Author: kwright
Date: Sat Jul 19 06:17:52 2014
New Revision: 1611848
URL: http://svn.apache.org/r1611848
Log:
Add comments about versioning
Modified:
manifoldcf/trunk/connectors/hdfs/connector/src/main/java/org/apache/manifoldcf/crawler/connectors/hdfs/HDFSRepositoryConnector.java
Modified:
manifoldcf/trunk/connectors/hdfs/connector/src/main/java/org/apache/manifoldcf/crawler/connectors/hdfs/HDFSRepositoryConnector.java
URL:
http://svn.apache.org/viewvc/manifoldcf/trunk/connectors/hdfs/connector/src/main/java/org/apache/manifoldcf/crawler/connectors/hdfs/HDFSRepositoryConnector.java?rev=1611848&r1=1611847&r2=1611848&view=diff
==============================================================================
---
manifoldcf/trunk/connectors/hdfs/connector/src/main/java/org/apache/manifoldcf/crawler/connectors/hdfs/HDFSRepositoryConnector.java
(original)
+++
manifoldcf/trunk/connectors/hdfs/connector/src/main/java/org/apache/manifoldcf/crawler/connectors/hdfs/HDFSRepositoryConnector.java
Sat Jul 19 06:17:52 2014
@@ -350,6 +350,10 @@ public class HDFSRepositoryConnector ext
FileStatus fileStatus = getObject(new Path(documentIdentifier));
if (fileStatus != null) {
if (fileStatus.isDirectory()) {
+ // If HDFS directory modify dates are transitive, as they are on
Unix,
+ // then getting the modify date of the current version is sufficient
+ // to detect any downstream changes we need to be aware of.
+ // (If this turns out to be a bad assumption, this should simply set
rval[i] ="").
long lastModified = fileStatus.getModificationTime();
rval[i] = new Long(lastModified).toString();
} else {
@@ -414,6 +418,8 @@ public class HDFSRepositoryConnector ext
}
if (fileStatus.isDirectory()) {
+ // Since we believe that downstream changes affect the current node's
version string,
+ // then we only have to add references when there are detected changes.
if (!scanOnly[i]) {
activities.noDocument(documentIdentifier,version);
/*