| 1 | package be.nikiroo.fanfix; |
| 2 | |
| 3 | import java.io.BufferedOutputStream; |
| 4 | import java.io.File; |
| 5 | import java.io.FileInputStream; |
| 6 | import java.io.FileOutputStream; |
| 7 | import java.io.IOException; |
| 8 | import java.io.InputStream; |
| 9 | import java.net.CookieHandler; |
| 10 | import java.net.CookieManager; |
| 11 | import java.net.CookiePolicy; |
| 12 | import java.net.CookieStore; |
| 13 | import java.net.HttpCookie; |
| 14 | import java.net.HttpURLConnection; |
| 15 | import java.net.URISyntaxException; |
| 16 | import java.net.URL; |
| 17 | import java.net.URLConnection; |
| 18 | import java.util.Date; |
| 19 | import java.util.Map; |
| 20 | import java.util.zip.GZIPInputStream; |
| 21 | |
| 22 | import javax.imageio.ImageIO; |
| 23 | |
| 24 | import be.nikiroo.fanfix.bundles.Config; |
| 25 | import be.nikiroo.fanfix.supported.BasicSupport; |
| 26 | import be.nikiroo.utils.IOUtils; |
| 27 | import be.nikiroo.utils.MarkableFileInputStream; |
| 28 | |
| 29 | /** |
| 30 | * This cache will manage Internet (and local) downloads, as well as put the |
| 31 | * downloaded files into a cache. |
| 32 | * <p> |
| 33 | * As long the cached resource is not too old, it will use it instead of |
| 34 | * retrieving the file again. |
| 35 | * |
| 36 | * @author niki |
| 37 | */ |
| 38 | public class Cache { |
| 39 | private File dir; |
| 40 | private String UA; |
| 41 | private long tooOldChanging; |
| 42 | private long tooOldStable; |
| 43 | private CookieManager cookies; |
| 44 | |
| 45 | /** |
| 46 | * Create a new {@link Cache} object. |
| 47 | * |
| 48 | * @param dir |
| 49 | * the directory to use as cache |
| 50 | * @param UA |
| 51 | * the User-Agent to use to download the resources |
| 52 | * @param hoursChanging |
| 53 | * the number of hours after which a cached file that is thought |
| 54 | * to change ~often is considered too old (or -1 for |
| 55 | * "never too old") |
| 56 | * @param hoursStable |
| 57 | * the number of hours after which a LARGE cached file that is |
| 58 | * thought to change rarely is considered too old (or -1 for |
| 59 | * "never too old") |
| 60 | * |
| 61 | * @throws IOException |
| 62 | * in case of I/O error |
| 63 | */ |
| 64 | public Cache(File dir, String UA, int hoursChanging, int hoursStable) |
| 65 | throws IOException { |
| 66 | this.dir = dir; |
| 67 | this.UA = UA; |
| 68 | this.tooOldChanging = 1000 * 60 * 60 * hoursChanging; |
| 69 | this.tooOldStable = 1000 * 60 * 60 * hoursStable; |
| 70 | |
| 71 | if (dir != null) { |
| 72 | if (!dir.exists()) { |
| 73 | dir.mkdirs(); |
| 74 | } |
| 75 | } |
| 76 | |
| 77 | if (dir == null || !dir.exists()) { |
| 78 | throw new IOException("Cannot create the cache directory: " |
| 79 | + (dir == null ? "null" : dir.getAbsolutePath())); |
| 80 | } |
| 81 | |
| 82 | cookies = new CookieManager(); |
| 83 | cookies.setCookiePolicy(CookiePolicy.ACCEPT_ALL); |
| 84 | CookieHandler.setDefault(cookies); |
| 85 | } |
| 86 | |
| 87 | /** |
| 88 | * Open a resource (will load it from the cache if possible, or save it into |
| 89 | * the cache after downloading if not). |
| 90 | * |
| 91 | * @param url |
| 92 | * the resource to open |
| 93 | * @param support |
| 94 | * the support to use to download the resource |
| 95 | * @param stable |
| 96 | * TRUE for more stable resources, FALSE when they often change |
| 97 | * |
| 98 | * @return the opened resource, NOT NULL |
| 99 | * |
| 100 | * @throws IOException |
