @param first | |
1153 | * takes the first result (as opposed to the last one, which will | |
1154 | * also always spend the input) | |
1155 | * | |
1156 | * @return the line | |
1157 | */ | |
1158 | static protected String getLine(InputStream in, String needle, | |
1159 | int relativeLine, boolean first) { | |
1160 | String rep = null; | |
1161 | ||
1162 | reset(in); | |
1163 | ||
1164 | List<String> lines = new ArrayList<String>(); | |
1165 | @SuppressWarnings("resource") | |
1166 | Scanner scan = new Scanner(in, "UTF-8"); | |
1167 | int index = -1; | |
1168 | scan.useDelimiter("\\n"); | |
1169 | while (scan.hasNext()) { | |
1170 | lines.add(scan.next()); | |
1171 | ||
1172 | if (index == -1) { | |
1173 | if (needle.startsWith("^")) { | |
1174 | if (lines.get(lines.size() - 1).startsWith( | |
1175 | needle.substring(1))) { | |
1176 | index = lines.size() - 1; | |
1177 | } | |
1178 | ||
1179 | } else { | |
1180 | if (lines.get(lines.size() - 1).contains(needle)) { | |
1181 | index = lines.size() - 1; | |
1182 | } | |
1183 | } | |
1184 | } | |
1185 | ||
1186 | if (index >= 0 && index + relativeLine < lines.size()) { | |
1187 | rep = lines.get(index + relativeLine); | |
1188 | if (first) { | |
1189 | break; | |
1190 | } | |
1191 | } | |
1192 | } | |
1193 | ||
1194 | return rep; | |
1195 | } | |
1196 | ||
1197 | /** | |
1198 | * Return the text between the key and the endKey (and optional subKey can | |
1199 | * be passed, in this case we will look for the key first, then take the | |
1200 | * text between the subKey and the endKey). | |
1201 | * <p> | |
1202 | * Will only match the first line with the given key if more than one are | |
1203 | * possible. Which also means that if the subKey or endKey is not found on | |
1204 | * that line, NULL will be returned. | |
1205 | * | |
1206 | * @param in | |
1207 | * the input | |
1208 | * @param key | |
1209 | * the key to match (also supports "^" at start to say | |
1210 | * "only if it starts with" the key) | |
1211 | * @param subKey | |
1212 | * the sub key or NULL if none | |
1213 | * @param endKey | |
1214 | * the end key or NULL for "up to the end" | |
1215 | * @return the text or NULL if not found | |
1216 | */ | |
1217 | static protected String getKeyLine(InputStream in, String key, | |
1218 | String subKey, String endKey) { | |
1219 | return getKeyText(getLine(in, key, 0), key, subKey, endKey); | |
1220 | } | |
1221 | ||
1222 | /** | |
1223 | * Return the text between the key and the endKey (and optional subKey can | |
1224 | * be passed, in this case we will look for the key first, then take the | |
1225 | * text between the subKey and the endKey). | |
1226 | * | |
1227 | * @param in | |
1228 | * the input | |
1229 | * @param key | |
1230 | * the key to match (also supports "^" at start to say | |
1231 | * "only if it starts with" the key) | |
1232 | * @param subKey | |
1233 | * the sub key or NULL if none | |
1234 | * @param endKey | |
1235 | * the end key or NULL for "up to the end" | |
1236 | * @return the text or NULL if not found | |
1237 | */ | |
1238 | static protected String getKeyText(String in, String key, String subKey, | |
1239 | String endKey) { | |
1240 | String result = null; | |
1241 | ||
1242 | String line = in; | |
1243 | if (line != null && line.contains(key)) { | |
1244 | line = line.substring(line.indexOf(key) + key.length()); | |
1245 | if (subKey == null || subKey.isEmpty() || line.contains(subKey)) { | |
1246 | if (subKey != null) { | |
1247 | line = line.substring(line.indexOf(subKey) | |
1248 | + subKey.length()); | |
1249 | } | |
1250 | if (endKey == null || line.contains(endKey)) { | |
1251 | if (endKey != null) { | |
1252 | line = line.substring(0, line.indexOf(endKey)); | |
1253 | result = line; | |
1254 | } | |
1255 | } | |
1256 | } | |
1257 | } | |
1258 | ||
1259 | return result; | |
1260 | } | |
1261 | ||
1262 | /** | |
1263 | * Return the text between the key and the endKey (optional subKeys can be | |
1264 | * passed, in this case we will look for the subKeys first, then take the | |
1265 | * text between the key and the endKey). | |
1266 | * | |
1267 | * @param in | |
1268 | * the input | |
1269 | * @param key | |
1270 | * the key to match | |
1271 | * @param endKey | |
1272 | * the end key or NULL for "up to the end" | |
1273 | * @param afters | |
1274 | * the sub-keys to find before checking for key/endKey | |
1275 | * | |
1276 | * @return the text or NULL if not found | |
1277 | */ | |
1278 | static protected String getKeyTextAfter(String in, String key, | |
1279 | String endKey, String... afters) { | |
1280 | ||
1281 | if (in != null && !in.isEmpty()) { | |
1282 | int pos = indexOfAfter(in, 0, afters); | |
1283 | if (pos < 0) { | |
1284 | return null; | |
1285 | } | |
1286 | ||
1287 | in = in.substring(pos); | |
1288 | } | |
1289 | ||
1290 | return getKeyText(in, key, null, endKey); | |
1291 | } | |
1292 | ||
1293 | /** | |
1294 | * Return the first index after all the given "afters" have been found in | |
1295 | * the {@link String}, or -1 if it was not possible. | |
1296 | * | |
1297 | * @param in | |
1298 | * the input | |
1299 | * @param startAt | |
1300 | * start at this position in the string | |
1301 | * @param afters | |
1302 | * the sub-keys to find before checking for key/endKey | |
1303 | * | |
1304 | * @return the text or NULL if not found | |
1305 | */ | |
1306 | static protected int indexOfAfter(String in, int startAt, String... afters) { | |
1307 | int pos = -1; | |
1308 | if (in != null && !in.isEmpty()) { | |
1309 | pos = startAt; | |
1310 | if (afters != null) { | |
1311 | for (int i = 0; pos >= 0 && i < afters.length; i++) { | |
1312 | String subKey = afters[i]; | |
1313 | if (!subKey.isEmpty()) { | |
1314 | pos = in.indexOf(subKey, pos); | |
1315 | if (pos >= 0) { | |
1316 | pos += subKey.length(); | |
1317 | } | |
1318 | } | |
1319 | } | |
1320 | } | |
1321 | } | |
1322 | ||
1323 | return pos; | |
1324 | } | |
1325 | } |