Merge branch 'master' into subtree
[fanfix.git] / supported / Text.java
1 package be.nikiroo.fanfix.supported;
2
3 import java.io.File;
4 import java.io.IOException;
5 import java.io.InputStream;
6 import java.net.URISyntaxException;
7 import java.net.URL;
8 import java.util.AbstractMap;
9 import java.util.ArrayList;
10 import java.util.List;
11 import java.util.Map.Entry;
12 import java.util.Scanner;
13
14 import org.jsoup.nodes.Document;
15
16 import be.nikiroo.fanfix.Instance;
17 import be.nikiroo.fanfix.bundles.Config;
18 import be.nikiroo.fanfix.data.Chapter;
19 import be.nikiroo.fanfix.data.MetaData;
20 import be.nikiroo.fanfix.data.Paragraph;
21 import be.nikiroo.utils.Image;
22 import be.nikiroo.utils.ImageUtils;
23 import be.nikiroo.utils.Progress;
24 import be.nikiroo.utils.streams.MarkableFileInputStream;
25
26 /**
27 * Support class for local stories encoded in textual format, with a few rules:
28 * <ul>
29 * <li>The title must be on the first line</li>
30 * <li>The author (preceded by nothing, "by " or "©") must be on the second
31 * line, possibly with the publication date in parenthesis (i.e., "
32 * <tt>By Unknown (3rd October 1998)</tt>")</li>
33 * <li>Chapters must be declared with "<tt>Chapter x</tt>" or "
34 * <tt>Chapter x: NAME OF THE CHAPTER</tt>", where "<tt>x</tt>" is the chapter
35 * number</li>
36 * <li>A description of the story must be given as chapter number 0</li>
37 * <li>A cover may be present, with the same filename but a PNG, JPEG or JPG
38 * extension</li>
39 * </ul>
40 *
41 * @author niki
42 */
43 class Text extends BasicSupport {
44 private File sourceFile;
45 private InputStream in;
46
47 protected File getSourceFile() {
48 return sourceFile;
49 }
50
51 protected InputStream getInput() {
52 if (in != null) {
53 try {
54 in.reset();
55 } catch (IOException e) {
56 Instance.getInstance().getTraceHandler().error(new IOException("Cannot reset the Text stream", e));
57 }
58
59 return in;
60 }
61
62 return null;
63 }
64
65 @Override
66 protected boolean isHtml() {
67 return false;
68 }
69
70 @Override
71 protected Document loadDocument(URL source) throws IOException {
72 try {
73 sourceFile = new File(source.toURI());
74 in = new MarkableFileInputStream(sourceFile);
75 } catch (URISyntaxException e) {
76 throw new IOException("Cannot load the text document: " + source);
77 }
78
79 return null;
80 }
81
82 @Override
83 protected MetaData getMeta() throws IOException {
84 MetaData meta = new MetaData();
85
86 meta.setTitle(getTitle());
87 meta.setAuthor(getAuthor());
88 meta.setDate(getDate());
89 meta.setTags(new ArrayList<String>());
90 meta.setSource(getType().getSourceName());
91 meta.setUrl(getSourceFile().toURI().toURL().toString());
92 meta.setPublisher("");
93 meta.setUuid(getSourceFile().toString());
94 meta.setLuid("");
95 meta.setLang(getLang()); // default is EN
96 meta.setSubject(getSourceFile().getParentFile().getName());
97 meta.setType(getType().toString());
98 meta.setImageDocument(false);
99 meta.setCover(getCover(getSourceFile()));
100
101 return meta;
102 }
103
104 private String getLang() {
105 @SuppressWarnings("resource")
106 Scanner scan = new Scanner(getInput(), "UTF-8");
107 scan.useDelimiter("\\n");
108 scan.next(); // Title
109 scan.next(); // Author (Date)
110 String chapter0 = scan.next(); // empty or Chapter 0
111 while (chapter0.isEmpty()) {
112 chapter0 = scan.next();
113 }
114
115 String lang = detectChapter(chapter0, 0);
116 if (lang == null) {
117 // No description??
