| 1 | package be.nikiroo.utils; |
| 2 | |
| 3 | import java.io.FileNotFoundException; |
| 4 | import java.io.IOException; |
| 5 | import java.io.InputStream; |
| 6 | import java.io.OutputStreamWriter; |
| 7 | import java.net.CookieHandler; |
| 8 | import java.net.CookieManager; |
| 9 | import java.net.CookiePolicy; |
| 10 | import java.net.CookieStore; |
| 11 | import java.net.HttpCookie; |
| 12 | import java.net.HttpURLConnection; |
| 13 | import java.net.URL; |
| 14 | import java.net.URLConnection; |
| 15 | import java.net.URLEncoder; |
| 16 | import java.util.Map; |
| 17 | import java.util.zip.GZIPInputStream; |
| 18 | |
| 19 | /** |
| 20 | * This class will help you download content from Internet Sites ({@link URL} |
| 21 | * based). |
| 22 | * <p> |
| 23 | * It allows you to control some options often required on web sites that do not |
| 24 | * want to simply serve HTML, but actively makes your life difficult with stupid |
| 25 | * checks. |
| 26 | * |
| 27 | * @author niki |
| 28 | */ |
| 29 | public class Downloader { |
| 30 | private String UA; |
| 31 | private CookieManager cookies; |
| 32 | private TraceHandler tracer = new TraceHandler(); |
| 33 | private Cache cache; |
| 34 | |
| 35 | /** |
| 36 | * Create a new {@link Downloader}. |
| 37 | * |
| 38 | * @param UA |
| 39 | * the User-Agent to use to download the resources -- note that |
| 40 | * some websites require one, some actively blacklist real UAs |
| 41 | * like the one from wget, some whitelist a couple of browsers |
| 42 | * only (!) |
| 43 | */ |
| 44 | public Downloader(String UA) { |
| 45 | this(UA, null); |
| 46 | } |
| 47 | |
| 48 | /** |
| 49 | * Create a new {@link Downloader}. |
| 50 | * |
| 51 | * @param UA |
| 52 | * the User-Agent to use to download the resources -- note that |
| 53 | * some websites require one, some actively blacklist real UAs |
| 54 | * like the one from wget, some whitelist a couple of browsers |
| 55 | * only (!) |
| 56 | * @param cache |
| 57 | * the {@link Cache} to use for all access (can be NULL) |
| 58 | */ |
| 59 | public Downloader(String UA, Cache cache) { |
| 60 | this.UA = UA; |
| 61 | |
| 62 | cookies = new CookieManager(null, CookiePolicy.ACCEPT_ALL); |
| 63 | CookieHandler.setDefault(cookies); |
| 64 | |
| 65 | setCache(cache); |
| 66 | } |
| 67 | |
| 68 | /** |
| 69 | * The traces handler for this {@link Cache}. |
| 70 | * |
| 71 | * @return the traces handler |
| 72 | */ |
| 73 | public TraceHandler getTraceHandler() { |
| 74 | return tracer; |
| 75 | } |
| 76 | |
| 77 | /** |
| 78 | * The traces handler for this {@link Cache}. |
| 79 | * |
| 80 | * @param tracer |
| 81 | * the new traces handler |
| 82 | */ |
| 83 | public void setTraceHandler(TraceHandler tracer) { |
| 84 | if (tracer == null) { |
| 85 | tracer = new TraceHandler(false, false, false); |
| 86 | } |
| 87 | |
| 88 | this.tracer = tracer; |
| 89 | } |
| 90 | |
| 91 | /** |
| 92 | * The {@link Cache} to use for all access (can be NULL). |
| 93 | * |
| 94 | * @return the cache |
| 95 | */ |
| 96 | public Cache getCache() { |
| 97 | return cache; |
| 98 | } |
| 99 | |
| 100 | /** |
| 101 | * The {@link Cache} to use for all access (can be NULL). |
| 102 | * |
| 103 | * @param cache |
| 104 | * the new cache |
| 105 | */ |
| 106 | public void setCache(Cache cache) { |
| 107 | this.cache = cache; |
| 108 | } |
| 109 | |
| 110 | /** |
| 111 | * Clear all the cookies currently in the jar. |
| 112 | * <p> |
| 113 | * As long as you don't, the cookies are kept. |
| 114 | */ |
| 115 | public void clearCookies() { |
| 116 | cookies.getCookieStore().removeAll(); |
| 117 | } |
| 118 | |
| 119 | /** |
| 120 | * Open the given {@link URL} and update the cookies. |
| 121 | * |
| 122 | * @param url |
