Author: markus
Date: Tue Oct 23 13:28:32 2012
New Revision: 1401288
URL: http://svn.apache.org/viewvc?rev=1401288&view=rev
Log:
NUTCH-1341 NotModified time set to now but page not modified
Modified:
nutch/trunk/CHANGES.txt
nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDbReducer.java
Modified: nutch/trunk/CHANGES.txt
URL:
http://svn.apache.org/viewvc/nutch/trunk/CHANGES.txt?rev=1401288&r1=1401287&r2=1401288&view=diff
==============================================================================
--- nutch/trunk/CHANGES.txt (original)
+++ nutch/trunk/CHANGES.txt Tue Oct 23 13:28:32 2012
@@ -2,6 +2,8 @@ Nutch Change Log
(trunk) Current Development:
+* NUTCH-1341 NotModified time set to now but page not modified (markus)
+
* NUTCH-1215 UpdateDB should not require segment as input (markus)
* NUTCH-1383 IndexingFiltersChecker to show error message instead of null
pointer exception (snagel)
Modified: nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDbReducer.java
URL:
http://svn.apache.org/viewvc/nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDbReducer.java?rev=1401288&r1=1401287&r2=1401288&view=diff
==============================================================================
--- nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDbReducer.java (original)
+++ nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDbReducer.java Tue Oct 23
13:28:32 2012
@@ -221,6 +221,10 @@ public class CrawlDbReducer implements R
// set the result status and signature
if (modified == FetchSchedule.STATUS_NOTMODIFIED) {
result.setStatus(CrawlDatum.STATUS_DB_NOTMODIFIED);
+
+ // NUTCH-1341 The page is not modified according to its signature,
let's reset lastModified as well
+ result.setModifiedTime(prevModifiedTime);
+
if (oldSet) result.setSignature(old.getSignature());
} else {
switch (fetch.getStatus()) {
