*) Snippet fetching:

Snippet are now fetched synchronous if the query parameter "fetchSnippet=" 
   is appended to the query string on the yacy search page. This is required 
   for the RSS feed.
   See: http://www.yacy-forum.de/viewtopic.php?t=4051
*) Small changes in the XSLT-stylesheet that is used to generate a html page from
   the RSS feed.

git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@3787 6c8d7289-2bf4-0310-a012-ef5d649a1542

htroot/xml/snippet.java

@@ -25,8 +25,8 @@ public class snippet {
         //get the timeout for snippet-fetching
         int mediasnippet_timeout = 15000;
         int textsnippet_timeout = 10000;
-        mediasnippet_timeout = Integer.parseInt((env.getConfig("timeout_text", "15000")));
+        mediasnippet_timeout = Integer.parseInt(env.getConfig("timeout_text", "15000"));
-        textsnippet_timeout = Integer.parseInt((env.getConfig("timeout_media", "10000")));
+        textsnippet_timeout = Integer.parseInt(env.getConfig("timeout_media", "10000"));
         	
         // getting url
         String urlString = post.get("url", "");

htroot/yacysearch.java

@@ -52,6 +52,7 @@ import java.net.MalformedURLException;
 import java.net.URLEncoder;
 import java.util.HashMap;
 import java.util.Iterator;
+import java.util.Set;
 import java.util.regex.PatternSyntaxException;
 import java.util.TreeSet;
 
@@ -69,6 +70,7 @@ import de.anomic.plasma.plasmaSearchPreOrder;
 import de.anomic.plasma.plasmaSearchQuery;
 import de.anomic.plasma.plasmaSearchRankingProfile;
 import de.anomic.plasma.plasmaSearchTimingProfile;
+import de.anomic.plasma.plasmaSnippetCache;
 import de.anomic.plasma.plasmaSwitchboard;
 import de.anomic.plasma.plasmaURL;
 import de.anomic.plasma.plasmaSearchResults;
@@ -266,9 +268,10 @@ public class yacysearch {
         final boolean globalsearch = (global) && (yacyonline) && (!samesearch);
         
         // do the search
+        TreeSet queryHashes = plasmaCondenser.words2hashes(query[0]);
         plasmaSearchQuery thisSearch = new plasmaSearchQuery(
         			querystring,
-        			plasmaCondenser.words2hashes(query[0]),
+        			queryHashes,
         			plasmaCondenser.words2hashes(query[1]),
                     maxDistance,
                     prefermask,
@@ -338,9 +341,47 @@ public class yacysearch {
                 if (result.hasSnippet()) {
                     prop.put("type_results_" + i + "_snippet", 1);
                     prop.putASIS("type_results_" + i + "_snippet_text", result.getSnippet().getLineMarked(results.getQuery().queryHashes));//FIXME: the ASIS should not be needed, if there is no html in .java
-                } else {
+                } else {                	
-                    prop.put("type_results_" + i + "_snippet", 0);
+                	if (post.containsKey("fetchSnippet")) {
-                    prop.put("type_results_" + i + "_snippet_text", "");
+                		/* fetch the snippet now */
+                        try {
+                        	// snippet fetch timeout
+                        	int textsnippet_timeout = Integer.parseInt(env.getConfig("timeout_media", "10000"));
+                        	
+                        	// boolean line_end_with_punctuation
+                        	boolean pre = post.get("pre", "false").equals("true");
+                        	
+//                        	 if 'remove' is set to true, then RWI references to URLs that do not have the snippet are removed
+                        	boolean remove = post.get("remove", "false").equals("true");
+                        	
+                        	URL resultURL = new URL(result.getUrl());                        	
+                        	plasmaSnippetCache.TextSnippet snippet = sb.snippetCache.retrieveTextSnippet(
+                        			resultURL, 
+                        			queryHashes, 
+                        			true, 
+                        			pre, 
+                        			260, 
+                        			textsnippet_timeout
+                        	);
+                        	                        	
+                            if (snippet.getErrorCode() < 11) {
+                                // no problems occurred
+                                //prop.put("text", (snippet.exists()) ? snippet.getLineMarked(queryHashes) : "unknown");
+                                prop.putASIS("type_results_" + i + "_snippet_text", (snippet.exists()) ? snippet.getLineMarked(queryHashes) : "unknown");
+                            } else {
+                                // problems with snippet fetch
+                               prop.put("type_results_" + i + "_snippet_text", (remove) ? sb.snippetCache.failConsequences(snippet, queryHashes) : snippet.getError());
+                            }          
+                            prop.put("type_results_" + i + "_snippet", 1);
+                        } catch (MalformedURLException e) {
+                    		prop.put("type_results_" + i + "_snippet", 0);
+                    		prop.put("type_results_" + i + "_snippet_text", "");
+                        }
+                	} else {
+                		/* no snippet available (will be fetched later via ajax) */
+                		prop.put("type_results_" + i + "_snippet", 0);
+                		prop.put("type_results_" + i + "_snippet_text", "");
+                	}
                 }
                 prop.put("type_results", results.numResults());
                 prop.put("references", results.getReferences());

