| 1 | package be.nikiroo.utils; |
| 2 | |
| 3 | import java.io.IOException; |
| 4 | import java.io.InputStream; |
| 5 | import java.io.OutputStreamWriter; |
| 6 | import java.net.CookieHandler; |
| 7 | import java.net.CookieManager; |
| 8 | import java.net.CookiePolicy; |
| 9 | import java.net.CookieStore; |
| 10 | import java.net.HttpCookie; |
| 11 | import java.net.HttpURLConnection; |
| 12 | import java.net.URL; |
| 13 | import java.net.URLConnection; |
| 14 | import java.net.URLEncoder; |
| 15 | import java.util.Map; |
| 16 | import java.util.zip.GZIPInputStream; |
| 17 | |
| 18 | /** |
| 19 | * This class will help you download content from Internet Sites ({@link URL} |
| 20 | * based). |
| 21 | * <p> |
| 22 | * It allows you to control some options often required on web sites that do not |
| 23 | * want to simply serve HTML, but actively makes your life difficult with stupid |
| 24 | * checks. |
| 25 | * |
| 26 | * @author niki |
| 27 | */ |
| 28 | public class Downloader { |
| 29 | private String UA; |
| 30 | private CookieManager cookies; |
| 31 | |
| 32 | /** |
| 33 | * Create a new {@link Downloader}. |
| 34 | * |
| 35 | * @param UA |
| 36 | * the User-Agent to use to download the resources -- note that |
| 37 | * some websites require one, some actively blacklist real UAs |
| 38 | * like the one from wget, some whitelist a couple of browsers |
| 39 | * only (!) |
| 40 | */ |
| 41 | public Downloader(String UA) { |
| 42 | this.UA = UA; |
| 43 | |
| 44 | cookies = new CookieManager(); |
| 45 | cookies.setCookiePolicy(CookiePolicy.ACCEPT_ALL); |
| 46 | CookieHandler.setDefault(cookies); |
| 47 | } |
| 48 | |
| 49 | /** |
| 50 | * Clear all the cookies currently in the jar. |
| 51 | * <p> |
| 52 | * As long as you don't, the cookies are kept. |
| 53 | */ |
| 54 | public void clearCookies() { |
| 55 | cookies.getCookieStore().removeAll(); |
| 56 | } |
| 57 | |
| 58 | /** |
| 59 | * Open the given {@link URL} and update the cookies. |
| 60 | * |
| 61 | * @param url |
| 62 | * the {@link URL} to open |
| 63 | * @return the {@link InputStream} of the opened page |
| 64 | * |
| 65 | * @throws IOException |
| 66 | * in case of I/O error |
| 67 | **/ |
| 68 | public InputStream open(URL url) throws IOException { |
| 69 | return open(url, url, url, null, null, null, null); |
| 70 | } |
| 71 | |
| 72 | /** |
| 73 | * Open the given {@link URL} and update the cookies. |
| 74 | * |
| 75 | * @param url |
| 76 | * the {@link URL} to open |
| 77 | * @param postParams |
| 78 | * the POST parameters |
| 79 | * @param getParams |
| 80 | * the GET parameters (priority over POST) |
| 81 | * @param oauth |
| 82 | * OAuth authorization (aka, "bearer XXXXXXX") |
| 83 | * |
| 84 | * @return the {@link InputStream} of the opened page |
| 85 | * |
| 86 | * @throws IOException |
| 87 | * in case of I/O error |
| 88 | */ |
| 89 | public InputStream open(URL url, URL currentReferer, |
| 90 | Map<String, String> cookiesValues, Map<String, String> postParams, |
| 91 | Map<String, String> getParams, String oauth) throws IOException { |
| 92 | return open(url, url, currentReferer, cookiesValues, postParams, |
| 93 | getParams, oauth); |
| 94 | } |
| 95 | |
| 96 | /** |
| 97 | * Trace information (info/error) generated by this class. |
| 98 | * <p> |
| 99 | * You can override it if you don't want the default sysout/syserr. |
| 100 | * |
| 101 | * @param message |
| 102 | * the message |
| 103 | * @param error |
| 104 | * TRUE for error messages, FALSE for information messages |
| 105 | */ |
| 106 | protected void trace(String message, boolean error) { |
| 107 | if (error) { |
| 108 | System.err.println(message); |
| 109 | } else { |
| 110 | System.out.println(message); |
| 111 | } |
| 112 | } |
| 113 | |
| 114 | /** |
| 115 | * Open the given {@link URL} and update the cookies. |
| 116 | * |
| 117 | * @param url |
| 118 | * the {@link URL} to open |
| 119 | * @param originalUrl |
| 120 | * the original {@link URL} before any redirection occurs |
| 121 | * @param postParams |
| 122 | * the POST parameters |
| 123 | * @param getParams |
| 124 | * the GET parameters (priority over POST) |
| 125 | * @param oauth |
| 126 | * OAuth authorisation (aka, "bearer XXXXXXX") |
| 127 | * @return the {@link InputStream} of the opened page |
| 128 | * |
| 129 | * @throws IOException |
| 130 | * in case of I/O error |
| 131 | */ |
| 132 | private InputStream open(URL url, final URL originalUrl, |
