Merge branch 'master' of ssh://git@gitorious.org/yacy/rc1.git

Conflicts:
	source/net/yacy/cora/federate/solr/connector/MirrorSolrConnector.java
pull/1/head
Michael Peter Christen 12 years ago
commit a1a4d9aa94

@ -55,8 +55,8 @@
<classpathentry kind="lib" path="lib/log4j-over-slf4j-1.6.4.jar"/>
<classpathentry kind="lib" path="lib/slf4j-api-1.6.4.jar"/>
<classpathentry kind="lib" path="lib/slf4j-jdk14-1.6.4.jar"/>
<classpathentry kind="lib" path="lib/apache-solr-core-4.0.0.jar"/>
<classpathentry kind="lib" path="lib/apache-solr-solrj-4.0.0.jar"/>
<classpathentry kind="lib" path="lib/apache-solr-core-4.0.0.jar" sourcepath="/Users/admin/Development/sourcecode/apache-solr-4.0.0-src/src/java"/>
<classpathentry kind="lib" path="lib/apache-solr-solrj-4.0.0.jar" sourcepath="/Users/admin/Development/sourcecode/apache-solr-4.0.0-src/src/java"/>
<classpathentry kind="lib" path="lib/lucene-analyzers-common-4.0.0.jar"/>
<classpathentry kind="lib" path="lib/lucene-analyzers-phonetic-4.0.0.jar"/>
<classpathentry kind="lib" path="lib/lucene-core-4.0.0.jar"/>

@ -490,6 +490,7 @@ public class CrawlProfile extends ConcurrentHashMap<String, String> implements M
public static String mustMatchFilterFullDomain(final MultiProtocolURI uri) {
String host = uri.getHost();
if (host == null) return uri.getProtocol() + ".*";
if (host.startsWith("www.")) host = host.substring(4);
String protocol = uri.getProtocol();
if ("http".equals(protocol) || "https".equals(protocol)) protocol = "https?+";

@ -236,6 +236,7 @@ public class RobotsTxt {
public void ensureExist(final MultiProtocolURI theURL, final Set<String> thisAgents, boolean concurrent) {
final String urlHostPort = getHostPort(theURL);
if (urlHostPort == null) return;
final BEncodedHeap robotsTable;
try {
robotsTable = this.tables.getHeap(WorkTables.TABLE_ROBOTS_NAME);
@ -371,6 +372,7 @@ public class RobotsTxt {
static final String getHostPort(final MultiProtocolURI theURL) {
final int port = getPort(theURL);
String host = theURL.getHost();
if (host == null) return null;
StringBuilder sb = new StringBuilder(host.length() + 6);
sb.append(host).append(':').append(Integer.toString(port));
return sb.toString();

Loading…
Cancel
Save