You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@nutch.apache.org by ma...@apache.org on 2015/04/17 20:04:52 UTC
svn commit: r1674371 - in /nutch/trunk: CHANGES.txt
src/java/org/apache/nutch/util/domain/DomainStatistics.java
Author: mattmann
Date: Fri Apr 17 18:04:52 2015
New Revision: 1674371
URL: http://svn.apache.org/r1674371
Log:
fix for NUTCH-1911 Imeprove DomainStatistics tool command line parsing contributed by Michael Joyce <ml...@gmail.com> this closes #21.
Modified:
nutch/trunk/CHANGES.txt
nutch/trunk/src/java/org/apache/nutch/util/domain/DomainStatistics.java
Modified: nutch/trunk/CHANGES.txt
URL: http://svn.apache.org/viewvc/nutch/trunk/CHANGES.txt?rev=1674371&r1=1674370&r2=1674371&view=diff
==============================================================================
--- nutch/trunk/CHANGES.txt (original)
+++ nutch/trunk/CHANGES.txt Fri Apr 17 18:04:52 2015
@@ -2,6 +2,8 @@ Nutch Change Log
Nutch Current Development 1.10-SNAPSHOT
+* NUTCH-1911 Imeprove DomainStatistics tool command line parsing (Michael Joyce via mattmann)
+
* NUTCH-1981 Upgrade to icu4j 55.1 (Marko Asplund via snagel)
* NUTCH-1960 JUnit test for dump method of CommonCrawlDataDumper (Giuseppe Totaro via mattmann)
Modified: nutch/trunk/src/java/org/apache/nutch/util/domain/DomainStatistics.java
URL: http://svn.apache.org/viewvc/nutch/trunk/src/java/org/apache/nutch/util/domain/DomainStatistics.java?rev=1674371&r1=1674370&r2=1674371&view=diff
==============================================================================
--- nutch/trunk/src/java/org/apache/nutch/util/domain/DomainStatistics.java (original)
+++ nutch/trunk/src/java/org/apache/nutch/util/domain/DomainStatistics.java Fri Apr 17 18:04:52 2015
@@ -67,8 +67,20 @@ public class DomainStatistics extends Co
public int run(String[] args) throws Exception {
if (args.length < 3) {
- System.out
- .println("usage: DomainStatistics inputDirs outDir host|domain|suffix|tld [numOfReducer]");
+ System.err.println("Usage: DomainStatistics inputDirs outDir mode [numOfReducer]");
+
+ System.err.println("\tinputDirs\tComma separated list of crawldb input directories");
+ System.err.println("\t\t\tE.g.: crawl/crawldb/current/");
+
+ System.err.println("\toutDir\t\tOutput directory where results should be dumped");
+
+ System.err.println("\tmode\t\tSet statistics gathering mode");
+ System.err.println("\t\t\t\thost\tGather statistics by host");
+ System.err.println("\t\t\t\tdomain\tGather statistics by domain");
+ System.err.println("\t\t\t\tsuffix\tGather statistics by suffix");
+ System.err.println("\t\t\t\ttld\tGather statistics by top level directory");
+
+ System.err.println("\t[numOfReducers]\tOptional number of reduce jobs to use. Defaults to 1.");
return 1;
}
String inputDir = args[0];