471147ea0548ede456aa66e01a6337c9ae4ef1ef
[fanfix.git] / src / be / nikiroo / fanfix / supported / BasicSupport.java
1 package be.nikiroo.fanfix.supported;
2
3 import java.awt.image.BufferedImage;
4 import java.io.BufferedReader;
5 import java.io.ByteArrayInputStream;
6 import java.io.File;
7 import java.io.IOException;
8 import java.io.InputStream;
9 import java.io.InputStreamReader;
10 import java.net.MalformedURLException;
11 import java.net.URL;
12 import java.util.ArrayList;
13 import java.util.Date;
14 import java.util.HashMap;
15 import java.util.List;
16 import java.util.Map;
17 import java.util.Map.Entry;
18 import java.util.Scanner;
19
20 import be.nikiroo.fanfix.Instance;
21 import be.nikiroo.fanfix.bundles.Config;
22 import be.nikiroo.fanfix.bundles.StringId;
23 import be.nikiroo.fanfix.data.Chapter;
24 import be.nikiroo.fanfix.data.MetaData;
25 import be.nikiroo.fanfix.data.Paragraph;
26 import be.nikiroo.fanfix.data.Paragraph.ParagraphType;
27 import be.nikiroo.fanfix.data.Story;
28 import be.nikiroo.utils.IOUtils;
29 import be.nikiroo.utils.Progress;
30 import be.nikiroo.utils.StringUtils;
31
32 /**
33 * This class is the base class used by the other support classes. It can be
34 * used outside of this package, and have static method that you can use to get
35 * access to the correct support class.
36 * <p>
37 * It will be used with 'resources' (usually web pages or files).
38 *
39 * @author niki
40 */
41 public abstract class BasicSupport {
42 /**
43 * The supported input types for which we can get a {@link BasicSupport}
44 * object.
45 *
46 * @author niki
47 */
48 public enum SupportType {
49 /** EPUB files created with this program */
50 EPUB,
51 /** Pure text file with some rules */
52 TEXT,
53 /** TEXT but with associated .info file */
54 INFO_TEXT,
55 /** My Little Pony fanfictions */
56 FIMFICTION,
57 /** Fanfictions from a lot of different universes */
58 FANFICTION,
59 /** Website with lots of Mangas */
60 MANGAFOX,
61 /** Furry website with comics support */
62 E621,
63 /** Furry website with stories */
64 YIFFSTAR,
65 /** Comics and images groups, mostly but not only NSFW */
66 E_HENTAI,
67 /** CBZ files */
68 CBZ,
69 /** HTML files */
70 HTML;
71
72 /**
73 * A description of this support type (more information than the
74 * {@link BasicSupport#getSourceName()}).
75 *
76 * @return the description
77 */
78 public String getDesc() {
79 String desc = Instance.getTrans().getStringX(StringId.INPUT_DESC,
80 this.name());
81
82 if (desc == null) {
83 desc = Instance.getTrans().getString(StringId.INPUT_DESC, this);
84 }
85
86 return desc;
87 }
88
89 /**
90 * The name of this support type (a short version).
91 *
92 * @return the name
93 */
94 public String getSourceName() {
95 BasicSupport support = BasicSupport.getSupport(this);
96 if (support != null) {
97 return support.getSourceName();
98 }
99
100 return null;
101 }
102
103 @Override
104 public String toString() {
105 return super.toString().toLowerCase();
106 }
107
108 /**
109 * Call {@link SupportType#valueOf(String.toUpperCase())}.
110 *
111 * @param typeName
112 * the possible type name
113 *
114 * @return NULL or the type
115 */
116 public static SupportType valueOfUC(String typeName) {
117 return SupportType.valueOf(typeName == null ? null : typeName
118 .toUpperCase());
119 }
120
121 /**
122 * Call {@link SupportType#valueOf(String.toUpperCase())} but return
123 * NULL for NULL instead of raising exception.
124 *
125 * @param typeName
126 * the possible type name
127 *
128 * @return NULL or the type
129 */
130 public static SupportType valueOfNullOkUC(String typeName) {
131 if (typeName == null) {
132 return null;
133 }
134
135 return SupportType.valueOfUC(typeName);
136 }
137
138 /**
139 * Call {@link SupportType#valueOf(String.toUpperCase())} but return
140 * NULL in case of error instead of raising an exception.
141 *
142 * @param typeName
143 * the possible type name
144 *
145 * @return NULL or the type
146 */
147 public static SupportType valueOfAllOkUC(String typeName) {
148 try {
149 return SupportType.valueOfUC(typeName);
150 } catch (Exception e) {
151 return null;
152 }
153 }
154 }
155
156 private InputStream in;
157 private SupportType type;
158 private URL currentReferer; // with only one 'r', as in 'HTTP'...
159
160 // quote chars
161 private char openQuote = Instance.getTrans().getCharacter(
162 StringId.OPEN_SINGLE_QUOTE);
163 private char closeQuote = Instance.getTrans().getCharacter(
164 StringId.CLOSE_SINGLE_QUOTE);
165 private char openDoubleQuote = Instance.getTrans().getCharacter(
166 StringId.OPEN_DOUBLE_QUOTE);
167 private char closeDoubleQuote = Instance.getTrans().getCharacter(
168 StringId.CLOSE_DOUBLE_QUOTE);
169
170 /**
171 * The name of this support class.
