Author: ab
Date: Wed Mar 9 11:06:43 2011
New Revision: 1079746
URL: http://svn.apache.org/viewvc?rev=1079746&view=rev
Log:
NUTCH-872 Change the default fetcher.parse to FALSE.
Modified:
nutch/branches/branch-1.3/CHANGES.txt
nutch/branches/branch-1.3/conf/nutch-default.xml
nutch/branches/branch-1.3/src/java/org/apache/nutch/fetcher/Fetcher.java
Modified: nutch/branches/branch-1.3/CHANGES.txt
URL:
http://svn.apache.org/viewvc/nutch/branches/branch-1.3/CHANGES.txt?rev=1079746&r1=1079745&r2=1079746&view=diff
==============================================================================
--- nutch/branches/branch-1.3/CHANGES.txt (original)
+++ nutch/branches/branch-1.3/CHANGES.txt Wed Mar 9 11:06:43 2011
@@ -2,6 +2,8 @@ Nutch Change Log
Release 1.3 - Current Development
+* NUTCH-872 Change the default fetcher.parse to FALSE (ab)
+
* NUTCH-564 External parser supports encoding attribute (Antony Bowesman,
mattmann)
* NUTCH-964 Upgraded Xerces to 2.91, ERROR conf.Configuration - Failed to set
setXIncludeAware (markus)
Modified: nutch/branches/branch-1.3/conf/nutch-default.xml
URL:
http://svn.apache.org/viewvc/nutch/branches/branch-1.3/conf/nutch-default.xml?rev=1079746&r1=1079745&r2=1079746&view=diff
==============================================================================
--- nutch/branches/branch-1.3/conf/nutch-default.xml (original)
+++ nutch/branches/branch-1.3/conf/nutch-default.xml Wed Mar 9 11:06:43 2011
@@ -653,8 +653,9 @@
<property>
<name>fetcher.parse</name>
- <value>true</value>
- <description>If true, fetcher will parse content.</description>
+ <value>false</value>
+ <description>If true, fetcher will parse content. Default is false, which
means
+ that a separate parsing step is required after fetching is
finished.</description>
</property>
<property>
Modified:
nutch/branches/branch-1.3/src/java/org/apache/nutch/fetcher/Fetcher.java
URL:
http://svn.apache.org/viewvc/nutch/branches/branch-1.3/src/java/org/apache/nutch/fetcher/Fetcher.java?rev=1079746&r1=1079745&r2=1079746&view=diff
==============================================================================
--- nutch/branches/branch-1.3/src/java/org/apache/nutch/fetcher/Fetcher.java
(original)
+++ nutch/branches/branch-1.3/src/java/org/apache/nutch/fetcher/Fetcher.java
Wed Mar 9 11:06:43 2011
@@ -1129,7 +1129,7 @@ public class Fetcher extends Configured
Path segment = new Path(args[0]);
int threads = getConf().getInt("fetcher.threads.fetch", 10);
- boolean parsing = true;
+ boolean parsing = false;
for (int i = 1; i < args.length; i++) { // parse command line
if (args[i].equals("-threads")) { // found -threads option
