src/be/nikiroo/fanfix/searchable/Fanfiction.java

   1 package be.nikiroo.fanfix.searchable;
   2
   3 import java.io.IOException;
   4 import java.io.InputStream;
   5 import java.net.URL;
   6 import java.net.URLEncoder;
   7 import java.text.SimpleDateFormat;
   8 import java.util.ArrayList;
   9 import java.util.Date;
  10 import java.util.HashMap;
  11 import java.util.List;
  12 import java.util.Map;
  13
  14 import org.jsoup.nodes.Document;
  15 import org.jsoup.nodes.Element;
  16 import org.jsoup.select.Elements;
  17
  18 import be.nikiroo.fanfix.Instance;
  19 import be.nikiroo.fanfix.bundles.StringId;
  20 import be.nikiroo.fanfix.data.MetaData;
  21 import be.nikiroo.fanfix.supported.SupportType;
  22 import be.nikiroo.utils.Image;
  23 import be.nikiroo.utils.StringUtils;
  24
  25 /**
  26  * A {@link BasicSearchable} for Fanfiction.NET.
  27  *
  28  * @author niki
  29  */
  30 class Fanfiction extends BasicSearchable {
  31         static private String BASE_URL = "http://fanfiction.net/";
  32
  33         /**
  34          * Create a new {@link Fanfiction}.
  35          *
  36          * @param type
  37          *            {@link SupportType#FANFICTION}
  38          */
  39         public Fanfiction(SupportType type) {
  40                 super(type);
  41         }
  42
  43         @Override
  44         public List<SearchableTag> getTags() throws IOException {
  45                 String storiesName = null;
  46                 String crossoversName = null;
  47                 Map<String, String> stories = new HashMap<String, String>();
  48                 Map<String, String> crossovers = new HashMap<String, String>();
  49
  50                 Document mainPage = load(BASE_URL, true);
  51                 Element menu = mainPage.getElementsByClass("dropdown").first();
  52                 if (menu != null) {
  53                         Element ul = menu.getElementsByClass("dropdown-menu").first();
  54                         if (ul != null) {
  55                                 Map<String, String> currentList = null;
  56                                 for (Element li : ul.getElementsByTag("li")) {
  57                                         if (li.hasClass("disabled")) {
  58                                                 if (storiesName == null) {
  59                                                         storiesName = li.text();
  60                                                         currentList = stories;
  61                                                 } else {
  62                                                         crossoversName = li.text();
  63                                                         currentList = crossovers;
  64                                                 }
  65                                         } else if (currentList != null) {
  66                                                 Element a = li.getElementsByTag("a").first();
  67                                                 if (a != null) {
  68                                                         currentList.put(a.absUrl("href"), a.text());
  69                                                 }
  70                                         }
  71                                 }
  72                         }
  73                 }
  74
  75                 List<SearchableTag> tags = new ArrayList<SearchableTag>();
  76
  77                 if (storiesName != null) {
  78                         SearchableTag tag = new SearchableTag(null, storiesName, false);
  79                         for (String id : stories.keySet()) {
  80                                 tag.add(new SearchableTag(id, stories.get(id), false, false));
  81                         }
  82                         tags.add(tag);
  83                 }
  84
  85                 if (crossoversName != null) {
  86                         SearchableTag tag = new SearchableTag(null, crossoversName, false);
  87                         for (String id : crossovers.keySet()) {
  88                                 tag.add(new SearchableTag(id, crossovers.get(id), false, false));
  89                         }
  90                         tags.add(tag);
  91                 }
  92
  93                 return tags;
  94         }
  95
  96         @Override
  97         public void fillTag(SearchableTag tag) throws IOException {
  98                 if (tag.getId() == null || tag.isComplete()) {
  99                         return;
 100                 }
 101
 102                 Document doc = load(tag.getId(), false);
 103                 Element list = doc.getElementById("list_output");
 104                 if (list != null) {
 105                         Element table = list.getElementsByTag("table").first();
 106                         if (table != null) {
 107                                 for (Element div : table.getElementsByTag("div")) {
 108                                         Element a = div.getElementsByTag("a").first();
 109                                         Element span = div.getElementsByTag("span").first();
 110
 111                                         if (a != null) {
 112                                                 String subid = a.absUrl("href");
 113                                                 boolean crossoverSubtag = subid
 114                                                                 .contains("/crossovers/");
 115
 116                                                 SearchableTag subtag = new SearchableTag(subid,
 117                                                                 a.text(), !crossoverSubtag, !crossoverSubtag);
 118
 119                                                 tag.add(subtag);
 120                                                 if (span != null) {
 121                                                         String nr = span.text();
 122                                                         if (nr.startsWith("(")) {
 123                                                                 nr = nr.substring(1);
 124                                                         }
 125                                                         if (nr.endsWith(")")) {
 126                                                                 nr = nr.substring(0, nr.length() - 1);
 127                                                         }
