fixed scraper

git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@2720 6c8d7289-2bf4-0310-a012-ef5d649a1542

source/de/anomic/htmlFilter/htmlFilterAbstractScraper.java

@@ -304,7 +304,7 @@ public abstract class htmlFilterAbstractScraper implements htmlFilterScraper {
     }
 
     //the 'missing' method that shall be implemented:
-    public abstract void scrapeText(char[] text, boolean insideTag);
+    public abstract void scrapeText(char[] text, String insideTag);
 
     // the other methods must take into account to construct the return value correctly
     public abstract void scrapeTag0(String tagname, Properties tagopts);

source/de/anomic/htmlFilter/htmlFilterContentScraper.java

@@ -119,10 +119,10 @@ public class htmlFilterContentScraper extends htmlFilterAbstractScraper implemen
         return (c == '.') || (c == '!') || (c == '?');
     }
     
-    public void scrapeText(char[] newtext, boolean insideTag) {
+    public void scrapeText(char[] newtext, String insideTag) {
         // System.out.println("SCRAPE: " + new String(newtext));
         serverCharBuffer b = super.stripAll(new serverCharBuffer(newtext, newtext.length + 1)).trim();
-        if (insideTag) {
+        if ((insideTag != null) && (!(insideTag.equals("a")))) {
             // texts inside tags sometimes have no punctuation at the line end
             // this is bad for the text sematics, because it is not possible for the
             // condenser to distinguish headlines from text beginnings.

source/de/anomic/htmlFilter/htmlFilterScraper.java

@@ -48,7 +48,7 @@ public interface htmlFilterScraper {
 
     public boolean isTag1(String tag);
 
-    public void scrapeText(char[] text, boolean insideTag);
+    public void scrapeText(char[] text, String insideTag);
 
     public void scrapeTag0(String tagname, Properties tagopts);
 

source/de/anomic/htmlFilter/htmlFilterWriter.java

@@ -186,7 +186,7 @@ public final class htmlFilterWriter extends Writer {
             // we are not collection tag text
             if (tag == null) {
                 // and this is not a tag opener/closer
-                if (scraper != null) scraper.scrapeText(content, false);
+                if (scraper != null) scraper.scrapeText(content, null);
                 if (transformer != null) return transformer.transformText(content);
                 return content;
             }
@@ -221,7 +221,7 @@ public final class htmlFilterWriter extends Writer {
         // we are collection tag text for the tag 'filterTag'
         if (tag == null) {
             // go on collecting content
-            if (scraper != null) scraper.scrapeText(content, true);
+            if (scraper != null) scraper.scrapeText(content, filterTag);
             if (transformer != null) {
                 filterCont.append(transformer.transformText(content));
             } else {