| 101 | * in case of I/O error |
| 102 | */ |
| 103 | public InputStream open(URL url, BasicSupport support, boolean stable) |
| 104 | throws IOException { |
| 105 | // MUST NOT return null |
| 106 | return open(url, support, stable, url); |
| 107 | } |
| 108 | |
| 109 | /** |
| 110 | * Open a resource (will load it from the cache if possible, or save it into |
| 111 | * the cache after downloading if not). |
| 112 | * <p> |
| 113 | * The cached resource will be assimilated to the given original {@link URL} |
| 114 | * |
| 115 | * @param url |
| 116 | * the resource to open |
| 117 | * @param support |
| 118 | * the support to use to download the resource |
| 119 | * @param stable |
| 120 | * TRUE for more stable resources, FALSE when they often change |
| 121 | * @param originalUrl |
| 122 | * the original {@link URL} used to locate the cached resource |
| 123 | * |
| 124 | * @return the opened resource, NOT NULL |
| 125 | * |
| 126 | * @throws IOException |
| 127 | * in case of I/O error |
| 128 | */ |
| 129 | public InputStream open(URL url, BasicSupport support, boolean stable, |
| 130 | URL originalUrl) throws IOException { |
| 131 | // MUST NOT return null |
| 132 | try { |
| 133 | InputStream in = load(originalUrl, false, stable); |
| 134 | if (in == null) { |
| 135 | try { |
| 136 | save(url, support, originalUrl); |
| 137 | } catch (IOException e) { |
| 138 | throw new IOException("Cannot save the url: " |
| 139 | + (url == null ? "null" : url.toString()), e); |
| 140 | } |
| 141 | |
| 142 | // Was just saved, can load old, so, will not be null |
| 143 | in = load(originalUrl, true, stable); |
| 144 | } |
| 145 | |
| 146 | return in; |
| 147 | } catch (IOException e) { |
| 148 | throw new IOException("Cannot open the url: " |
| 149 | + (url == null ? "null" : url.toString()), e); |
| 150 | } |
| 151 | } |
| 152 | |
| 153 | /** |
| 154 | * Refresh the resource into cache if needed. |
| 155 | * |
| 156 | * @param url |
| 157 | * the resource to open |
| 158 | * @param support |
| 159 | * the support to use to download the resource |
| 160 | * @param stable |
| 161 | * TRUE for more stable resources, FALSE when they often change |
| 162 | * |
| 163 | * @throws IOException |
| 164 | * in case of I/O error |
| 165 | */ |
| 166 | public void refresh(URL url, BasicSupport support, boolean stable) |
| 167 | throws IOException { |
| 168 | File cached = getCached(url); |
| 169 | if (cached.exists() && !isOld(cached, stable)) { |
| 170 | return; |
| 171 | } |
| 172 | |
| 173 | open(url, support, stable).close(); |
| 174 | } |
| 175 | |
| 176 | /** |
| 177 | * Check the resource to see if it is in the cache. |
| 178 | * |
| 179 | * @param url |
| 180 | * the resource to check |
| 181 | * |
| 182 | * @return TRUE if it is |
| 183 | * |
| 184 | */ |
| 185 | public boolean check(URL url) { |
| 186 | return getCached(url).exists(); |
| 187 | } |
| 188 | |
| 189 | /** |
| 190 | * Open a resource (will load it from the cache if possible, or save it into |
| 191 | * the cache after downloading if not) as an Image, then save it where |
| 192 | * requested. |
| 193 | * <p> |
| 194 | * This version will not always work properly if the original file was not |
| 195 | * downloaded before. |
| 196 | * |
| 197 | * @param url |
| 198 | * the resource to open |
| 199 | * |
| 200 | * @return the opened resource image |
| 201 | * |
| 202 | * @throws IOException |
| 203 | * in case of I/O error |
| 204 | */ |