| 123 | * the {@link URL} to open |
| 124 | * @return the {@link InputStream} of the opened page |
| 125 | * |
| 126 | * @throws IOException |
| 127 | * in case of I/O error |
| 128 | **/ |
| 129 | public InputStream open(URL url) throws IOException { |
| 130 | return open(url, false); |
| 131 | } |
| 132 | |
| 133 | /** |
| 134 | * Open the given {@link URL} and update the cookies. |
| 135 | * |
| 136 | * @param url |
| 137 | * the {@link URL} to open |
| 138 | * @param stable |
| 139 | * stable a stable file (that doesn't change too often) -- |
| 140 | * parameter used to check if the file is too old to keep or not |
| 141 | * in the cache (default is false) |
| 142 | * |
| 143 | * @return the {@link InputStream} of the opened page |
| 144 | * |
| 145 | * @throws IOException |
| 146 | * in case of I/O error |
| 147 | **/ |
| 148 | public InputStream open(URL url, boolean stable) throws IOException { |
| 149 | return open(url, url, url, null, null, null, null, stable); |
| 150 | } |
| 151 | |
| 152 | /** |
| 153 | * Open the given {@link URL} and update the cookies. |
| 154 | * |
| 155 | * @param url |
| 156 | * the {@link URL} to open |
| 157 | * @param currentReferer |
| 158 | * the current referer, for websites that needs this info |
| 159 | * @param cookiesValues |
| 160 | * the cookies |
| 161 | * @param postParams |
| 162 | * the POST parameters |
| 163 | * @param getParams |
| 164 | * the GET parameters (priority over POST) |
| 165 | * @param oauth |
| 166 | * OAuth authorization (aka, "bearer XXXXXXX") |
| 167 | * |
| 168 | * @return the {@link InputStream} of the opened page |
| 169 | * |
| 170 | * @throws IOException |
| 171 | * in case of I/O error |
| 172 | */ |
| 173 | public InputStream open(URL url, URL currentReferer, |
| 174 | Map<String, String> cookiesValues, Map<String, String> postParams, |
| 175 | Map<String, String> getParams, String oauth) throws IOException { |
| 176 | return open(url, currentReferer, cookiesValues, postParams, getParams, |
| 177 | oauth, false); |
| 178 | } |
| 179 | |
| 180 | /** |
| 181 | * Open the given {@link URL} and update the cookies. |
| 182 | * |
| 183 | * @param url |
| 184 | * the {@link URL} to open |
| 185 | * @param currentReferer |
| 186 | * the current referer, for websites that needs this info |
| 187 | * @param cookiesValues |
| 188 | * the cookies |
| 189 | * @param postParams |
| 190 | * the POST parameters |
| 191 | * @param getParams |
| 192 | * the GET parameters (priority over POST) |
| 193 | * @param oauth |
| 194 | * OAuth authorization (aka, "bearer XXXXXXX") |
| 195 | * @param stable |
| 196 | * stable a stable file (that doesn't change too often) -- |
| 197 | * parameter used to check if the file is too old to keep or not |
| 198 | * in the cache (default is false) |
| 199 | * |
| 200 | * @return the {@link InputStream} of the opened page |
| 201 | * |
| 202 | * @throws IOException |
| 203 | * in case of I/O error |
| 204 | */ |
| 205 | public InputStream open(URL url, URL currentReferer, |
| 206 | Map<String, String> cookiesValues, Map<String, String> postParams, |
| 207 | Map<String, String> getParams, String oauth, boolean stable) |
| 208 | throws IOException { |
| 209 | return open(url, url, currentReferer, cookiesValues, postParams, |
| 210 | getParams, oauth, stable); |
| 211 | } |
| 212 | |
| 213 | /** |
| 214 | * Open the given {@link URL} and update the cookies. |
| 215 | * |
| 216 | * @param url |
| 217 | * the {@link URL} to open |
| 218 | * @param originalUrl |
| 219 | * the original {@link URL} before any redirection occurs, which |
| 220 | * is also used for the cache ID if needed (so we can retrieve |
| 221 | * the content with this URL if needed) |
| 222 | * @param currentReferer |