172 *
173 * @return the name
174 */
175 protected abstract String getSourceName();
176
177 /**
178 * Check if the given resource is supported by this {@link BasicSupport}.
179 *
180 * @param url
181 * the resource to check for
182 *
183 * @return TRUE if it is
184 */
185 protected abstract boolean supports(URL url);
186
187 /**
188 * Return TRUE if the support will return HTML encoded content values for
189 * the chapters content.
190 *
191 * @return TRUE for HTML
192 */
193 protected abstract boolean isHtml();
194
195 protected abstract MetaData getMeta(URL source, InputStream in)
196 throws IOException;
197
198 /**
199 * Return the story description.
200 *
201 * @param source
202 * the source of the story
203 * @param in
204 * the input (the main resource)
205 *
206 * @return the description
207 *
208 * @throws IOException
209 * in case of I/O error
210 */
211 protected abstract String getDesc(URL source, InputStream in)
212 throws IOException;
213
214 /**
215 * Return the list of chapters (name and resource).
216 *
217 * @param source
218 * the source of the story
219 * @param in
220 * the input (the main resource)
221 * @param pg
222 * the optional progress reporter
223 *
224 * @return the chapters
225 *
226 * @throws IOException
227 * in case of I/O error
228 */
229 protected abstract List<Entry<String, URL>> getChapters(URL source,
230 InputStream in, Progress pg) throws IOException;
231
232 /**
233 * Return the content of the chapter (possibly HTML encoded, if
234 * {@link BasicSupport#isHtml()} is TRUE).
235 *
236 * @param source
237 * the source of the story
238 * @param in
239 * the input (the main resource)
240 * @param number
241 * the chapter number
242 * @param pg
243 * the optional progress reporter
244 *
245 * @return the content
246 *
247 * @throws IOException
248 * in case of I/O error
249 */
250 protected abstract String getChapterContent(URL source, InputStream in,
251 int number, Progress pg) throws IOException;
252
253 /**
254 * Log into the support (can be a no-op depending upon the support).
255 *
256 * @throws IOException
257 * in case of I/O error
258 */
259 public void login() throws IOException {
260
261 }
262
263 /**
264 * Return the list of cookies (values included) that must be used to
265 * correctly fetch the resources.
266 * <p>
267 * You are expected to call the super method implementation if you override
268 * it.
269 *
270 * @return the cookies
271 *
272 * @throws IOException
273 * in case of I/O error
274 */
275 public Map<String, String> getCookies() throws IOException {
276 return new HashMap<String, String>();
277 }
278
279 /**
280 * Return the canonical form of the main {@link URL}.
281 *
282 * @param source
283 * the source {@link URL}
284 *
285 * @return the canonical form of this {@link URL}
286 *
287 * @throws IOException
288 * in case of I/O error
289 */
290 public URL getCanonicalUrl(URL source) throws IOException {
291 return source;
292 }
293
294 /**
295 * Process the given story resource into a partially filled {@link Story}
296 * object containing the name and metadata, except for the description.
297 *
298 * @param url
299 * the story resource
300 *
301 * @return the {@link Story}
302 *
303 * @throws IOException
304 * in case of I/O error
305 */
306 public Story processMeta(URL url) throws IOException {
307 return processMeta(url, true, false, null);
308 }
309
310 /**
311 * Process the given story resource into a partially filled {@link Story}
312 * object containing the name and metadata.
313 *
314 * @param url
315 * the story resource
316 *
317 * @param close
318 * close "this" and "in" when done
319 * @param pg
320 * the optional progress reporter
321 *
322 * @return the {@link Story}
323 *
324 * @throws IOException
325 * in case of I/O error
326 */
327 protected Story processMeta(URL url, boolean close, boolean getDesc,
328 Progress pg) throws IOException {
329 if (pg == null) {
330 pg = new Progress();
331 } else {
332 pg.setMinMax(0, 100);
333 }
334
335 login();
336 pg.setProgress(10);
337
338 url = getCanonicalUrl(url);
339
340 setCurrentReferer(url);
341
342 in = openInput(url);
343 if (in == null) {
344 return null;
345 }
346
347 try {
348 preprocess(url, getInput());
349 pg.setProgress(30);
350
351 Story story = new Story();
352 MetaData meta = getMeta(url, getInput());
353 if (meta.getCreationDate() == null
354 || meta.getCreationDate().isEmpty()) {
355 meta.setCreationDate(StringUtils.fromTime(new Date().getTime()));
356 }
357 story.setMeta(meta);
358
359 pg.setProgress(50);
360
361 if (meta != null && meta.getCover() == null) {
362 meta.setCover(getDefaultCover(meta.getSubject()));
363 }
364
365 pg.setProgress(60);
366
367 if (getDesc) {
368 String descChapterName = Instance.getTrans().getString(
369 StringId.DESCRIPTION);
370 story.getMeta().setResume(
371 makeChapter(url, 0, descChapterName,
372 getDesc(url, getInput()), null));
373 }
374
375 pg.setProgress(100);
376 return story;
377 } finally {
378 if (close) {
379 try {
380 close();
381 } catch (IOException e) {
382 Instance.syserr(e);
383 }
384
385 if (in != null) {
386 in.close();
387 }
388 }
389
390 setCurrentReferer(null);
391 }
392 }
393
394 /**
395 * Process the given story resource into a fully filled {@link Story}
396 * object.