| 205 | public void saveAsImage(URL url, File target) throws IOException { |
| 206 | URL cachedUrl = new URL(url.toString()); |
| 207 | File cached = getCached(cachedUrl); |
| 208 | |
| 209 | if (!cached.exists() || isOld(cached, true)) { |
| 210 | InputStream imageIn = open(url, null, true); |
| 211 | ImageIO.write(IOUtils.toImage(imageIn), Instance.getConfig() |
| 212 | .getString(Config.IMAGE_FORMAT_CONTENT).toLowerCase(), |
| 213 | cached); |
| 214 | } |
| 215 | |
| 216 | IOUtils.write(new FileInputStream(cached), target); |
| 217 | } |
| 218 | |
| 219 | /** |
| 220 | * Manually add this item to the cache. |
| 221 | * |
| 222 | * @param in |
| 223 | * the input data |
| 224 | * @param uniqueID |
| 225 | * a unique ID for this resource |
| 226 | * |
| 227 | * @return the resulting {@link File} |
| 228 | * |
| 229 | * @throws IOException |
| 230 | * in case of I/O error |
| 231 | */ |
| 232 | public File addToCache(InputStream in, String uniqueID) throws IOException { |
| 233 | File file = getCached(new File(uniqueID).toURI().toURL()); |
| 234 | IOUtils.write(in, file); |
| 235 | return file; |
| 236 | } |
| 237 | |
| 238 | /** |
| 239 | * Clean the cache (delete the cached items). |
| 240 | * |
| 241 | * @param onlyOld |
| 242 | * only clean the files that are considered too old |
| 243 | * |
| 244 | * @return the number of cleaned items |
| 245 | */ |
| 246 | public int cleanCache(boolean onlyOld) { |
| 247 | int num = 0; |
| 248 | for (File file : dir.listFiles()) { |
| 249 | if (!onlyOld || isOld(file, true)) { |
| 250 | if (file.delete()) { |
| 251 | num++; |
| 252 | } else { |
| 253 | System.err.println("Cannot delete temporary file: " |
| 254 | + file.getAbsolutePath()); |
| 255 | } |
| 256 | } |
| 257 | } |
| 258 | return num; |
| 259 | } |
| 260 | |
| 261 | /** |
| 262 | * Open a resource from the cache if it exists. |
| 263 | * |
| 264 | * @param url |
| 265 | * the resource to open |
| 266 | * |
| 267 | * @return the opened resource if found, NULL i not |
| 268 | * |
| 269 | * @throws IOException |
| 270 | * in case of I/O error |
| 271 | */ |
| 272 | private InputStream load(URL url, boolean allowOld, boolean stable) |
| 273 | throws IOException { |
| 274 | File cached = getCached(url); |
| 275 | if (cached.exists() && !isOld(cached, stable)) { |
| 276 | return new MarkableFileInputStream(new FileInputStream(cached)); |
| 277 | } |
| 278 | |
| 279 | return null; |
| 280 | } |
| 281 | |
| 282 | /** |
| 283 | * Save the given resource to the cache. |
| 284 | * |
| 285 | * @param url |
| 286 | * the resource |
| 287 | * @param support |
| 288 | * the {@link BasicSupport} used to download it |
| 289 | * @param originalUrl |
| 290 | * the original {@link URL} used to locate the cached resource |
| 291 | * |
| 292 | * @throws IOException |
| 293 | * in case of I/O error |
| 294 | * @throws URISyntaxException |
| 295 | */ |
| 296 | private void save(URL url, BasicSupport support, URL originalUrl) |
| 297 | throws IOException { |
| 298 | URLConnection conn = url.openConnection(); |
| 299 | |
| 300 | conn.setRequestProperty("User-Agent", UA); |
| 301 | conn.setRequestProperty("Cookie", generateCookies(support)); |
| 302 | conn.setRequestProperty("Accept-Encoding", "gzip"); |
| 303 | if (support != null && support.getCurrentReferer() != null) { |
| 304 | conn.setRequestProperty("Referer", support.getCurrentReferer() |
| 305 | .toString()); |
| 306 | conn.setRequestProperty("Host", support.getCurrentReferer() |
| 307 | .getHost()); |
| 308 | } |
