separate support name and BasicSupport
[fanfix.git] / src / be / nikiroo / fanfix / supported / BasicSupport.java
1 package be.nikiroo.fanfix.supported;
2
3 import java.io.IOException;
4 import java.io.InputStream;
5 import java.net.URL;
6 import java.util.ArrayList;
7 import java.util.Date;
8 import java.util.HashMap;
9 import java.util.List;
10 import java.util.Map;
11 import java.util.Map.Entry;
12
13 import org.jsoup.helper.DataUtil;
14 import org.jsoup.nodes.Document;
15 import org.jsoup.nodes.Element;
16 import org.jsoup.nodes.Node;
17
18 import be.nikiroo.fanfix.Instance;
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.Story;
23 import be.nikiroo.utils.Progress;
24 import be.nikiroo.utils.StringUtils;
25
26 /**
27 * This class is the base class used by the other support classes. It can be
28 * used outside of this package, and have static method that you can use to get
29 * access to the correct support class.
30 * <p>
31 * It will be used with 'resources' (usually web pages or files).
32 *
33 * @author niki
34 */
35 public abstract class BasicSupport {
36 private Document sourceNode;
37 private URL source;
38 private SupportType type;
39 private URL currentReferer; // with only one 'r', as in 'HTTP'...
40
41 /**
42 * Check if the given resource is supported by this {@link BasicSupport}.
43 *
44 * @param url
45 * the resource to check for
46 *
47 * @return TRUE if it is
48 */
49 protected abstract boolean supports(URL url);
50
51 /**
52 * Return TRUE if the support will return HTML encoded content values for
53 * the chapters content.
54 *
55 * @return TRUE for HTML
56 */
57 protected abstract boolean isHtml();
58
59 /**
60 * Return the {@link MetaData} of this story.
61 *
62 * @return the associated {@link MetaData}, never NULL
63 *
64 * @throws IOException
65 * in case of I/O error
66 */
67 protected abstract MetaData getMeta() throws IOException;
68
69 /**
70 * Return the story description.
71 *
72 * @return the description
73 *
74 * @throws IOException
75 * in case of I/O error
76 */
77 protected abstract String getDesc() throws IOException;
78
79 /**
80 * Return the list of chapters (name and resource).
81 * <p>
82 * Can be NULL if this {@link BasicSupport} do no use chapters.
83 *
84 * @param pg
85 * the optional progress reporter
86 *
87 * @return the chapters or NULL
88 *
89 * @throws IOException
90 * in case of I/O error
91 */
92 protected abstract List<Entry<String, URL>> getChapters(Progress pg)
93 throws IOException;
94
95 /**
96 * Return the content of the chapter (possibly HTML encoded, if
97 * {@link BasicSupport#isHtml()} is TRUE).
98 *
99 * @param chapUrl
100 * the chapter {@link URL}
101 * @param number
102 * the chapter number
103 * @param pg
104 * the optional progress reporter
105 *
106 * @return the content
107 *
108 * @throws IOException
109 * in case of I/O error
110 */
111 protected abstract String getChapterContent(URL chapUrl, int number,
112 Progress pg) throws IOException;
113
114 /**
115 * Return the list of cookies (values included) that must be used to
116 * correctly fetch the resources.
117 * <p>
118 * You are expected to call the super method implementation if you override
119 * it.
120 *
121 * @return the cookies
122 */
123 public Map<String, String> getCookies() {
124 return new HashMap<String, String>();
125 }
126
127 /**
128 * OAuth authorisation (aka, "bearer XXXXXXX").
129 *
130 * @return the OAuth string
131 */
132 public String getOAuth() {
133 return null;
134 }
135
136 /**
137 * Return the canonical form of the main {@link URL}.
138 *
139 * @param source
140 * the source {@link URL}, which can be NULL
141 *
142 * @return the canonical form of this {@link URL} or NULL if the source was
143 * NULL
144 */
145 protected URL getCanonicalUrl(URL source) {
146 return source;
147 }
148
149 /**
150 * The main {@link Node} for this {@link Story}.
151 *
152 * @return the node
153 */
154 protected Element getSourceNode() {
155 return sourceNode;
156 }
157
158 /**
159 * The main {@link URL} for this {@link Story}.
160 *
161 * @return the URL
162 */
163 protected URL getSource() {
164 return source;
165 }
166
167 /**
168 * The current referer {@link URL} (only one 'r', as in 'HTML'...), i.e.,
169 * the current {@link URL} we work on.