118 lang = detectChapter(chapter0, 1);
119 }
120
121 if (lang == null) {
122 lang = "en";
123 } else {
124 lang = lang.toLowerCase();
125 }
126
127 return lang;
128 }
129
130 private String getTitle() {
131 @SuppressWarnings("resource")
132 Scanner scan = new Scanner(getInput(), "UTF-8");
133 scan.useDelimiter("\\n");
134 return scan.next();
135 }
136
137 private String getAuthor() {
138 @SuppressWarnings("resource")
139 Scanner scan = new Scanner(getInput(), "UTF-8");
140 scan.useDelimiter("\\n");
141 scan.next();
142 String authorDate = scan.next();
143
144 String author = authorDate;
145 int pos = authorDate.indexOf('(');
146 if (pos >= 0) {
147 author = authorDate.substring(0, pos);
148 }
149
150 return bsHelper.fixAuthor(author);
151 }
152
153 private String getDate() {
154 @SuppressWarnings("resource")
155 Scanner scan = new Scanner(getInput(), "UTF-8");
156 scan.useDelimiter("\\n");
157 scan.next();
158 String authorDate = scan.next();
159
160 String date = "";
161 int pos = authorDate.indexOf('(');
162 if (pos >= 0) {
163 date = authorDate.substring(pos + 1).trim();
164 pos = date.lastIndexOf(')');
165 if (pos >= 0) {
166 date = date.substring(0, pos).trim();
167 }
168 }
169
170 return date;
171 }
172
173 @Override
174 protected String getDesc() throws IOException {
175 String content = getChapterContent(null, 0, null).trim();
176 if (!content.isEmpty()) {
177 Chapter desc = bsPara.makeChapter(this, null, 0, "Description",
178 content, isHtml(), null);
179 StringBuilder builder = new StringBuilder();
180 for (Paragraph para : desc) {
181 if (builder.length() > 0) {
182 builder.append("\n");
183 }
184 builder.append(para.getContent());
185 }
186 }
187
188 return content;
189 }
190
191 private Image getCover(File sourceFile) {
192 String path = sourceFile.getName();
193
194 for (String ext : new String[] { ".txt", ".text", ".story" }) {
195 if (path.endsWith(ext)) {
196 path = path.substring(0, path.length() - ext.length());
197 }
198 }
199
200 Image cover = bsImages.getImage(this, sourceFile.getParentFile(), path);
201 if (cover != null) {
202 try {
203 File tmp = Instance.getInstance().getTempFiles().createTempFile("test_cover_image");
204 ImageUtils.getInstance().saveAsImage(cover, tmp, "png");
205 tmp.delete();
206 } catch (IOException e) {
207 cover = null;
208 }
209 }
210
211 return cover;
212 }
213
214 @Override
215 protected List<Entry<String, URL>> getChapters(Progress pg)
216 throws IOException {
217 List<Entry<String, URL>> chaps = new ArrayList<Entry<String, URL>>();
218 @SuppressWarnings("resource")
219 Scanner scan = new Scanner(getInput(), "UTF-8");
220 scan.useDelimiter("\\n");
221 boolean prevLineEmpty = false;
222 while (scan.hasNext()) {
223 String line = scan.next();
224 if (prevLineEmpty && detectChapter(line, chaps.size() + 1) != null) {
225 String chapName = Integer.toString(chaps.size() + 1);
226 int pos = line.indexOf(':');
227 if (pos >= 0 && pos + 1 < line.length()) {
228 chapName = line.substring(pos + 1).trim();
229 }
230
231 chaps.add(new AbstractMap.SimpleEntry<String, URL>(//
232 chapName, //
233 getSourceFile().toURI().toURL()));
234 }
235
236 prevLineEmpty = line.trim().isEmpty();
237 }
238
239 return chaps;
240 }
241
242 @Override
243 protected String getChapterContent(URL source, int number, Progress pg)
244 throws IOException {
245 StringBuilder builder = new StringBuilder();
246 @SuppressWarnings("resource")
247 Scanner scan = new Scanner(getInput(), "UTF-8");