| 309 | |
| 310 | conn.connect(); |
| 311 | |
| 312 | // Check if redirect |
| 313 | if (conn instanceof HttpURLConnection |
| 314 | && ((HttpURLConnection) conn).getResponseCode() / 100 == 3) { |
| 315 | String newUrl = conn.getHeaderField("Location"); |
| 316 | save(new URL(newUrl), support, originalUrl); |
| 317 | return; |
| 318 | } |
| 319 | |
| 320 | InputStream in = conn.getInputStream(); |
| 321 | if ("gzip".equals(conn.getContentEncoding())) { |
| 322 | in = new GZIPInputStream(in); |
| 323 | } |
| 324 | |
| 325 | try { |
| 326 | File cached = getCached(originalUrl); |
| 327 | BufferedOutputStream out = new BufferedOutputStream( |
| 328 | new FileOutputStream(cached)); |
| 329 | try { |
| 330 | byte[] buf = new byte[4096]; |
| 331 | int len; |
| 332 | while ((len = in.read(buf)) > 0) { |
| 333 | out.write(buf, 0, len); |
| 334 | } |
| 335 | } finally { |
| 336 | out.close(); |
| 337 | } |
| 338 | } finally { |
| 339 | in.close(); |
| 340 | } |
| 341 | } |
| 342 | |
| 343 | /** |
| 344 | * Check if the {@link File} is too old according to |
| 345 | * {@link Cache#tooOldChanging}. |
| 346 | * |
| 347 | * @param file |
| 348 | * the file to check |
| 349 | * @param stable |
| 350 | * TRUE to denote files that are not supposed to change too often |
| 351 | * |
| 352 | * @return TRUE if it is |
| 353 | */ |
| 354 | private boolean isOld(File file, boolean stable) { |
| 355 | long max = tooOldChanging; |
| 356 | if (stable) { |
| 357 | max = tooOldStable; |
| 358 | } |
| 359 | |
| 360 | if (max < 0) { |
| 361 | return false; |
| 362 | } |
| 363 | |
| 364 | long time = new Date().getTime() - file.lastModified(); |
| 365 | if (time < 0) { |
| 366 | System.err.println("Timestamp in the future for file: " |
| 367 | + file.getAbsolutePath()); |
| 368 | } |
| 369 | |
| 370 | return time < 0 || time > max; |
| 371 | } |
| 372 | |
| 373 | /** |
| 374 | * Get the cache resource from the cache if it is present for this |
| 375 | * {@link URL}. |
| 376 | * |
| 377 | * @param url |
| 378 | * the url |
| 379 | * @return the cached version if present, NULL if not |
| 380 | */ |
| 381 | private File getCached(URL url) { |
| 382 | String name = url.getHost(); |
| 383 | if (name == null || name.length() == 0) { |
| 384 | name = url.getFile(); |
| 385 | } else { |
| 386 | name = url.toString(); |
| 387 | } |
| 388 | |
| 389 | name = name.replace('/', '_').replace(':', '_'); |
| 390 | |
| 391 | return new File(dir, name); |
| 392 | } |
| 393 | |
| 394 | /** |
| 395 | * Generate the cookie {@link String} from the local {@link CookieStore} so |
| 396 | * it is ready to be passed. |
| 397 | * |
| 398 | * @return the cookie |
| 399 | */ |
| 400 | private String generateCookies(BasicSupport support) { |
| 401 | StringBuilder builder = new StringBuilder(); |
| 402 | for (HttpCookie cookie : cookies.getCookieStore().getCookies()) { |
| 403 | if (builder.length() > 0) { |
| 404 | builder.append(';'); |
| 405 | } |
| 406 | |
| 407 | // TODO: check if format is ok |
| 408 | builder.append(cookie.toString()); |
| 409 | } |
| 410 | |
| 411 | if (support != null) { |
| 412 | for (Map.Entry<String, String> set : support.getCookies() |
| 413 | .entrySet()) { |
| 414 | if (builder.length() > 0) { |
| 415 | builder.append(';'); |
| 416 | } |
| 417 | builder.append(set.getKey()); |
| 418 | builder.append('='); |
| 419 | builder.append(set.getValue()); |
| 420 | } |
| 421 | } |
| 422 | |
| 423 | return builder.toString(); |
| 424 | } |
| 425 | } |