170 *
171 * @return the referer
172 */
173 public URL getCurrentReferer() {
174 return currentReferer;
175 }
176
177 /**
178 * The current referer {@link URL} (only one 'r', as in 'HTML'...), i.e.,
179 * the current {@link URL} we work on.
180 *
181 * @param currentReferer
182 * the new referer
183 */
184 protected void setCurrentReferer(URL currentReferer) {
185 this.currentReferer = currentReferer;
186 }
187
188 /**
189 * The support type.
190 *
191 * @return the type
192 */
193 public SupportType getType() {
194 return type;
195 }
196
197 /**
198 * The support type.
199 *
200 * @param type
201 * the new type
202 */
203 protected void setType(SupportType type) {
204 this.type = type;
205 }
206
207 /**
208 * Open an input link that will be used for the support.
209 * <p>
210 * Can return NULL, in which case you are supposed to work without a source
211 * node.
212 *
213 * @param source
214 * the source {@link URL}
215 *
216 * @return the {@link InputStream}
217 *
218 * @throws IOException
219 * in case of I/O error
220 */
221 protected Document loadDocument(URL source) throws IOException {
222 String url = getCanonicalUrl(source).toString();
223 return DataUtil.load(Instance.getCache().open(source, this, false),
224 "UTF-8", url.toString());
225 }
226
227 /**
228 * Log into the support (can be a no-op depending upon the support).
229 *
230 * @throws IOException
231 * in case of I/O error
232 */
233 protected void login() throws IOException {
234 }
235
236 /**
237 * Now that we have processed the {@link Story}, close the resources if any.
238 */
239 protected void close() {
240 setCurrentReferer(null);
241 }
242
243 /**
244 * Process the given story resource into a partially filled {@link Story}
245 * object containing the name and metadata.
246 *
247 * @param getDesc
248 * retrieve the description of the story, or not
249 * @param pg
250 * the optional progress reporter
251 *
252 * @return the {@link Story}, never NULL
253 *
254 * @throws IOException
255 * in case of I/O error
256 */
257 protected Story processMeta(boolean getDesc, Progress pg)
258 throws IOException {
259 if (pg == null) {
260 pg = new Progress();
261 } else {
262 pg.setMinMax(0, 100);
263 }
264
265 pg.setProgress(30);
266
267 Story story = new Story();
268 MetaData meta = getMeta();
269 if (meta.getCreationDate() == null || meta.getCreationDate().isEmpty()) {
270 meta.setCreationDate(StringUtils.fromTime(new Date().getTime()));
271 }
272 story.setMeta(meta);
273
274 pg.setProgress(50);
275
276 if (meta.getCover() == null) {
277 meta.setCover(BasicSupportHelper.getDefaultCover(meta.getSubject()));
278 }
279
280 pg.setProgress(60);
281
282 if (getDesc) {
283 String descChapterName = Instance.getTrans().getString(
284 StringId.DESCRIPTION);
285 story.getMeta().setResume(
286 BasicSupportPara.makeChapter(this, source, 0,
287 descChapterName, //
288 getDesc(), isHtml(), null));
289 }
290
291 pg.done();
292 return story;
293 }
294
295 /**
296 * Process the given story resource into a fully filled {@link Story}
297 * object.
298 *
299 * @param pg
300 * the optional progress reporter
301 *
302 * @return the {@link Story}, never NULL
303 *
304 * @throws IOException
305 * in case of I/O error
306 */
307 // ADD final when BasicSupport_Deprecated is gone
308 public Story process(Progress pg) throws IOException {
309 setCurrentReferer(source);
310 login();
311 sourceNode = loadDocument(source);
312
313 try {
314 return doProcess(pg);
315 } finally {
316 close();
317 }
318 }
319
320 /**
321 * Actual processing step, without the calls to other methods.
322 * <p>
323 * Will convert the story resource into a fully filled {@link Story} object.
