| 1 | package be.nikiroo.gofetch.support; |
| 2 | |
| 3 | import java.io.IOException; |
| 4 | import java.io.InputStream; |
| 5 | import java.net.URL; |
| 6 | import java.net.URLConnection; |
| 7 | import java.util.ArrayList; |
| 8 | import java.util.List; |
| 9 | import java.util.zip.GZIPInputStream; |
| 10 | |
| 11 | import org.jsoup.helper.StringUtil; |
| 12 | import org.jsoup.nodes.Element; |
| 13 | import org.jsoup.nodes.Node; |
| 14 | import org.jsoup.nodes.TextNode; |
| 15 | import org.jsoup.select.Elements; |
| 16 | import org.jsoup.select.NodeTraversor; |
| 17 | import org.jsoup.select.NodeVisitor; |
| 18 | |
| 19 | import be.nikiroo.gofetch.data.Story; |
| 20 | |
| 21 | public abstract class BasicSupport { |
| 22 | public enum Type { |
| 23 | SLASHDOT, PIPEDOT, LWN, LEMONDE, |
| 24 | } |
| 25 | |
| 26 | public interface QuoteProcessor { |
| 27 | public boolean detectQuote(Node node); |
| 28 | |
| 29 | public String processText(String text); |
| 30 | |
| 31 | public boolean ignoreNode(Node node); |
| 32 | |
| 33 | /** |
| 34 | * Manually process this node if so desired. |
| 35 | * |
| 36 | * @param node |
| 37 | * the node to optionally process |
| 38 | * |
| 39 | * @return NULL if not processed, a {@link String} (may be empty) if we |
| 40 | * must not process it any further |
| 41 | */ |
| 42 | public String manualProcessing(Node node); |
| 43 | } |
| 44 | |
| 45 | static private String preselector; |
| 46 | |
| 47 | private Type type; |
| 48 | |
| 49 | /** |
| 50 | * List all the recent items, but only assure the ID and internal URL to |
| 51 | * fetch it later on (until it has been fetched, the rest of the |
| 52 | * {@link Story} is not confirmed). |
| 53 | * |
| 54 | * @return the list of new stories |
| 55 | * |
| 56 | * @throws IOException |
| 57 | * in case of I/O |
| 58 | */ |
| 59 | abstract public List<Story> list() throws IOException; |
| 60 | |
| 61 | /** |
| 62 | * Fetch the full article content as well as all the comments associated to |
| 63 | * this {@link Story}, if any (can be empty, but not NULL). |
| 64 | * |
| 65 | * @param story |
| 66 | * the story to fetch the comments of |
| 67 | * |
| 68 | * @throws IOException |
| 69 | * in case of I/O error |
| 70 | */ |
| 71 | abstract public void fetch(Story story) throws IOException; |
| 72 | |
| 73 | abstract public String getDescription(); |
| 74 | |
| 75 | public String getSelector() { |
| 76 | return getSelector(type); |
| 77 | } |
| 78 | |
| 79 | public Type getType() { |
| 80 | return type; |
| 81 | } |
| 82 | |
| 83 | protected void setType(Type type) { |
| 84 | this.type = type; |
| 85 | } |
| 86 | |
| 87 | /** |
| 88 | * @param preselector |
| 89 | * the preselector to set |
| 90 | */ |
| 91 | static public void setPreselector(String preselector) { |
| 92 | BasicSupport.preselector = preselector; |
| 93 | } |
| 94 | |
| 95 | static public BasicSupport getSupport(Type type) { |
| 96 | BasicSupport support = null; |
| 97 | |
| 98 | if (type != null) { |
| 99 | switch (type) { |
| 100 | case SLASHDOT: |
| 101 | support = new Slashdot(); |
| 102 | break; |
| 103 | case PIPEDOT: |
| 104 | support = new Pipedot(); |
| 105 | break; |
| 106 | case LWN: |
| 107 | support = new LWN(); |
| 108 | break; |
| 109 | case LEMONDE: |
| 110 | support = new LeMonde(); |
| 111 | break; |
| 112 | } |
| 113 | |
| 114 | if (support != null) { |
| 115 | support.setType(type); |
| 116 | } |
| 117 | } |
| 118 | |
| 119 | return support; |
| 120 | } |
| 121 | |
| 122 | static public String getSelector(Type type) { |
| 123 | return preselector + "/" + type + "/"; |
| 124 | } |
| 125 | |
| 126 | // TODO: check Downloader.java? |
| 127 | static protected InputStream open(URL url) throws IOException { |
| 128 | URLConnection conn = url.openConnection(); |
| 129 | conn.connect(); |
| 130 | InputStream in = conn.getInputStream(); |
| 131 | if ("gzip".equals(conn.getContentEncoding())) { |
| 132 | in = new GZIPInputStream(in); |
| 133 | } |
| 134 | |
| 135 | return in; |
| 136 | } |
| 137 | |
| 138 | /** |
| 139 | * Get the first {@link Element} of the given class, or an empty span |
| 140 | * {@link Element} if none found. |
| 141 | * |
| 142 | * @param element |
| 143 | * the element to look in |
| 144 | * @param className |
| 145 | * the class to look for |
| 146 | * |