397 *
398 * @param url
399 * the story resource
400 * @param pg
401 * the optional progress reporter
402 *
403 * @return the {@link Story}
404 *
405 * @throws IOException
406 * in case of I/O error
407 */
408 public Story process(URL url, Progress pg) throws IOException {
409 if (pg == null) {
410 pg = new Progress();
411 } else {
412 pg.setMinMax(0, 100);
413 }
414
415 url = getCanonicalUrl(url);
416 pg.setProgress(1);
417 try {
418 Progress pgMeta = new Progress();
419 pg.addProgress(pgMeta, 10);
420 Story story = processMeta(url, false, true, pgMeta);
421 if (!pgMeta.isDone()) {
422 pgMeta.setProgress(pgMeta.getMax()); // 10%
423 }
424
425 if (story == null) {
426 pg.setProgress(90);
427 return null;
428 }
429
430 pg.setName("Retrieving " + story.getMeta().getTitle());
431
432 setCurrentReferer(url);
433
434 Progress pgGetChapters = new Progress();
435 pg.addProgress(pgGetChapters, 10);
436 story.setChapters(new ArrayList<Chapter>());
437 List<Entry<String, URL>> chapters = getChapters(url, getInput(),
438 pgGetChapters);
439 if (!pgGetChapters.isDone()) {
440 pgGetChapters.setProgress(pgGetChapters.getMax()); // 20%
441 }
442
443 if (chapters != null) {
444 Progress pgChaps = new Progress("Extracting chapters", 0,
445 chapters.size() * 300);
446 pg.addProgress(pgChaps, 80);
447
448 long words = 0;
449 int i = 1;
450 for (Entry<String, URL> chap : chapters) {
451 pgChaps.setName("Extracting chapter " + i);
452 setCurrentReferer(chap.getValue());
453 InputStream chapIn = Instance.getCache().open(
454 chap.getValue(), this, true);
455 pgChaps.setProgress(i * 100);
456 try {
457 Progress pgGetChapterContent = new Progress();
458 Progress pgMakeChapter = new Progress();
459 pgChaps.addProgress(pgGetChapterContent, 100);
460 pgChaps.addProgress(pgMakeChapter, 100);
461
462 String content = getChapterContent(url, chapIn, i,
463 pgGetChapterContent);
464 if (!pgGetChapterContent.isDone()) {
465 pgGetChapterContent.setProgress(pgGetChapterContent
466 .getMax());
467 }
468
469 Chapter cc = makeChapter(url, i, chap.getKey(),
470 content, pgMakeChapter);
471 if (!pgMakeChapter.isDone()) {
472 pgMakeChapter.setProgress(pgMakeChapter.getMax());
473 }
474
475 words += cc.getWords();
476 story.getChapters().add(cc);
477 if (story.getMeta() != null) {
478 story.getMeta().setWords(words);
479 }
480 } finally {
481 chapIn.close();
482 }
483
484 i++;
485 }
486
487 pgChaps.setName("Extracting chapters");
488 } else {
489 pg.setProgress(80);
490 }
491
492 return story;
493
494 } finally {
495 try {
496 close();
497 } catch (IOException e) {
498 Instance.syserr(e);
499 }
500
501 if (in != null) {
502 in.close();
503 }
504
505 setCurrentReferer(null);
506 }
507 }
508
509 /**
510 * The support type.
511 *
512 * @return the type
513 */
514 public SupportType getType() {
515 return type;
516 }
517
518 /**
519 * The current referer {@link URL} (only one 'r', as in 'HTML'...), i.e.,
520 * the current {@link URL} we work on.
521 *
522 * @return the referer
523 */
524 public URL getCurrentReferer() {
525 return currentReferer;
526 }
527
528 /**
529 * The current referer {@link URL} (only one 'r', as in 'HTML'...), i.e.,
530 * the current {@link URL} we work on.
531 *
532 * @param currentReferer
533 * the new referer
534 */
535 protected void setCurrentReferer(URL currentReferer) {
536 this.currentReferer = currentReferer;
537 }
538
539 /**
540 * The support type.
541 *
542 * @param type
543 * the new type
544 *
545 * @return this
546 */
547 protected BasicSupport setType(SupportType type) {
548 this.type = type;
549 return this;
550 }
551
552 /**
553 * Prepare the support if needed before processing.
554 *
555 * @param source
556 * the source of the story
557 * @param in
558 * the input (the main resource)
559 *
560 * @throws IOException
561 * on I/O error
562 */
563 protected void preprocess(URL source, InputStream in) throws IOException {
564 }
565
566 /**
567 * Now that we have processed the {@link Story}, close the resources if any.
568 *
569 * @throws IOException
570 * on I/O error
571 */
572 protected void close() throws IOException {
573 }
574
575 /**
576 * Create a {@link Chapter} object from the given information, formatting
577 * the content as it should be.
