X-Git-Url: http://git.nikiroo.be/?a=blobdiff_plain;f=src%2Fbe%2Fnikiroo%2Fgofetch%2Fsupport%2FBasicSupport.java;h=f3348e366f1ceeece3c6bec39ea01f031ffe8c95;hb=202173602397b0793542c7a90f9d86013e153067;hp=1db066b3d107c8fefb26218675fa8c5c0767b815;hpb=27008a8782c0ed96e07c8dc39ff0ed1f5163a9d0;p=gofetch.git

diff --git a/src/be/nikiroo/gofetch/support/BasicSupport.java b/src/be/nikiroo/gofetch/support/BasicSupport.java
index 1db066b..f3348e3 100644
--- a/src/be/nikiroo/gofetch/support/BasicSupport.java
+++ b/src/be/nikiroo/gofetch/support/BasicSupport.java
@@ -20,21 +20,103 @@ import be.nikiroo.gofetch.data.Story;
 
 public abstract class BasicSupport {
 	public enum Type {
-		SLASHDOT, PIPEDOT, LWN,
+		SLASHDOT, PIPEDOT, LWN, LEMONDE,
 	}
 
-	public interface QuoteProcessor {
+	/**
+	 * Used to process an element into lines.
+	 * 
+	 * @author niki
+	 */
+	public interface ElementProcessor {
+		/**
+		 * Detect if this node is a quote and should be trated as such.
+		 * 
+		 * @param node
+		 *            the node to check
+		 * @return TRUE if it is
+		 */
 		public boolean detectQuote(Node node);
 
+		/**
+		 * Process text content (will be called on each text element, allowing
+		 * you to modify it if needed).
+		 * 
+		 * @param text
+		 *            the text to process
+		 * @return
+		 */
 		public String processText(String text);
 
+		/**
+		 * Ignore this node.
+		 * 
+		 * @param node
+		 *            the node to ignore
+		 * @return TRUE if it has to be ignored
+		 */
 		public boolean ignoreNode(Node node);
+
+		/**
+		 * Manually process this node (and return the manual processing value)
+		 * if so desired.
+		 * <p>
+		 * If the node is manually processed, it and its children will not be
+		 * automatically processed.
+		 * 
+		 * @param node
+		 *            the node to optionally process
+		 * 
+		 * @return NULL if not processed (will thus be automatically processed
+		 *         as usual), a {@link String} (may be empty) if we process it
+		 *         manually -- the given {@link String} will be used instead of
+		 *         the usual automatic processing if not NULL
+		 */
+		public String manualProcessing(Node node);
+	}
+
+	/**
+	 * A default {@link ElementProcessor} (will not detect or process anything
+	 * manually).
+	 * 
+	 * @author niki
+	 */
+	public class BasicElementProcessor implements ElementProcessor {
+		@Override
+		public boolean detectQuote(Node node) {
+			return false;
+		}
+
+		@Override
+		public String processText(String text) {
+			return text;
+		}
+
+		@Override
+		public boolean ignoreNode(Node node) {
+			return false;
+		}
+
+		@Override
+		public String manualProcessing(Node node) {
+			return null;
+		}
 	}
 
 	static private String preselector;
 
 	private Type type;
 
+	/**
+	 * List all the recent items, but only assure the ID and internal URL to
+	 * fetch it later on (until it has been fetched, the rest of the
+	 * {@link Story} is not confirmed).
+	 * 
+	 * @return the list of new stories
+	 * 
+	 * @throws IOException
+	 *             in case of I/O
+	 */
 	abstract public List<Story> list() throws IOException;
 
 	/**
@@ -71,6 +153,15 @@ public abstract class BasicSupport {
 		BasicSupport.preselector = preselector;
 	}
 
+	/**
+	 * Return a {@link BasicSupport} that is compatible with the given
+	 * {@link Type} if it exists (or NULL if not).
+	 * 
+	 * @param type
+	 *            the type
+	 * 
+	 * @return a compatible {@link BasicSupport} if it exists (or NULL if not)
+	 */
 	static public BasicSupport getSupport(Type type) {
 		BasicSupport support = null;
 
@@ -85,6 +176,9 @@ public abstract class BasicSupport {
 			case LWN:
 				support = new LWN();
 				break;
+			case LEMONDE:
+				support = new LeMonde();
+				break;
 			}
 
 			if (support != null) {
@@ -151,8 +245,20 @@ public abstract class BasicSupport {
 		return new Element("span");
 	}
 
+	/**
+	 * Process the given element into text (each line is a text paragraph and
+	 * can be prepended with ">" signs to indicate a quote or sub-quote or
+	 * sub-sub-quote...).
+	 * 
+	 * @param element
+	 *            the element to process
+	 * @param elementProcessor
+	 *            the element processor, must not be NULL
+	 * 
+	 * @return text lines, each line is a paragraph
+	 */
 	static protected List<String> toLines(Element element,
-			final QuoteProcessor quoteProcessor) {
+			final ElementProcessor elementProcessor) {
 		final List<String> lines = new ArrayList<String>();
 		final StringBuilder currentLine = new StringBuilder();
 		final List<Integer> quoted = new ArrayList<Integer>();
@@ -162,8 +268,18 @@ public abstract class BasicSupport {
 			new NodeTraversor(new NodeVisitor() {
 				@Override
 				public void head(Node node, int depth) {
-					if (quoteProcessor.ignoreNode(node)
-							|| ignoredNodes.contains(node.parentNode())) {
+					String manual = null;
+					boolean ignore = elementProcessor.ignoreNode(node)
+							|| ignoredNodes.contains(node.parentNode());
+					if (!ignore) {
+						manual = elementProcessor.manualProcessing(node);
+						if (manual != null) {
+							currentLine.append(manual);
+							ignore = true;
+						}
+					}
+
+					if (ignore) {
 						ignoredNodes.add(node);
 						return;
 					}
@@ -174,7 +290,7 @@ public abstract class BasicSupport {
 					}
 					prep += " ";
 
-					boolean enterQuote = quoteProcessor.detectQuote(node);
+					boolean enterQuote = elementProcessor.detectQuote(node);
 					boolean leaveQuote = quoted.contains(depth);
 
 					if (enterQuote) {
@@ -209,7 +325,7 @@ public abstract class BasicSupport {
 						String line = StringUtil.normaliseWhitespace(textNode
 								.getWholeText());
 
-						currentLine.append(quoteProcessor.processText(line));
+						currentLine.append(elementProcessor.processText(line));
 						currentLine.append(" ");
 					}
 				}