 128                                                         nr = nr.trim();
 129
 130                                                         // TODO: fix toNumber/fromNumber
 131                                                         nr = nr.replaceAll("\\.[0-9]*", "");
 132
 133                                                         subtag.setCount(StringUtils.toNumber(nr));
 134                                                 }
 135                                         }
 136                                 }
 137                         }
 138                 }
 139
 140                 tag.setComplete(true);
 141         }
 142
 143         @Override
 144         public List<MetaData> search(String search, int page) throws IOException {
 145                 String encoded = URLEncoder.encode(search.toLowerCase(), "utf-8");
 146                 return getStories(BASE_URL + "search/?ready=1&type=story&keywords="
 147                                 + encoded + "&ppage=" + page, null, null);
 148         }
 149
 150         @Override
 151         public List<MetaData> search(SearchableTag tag, int page)
 152                         throws IOException {
 153                 List<MetaData> metas = new ArrayList<MetaData>();
 154
 155                 String url = tag.getId();
 156                 if (url != null) {
 157                         if (page > 1) {
 158                                 int pos = url.indexOf("&p=");
 159                                 if (pos >= 0) {
 160                                         url = url.replaceAll("(.*\\&p=)[0-9]*(.*)", "$1\\" + page
 161                                                         + "$2");
 162                                 } else {
 163                                         url += "&p=" + page;
 164                                 }
 165                         }
 166
 167                         Document doc = load(url, false);
 168
 169                         // Update the pages number if needed
 170                         if (tag.getPages() < 0 && tag.isLeaf()) {
 171                                 tag.setPages(getPages(doc));
 172                         }
 173
 174                         // Find out the full subjects (including parents)
 175                         String subjects = "";
 176                         for (SearchableTag t = tag; t != null; t = t.getParent()) {
 177                                 if (!subjects.isEmpty()) {
 178                                         subjects += ", ";
 179                                 }
 180                                 subjects += t.getName();
 181                         }
 182
 183                         metas = getStories(url, doc, subjects);
 184                 }
 185
 186                 return metas;
 187         }
 188
 189         /**
 190          * Return the number of pages in this stories result listing.
 191          *
 192          * @param doc
 193          *            the document
 194          *
 195          * @return the number of pages or -1 if unknown
 196          */
 197         private int getPages(Document doc) {
 198                 int pages = -1;
 199
 200                 if (doc != null) {
 201                         Element center = doc.getElementsByTag("center").first();
 202                         if (center != null) {
 203                                 for (Element a : center.getElementsByTag("a")) {
 204                                         if (a.absUrl("href").contains("&p=")) {
 205                                                 int thisLinkPages = -1;
 206                                                 try {
 207                                                         String[] tab = a.absUrl("href").split("=");
 208                                                         tab = tab[tab.length - 1].split("&");
 209                                                         thisLinkPages = Integer
 210                                                                         .parseInt(tab[tab.length - 1]);
 211                                                 } catch (Exception e) {
 212                                                 }
 213
 214                                                 pages = Math.max(pages, thisLinkPages);
 215                                         }
 216                                 }
 217                         }
 218                 }
 219
 220                 return pages;
 221         }
 222
 223         /**
 224          * Fetch the stories from the given page.
 225          *
 226          * @param sourceUrl
 227          *            the url of the document
 228          * @param doc
 229          *            the document to use (if NULL, will be loaded from
 230          *            <tt>sourceUrl</tt>)
 231          * @param mainSubject
 232          *            the main subject (the anime/book/movie item related to the
 233          *            stories, like "MLP" or "Doctor Who"), or NULL if none
 234          *
 235          * @return the stories found in it
 236          *
 237          * @throws IOException
 238          *             in case of I/O errors
 239          */
 240         private List<MetaData> getStories(String sourceUrl, Document doc,
 241                         String mainSubject) throws IOException {
 242                 List<MetaData> metas = new ArrayList<MetaData>();
 243
 244                 if (doc == null) {
 245                         doc = load(sourceUrl, false);
 246                 }
 247
 248                 for (Element story : doc.getElementsByClass("z-list")) {
 249                         MetaData meta = new MetaData();
 250                         meta.setImageDocument(false);
 251                         meta.setSource(getType().getSourceName());
 252
 253                         // Title, URL, Cover
 254                         Element stitle = story.getElementsByClass("stitle").first();
 255                         if (stitle != null) {
 256                                 meta.setTitle(stitle.text());
 257                                 meta.setUrl(stitle.absUrl("href"));
 258                                 Element cover = stitle.getElementsByTag("img").first();
 259                                 if (cover != null) {
 260                                         // note: see data-original if needed?