578 *
579 * @param number
580 * the chapter number
581 * @param name
582 * the chapter name
583 * @param content
584 * the chapter content
585 * @param pg
586 * the optional progress reporter
587 *
588 * @return the {@link Chapter}
589 *
590 * @throws IOException
591 * in case of I/O error
592 */
593 protected Chapter makeChapter(URL source, int number, String name,
594 String content, Progress pg) throws IOException {
595 // Chapter name: process it correctly, then remove the possible
596 // redundant "Chapter x: " in front of it
597 String chapterName = processPara(name).getContent().trim();
598 for (String lang : Instance.getConfig().getString(Config.CHAPTER)
599 .split(",")) {
600 String chapterWord = Instance.getConfig().getStringX(
601 Config.CHAPTER, lang);
602 if (chapterName.startsWith(chapterWord)) {
603 chapterName = chapterName.substring(chapterWord.length())
604 .trim();
605 break;
606 }
607 }
608
609 if (chapterName.startsWith(Integer.toString(number))) {
610 chapterName = chapterName.substring(
611 Integer.toString(number).length()).trim();
612 }
613
614 if (chapterName.startsWith(":")) {
615 chapterName = chapterName.substring(1).trim();
616 }
617 //
618
619 Chapter chap = new Chapter(number, chapterName);
620
621 if (content != null) {
622 List<Paragraph> paras = makeParagraphs(source, content, pg);
623 long words = 0;
624 for (Paragraph para : paras) {
625 words += para.getWords();
626 }
627 chap.setParagraphs(paras);
628 chap.setWords(words);
629 }
630
631 return chap;
632
633 }
634
635 /**
636 * Convert the given content into {@link Paragraph}s.
637 *
638 * @param source
639 * the source URL of the story
640 * @param content
641 * the textual content
642 * @param pg
643 * the optional progress reporter
644 *
645 * @return the {@link Paragraph}s
646 *
647 * @throws IOException
648 * in case of I/O error
649 */
650 protected List<Paragraph> makeParagraphs(URL source, String content,
651 Progress pg) throws IOException {
652 if (pg == null) {
653 pg = new Progress();
654 }
655
656 if (isHtml()) {
657 // Special <HR> processing:
658 content = content.replaceAll("(<hr [^>]*>)|(<hr/>)|(<hr>)",
659 "<br/>* * *<br/>");
660 }
661
662 List<Paragraph> paras = new ArrayList<Paragraph>();
663
664 if (content != null && !content.trim().isEmpty()) {
665 if (isHtml()) {
666 String[] tab = content.split("(<p>|</p>|<br>|<br/>)");
667 pg.setMinMax(0, tab.length);
668 int i = 1;
669 for (String line : tab) {
670 if (line.startsWith("[") && line.endsWith("]")) {
671 pg.setName("Extracting image " + i);
672 }
673 paras.add(makeParagraph(source, line.trim()));
674 pg.setProgress(i++);
675 }
676 pg.setName(null);
677 } else {
678 List<String> lines = new ArrayList<String>();
679 BufferedReader buff = null;
680 try {
681 buff = new BufferedReader(
682 new InputStreamReader(new ByteArrayInputStream(
683 content.getBytes("UTF-8")), "UTF-8"));
684 for (String line = buff.readLine(); line != null; line = buff
685 .readLine()) {
686 lines.add(line.trim());
687 }
688 } finally {
689 if (buff != null) {
690 buff.close();
691 }
692 }
693
694 pg.setMinMax(0, lines.size());
695 int i = 0;
696 for (String line : lines) {
697 if (line.startsWith("[") && line.endsWith("]")) {
698 pg.setName("Extracting image " + i);
699 }
700 paras.add(makeParagraph(source, line));
701 pg.setProgress(i++);
702 }
703 pg.setName(null);
704 }
705
706 // Check quotes for "bad" format
707 List<Paragraph> newParas = new ArrayList<Paragraph>();
708 for (Paragraph para : paras) {
709 newParas.addAll(requotify(para));
710 }
711 paras = newParas;
712
713 // Remove double blanks/brks
714 fixBlanksBreaks(paras);
715 }
716
717 return paras;
718 }
719
720 /**
721 * Convert the given line into a single {@link Paragraph}.
722 *
723 * @param source
724 * the source URL of the story
725 * @param line
726 * the textual content of the paragraph
727 *
728 * @return the {@link Paragraph}
729 */
730 private Paragraph makeParagraph(URL source, String line) {
731 URL image = null;
732 if (line.startsWith("[") && line.endsWith("]")) {
733 image = getImageUrl(this, source,
734 line.substring(1, line.length() - 1).trim());
735 }
736
737 if (image != null) {
738 return new Paragraph(image);
739 } else {
740 return processPara(line);
741 }
742 }
743
744 /**
745 * Fix the {@link ParagraphType#BLANK}s and {@link ParagraphType#BREAK}s of
746 * those {@link Paragraph}s.
747 * <p>
748 * The resulting list will not contain a starting or trailing blank/break
749 * nor 2 blanks or breaks following each other.
750 *
751 * @param paras
752 * the list of {@link Paragraph}s to fix
753 */
754 protected void fixBlanksBreaks(List<Paragraph> paras) {
755 boolean space = false;
756 boolean brk = true;
757 for (int i = 0; i < paras.size(); i++) {
758 Paragraph para = paras.get(i);
759 boolean thisSpace = para.getType() == ParagraphType.BLANK;
760 boolean thisBrk = para.getType() == ParagraphType.BREAK;
761
762 if (i > 0 && space && thisBrk) {
763 paras.remove(i - 1);
764 i--;
765 } else if ((space || brk) && (thisSpace || thisBrk)) {
766 paras.remove(i);
767 i--;
768 }
769
770 space = thisSpace;
771 brk = thisBrk;
772 }
773
774 // Remove blank/brk at start
775 if (paras.size() > 0
776 && (paras.get(0).getType() == ParagraphType.BLANK || paras.get(
777 0).getType() == ParagraphType.BREAK)) {
778 paras.remove(0);
779 }
780
781 // Remove blank/brk at end
782 int last = paras.size() - 1;
783 if (paras.size() > 0
784 && (paras.get(last).getType() == ParagraphType.BLANK || paras
785 .get(last).getType() == ParagraphType.BREAK)) {
786 paras.remove(last);
787 }
788 }
789
790 /**
791 * Get the default cover related to this subject (see <tt>.info</tt> files).