248 scan.useDelimiter("\\n");
249 boolean inChap = false;
250 while (scan.hasNext()) {
251 String line = scan.next();
252 if (!inChap && detectChapter(line, number) != null) {
253 inChap = true;
254 } else if (detectChapter(line, number + 1) != null) {
255 break;
256 } else if (inChap) {
257 builder.append(line);
258 builder.append("\n");
259 }
260 }
261
262 return builder.toString();
263 }
264
265 @Override
266 protected void close() {
267 InputStream in = getInput();
268 if (in != null) {
269 try {
270 in.close();
271 } catch (IOException e) {
272 Instance.getInstance().getTraceHandler()
273 .error(new IOException("Cannot close the text source file input", e));
274 }
275 }
276
277 super.close();
278 }
279
280 @Override
281 protected boolean supports(URL url) {
282 return supports(url, false);
283 }
284
285 /**
286 * Check if we supports this {@link URL}, that is, if the info file can be
287 * found OR not found.
288 * <p>
289 * It must also be a file, not another kind of URL.
290 *
291 * @param url
292 * the {@link URL} to check
293 * @param info
294 * TRUE to require the info file, FALSE to forbid the info file
295 *
296 * @return TRUE if it is supported
297 */
298 protected boolean supports(URL url, boolean info) {
299 if (!"file".equals(url.getProtocol())) {
300 return false;
301 }
302
303 boolean infoPresent = false;
304 File file;
305 try {
306 file = new File(url.toURI());
307 file = assureNoTxt(file);
308 file = new File(file.getPath() + ".info");
309 } catch (URISyntaxException e) {
310 Instance.getInstance().getTraceHandler().error(e);
311 file = null;
312 }
313
314 infoPresent = (file != null && file.exists());
315
316 return infoPresent == info;
317 }
318
319 /**
320 * Remove the ".txt" extension if it is present.
321 *
322 * @param file
323 * the file to process
324 *
325 * @return the same file or a copy of it without the ".txt" extension if it
326 * was present
327 */
328 protected File assureNoTxt(File file) {
329 if (file.getName().endsWith(".txt")) {
330 file = new File(file.getPath().substring(0,
331 file.getPath().length() - 4));
332 }
333
334 return file;
335 }
336
337 /**
338 * Check if the given line looks like the given starting chapter in a
339 * supported language, and return the language if it does (or NULL if not).
340 *
341 * @param line
342 * the line to check
343 * @param number
344 * the specific chapter number to check for
345 *
346 * @return the language or NULL
347 */
348 static private String detectChapter(String line, int number) {
349 line = line.toUpperCase();
350 for (String lang : Instance.getInstance().getConfig().getList(Config.CONF_CHAPTER)) {
351 String chapter = Instance.getInstance().getConfig().getStringX(Config.CONF_CHAPTER, lang);
352 if (chapter != null && !chapter.isEmpty()) {
353 chapter = chapter.toUpperCase() + " ";
354 if (line.startsWith(chapter)) {
355 // We want "[CHAPTER] [number]: [name]", with ": [name]"
356 // optional
357 String test = line.substring(chapter.length()).trim();
358
359 String possibleNum = test.trim();
360 if (possibleNum.indexOf(':') > 0) {
361 possibleNum = possibleNum.substring(0,
362 possibleNum.indexOf(':')).trim();
363 }
364
365 if (test.startsWith(Integer.toString(number))) {
366 test = test
367 .substring(Integer.toString(number).length())
368 .trim();
369 if (test.isEmpty() || test.startsWith(":")) {
370 return lang;
371 }
372 }
373 }
374 }
375 }
376
377 return null;
378 }
379 }