You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@manifoldcf.apache.org by kw...@apache.org on 2014/06/05 14:51:43 UTC
svn commit: r1600639 - in
/manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents:
incrementalingest/IncrementalIngester.java
interfaces/DocumentIngestStatus.java interfaces/IIncrementalIngester.java
Author: kwright
Date: Thu Jun 5 12:51:43 2014
New Revision: 1600639
URL: http://svn.apache.org/r1600639
Log:
More IIncrementalIngester modifications
Modified:
manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/incrementalingest/IncrementalIngester.java
manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/DocumentIngestStatus.java
manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/IIncrementalIngester.java
Modified: manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/incrementalingest/IncrementalIngester.java
URL: http://svn.apache.org/viewvc/manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/incrementalingest/IncrementalIngester.java?rev=1600639&r1=1600638&r2=1600639&view=diff
==============================================================================
--- manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/incrementalingest/IncrementalIngester.java (original)
+++ manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/incrementalingest/IncrementalIngester.java Thu Jun 5 12:51:43 2014
@@ -555,6 +555,54 @@ public class IncrementalIngester extends
IOutputActivity activities)
throws ManifoldCFException, ServiceInterruption
{
+ return documentIngest(new String[0],
+ outputConnectionName,
+ identifierClass, identifierHash,
+ documentVersion,
+ new String[0],
+ outputVersion,
+ parameterVersion,
+ authorityName,
+ data,
+ ingestTime, documentURI,
+ activities);
+ }
+
+ /** Ingest a document.
+ * This ingests the document, and notes it. If this is a repeat ingestion of the document, this
+ * method also REMOVES ALL OLD METADATA. When complete, the index will contain only the metadata
+ * described by the RepositoryDocument object passed to this method.
+ * ServiceInterruption is thrown if the document ingestion must be rescheduled.
+ *@param transformationConnectionNames are the names of the transformation connections associated with this action.
+ *@param outputConnectionName is the name of the output connection associated with this action.
+ *@param identifierClass is the name of the space in which the identifier hash should be interpreted.
+ *@param identifierHash is the hashed document identifier.
+ *@param documentVersion is the document version.
+ *@param parameterVersion is the forced parameter version.
+ *@param transformationVersions are the version strings for the transformations to be performed on the document.
+ *@param outputVersion is the output version string constructed from the output specification by the output connector.
+ *@param authorityName is the name of the authority associated with the document, if any.
+ *@param data is the document data. The data is closed after ingestion is complete.
+ *@param ingestTime is the time at which the ingestion took place, in milliseconds since epoch.
+ *@param documentURI is the URI of the document, which will be used as the key of the document in the index.
+ *@param activities is an object providing a set of methods that the implementer can use to perform the operation.
+ *@return true if the ingest was ok, false if the ingest is illegal (and should not be repeated).
+ */
+ public boolean documentIngest(
+ String[] transformationConnectionNames,
+ String outputConnectionName,
+ String identifierClass, String identifierHash,
+ String documentVersion,
+ String[] transformationVersions,
+ String outputVersion,
+ String parameterVersion,
+ String authorityName,
+ RepositoryDocument data,
+ long ingestTime, String documentURI,
+ IOutputActivity activities)
+ throws ManifoldCFException, ServiceInterruption
+ {
+ // MHL
IOutputConnection connection = connectionManager.load(outputConnectionName);
String docKey = makeKey(identifierClass,identifierHash);
@@ -1803,7 +1851,7 @@ public class IncrementalIngester extends
String lastOutputVersion = (String)row.getValue(lastOutputVersionField);
String authorityName = (String)row.getValue(authorityNameField);
String paramVersion = (String)row.getValue(forcedParamsField);
- rval[position.intValue()] = new DocumentIngestStatus(lastVersion,lastOutputVersion,authorityName,paramVersion);
+ rval[position.intValue()] = new DocumentIngestStatus(lastVersion,new String[0],lastOutputVersion,authorityName,paramVersion);
}
}
}
Modified: manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/DocumentIngestStatus.java
URL: http://svn.apache.org/viewvc/manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/DocumentIngestStatus.java?rev=1600639&r1=1600638&r2=1600639&view=diff