792 *
793 * @param subject
794 * the subject
795 *
796 * @return the cover if any, or NULL
797 */
798 static BufferedImage getDefaultCover(String subject) {
799 if (subject != null && !subject.isEmpty()
800 && Instance.getCoverDir() != null) {
801 try {
802 File fileCover = new File(Instance.getCoverDir(), subject);
803 return getImage(null, fileCover.toURI().toURL(), subject);
804 } catch (MalformedURLException e) {
805 }
806 }
807
808 return null;
809 }
810
811 /**
812 * Return the list of supported image extensions.
813 *
814 * @param emptyAllowed
815 * TRUE to allow an empty extension on first place, which can be
816 * used when you may already have an extension in your input but
817 * are not sure about it
818 *
819 * @return the extensions
820 */
821 static String[] getImageExt(boolean emptyAllowed) {
822 if (emptyAllowed) {
823 return new String[] { "", ".png", ".jpg", ".jpeg", ".gif", ".bmp" };
824 } else {
825 return new String[] { ".png", ".jpg", ".jpeg", ".gif", ".bmp" };
826 }
827 }
828
829 /**
830 * Check if the given resource can be a local image or a remote image, then
831 * refresh the cache with it if it is.
832 *
833 * @param source
834 * the story source
835 * @param line
836 * the resource to check
837 *
838 * @return the image if found, or NULL
839 *
840 */
841 static BufferedImage getImage(BasicSupport support, URL source, String line) {
842 URL url = getImageUrl(support, source, line);
843 if (url != null) {
844 InputStream in = null;
845 try {
846 in = Instance.getCache().open(url, getSupport(url), true);
847 return IOUtils.toImage(in);
848 } catch (IOException e) {
849 } finally {
850 if (in != null) {
851 try {
852 in.close();
853 } catch (IOException e) {
854 }
855 }
856 }
857 }
858
859 return null;
860 }
861
862 /**
863 * Check if the given resource can be a local image or a remote image, then
864 * refresh the cache with it if it is.
865 *
866 * @param source
867 * the story source
868 * @param line
869 * the resource to check
870 *
871 * @return the image URL if found, or NULL
872 *
873 */
874 static URL getImageUrl(BasicSupport support, URL source, String line) {
875 URL url = null;
876
877 if (line != null) {
878 // try for files
879 if (source != null) {
880 try {
881
882 String relPath = null;
883 String absPath = null;
884 try {
885 String path = new File(source.getFile()).getParent();
886 relPath = new File(new File(path), line.trim())
887 .getAbsolutePath();
888 } catch (Exception e) {
889 // Cannot be converted to path (one possibility to take
890 // into account: absolute path on Windows)
891 }
892 try {
893 absPath = new File(line.trim()).getAbsolutePath();
894 } catch (Exception e) {
895 // Cannot be converted to path (at all)
896 }
897
898 for (String ext : getImageExt(true)) {
899 if (absPath != null && new File(absPath + ext).exists()) {
900 url = new File(absPath + ext).toURI().toURL();
901 } else if (relPath != null
902 && new File(relPath + ext).exists()) {
903 url = new File(relPath + ext).toURI().toURL();
904 }
905 }
906 } catch (Exception e) {
907 // Should not happen since we control the correct arguments
908 }
909 }
910
911 if (url == null) {
912 // try for URLs
913 try {
914 for (String ext : getImageExt(true)) {
915 if (Instance.getCache().check(new URL(line + ext))) {
916 url = new URL(line + ext);
917 break;
918 }
919 }
920
921 // try out of cache
922 if (url == null) {
923 for (String ext : getImageExt(true)) {
924 try {
925 url = new URL(line + ext);
926 Instance.getCache().refresh(url, support, true);
927 break;
928 } catch (IOException e) {
929 // no image with this ext
930 url = null;
931 }
932 }
933 }
934 } catch (MalformedURLException e) {
935 // Not an url
936 }
937 }
938
939 // refresh the cached file
940 if (url != null) {
941 try {
942 Instance.getCache().refresh(url, support, true);
943 } catch (IOException e) {
944 // woops, broken image
945 url = null;
946 }
947 }
948 }
949
950 return url;
951 }
952
953 /**
954 * Open the input file that will be used through the support.
955 *
956 * @param source
957 * the source {@link URL}
958 *
959 * @return the {@link InputStream}
960 *
961 * @throws IOException
962 * in case of I/O error
963 */
964 protected InputStream openInput(URL source) throws IOException {
965 return Instance.getCache().open(source, this, false);
966 }
967
968 /**
969 * Reset the given {@link InputStream} and return it.
970 *
971 * @param in
972 * the {@link InputStream} to reset
973 *
974 * @return the same {@link InputStream} after reset
975 */
976 protected InputStream reset(InputStream in) {
977 try {
978 in.reset();
979 } catch (IOException e) {
980 }
981 return in;
982 }
983
984 /**
985 * Reset then return {@link BasicSupport#in}.
