Initial commit that parses exported or storage foxml.

.gitignore

@@ -0,0 +1,2 @@
+target/
+.idea

pom.xml

@@ -0,0 +1,64 @@
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+  xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+  <modelVersion>4.0.0</modelVersion>
+
+  <groupId>org.fcrepo.migration</groupId>
+  <artifactId>migration-utils</artifactId>
+  <version>1.0-SNAPSHOT</version>
+  <packaging>jar</packaging>
+
+  <name>migration-utils</name>
+  <url>http://maven.apache.org</url>
+
+  <properties>
+    <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
+  </properties>
+
+  <dependencies>
+    <dependency>
+      <groupId>commons-io</groupId>
+      <artifactId>commons-io</artifactId>
+      <version>2.4</version>
+    </dependency>
+    <dependency>
+      <groupId>commons-codec</groupId>
+      <artifactId>commons-codec</artifactId>
+      <version>1.10</version>
+    </dependency>
+    <dependency>
+      <groupId>org.apache.lucene</groupId>
+      <artifactId>lucene-core</artifactId>
+      <version>4.10.3</version>
+    </dependency>
+    <dependency>
+      <groupId>org.apache.lucene</groupId>
+      <artifactId>lucene-analyzers-common</artifactId>
+      <version>4.10.3</version>
+    </dependency>
+    <dependency>
+      <groupId>org.apache.httpcomponents</groupId>
+      <artifactId>httpclient</artifactId>
+      <version>4.3.6</version>
+    </dependency>
+
+    <dependency>
+      <groupId>junit</groupId>
+      <artifactId>junit</artifactId>
+      <version>4.8.1</version>
+      <scope>test</scope>
+    </dependency>
+  </dependencies>
+
+  <build>
+    <plugins>
+      <plugin>
+        <groupId>org.apache.maven.plugins</groupId>
+        <artifactId>maven-compiler-plugin</artifactId>
+        <configuration>
+          <source>1.7</source>
+          <target>1.7</target>
+        </configuration>
+      </plugin>
+    </plugins>
+  </build>
+</project>

readme.md

@@ -0,0 +1,39 @@
+A framework to support migration of data from Fedora 3 to Fedora 4 repositories.
+
+# Overview
+
+The basic program allows for a configuration to define one or more Fedora Object Handlers and
+a repository source.  The handlers will in turn be provided information about each each object
+in the repository under the theory that one ore more Handler implementations may be written to
+achieve whatever complex data migration or analysis is desired.
+
+# Status
+
+Currently this application is far from ready for use, but the basic framework is presented to allow:
+
+1.  testing to ensure that it does indeed completely and accurately parse FOXML from various fedora versions and contexts
+2.  the framework to be discussed and improved before complex mappings and migration scenarios are implemented
+
+# Usage
+
+To get basic output for a directory of FOXML that have been produced using the REST API's export with the "archive" context:
+
+``` mvn clean compile exec:java -Dexec.mainClass=org.fcrepo.migration.Migrator -Dexec.args="path/to/exported/foxml" ```
+
+To get basic output for a directory of FOXML read from fedora's storage:
+
+``` mvn clean compile exec:java -Dexec.mainClass=org.fcrepo.migration.Migrator -Dexec.args="path/to/fedora/data/objectStore path/to/fedora/data/datastreamStore work/directory" ```
+
+# Development
+
+To do something more sophisticated (like actually migrating or analyzing content) implement
+FedoraObjectHandler and update Migrator before running.
+
+# Additional Documentation
+ * [wiki](https://wiki.duraspace.org/display/FF/Fedora+3+to+4+Data+Migration)
+
+
+
+
+
+

src/main/java/org/fcrepo/migration/ContentDigest.java

@@ -0,0 +1,18 @@
+package org.fcrepo.migration;
+
+/**
+ * An interface defining access to information about a fedora datastream's
+ * content digest.
+ */
+public interface ContentDigest {
+
+    /**
+     * Gets the type: one of several defined in the fedora foxml schema.
+     */
+    public String getType();
+
+    /**
+     * Gets the value of the content digest.
+     */
+    public String getDigest();
+}

src/main/java/org/fcrepo/migration/DatastreamInfo.java

@@ -0,0 +1,40 @@
+package org.fcrepo.migration;
+
+/**
+ * An interface defining access to information about a fedora datastream.
+ */
+public interface DatastreamInfo {
+
+    /**
+     * Gets the information about the object to which this datastream
+     * belongs.
+     */
+    public ObjectInfo getObjectInfo();
+
+    /**
+     * Gets the identifier for this datastream (Unique within an object).
+     */
+    public String getDatastreamId();
+
+    /**
+     * Gets the control group for this datastream.  This is expected to be
+     * "M", "X", "R" or "E".
+     */
+    public String getControlGroup();
+
+    /**
+     * Gets the fedora URI for this datastream.
+     */
+    public String getFedoraURI();
+
+    /**
+     * Gets the state for this datastream.  This is expected to be "A", "I" or
+     * "D".
+     */
+    public String getState();
+
+    /**
+     * Returns true if this datastream was/is versionable.
+     */
+    public boolean getVersionable();
+}