| 223 | * the current referer, for websites that needs this info |
| 224 | * @param cookiesValues |
| 225 | * the cookies |
| 226 | * @param postParams |
| 227 | * the POST parameters |
| 228 | * @param getParams |
| 229 | * the GET parameters (priority over POST) |
| 230 | * @param oauth |
| 231 | * OAuth authorisation (aka, "bearer XXXXXXX") |
| 232 | * @param stable |
| 233 | * a stable file (that doesn't change too often) -- parameter |
| 234 | * used to check if the file is too old to keep or not in the |
| 235 | * cache |
| 236 | * |
| 237 | * @return the {@link InputStream} of the opened page |
| 238 | * |
| 239 | * @throws IOException |
| 240 | * in case of I/O error |
| 241 | */ |
| 242 | public InputStream open(URL url, final URL originalUrl, URL currentReferer, |
| 243 | Map<String, String> cookiesValues, Map<String, String> postParams, |
| 244 | Map<String, String> getParams, String oauth, boolean stable) |
| 245 | throws IOException { |
| 246 | |
| 247 | tracer.trace("Request: " + url); |
| 248 | |
| 249 | if (cache != null) { |
| 250 | InputStream in = cache.load(originalUrl, false, stable); |
| 251 | if (in != null) { |
| 252 | tracer.trace("Use the cache: " + url); |
| 253 | tracer.trace("Original URL : " + originalUrl); |
| 254 | return in; |
| 255 | } |
| 256 | } |
| 257 | |
| 258 | tracer.trace("Download: " + url); |
| 259 | |
| 260 | URLConnection conn = openConnectionWithCookies(url, currentReferer, |
| 261 | cookiesValues); |
| 262 | |
| 263 | // Priority: GET over POST |
| 264 | Map<String, String> params = getParams; |
| 265 | if (getParams == null) { |
| 266 | params = postParams; |
| 267 | } |
| 268 | |
| 269 | StringBuilder requestData = null; |
| 270 | if ((params != null || oauth != null) |
| 271 | && conn instanceof HttpURLConnection) { |
| 272 | if (params != null) { |
| 273 | requestData = new StringBuilder(); |
| 274 | for (Map.Entry<String, String> param : params.entrySet()) { |
| 275 | if (requestData.length() != 0) |
| 276 | requestData.append('&'); |
| 277 | requestData.append(URLEncoder.encode(param.getKey(), |
| 278 | "UTF-8")); |
| 279 | requestData.append('='); |
| 280 | requestData.append(URLEncoder.encode( |
| 281 | String.valueOf(param.getValue()), "UTF-8")); |
| 282 | } |
| 283 | |
| 284 | if (getParams == null && postParams != null) { |
| 285 | ((HttpURLConnection) conn).setRequestMethod("POST"); |
| 286 | } |
| 287 | |
| 288 | conn.setRequestProperty("Content-Type", |
| 289 | "application/x-www-form-urlencoded"); |
| 290 | conn.setRequestProperty("Content-Length", |
| 291 | Integer.toString(requestData.length())); |
| 292 | } |
| 293 | |
| 294 | if (oauth != null) { |
| 295 | conn.setRequestProperty("Authorization", oauth); |
| 296 | } |
| 297 | |
| 298 | if (requestData != null) { |
| 299 | conn.setDoOutput(true); |
| 300 | OutputStreamWriter writer = new OutputStreamWriter( |
| 301 | conn.getOutputStream()); |
| 302 | try { |
| 303 | writer.write(requestData.toString()); |
| 304 | writer.flush(); |
| 305 | } finally { |
| 306 | writer.close(); |
| 307 | } |
| 308 | } |
| 309 | } |
| 310 | |
| 311 | // Manual redirection, much better for POST data |
| 312 | if (conn instanceof HttpURLConnection) { |
| 313 | ((HttpURLConnection) conn).setInstanceFollowRedirects(false); |
| 314 | } |
| 315 | |
| 316 | conn.connect(); |
| 317 | |
| 318 | // Check if redirect |
| 319 | // BEWARE! POST data cannot be redirected (some webservers complain) for |
| 320 | // HTTP codes 302 and 303 |
| 321 | if (conn instanceof HttpURLConnection) { |
| 322 | int repCode = 0; |
| 323 | try { |
| 324 | // Can fail in some circumstances |
| 325 | repCode = ((HttpURLConnection) conn).getResponseCode(); |
| 326 | } catch (IOException e) { |
| 327 | } |
| 328 | |
