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(toNumber(nr));
 134                                                 }
 135                                         }
 136                                 }
 137                         }
 138                 }
 139
 140                 tag.setComplete(true);
 141         }
 142
 143         /**
 144          * @deprecated use {@link StringUtils} when updated
 145          */
 146         @Deprecated
 147         private static long toNumber(String value) {
 148                 // TODO: use StringUtils instead after update
 149                 long count = 0l;
 150                 if (value != null) {
 151                         try {
 152                                 if (value.toLowerCase().endsWith("m")) {
 153                                         count = Long.parseLong(value.substring(0,
 154                                                         value.length() - 1).trim());
 155                                         count *= 1000000;
 156                                 } else if (value.toLowerCase().endsWith("k")) {
 157                                         count = Long.parseLong(value.substring(0,
 158                                                         value.length() - 1).trim());
 159                                         count *= 1000;
 160                                 } else {
 161                                         count = Long.parseLong(value);
 162                                 }
 163                         } catch (NumberFormatException pe) {
 164                         }
 165                 }
 166
 167                 return count;
 168         }
 169
 170         @Override
 171         public List<MetaData> search(String search) throws IOException {
 172                 String encoded = URLEncoder.encode(search.toLowerCase(), "utf-8");
 173                 return getStories(BASE_URL + "search/?ready=1&type=story&keywords="
 174                                 + encoded, null, null);
 175         }
 176
 177         @Override
 178         public List<MetaData> search(SearchableTag tag, int page)
 179                         throws IOException {
 180                 List<MetaData> metas = new ArrayList<MetaData>();
 181
 182                 String url = tag.getId();
 183                 if (url != null) {
 184                         if (page > 1) {
 185                                 int pos = url.indexOf("&p=");
 186                                 if (pos >= 0) {
 187                                         url = url.replaceAll("(.*\\&p=)[0-9]*(.*)", "$1\\" + page
 188                                                         + "$2");
 189                                 } else {
 190                                         url += "&p=" + page;
 191                                 }
 192                         }
 193
 194                         Document doc = load(url, false);
 195
 196                         // Update the pages number if needed
 197                         if (tag.getPages() < 0 && tag.isLeaf()) {
 198                                 tag.setPages(getPages(doc));
 199                         }
 200
 201                         // Find out the full subjects (including parents)
 202                         String subjects = "";
 203                         for (SearchableTag t = tag; t != null; t = t.getParent()) {
 204                                 if (!subjects.isEmpty()) {
 205                                         subjects += ", ";
 206                                 }
 207                                 subjects += t.getName();
 208                         }
 209
 210                         metas = getStories(url, doc, subjects);
 211                 }
 212
 213                 return metas;
 214         }
 215
 216         /**
 217          * Return the number of pages in this stories result listing.
 218          *
 219          * @param doc
 220          *            the document
 221          *
 222          * @return the number of pages or -1 if unknown
 223          *
 224          * @throws IOException
 225          *             in case of I/O errors
 226          */
 227         private int getPages(Document doc) throws IOException {
 228                 int pages = -1;
 229
 230                 if (doc != null) {
 231                         Element center = doc.getElementsByTag("center").first();
 232                         if (center != null) {
 233                                 for (Element a : center.getElementsByTag("a")) {
 234                                         if (a.absUrl("href").contains("&p=")) {
 235                                                 int thisLinkPages = -1;
 236                                                 try {
 237                                                         String[] tab = a.absUrl("href").split("=");
 238                                                         tab = tab[tab.length - 1].split("&");
 239                                                         thisLinkPages = Integer
 240                                                                         .parseInt(tab[tab.length - 1]);
 241                                                 } catch (Exception e) {
 242                                                 }
 243
 244                                                 pages = Math.max(pages, thisLinkPages);
 245                                         }
 246                                 }
 247                         }
 248                 }
 249
 250                 return pages;
 251         }
 252
 253         /**
 254          * Fetch the stories from the given page.
 255          *
 256          * @param sourceUrl
 257          *            the url of the document
 258          * @param doc
 259          *            the document to use (if NULL, will be loaded from
 260          *            <tt>sourceUrl</tt>)
 261          * @param mainSubject
 262          *            the main subject (the anime/book/movie item related to the
 263          *            stories, like "MLP" or "Doctor Who"), or NULL if none
 264          *
 265          * @return the stories found in it
 266          *
 267          * @throws IOException
 268          *             in case of I/O errors
 269          */
 270         private List<MetaData> getStories(String sourceUrl, Document doc,
 271                         String mainSubject) throws IOException {
 272                 List<MetaData> metas = new ArrayList<MetaData>();
 273
 274                 if (doc == null) {
 275                         doc = load(sourceUrl, false);
 276                 }
 277
 278                 for (Element story : doc.getElementsByClass("z-list")) {
 279                         MetaData meta = new MetaData();
 280                         meta.setImageDocument(false);
 281                         meta.setSource(getType().getSourceName());
 282
 283                         // Title, URL, Cover
 284                         Element stitle = story.getElementsByClass("stitle").first();
 285                         if (stitle != null) {
 286                                 meta.setTitle(stitle.text());
 287                                 meta.setUrl(stitle.absUrl("href"));
 288                                 Element cover = stitle.getElementsByTag("img").first();
 289                                 if (cover != null) {
 290                                         // note: see data-original if needed?