htroot/yacysearch.rss

@@ -3,6 +3,10 @@
 <rss version="2.0"
 	xmlns:yacyTopwords="http://www.yacy.net/yacy/topwords"
 	xmlns:opensearch="http://a9.com/-/spec/opensearch/1.1/">
+	<!--
+		YACY P2P WEB SEARCH - Results
+		Hint: append the query-parameter "fetchSnippet=" to embed snippets
+	-->
 	<channel>
 		<title>YaCy P2P-Search for #[former]#</title>
 		<description>Search for #[former]#</description>
@@ -11,21 +15,21 @@
 			<title>Search for #[former]#</title>
 		</image>
 		<opensearch:totalResults>#[type_results]#</opensearch:totalResults>
-    		<opensearch:startIndex>1</opensearch:startIndex>
+		<opensearch:startIndex>1</opensearch:startIndex>
-    		<opensearch:itemsPerPage>#[type_results]#</opensearch:itemsPerPage>
+		<opensearch:itemsPerPage>#[type_results]#</opensearch:itemsPerPage>
-    		<opensearch:link rel="search" href="opensearchdescription.xml" type="application/opensearchdescription+xml"/>
+		<opensearch:link rel="search" href="opensearchdescription.xml" type="application/opensearchdescription+xml"/>
-    		<opensearch:Query role="request" searchTerms="#[former]#" />
+		<opensearch:Query role="request" searchTerms="#[former]#" />
+		
 		#(type)#
-		#{results}#
+		#{results}#<item>
-		<item>
 			<title><![CDATA[#[description]#]]></title>
 			<link>#[url]#</link>
 			#(snippet)#::<description><![CDATA[#[text]#]]></description>#(/snippet)#
 			<pubDate>#[date]#</pubDate>
 			<guid>#[urlhash]#</guid>
 		</item>
-		#{/results}#
+		
-		<yacyTopwords:topwords>
+		#{/results}#<yacyTopwords:topwords>
             #(combine)#
             ::
             #{words}#
@@ -34,7 +38,7 @@
             </yacyTopwords:item>
             #{/words}#
             #(/combine)#
-        </yacyTopwords:topwords>
+		</yacyTopwords:topwords>
         ::
 		#(/type)#
 	</channel>

htroot/yacysearch.xsl

@@ -1,6 +1,7 @@
 <?xml version="1.0" encoding="utf-8" ?>
 <?xml-stylesheet type='text/xsl' href='/rss.xsl' version='1.0'?>
 <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:dc="http://purl.org/dc/elements/1.1/" version="1.0">
+	<xsl:output method="html"/>
 	<xsl:template match='/rss'>
 		<html>
 			<head>
@@ -9,6 +10,7 @@
 				<link rel="shortcut icon" href="favicon.ico" />
 				<style type="text/css">
 					@import "/env/style.css";
+					@import "/env/base.css";
 				</style>
 			</head>
 			<body>
@@ -24,11 +26,12 @@
 	</xsl:template>
 	
 	<xsl:template match='item'>
-		<p>
+		<div class="searchresults">
-			<b><xsl:value-of select='title'/></b><br/>
+			<h4 class="linktitle"><a href="{link}" ><xsl:value-of select='title'/></a></h4>
-			<a href="{link}" ><xsl:value-of select='link' /></a><br/>
+			<p class="snippet"><span class="snippetLoaded"><xsl:value-of select='description'/></span></p>
-			<xsl:value-of select='pubDate' /><br/>
+			<p class="url"><a href="{link}" ><xsl:value-of select='link' /></a></p>
-		</p>
+			<p class="urlinfo"><xsl:value-of select='pubDate' /></p>
+		</div>
 	</xsl:template>
 	
 </xsl:stylesheet>