==============================================================================
--- manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/DocumentIngestStatus.java (original)
+++ manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/DocumentIngestStatus.java Thu Jun 5 12:51:43 2014
@@ -32,15 +32,18 @@ public class DocumentIngestStatus
public static final String _rcsid = "@(#)$Id: DocumentIngestStatus.java 988245 2010-08-23 18:39:35Z kwright $";
protected final String documentVersionString;
+ protected final String[] transformationVersionStrings;
protected final String outputVersionString;
protected final String documentAuthorityNameString;
protected final String parameterVersionString;
/** Constructor */
- public DocumentIngestStatus(String documentVersionString, String outputVersionString,
+ public DocumentIngestStatus(String documentVersionString,
+ String[] transformationVersionStrings, String outputVersionString,
String documentAuthorityNameString, String parameterVersionString)
{
this.documentVersionString = documentVersionString;
+ this.transformationVersionStrings = transformationVersionStrings;
this.outputVersionString = outputVersionString;
this.documentAuthorityNameString = documentAuthorityNameString;
this.parameterVersionString = parameterVersionString;
@@ -52,6 +55,12 @@ public class DocumentIngestStatus
return documentVersionString;
}
+ /** Get the transformation version strings */
+ public String[] getTransformationVersions()
+ {
+ return transformationVersionStrings;
+ }
+
/** Get the output version */
public String getOutputVersion()
{
Modified: manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/IIncrementalIngester.java
URL: http://svn.apache.org/viewvc/manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/IIncrementalIngester.java?rev=1600639&r1=1600638&r2=1600639&view=diff
==============================================================================
--- manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/IIncrementalIngester.java (original)
+++ manifoldcf/branches/CONNECTORS-946/framework/agents/src/main/java/org/apache/manifoldcf/agents/interfaces/IIncrementalIngester.java Thu Jun 5 12:51:43 2014
@@ -252,6 +252,40 @@ public interface IIncrementalIngester
IOutputActivity activities)
throws ManifoldCFException, ServiceInterruption;
+ /** Ingest a document.
+ * This ingests the document, and notes it. If this is a repeat ingestion of the document, this
+ * method also REMOVES ALL OLD METADATA. When complete, the index will contain only the metadata
+ * described by the RepositoryDocument object passed to this method.
+ * ServiceInterruption is thrown if the document ingestion must be rescheduled.
+ *@param transformationConnectionNames are the names of the transformation connections associated with this action.
+ *@param outputConnectionName is the name of the output connection associated with this action.
+ *@param identifierClass is the name of the space in which the identifier hash should be interpreted.
+ *@param identifierHash is the hashed document identifier.
+ *@param documentVersion is the document version.
+ *@param parameterVersion is the forced parameter version.
+ *@param transformationVersions are the version strings for the transformations to be performed on the document.
+ *@param outputVersion is the output version string constructed from the output specification by the output connector.
+ *@param authorityName is the name of the authority associated with the document, if any.
+ *@param data is the document data. The data is closed after ingestion is complete.
+ *@param ingestTime is the time at which the ingestion took place, in milliseconds since epoch.
+ *@param documentURI is the URI of the document, which will be used as the key of the document in the index.
+ *@param activities is an object providing a set of methods that the implementer can use to perform the operation.
+ *@return true if the ingest was ok, false if the ingest is illegal (and should not be repeated).
+ */
+ public boolean documentIngest(
+ String[] transformationConnectionNames,
+ String outputConnectionName,
+ String identifierClass, String identifierHash,
+ String documentVersion,
+ String[] transformationVersions,
+ String outputVersion,
+ String parameterVersion,
+ String authorityName,
+ RepositoryDocument data,
+ long ingestTime, String documentURI,
+ IOutputActivity activities)
+ throws ManifoldCFException, ServiceInterruption;
+
/** Note the fact that we checked a document (and found that it did not need to be ingested, because the
* versions agreed).
*@param outputConnectionName is the name of the output connection associated with this action.