986 *
987 * @return {@link BasicSupport#in}
988 */
989 protected InputStream getInput() {
990 return reset(in);
991 }
992
993 /**
994 * Fix the author name if it is prefixed with some "by" {@link String}.
995 *
996 * @param author
997 * the author with a possible prefix
998 *
999 * @return the author without prefixes
1000 */
1001 protected String fixAuthor(String author) {
1002 if (author != null) {
1003 for (String suffix : new String[] { " ", ":" }) {
1004 for (String byString : Instance.getConfig()
1005 .getString(Config.BYS).split(",")) {
1006 byString += suffix;
1007 if (author.toUpperCase().startsWith(byString.toUpperCase())) {
1008 author = author.substring(byString.length()).trim();
1009 }
1010 }
1011 }
1012
1013 // Special case (without suffix):
1014 if (author.startsWith("©")) {
1015 author = author.substring(1);
1016 }
1017 }
1018
1019 return author;
1020 }
1021
1022 /**
1023 * Check quotes for bad format (i.e., quotes with normal paragraphs inside)
1024 * and requotify them (i.e., separate them into QUOTE paragraphs and other
1025 * paragraphs (quotes or not)).
1026 *
1027 * @param para
1028 * the paragraph to requotify (not necessarily a quote)
1029 *
1030 * @return the correctly (or so we hope) quotified paragraphs
1031 */
1032 protected List<Paragraph> requotify(Paragraph para) {
1033 List<Paragraph> newParas = new ArrayList<Paragraph>();
1034
1035 if (para.getType() == ParagraphType.QUOTE
1036 && para.getContent().length() > 2) {
1037 String line = para.getContent();
1038 boolean singleQ = line.startsWith("" + openQuote);
1039 boolean doubleQ = line.startsWith("" + openDoubleQuote);
1040
1041 // Do not try when more than one quote at a time
1042 // (some stories are not easily readable if we do)
1043 if (singleQ
1044 && line.indexOf(closeQuote, 1) < line
1045 .lastIndexOf(closeQuote)) {
1046 newParas.add(para);
1047 return newParas;
1048 }
1049 if (doubleQ
1050 && line.indexOf(closeDoubleQuote, 1) < line
1051 .lastIndexOf(closeDoubleQuote)) {
1052 newParas.add(para);
1053 return newParas;
1054 }
1055 //
1056
1057 if (!singleQ && !doubleQ) {
1058 line = openDoubleQuote + line + closeDoubleQuote;
1059 newParas.add(new Paragraph(ParagraphType.QUOTE, line, para
1060 .getWords()));
1061 } else {
1062 char open = singleQ ? openQuote : openDoubleQuote;
1063 char close = singleQ ? closeQuote : closeDoubleQuote;
1064
1065 int posDot = -1;
1066 boolean inQuote = false;
1067 int i = 0;
1068 for (char car : line.toCharArray()) {
1069 if (car == open) {
1070 inQuote = true;
1071 } else if (car == close) {
1072 inQuote = false;
1073 } else if (car == '.' && !inQuote) {
1074 posDot = i;
1075 break;
1076 }
1077 i++;
1078 }
1079
1080 if (posDot >= 0) {
1081 String rest = line.substring(posDot + 1).trim();
1082 line = line.substring(0, posDot + 1).trim();
1083 long words = 1;
1084 for (char car : line.toCharArray()) {
1085 if (car == ' ') {
1086 words++;
1087 }
1088 }
1089 newParas.add(new Paragraph(ParagraphType.QUOTE, line, words));
1090 if (!rest.isEmpty()) {
1091 newParas.addAll(requotify(processPara(rest)));
1092 }
1093 } else {
1094 newParas.add(para);
1095 }
1096 }
1097 } else {
1098 newParas.add(para);
1099 }
1100
1101 return newParas;
1102 }
1103
1104 /**
1105 * Process a {@link Paragraph} from a raw line of text.
1106 * <p>
1107 * Will also fix quotes and HTML encoding if needed.
1108 *
1109 * @param line
1110 * the raw line
1111 *
1112 * @return the processed {@link Paragraph}
1113 */
1114 protected Paragraph processPara(String line) {
1115 line = ifUnhtml(line).trim();
1116
1117 boolean space = true;
1118 boolean brk = true;
1119 boolean quote = false;
1120 boolean tentativeCloseQuote = false;
1121 char prev = '\0';
1122 int dashCount = 0;
1123 long words = 1;
1124
1125 StringBuilder builder = new StringBuilder();
1126 for (char car : line.toCharArray()) {
1127 if (car != '-') {
1128 if (dashCount > 0) {
1129 // dash, ndash and mdash: - – —
1130 // currently: always use mdash
1131 builder.append(dashCount == 1 ? '-' : '—');
1132 }
1133 dashCount = 0;
1134 }
1135
1136 if (tentativeCloseQuote) {
1137 tentativeCloseQuote = false;
1138 if (Character.isLetterOrDigit(car)) {
1139 builder.append("'");
1140 } else {
1141 // handle double-single quotes as double quotes
1142 if (prev == car) {
1143 builder.append(closeDoubleQuote);
1144 continue;
1145 } else {
1146 builder.append(closeQuote);
1147 }
1148 }
1149 }
1150
1151 switch (car) {
1152 case ' ': // note: unbreakable space
1153 case ' ':
1154 case '\t':
1155 case '\n': // just in case
1156 case '\r': // just in case
1157 if (builder.length() > 0
1158 && builder.charAt(builder.length() - 1) != ' ') {
1159 words++;
1160 }
1161 builder.append(' ');
1162 break;
1163
1164 case '\'':
1165 if (space || (brk && quote)) {
1166 quote = true;
1167 // handle double-single quotes as double quotes
1168 if (prev == car) {
1169 builder.deleteCharAt(builder.length() - 1);
1170 builder.append(openDoubleQuote);
1171 } else {
1172 builder.append(openQuote);
1173 }
1174 } else if (prev == ' ' || prev == car) {
1175 // handle double-single quotes as double quotes
1176 if (prev == car) {
1177 builder.deleteCharAt(builder.length() - 1);
1178 builder.append(openDoubleQuote);
1179 } else {
1180 builder.append(openQuote);
1181 }
1182 } else {
1183 // it is a quote ("I'm off") or a 'quote' ("This
1184 // 'good' restaurant"...)