| 147 | * @return the value or an empty span {@link Element} |
| 148 | */ |
| 149 | static protected Element firstOrEmpty(Element element, String className) { |
| 150 | Elements subElements = element.getElementsByClass(className); |
| 151 | if (subElements.size() > 0) { |
| 152 | return subElements.get(0); |
| 153 | } |
| 154 | |
| 155 | return new Element("span"); |
| 156 | } |
| 157 | |
| 158 | /** |
| 159 | * Get the first {@link Element} of the given tag, or an empty span |
| 160 | * {@link Element} if none found. |
| 161 | * |
| 162 | * @param element |
| 163 | * the element to look in |
| 164 | * @param tagName |
| 165 | * the tag to look for |
| 166 | * |
| 167 | * @return the value or an empty span {@link Element} |
| 168 | */ |
| 169 | static protected Element firstOrEmptyTag(Element element, String tagName) { |
| 170 | Elements subElements = element.getElementsByTag(tagName); |
| 171 | if (subElements.size() > 0) { |
| 172 | return subElements.get(0); |
| 173 | } |
| 174 | |
| 175 | return new Element("span"); |
| 176 | } |
| 177 | |
| 178 | static protected List<String> toLines(Element element, |
| 179 | final QuoteProcessor quoteProcessor) { |
| 180 | final List<String> lines = new ArrayList<String>(); |
| 181 | final StringBuilder currentLine = new StringBuilder(); |
| 182 | final List<Integer> quoted = new ArrayList<Integer>(); |
| 183 | final List<Node> ignoredNodes = new ArrayList<Node>(); |
| 184 | |
| 185 | if (element != null) { |
| 186 | new NodeTraversor(new NodeVisitor() { |
| 187 | @Override |
| 188 | public void head(Node node, int depth) { |
| 189 | String manual = null; |
| 190 | boolean ignore = quoteProcessor.ignoreNode(node) |
| 191 | || ignoredNodes.contains(node.parentNode()); |
| 192 | if (!ignore) { |
| 193 | manual = quoteProcessor.manualProcessing(node); |
| 194 | if (manual != null) { |
| 195 | currentLine.append(manual); |
| 196 | ignore = true; |
| 197 | } |
| 198 | } |
| 199 | |
| 200 | if (ignore) { |
| 201 | ignoredNodes.add(node); |
| 202 | return; |
| 203 | } |
| 204 | |
| 205 | String prep = ""; |
| 206 | for (int i = 0; i < quoted.size(); i++) { |
| 207 | prep += ">"; |
| 208 | } |
| 209 | prep += " "; |
| 210 | |
| 211 | boolean enterQuote = quoteProcessor.detectQuote(node); |
| 212 | boolean leaveQuote = quoted.contains(depth); |
| 213 | |
| 214 | if (enterQuote) { |
| 215 | quoted.add(depth); |
| 216 | } |
| 217 | |
| 218 | if (leaveQuote) { |
| 219 | quoted.remove(Integer.valueOf(depth)); |
| 220 | } |
| 221 | |
| 222 | if (enterQuote || leaveQuote) { |
| 223 | if (currentLine.length() > 0) { |
| 224 | if (currentLine.charAt(currentLine.length() - 1) == '\n') { |
| 225 | currentLine.setLength(currentLine.length() - 1); |
| 226 | } |
| 227 | for (String l : currentLine.toString().split("\n")) { |
| 228 | lines.add(prep + l); |
| 229 | } |
| 230 | } |
| 231 | currentLine.setLength(0); |
| 232 | } |
| 233 | |
| 234 | if (node instanceof Element) { |
| 235 | Element element = (Element) node; |
| 236 | boolean block = element.isBlock() |
| 237 | || element.tagName().equalsIgnoreCase("br"); |
| 238 | if (block && currentLine.length() > 0) { |
| 239 | currentLine.append("\n"); |
| 240 | } |
| 241 | } else if (node instanceof TextNode) { |
| 242 | TextNode textNode = (TextNode) node; |
| 243 | String line = StringUtil.normaliseWhitespace(textNode |
| 244 | .getWholeText()); |
| 245 | |
| 246 | currentLine.append(quoteProcessor.processText(line)); |
| 247 | currentLine.append(" "); |
| 248 | } |
| 249 | } |
| 250 | |
| 251 | @Override |
| 252 | public void tail(Node node, int depth) { |
| 253 | } |
| 254 | }).traverse(element); |
| 255 | } |
| 256 | |
| 257 | if (currentLine.length() > 0) { |
| 258 | String prep = ""; |
| 259 | for (int i = 0; i < quoted.size(); i++) { |
| 260 | prep += ">"; |
| 261 | } |
| 262 | prep += " "; |
| 263 | if (currentLine.length() > 0) { |
| 264 | if (currentLine.charAt(currentLine.length() - 1) == '\n') { |
| 265 | currentLine.setLength(currentLine.length() - 1); |
| 266 | } |
| 267 | for (String l : currentLine.toString().split("\n")) { |
| 268 | lines.add(prep + l); |
| 269 | } |
| 270 | } |
| 271 | } |
| 272 | |
| 273 | for (int i = 0; i < lines.size(); i++) { |
| 274 | lines.set(i, lines.get(i).replace(" ", " ").trim()); |
| 275 | } |
| 276 | |
| 277 | return lines; |
| 278 | } |
| 279 | } |