src/main/java/org/fcrepo/migration/DatastreamVersion.java

@@ -0,0 +1,69 @@
+package org.fcrepo.migration;
+
+import java.io.IOException;
+import java.io.InputStream;
+
+/**
+ * An interface defining access to information about a version of a
+ * fedora datastream.
+ */
+public interface DatastreamVersion {
+
+    /**
+     * Gets the information about the datastream for which this is
+     * a version.  (which in turn can be queried to get information about
+     * the object).
+     */
+    public DatastreamInfo getDatastreamInfo();
+
+    /**
+     * Gets the id for this version.
+     */
+    public String getVersionId();
+
+    /**
+     * Gets the mime type for this version.
+     */
+    public String getMimeType();
+
+    /**
+     * Gets the label for this version.
+     */
+    public String getLabel();
+
+    /**
+     * Gets the date when this version was created.
+     */
+    public String getCreated();
+
+    /**
+     * Gets the altIDs value for this version.
+     */
+    public String getAltIds();
+
+    /**
+     * Gets the format URI for this version.
+     */
+    public String getFormatUri();
+
+    /**
+     * Gets the size (in bytes) for the content of this datastream
+     * version.
+     */
+    public long getSize();
+
+    /**
+     * Gets the content digest (if available) for this version.
+     */
+    public ContentDigest getContentDigest();
+
+    /**
+     * Gets access to the content of this datastream.  When text, the
+     * encoding can be expected to be UTF-8.
+     * @throws IllegalStateException if invoked outside of the call
+     *         to @{link FedoraObjectHandler#processDatastreamVersion}
+     * @throws IOException when unable to access the stream
+     */
+    public InputStream getContent() throws IOException;
+
+}

src/main/java/org/fcrepo/migration/DefaultContentDigest.java

@@ -0,0 +1,27 @@
+package org.fcrepo.migration;
+
+/**
+ * A default implementation of ContentDigest that accepts
+ * values at construction time.
+ */
+public class DefaultContentDigest implements ContentDigest {
+
+    private String type;
+
+    private String digest;
+
+    public DefaultContentDigest(String type, String digest) {
+        this.type = type;
+        this.digest = digest;
+    }
+
+    @Override
+    public String getType() {
+        return type;
+    }
+
+    @Override
+    public String getDigest() {
+        return digest;
+    }
+}

src/main/java/org/fcrepo/migration/DefaultObjectInfo.java

@@ -0,0 +1,30 @@
+package org.fcrepo.migration;
+
+import java.io.InputStream;
+
+/**
+ * A default implementation of ObjectInfo that accepts
+ * values at construction time.
+ */
+public class DefaultObjectInfo implements ObjectInfo {
+
+    private String pid;
+
+    private String uri;
+
+    public DefaultObjectInfo(String pid, String uri) {
+        this.pid = pid;
+        this.uri = uri;
+    }
+
+    @Override
+    public String getPid() {
+        return pid;
+    }
+
+    @Override
+    public String getFedoraURI() {
+        return uri;
+    }
+
+}

src/main/java/org/fcrepo/migration/FedoraObjectHandler.java

@@ -0,0 +1,44 @@
+package org.fcrepo.migration;
+
+import java.io.IOException;
+
+/**
+ * An interface with methods that are meant to be invoked when processing
+ * a fedora 3 object such that every bit of information in that fedora 3
+ * object is exposed to the instance implementing this interface.
+ *
+ * Instances of this class are expected to be used for a single fedora
+ * object, and method calls should not require implementations to maintain
+ * state.
+ */
+public interface FedoraObjectHandler {
+
+    public void beginObject(ObjectInfo object);
+
+    /**
+     * Invoked to allow processing of properties by this FedoraObjectHandler.
+     * @param properties the properties for the object
+     */
+    public void processObjectProperties(ObjectProperties properties);
+
+    /**
+     * Invoked to allow processing of a datastream by this FedoraObjectHandler.
+     * @param dsVersion an encapsulation of the datastream version.  References to this object must
+     *                  not be retained or referenced outside of implementations of this method as
+     *                  attached resources (cached files, references to streams) may be updated and
+     *                  no longer valid.
+     */
+    public void processDatastreamVersion(DatastreamVersion dsVersion);
+
+    /**
+     * A hook called after the object has been completely processed.  This may be useful for any cleanup or
+     * finalization routines.  Furthermore, once this method invokation is complete, any references
+     * provided to prior calls will no longer be in scope.
+     */
+    public void completeObject(ObjectInfo object);
+
+    /**
+     * Invoked if processing of the object failed for some reason.
+     */
+    public void abortObject(ObjectInfo object);
+}

src/main/java/org/fcrepo/migration/FedoraObjectProcessor.java

@@ -0,0 +1,15 @@
+package org.fcrepo.migration;
+
+import javax.xml.stream.XMLStreamException;
+
+/**
+ * A class that encapsulates an object for processing.  This class represents a single object and
+ * exposes methods to query basic information about it and then to process it with an arbitrary
+ * FedoraObjectHandler.
+ */
+public interface FedoraObjectProcessor {
+
+    public ObjectInfo getObjectInfo();
+
+    public void processObject(FedoraObjectHandler handler) throws XMLStreamException;
+}