1185 tentativeCloseQuote = true;
1186 }
1187 break;
1188
1189 case '"':
1190 if (space || (brk && quote)) {
1191 quote = true;
1192 builder.append(openDoubleQuote);
1193 } else if (prev == ' ') {
1194 builder.append(openDoubleQuote);
1195 } else {
1196 builder.append(closeDoubleQuote);
1197 }
1198 break;
1199
1200 case '-':
1201 if (space) {
1202 quote = true;
1203 } else {
1204 dashCount++;
1205 }
1206 space = false;
1207 break;
1208
1209 case '*':
1210 case '~':
1211 case '/':
1212 case '\\':
1213 case '<':
1214 case '>':
1215 case '=':
1216 case '+':
1217 case '_':
1218 case '–':
1219 case '—':
1220 space = false;
1221 builder.append(car);
1222 break;
1223
1224 case '‘':
1225 case '`':
1226 case '‹':
1227 case '﹁':
1228 case '〈':
1229 case '「':
1230 if (space || (brk && quote)) {
1231 quote = true;
1232 builder.append(openQuote);
1233 } else {
1234 // handle double-single quotes as double quotes
1235 if (prev == car) {
1236 builder.deleteCharAt(builder.length() - 1);
1237 builder.append(openDoubleQuote);
1238 } else {
1239 builder.append(openQuote);
1240 }
1241 }
1242 space = false;
1243 brk = false;
1244 break;
1245
1246 case '’':
1247 case '›':
1248 case '﹂':
1249 case '〉':
1250 case '」':
1251 space = false;
1252 brk = false;
1253 // handle double-single quotes as double quotes
1254 if (prev == car) {
1255 builder.deleteCharAt(builder.length() - 1);
1256 builder.append(closeDoubleQuote);
1257 } else {
1258 builder.append(closeQuote);
1259 }
1260 break;
1261
1262 case '«':
1263 case '“':
1264 case '﹃':
1265 case '《':
1266 case '『':
1267 if (space || (brk && quote)) {
1268 quote = true;
1269 builder.append(openDoubleQuote);
1270 } else {
1271 builder.append(openDoubleQuote);
1272 }
1273 space = false;
1274 brk = false;
1275 break;
1276
1277 case '»':
1278 case '”':
1279 case '﹄':
1280 case '》':
1281 case '』':
1282 space = false;
1283 brk = false;
1284 builder.append(closeDoubleQuote);
1285 break;
1286
1287 default:
1288 space = false;
1289 brk = false;
1290 builder.append(car);
1291 break;
1292 }
1293
1294 prev = car;
1295 }
1296
1297 if (tentativeCloseQuote) {
1298 tentativeCloseQuote = false;
1299 builder.append(closeQuote);
1300 }
1301
1302 line = builder.toString().trim();
1303
1304 ParagraphType type = ParagraphType.NORMAL;
1305 if (space) {
1306 type = ParagraphType.BLANK;
1307 } else if (brk) {
1308 type = ParagraphType.BREAK;
1309 } else if (quote) {
1310 type = ParagraphType.QUOTE;
1311 }
1312
1313 return new Paragraph(type, line, words);
1314 }
1315
1316 /**
1317 * Remove the HTML from the input <b>if</b> {@link BasicSupport#isHtml()} is
1318 * true.
1319 *
1320 * @param input
1321 * the input
1322 *
1323 * @return the no html version if needed
1324 */
1325 private String ifUnhtml(String input) {
1326 if (isHtml() && input != null) {
1327 return StringUtils.unhtml(input);
1328 }
1329
1330 return input;
1331 }
1332
1333 /**
1334 * Return a {@link BasicSupport} implementation supporting the given
1335 * resource if possible.
1336 *
1337 * @param url
1338 * the story resource
1339 *
1340 * @return an implementation that supports it, or NULL
1341 */
1342 public static BasicSupport getSupport(URL url) {
1343 if (url == null) {
1344 return null;
1345 }
1346
1347 // TEXT and INFO_TEXT always support files (not URLs though)
1348 for (SupportType type : SupportType.values()) {
1349 if (type != SupportType.TEXT && type != SupportType.INFO_TEXT) {
1350 BasicSupport support = getSupport(type);
1351 if (support != null && support.supports(url)) {
1352 return support;
1353 }
1354 }
1355 }
1356
1357 for (SupportType type : new SupportType[] { SupportType.INFO_TEXT,
1358 SupportType.TEXT }) {
1359 BasicSupport support = getSupport(type);
1360 if (support != null && support.supports(url)) {
1361 return support;
1362 }
1363 }
1364
1365 return null;
1366 }
1367
1368 /**
1369 * Return a {@link BasicSupport} implementation supporting the given type.