 291                                         String coverUrl = cover.absUrl("src");
 292
 293                                         try {
 294                                                 InputStream in = Instance.getCache().open(
 295                                                                 new URL(coverUrl), getSupport(), true);
 296                                                 try {
 297                                                         meta.setCover(new Image(in));
 298                                                 } finally {
 299                                                         in.close();
 300                                                 }
 301                                         } catch (Exception e) {
 302                                                 Instance.getTraceHandler()
 303                                                                 .error(new Exception(
 304                                                                                 "Cannot download cover for Fanfiction story in search mode",
 305                                                                                 e));
 306                                         }
 307                                 }
 308                         }
 309
 310                         // Author
 311                         Elements as = story.getElementsByTag("a");
 312                         if (as.size() > 1) {
 313                                 meta.setAuthor(as.get(1).text());
 314                         }
 315
 316                         // Tags (concatenated text), published date, updated date, Resume
 317                         String tags = "";
 318                         List<String> tagList = new ArrayList<String>();
 319                         Elements divs = story.getElementsByTag("div");
 320                         if (divs.size() > 1 && divs.get(1).childNodeSize() > 0) {
 321                                 String resume = divs.get(1).text();
 322                                 if (divs.size() > 2) {
 323                                         tags = divs.get(2).text();
 324                                         resume = resume.substring(0,
 325                                                         resume.length() - tags.length()).trim();
 326
 327                                         for (Element d : divs.get(2).getElementsByAttribute(
 328                                                         "data-xutime")) {
 329                                                 String secs = d.attr("data-xutime");
 330                                                 try {
 331                                                         String date = new SimpleDateFormat("yyyy-MM-dd")
 332                                                                         .format(new Date(
 333                                                                                         Long.parseLong(secs) * 1000));
 334                                                         // (updated, ) published
 335                                                         if (meta.getDate() != null) {
 336                                                                 tagList.add("Updated: " + meta.getDate());
 337                                                         }
 338                                                         meta.setDate(date);
 339                                                 } catch (Exception e) {
 340                                                 }
 341                                         }
 342                                 }
 343
 344                                 meta.setResume(getSupport().makeChapter(new URL(sourceUrl), 0,
 345                                                 Instance.getTrans().getString(StringId.DESCRIPTION),
 346                                                 resume));
 347                         }
 348
 349                         // How are the tags ordered?
 350                         // We have "Rated: xx", then the language, then all other tags
 351                         // If the subject(s) is/are present, they are before "Rated: xx"
 352
 353                         // ////////////
 354                         // Examples: //
 355                         // ////////////
 356
 357                         // Search (Luna) Tags: [Harry Potter, Rated: T, English, Chapters:
 358                         // 1, Words: 270, Reviews: 2, Published: 2/19/2013, Luna L.]
 359
 360                         // Normal (MLP) Tags: [Rated: T, Spanish, Drama/Suspense, Chapters:
 361                         // 2, Words: 8,686, Reviews: 1, Favs: 1, Follows: 1, Updated: 4/7,
 362                         // Published: 4/2]
 363
 364                         // Crossover (MLP/Who) Tags: [Rated: K+, English, Adventure/Romance,
 365                         // Chapters: 8, Words: 7,788, Reviews: 2, Favs: 2, Follows: 1,
 366                         // Published: 9/1/2016]
 367
 368                         boolean rated = false;
 369                         boolean isLang = false;
 370                         String subject = mainSubject == null ? "" : mainSubject;
 371                         String[] tab = tags.split("  *-  *");
 372                         for (int i = 0; i < tab.length; i++) {
 373                                 String tag = tab[i];
 374                                 if (tag.startsWith("Rated: ")) {
 375                                         rated = true;
 376                                 }
 377
 378                                 if (!rated) {
 379                                         if (!subject.isEmpty()) {
 380                                                 subject += ", ";
 381                                         }
 382                                         subject += tag;
 383                                 } else if (isLang) {
 384                                         meta.setLang(tag);
 385                                         isLang = false;
 386                                 } else {
 387                                         if (tag.contains(":")) {
 388                                                 // Handle special tags:
 389                                                 if (tag.startsWith("Words: ")) {
 390                                                         try {
 391                                                                 meta.setWords(Long.parseLong(tag
 392                                                                                 .substring("Words: ".length())
 393                                                                                 .replace(",", "").trim()));
 394                                                         } catch (Exception e) {
 395                                                         }
 396                                                 } else if (tag.startsWith("Rated: ")) {
 397                                                         tagList.add(tag);
 398                                                 }
 399                                         } else {
 400                                                 // Normal tags are "/"-separated
 401                                                 for (String t : tag.split("/")) {
 402                                                         tagList.add(t);
 403                                                 }
 404                                         }
 405
 406                                         if (tag.startsWith("Rated: ")) {
 407                                                 isLang = true;
 408                                         }
 409                                 }
 410                         }
 411
 412                         meta.setSubject(subject);
 413                         meta.setTags(tagList);
 414
 415                         metas.add(meta);
 416                 }
 417
 418                 return metas;
 419         }
 420 }