Author: ab
Date: Thu Aug 4 14:41:18 2005
New Revision: 227501
URL: http://svn.apache.org/viewcvs?rev=227501&view=rev
Log:
Apply patches in NUTCH-65, submitted by Lutischan Ferenc.
Modified:
lucene/nutch/trunk/src/plugin/index-more/src/java/org/apache/nutch/indexer/more/MoreIndexingFilter.java
Modified:
lucene/nutch/trunk/src/plugin/index-more/src/java/org/apache/nutch/indexer/more/MoreIndexingFilter.java
URL:
http://svn.apache.org/viewcvs/lucene/nutch/trunk/src/plugin/index-more/src/java/org/apache/nutch/indexer/more/MoreIndexingFilter.java?rev=227501&r1=227500&r2=227501&view=diff
==============================================================================
---
lucene/nutch/trunk/src/plugin/index-more/src/java/org/apache/nutch/indexer/more/MoreIndexingFilter.java
(original)
+++
lucene/nutch/trunk/src/plugin/index-more/src/java/org/apache/nutch/indexer/more/MoreIndexingFilter.java
Thu Aug 4 14:41:18 2005
@@ -48,6 +48,7 @@
import java.text.SimpleDateFormat;
import java.util.Date;
+import java.util.Locale;
import java.util.TimeZone;
import java.util.Enumeration;
import java.util.Properties;
@@ -131,14 +132,21 @@
long time = -1;
try {
time = HttpDateFormat.toLong(date);
- } catch (ParseException e) {
+ } catch (ParseException e) {
// try to parse it as date in alternative format
+ String date2 = date;
try {
- DateFormat df = new SimpleDateFormat("EEE MMM dd HH:mm:ss yyyy zzz");
- Date d = df.parse(date);
- time = d.getTime();
+ if (date.length() > 25 ) date2 = date.substring(0, 25);
+ DateFormat df = new SimpleDateFormat("EEE, dd MMM yyyy HH:mm:ss",
Locale.US);
+ time = df.parse(date2).getTime();
} catch (Exception e1) {
- LOG.warning(url+": can't parse erroneous date: "+date);
+ try {
+ if (date.length() > 24 ) date2 = date.substring(0, 24);
+ DateFormat df = new SimpleDateFormat("EEE MMM dd HH:mm:ss yyyy",
Locale.US);
+ time = df.parse(date2).getTime();
+ } catch (Exception e2) {
+ LOG.warning(url + ": can't parse erroneous date: " + date);
+ }
}
}
return time;