 261                                         String coverUrl = cover.absUrl("src");
 262
 263                                         try {
 264                                                 InputStream in = Instance.getCache().open(
 265                                                                 new URL(coverUrl), getSupport(), true);
 266                                                 try {
 267                                                         meta.setCover(new Image(in));
 268                                                 } finally {
 269                                                         in.close();
 270                                                 }
 271                                         } catch (Exception e) {
 272                                                 Instance.getTraceHandler()
 273                                                                 .error(new Exception(
 274                                                                                 "Cannot download cover for Fanfiction story in search mode",
 275                                                                                 e));
 276                                         }
 277                                 }
 278                         }
 279
 280                         // Author
 281                         Elements as = story.getElementsByTag("a");
 282                         if (as.size() > 1) {
 283                                 meta.setAuthor(as.get(1).text());
 284                         }
 285
 286                         // Tags (concatenated text), published date, updated date, Resume
 287                         String tags = "";
 288                         List<String> tagList = new ArrayList<String>();
 289                         Elements divs = story.getElementsByTag("div");
 290                         if (divs.size() > 1 && divs.get(1).childNodeSize() > 0) {
 291                                 String resume = divs.get(1).text();
 292                                 if (divs.size() > 2) {
 293                                         tags = divs.get(2).text();
 294                                         resume = resume.substring(0,
 295                                                         resume.length() - tags.length()).trim();
 296
 297                                         for (Element d : divs.get(2).getElementsByAttribute(
 298                                                         "data-xutime")) {
 299                                                 String secs = d.attr("data-xutime");
 300                                                 try {
 301                                                         String date = new SimpleDateFormat("yyyy-MM-dd")
 302                                                                         .format(new Date(
 303                                                                                         Long.parseLong(secs) * 1000));
 304                                                         // (updated, ) published
 305                                                         if (meta.getDate() != null) {
 306                                                                 tagList.add("Updated: " + meta.getDate());
 307                                                         }
 308                                                         meta.setDate(date);
 309                                                 } catch (Exception e) {
 310                                                 }
 311                                         }
 312                                 }
 313
 314                                 meta.setResume(getSupport().makeChapter(new URL(sourceUrl), 0,
 315                                                 Instance.getTrans().getString(StringId.DESCRIPTION),
 316                                                 resume));
 317                         }
 318
 319                         // How are the tags ordered?
 320                         // We have "Rated: xx", then the language, then all other tags
 321                         // If the subject(s) is/are present, they are before "Rated: xx"
 322
 323                         // ////////////
 324                         // Examples: //
 325                         // ////////////
 326
 327                         // Search (Luna) Tags: [Harry Potter, Rated: T, English, Chapters:
 328                         // 1, Words: 270, Reviews: 2, Published: 2/19/2013, Luna L.]
 329
 330                         // Normal (MLP) Tags: [Rated: T, Spanish, Drama/Suspense, Chapters:
 331                         // 2, Words: 8,686, Reviews: 1, Favs: 1, Follows: 1, Updated: 4/7,
 332                         // Published: 4/2]
 333
 334                         // Crossover (MLP/Who) Tags: [Rated: K+, English, Adventure/Romance,
 335                         // Chapters: 8, Words: 7,788, Reviews: 2, Favs: 2, Follows: 1,
 336                         // Published: 9/1/2016]
 337
 338                         boolean rated = false;
 339                         boolean isLang = false;
 340                         String subject = mainSubject == null ? "" : mainSubject;
 341                         String[] tab = tags.split("  *-  *");
 342                         for (int i = 0; i < tab.length; i++) {
 343                                 String tag = tab[i];
 344                                 if (tag.startsWith("Rated: ")) {
 345                                         rated = true;
 346                                 }
 347
 348                                 if (!rated) {
 349                                         if (!subject.isEmpty()) {
 350                                                 subject += ", ";
 351                                         }
 352                                         subject += tag;
 353                                 } else if (isLang) {
 354                                         meta.setLang(tag);
 355                                         isLang = false;
 356                                 } else {
 357                                         if (tag.contains(":")) {
 358                                                 // Handle special tags:
 359                                                 if (tag.startsWith("Words: ")) {
 360                                                         try {
 361                                                                 meta.setWords(Long.parseLong(tag
 362                                                                                 .substring("Words: ".length())
 363                                                                                 .replace(",", "").trim()));
 364                                                         } catch (Exception e) {
 365                                                         }
 366                                                 } else if (tag.startsWith("Rated: ")) {
 367                                                         tagList.add(tag);
 368                                                 }
 369                                         } else {
 370                                                 // Normal tags are "/"-separated
 371                                                 for (String t : tag.split("/")) {
 372                                                         tagList.add(t);
 373                                                 }
 374                                         }
 375
 376                                         if (tag.startsWith("Rated: ")) {
 377                                                 isLang = true;
 378                                         }
 379                                 }
 380                         }
 381
 382                         meta.setSubject(subject);
 383                         meta.setTags(tagList);
 384
 385                         metas.add(meta);
 386                 }
 387
 388                 return metas;
 389         }
 390 }