| 329 | if (repCode / 100 == 3) { |
| 330 | String newUrl = conn.getHeaderField("Location"); |
| 331 | return open(new URL(newUrl), originalUrl, currentReferer, |
| 332 | cookiesValues, // |
| 333 | (repCode == 302 || repCode == 303) ? null : postParams, // |
| 334 | getParams, oauth, stable); |
| 335 | } |
| 336 | } |
| 337 | |
| 338 | try { |
| 339 | InputStream in = conn.getInputStream(); |
| 340 | if ("gzip".equals(conn.getContentEncoding())) { |
| 341 | in = new GZIPInputStream(in); |
| 342 | } |
| 343 | |
| 344 | if (in == null) { |
| 345 | throw new IOException("No InputStream!"); |
| 346 | } |
| 347 | |
| 348 | if (cache != null) { |
| 349 | String size = conn.getContentLengthLong() < 0 ? "unknown size" |
| 350 | : StringUtils.formatNumber(conn.getContentLengthLong()) |
| 351 | + "bytes"; |
| 352 | tracer.trace("Save to cache (" + size + "): " + originalUrl); |
| 353 | try { |
| 354 | try { |
| 355 | long bytes = cache.save(in, originalUrl); |
| 356 | tracer.trace("Saved to cache: " |
| 357 | + StringUtils.formatNumber(bytes) + "bytes"); |
| 358 | } finally { |
| 359 | in.close(); |
| 360 | } |
| 361 | in = cache.load(originalUrl, true, true); |
| 362 | } catch (IOException e) { |
| 363 | tracer.error(new IOException( |
| 364 | "Cannot save URL to cache, will ignore cache: " |
| 365 | + url, e)); |
| 366 | } |
| 367 | } |
| 368 | |
| 369 | return in; |
| 370 | } catch (IOException e) { |
| 371 | throw new IOException(String.format( |
| 372 | "Cannot find %s (current URL: %s)", originalUrl, url), e); |
| 373 | } |
| 374 | } |
| 375 | |
| 376 | /** |
| 377 | * Open a connection on the given {@link URL}, and manage the cookies that |
| 378 | * come with it. |
| 379 | * |
| 380 | * @param url |
| 381 | * the {@link URL} to open |
| 382 | * |
| 383 | * @return the connection |
| 384 | * |
| 385 | * @throws IOException |
| 386 | * in case of I/O error |
| 387 | */ |
| 388 | private URLConnection openConnectionWithCookies(URL url, |
| 389 | URL currentReferer, Map<String, String> cookiesValues) |
| 390 | throws IOException { |
| 391 | URLConnection conn = url.openConnection(); |
| 392 | |
| 393 | String cookies = generateCookies(cookiesValues); |
| 394 | if (cookies != null && !cookies.isEmpty()) { |
| 395 | conn.setRequestProperty("Cookie", cookies); |
| 396 | } |
| 397 | |
| 398 | conn.setRequestProperty("User-Agent", UA); |
| 399 | conn.setRequestProperty("Accept-Encoding", "gzip"); |
| 400 | conn.setRequestProperty("Accept", "*/*"); |
| 401 | conn.setRequestProperty("Charset", "utf-8"); |
| 402 | |
| 403 | if (currentReferer != null) { |
| 404 | conn.setRequestProperty("Referer", currentReferer.toString()); |
| 405 | conn.setRequestProperty("Host", currentReferer.getHost()); |
| 406 | } |
| 407 | |
| 408 | return conn; |
| 409 | } |
| 410 | |
| 411 | /** |
| 412 | * Generate the cookie {@link String} from the local {@link CookieStore} so |
| 413 | * it is ready to be passed. |
| 414 | * |
| 415 | * @return the cookie |
| 416 | */ |
| 417 | private String generateCookies(Map<String, String> cookiesValues) { |
| 418 | StringBuilder builder = new StringBuilder(); |
| 419 | for (HttpCookie cookie : cookies.getCookieStore().getCookies()) { |
| 420 | if (builder.length() > 0) { |
| 421 | builder.append(';'); |
| 422 | } |
| 423 | |
| 424 | builder.append(cookie.toString()); |
| 425 | } |
| 426 | |
| 427 | if (cookiesValues != null) { |
| 428 | for (Map.Entry<String, String> set : cookiesValues.entrySet()) { |
| 429 | if (builder.length() > 0) { |
| 430 | builder.append(';'); |
| 431 | } |
| 432 | builder.append(set.getKey()); |
| 433 | builder.append('='); |
| 434 | builder.append(set.getValue()); |
| 435 | } |
| 436 | } |
| 437 | |
| 438 | return builder.toString(); |
| 439 | } |
| 440 | } |