| 133 | URL currentReferer, Map<String, String> cookiesValues, |
| 134 | Map<String, String> postParams, Map<String, String> getParams, |
| 135 | String oauth) throws IOException { |
| 136 | |
| 137 | trace("Download: " + url, false); |
| 138 | |
| 139 | URLConnection conn = openConnectionWithCookies(url, currentReferer, |
| 140 | cookiesValues); |
| 141 | |
| 142 | // Priority: GET over POST |
| 143 | Map<String, String> params = getParams; |
| 144 | if (getParams == null) { |
| 145 | params = postParams; |
| 146 | } |
| 147 | |
| 148 | if ((params != null || oauth != null) |
| 149 | && conn instanceof HttpURLConnection) { |
| 150 | StringBuilder requestData = null; |
| 151 | if (params != null) { |
| 152 | requestData = new StringBuilder(); |
| 153 | for (Map.Entry<String, String> param : params.entrySet()) { |
| 154 | if (requestData.length() != 0) |
| 155 | requestData.append('&'); |
| 156 | requestData.append(URLEncoder.encode(param.getKey(), |
| 157 | "UTF-8")); |
| 158 | requestData.append('='); |
| 159 | requestData.append(URLEncoder.encode( |
| 160 | String.valueOf(param.getValue()), "UTF-8")); |
| 161 | } |
| 162 | |
| 163 | conn.setDoOutput(true); |
| 164 | |
| 165 | if (getParams == null && postParams != null) { |
| 166 | ((HttpURLConnection) conn).setRequestMethod("POST"); |
| 167 | } |
| 168 | |
| 169 | conn.setRequestProperty("Content-Type", |
| 170 | "application/x-www-form-urlencoded"); |
| 171 | conn.setRequestProperty("charset", "utf-8"); |
| 172 | } |
| 173 | |
| 174 | if (oauth != null) { |
| 175 | conn.setRequestProperty("Authorization", oauth); |
| 176 | } |
| 177 | |
| 178 | if (requestData != null) { |
| 179 | OutputStreamWriter writer = new OutputStreamWriter( |
| 180 | conn.getOutputStream()); |
| 181 | |
| 182 | writer.write(requestData.toString()); |
| 183 | writer.flush(); |
| 184 | writer.close(); |
| 185 | } |
| 186 | } |
| 187 | |
| 188 | conn.connect(); |
| 189 | |
| 190 | // Check if redirect |
| 191 | if (conn instanceof HttpURLConnection |
| 192 | && ((HttpURLConnection) conn).getResponseCode() / 100 == 3) { |
| 193 | String newUrl = conn.getHeaderField("Location"); |
| 194 | return open(new URL(newUrl), originalUrl, currentReferer, |
| 195 | cookiesValues, postParams, getParams, oauth); |
| 196 | } |
| 197 | |
| 198 | InputStream in = conn.getInputStream(); |
| 199 | if ("gzip".equals(conn.getContentEncoding())) { |
| 200 | in = new GZIPInputStream(in); |
| 201 | } |
| 202 | |
| 203 | return in; |
| 204 | } |
| 205 | |
| 206 | /** |
| 207 | * Open a connection on the given {@link URL}, and manage the cookies that |
| 208 | * come with it. |
| 209 | * |
| 210 | * @param url |
| 211 | * the {@link URL} to open |
| 212 | * |
| 213 | * @return the connection |
| 214 | * |
| 215 | * @throws IOException |
| 216 | * in case of I/O error |
| 217 | */ |
| 218 | private URLConnection openConnectionWithCookies(URL url, |
| 219 | URL currentReferer, Map<String, String> cookiesValues) |
| 220 | throws IOException { |
| 221 | URLConnection conn = url.openConnection(); |
| 222 | |
| 223 | conn.setRequestProperty("User-Agent", UA); |
| 224 | conn.setRequestProperty("Cookie", generateCookies(cookiesValues)); |
| 225 | conn.setRequestProperty("Accept-Encoding", "gzip"); |
| 226 | if (currentReferer != null) { |
| 227 | conn.setRequestProperty("Referer", currentReferer.toString()); |
| 228 | conn.setRequestProperty("Host", currentReferer.getHost()); |
| 229 | } |
| 230 | |
| 231 | return conn; |
| 232 | } |
| 233 | |
| 234 | /** |
| 235 | * Generate the cookie {@link String} from the local {@link CookieStore} so |
| 236 | * it is ready to be passed. |
| 237 | * |
| 238 | * @return the cookie |
| 239 | */ |
| 240 | private String generateCookies(Map<String, String> cookiesValues) { |
| 241 | StringBuilder builder = new StringBuilder(); |
| 242 | for (HttpCookie cookie : cookies.getCookieStore().getCookies()) { |
| 243 | if (builder.length() > 0) { |
| 244 | builder.append(';'); |
| 245 | } |
| 246 | |
| 247 | // TODO: check if format is ok |
| 248 | builder.append(cookie.toString()); |
| 249 | } |
| 250 | |
| 251 | if (cookiesValues != null) { |
| 252 | for (Map.Entry<String, String> set : cookiesValues.entrySet()) { |
| 253 | if (builder.length() > 0) { |
| 254 | builder.append(';'); |
| 255 | } |
| 256 | builder.append(set.getKey()); |
| 257 | builder.append('='); |
| 258 | builder.append(set.getValue()); |
| 259 | } |
| 260 | } |
| 261 | |
| 262 | return builder.toString(); |
| 263 | } |
| 264 | } |