324 *
325 * @param pg
326 * the optional progress reporter
327 *
328 * @return the {@link Story}, never NULL
329 *
330 * @throws IOException
331 * in case of I/O error
332 */
333 protected Story doProcess(Progress pg) throws IOException {
334 if (pg == null) {
335 pg = new Progress();
336 } else {
337 pg.setMinMax(0, 100);
338 }
339
340 pg.setProgress(1);
341 Progress pgMeta = new Progress();
342 pg.addProgress(pgMeta, 10);
343 Story story = processMeta(true, pgMeta);
344 pgMeta.done(); // 10%
345
346 pg.setName("Retrieving " + story.getMeta().getTitle());
347
348 Progress pgGetChapters = new Progress();
349 pg.addProgress(pgGetChapters, 10);
350 story.setChapters(new ArrayList<Chapter>());
351 List<Entry<String, URL>> chapters = getChapters(pgGetChapters);
352 pgGetChapters.done(); // 20%
353
354 if (chapters != null) {
355 Progress pgChaps = new Progress("Extracting chapters", 0,
356 chapters.size() * 300);
357 pg.addProgress(pgChaps, 80);
358
359 long words = 0;
360 int i = 1;
361 for (Entry<String, URL> chap : chapters) {
362 pgChaps.setName("Extracting chapter " + i);
363 URL chapUrl = chap.getValue();
364 String chapName = chap.getKey();
365 if (chapUrl != null) {
366 setCurrentReferer(chapUrl);
367 }
368
369 pgChaps.setProgress(i * 100);
370 Progress pgGetChapterContent = new Progress();
371 Progress pgMakeChapter = new Progress();
372 pgChaps.addProgress(pgGetChapterContent, 100);
373 pgChaps.addProgress(pgMakeChapter, 100);
374
375 String content = getChapterContent(chapUrl, i,
376 pgGetChapterContent);
377 pgGetChapterContent.done();
378 Chapter cc = BasicSupportPara.makeChapter(this, chapUrl, i,
379 chapName, content, isHtml(), pgMakeChapter);
380 pgMakeChapter.done();
381
382 words += cc.getWords();
383 story.getChapters().add(cc);
384 story.getMeta().setWords(words);
385
386 i++;
387 }
388
389 pgChaps.setName("Extracting chapters");
390 pgChaps.done();
391 }
392
393 pg.done();
394
395 return story;
396 }
397
398 /**
399 * Return a {@link BasicSupport} implementation supporting the given
400 * resource if possible.
401 *
402 * @param url
403 * the story resource
404 *
405 * @return an implementation that supports it, or NULL
406 */
407 public static BasicSupport getSupport(URL url) {
408 if (url == null) {
409 return null;
410 }
411
412 // TEXT and INFO_TEXT always support files (not URLs though)
413 for (SupportType type : SupportType.values()) {
414 if (type != SupportType.TEXT && type != SupportType.INFO_TEXT) {
415 BasicSupport support = getSupport(type, url);
416 if (support != null && support.supports(url)) {
417 return support;
418 }
419 }
420 }
421
422 for (SupportType type : new SupportType[] { SupportType.INFO_TEXT,
423 SupportType.TEXT }) {
424 BasicSupport support = getSupport(type, url);
425 if (support != null && support.supports(url)) {
426 return support;
427 }
428 }
429
430 return null;
431 }
432
433 /**
434 * Return a {@link BasicSupport} implementation supporting the given type.
435 *
436 * @param type
437 * the type
438 * @param url
439 * the {@link URL} to support (can be NULL to get an
440 * "abstract support"; if not NULL, will be used as the source
441 * URL)
442 *
443 * @return an implementation that supports it, or NULL
444 */
445 public static BasicSupport getSupport(SupportType type, URL url) {
446 BasicSupport support = null;
447
448 switch (type) {
449 case EPUB:
450 support = new Epub();
451 break;
452 case INFO_TEXT:
453 support = new InfoText();
454 break;
455 case FIMFICTION:
456 try {
457 // Can fail if no client key or NO in options
458 support = new FimfictionApi();
459 } catch (IOException e) {
460 support = new Fimfiction();
461 }
462 break;
463 case FANFICTION:
464 support = new Fanfiction();
465 break;
466 case TEXT:
467 support = new Text();
468 break;
469 case MANGAFOX:
470 support = new MangaFox();
471 break;
472 case E621:
473 support = new E621();
474 break;
475 case YIFFSTAR:
476 support = new YiffStar();
477 break;
478 case E_HENTAI:
479 support = new EHentai();
480 break;
481 case MANGA_LEL:
482 support = new MangaLel();
483 break;
484 case CBZ:
485 support = new Cbz();
486 break;
487 case HTML:
488 support = new Html();
489 break;
490 }
491
492 if (support != null) {
493 support.setType(type);
494 support.source = support.getCanonicalUrl(url);
495 }
496
497 return support;
498 }
499 }