1370 *
1371 * @param type
1372 * the type
1373 *
1374 * @return an implementation that supports it, or NULL
1375 */
1376 public static BasicSupport getSupport(SupportType type) {
1377 switch (type) {
1378 case EPUB:
1379 return new Epub().setType(type);
1380 case INFO_TEXT:
1381 return new InfoText().setType(type);
1382 case FIMFICTION:
1383 return new Fimfiction().setType(type);
1384 case FANFICTION:
1385 return new Fanfiction().setType(type);
1386 case TEXT:
1387 return new Text().setType(type);
1388 case MANGAFOX:
1389 return new MangaFox().setType(type);
1390 case E621:
1391 return new E621().setType(type);
1392 case YIFFSTAR:
1393 return new YiffStar().setType(type);
1394 case E_HENTAI:
1395 return new EHentai().setType(type);
1396 case CBZ:
1397 return new Cbz().setType(type);
1398 case HTML:
1399 return new Html().setType(type);
1400 }
1401
1402 return null;
1403 }
1404
1405 /**
1406 * Return the first line from the given input which correspond to the given
1407 * selectors.
1408 *
1409 * @param in
1410 * the input
1411 * @param needle
1412 * a string that must be found inside the target line (also
1413 * supports "^" at start to say "only if it starts with" the
1414 * needle)
1415 * @param relativeLine
1416 * the line to return based upon the target line position (-1 =
1417 * the line before, 0 = the target line...)
1418 *
1419 * @return the line
1420 */
1421 static String getLine(InputStream in, String needle, int relativeLine) {
1422 return getLine(in, needle, relativeLine, true);
1423 }
1424
1425 /**
1426 * Return a line from the given input which correspond to the given
1427 * selectors.
1428 *
1429 * @param in
1430 * the input
1431 * @param needle
1432 * a string that must be found inside the target line (also
1433 * supports "^" at start to say "only if it starts with" the
1434 * needle)
1435 * @param relativeLine
1436 * the line to return based upon the target line position (-1 =
1437 * the line before, 0 = the target line...)
1438 * @param first
1439 * takes the first result (as opposed to the last one, which will
1440 * also always spend the input)
1441 *
1442 * @return the line
1443 */
1444 static String getLine(InputStream in, String needle, int relativeLine,
1445 boolean first) {
1446 String rep = null;
1447
1448 try {
1449 in.reset();
1450 } catch (IOException e) {
1451 Instance.syserr(e);
1452 }
1453
1454 List<String> lines = new ArrayList<String>();
1455 @SuppressWarnings("resource")
1456 Scanner scan = new Scanner(in, "UTF-8");
1457 int index = -1;
1458 scan.useDelimiter("\\n");
1459 while (scan.hasNext()) {
1460 lines.add(scan.next());
1461
1462 if (index == -1) {
1463 if (needle.startsWith("^")) {
1464 if (lines.get(lines.size() - 1).startsWith(
1465 needle.substring(1))) {
1466 index = lines.size() - 1;
1467 }
1468
1469 } else {
1470 if (lines.get(lines.size() - 1).contains(needle)) {
1471 index = lines.size() - 1;
1472 }
1473 }
1474 }
1475
1476 if (index >= 0 && index + relativeLine < lines.size()) {
1477 rep = lines.get(index + relativeLine);
1478 if (first) {
1479 break;
1480 }
1481 }
1482 }
1483
1484 return rep;
1485 }
1486
1487 /**
1488 * Return the text between the key and the endKey (and optional subKey can
1489 * be passed, in this case we will look for the key first, then take the
1490 * text between the subKey and the endKey).
1491 * <p>
1492 * Will only match the first line with the given key if more than one are
1493 * possible. Which also means that if the subKey or endKey is not found on
1494 * that line, NULL will be returned.
1495 *
1496 * @param in
1497 * the input
1498 * @param key
1499 * the key to match (also supports "^" at start to say
1500 * "only if it starts with" the key)
1501 * @param subKey
1502 * the sub key or NULL if none
1503 * @param endKey
1504 * the end key or NULL for "up to the end"
1505 * @return the text or NULL if not found
1506 */
1507 static String getKeyLine(InputStream in, String key, String subKey,
1508 String endKey) {
1509 String result = null;
1510
1511 String line = getLine(in, key, 0);
1512 if (line != null && line.contains(key)) {
1513 line = line.substring(line.indexOf(key) + key.length());
1514 if (subKey == null || subKey.isEmpty() || line.contains(subKey)) {
1515 if (subKey != null) {
1516 line = line.substring(line.indexOf(subKey)
1517 + subKey.length());
1518 }
1519 if (endKey == null || line.contains(endKey)) {
1520 if (endKey != null) {
1521 line = line.substring(0, line.indexOf(endKey));
1522 result = line;
1523 }
1524 }
1525 }
1526 }
1527
1528 return result;
1529 }
1530 }