gCubeSystem
/
metadata-profile-discovery
15
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Added guava cache. New DocumentBuilder a Xpath parser

This commit is contained in:
Francesco Mangiacrapa 2022-03-03 16:14:07 +01:00
parent 97c9a716a0
commit 471536ae64
7 changed files with 284 additions and 61 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
.classpath
View File

@ -6,6 +6,12 @@
<attribute name="maven.pomderived" value="true"/>
</attributes>
</classpathentry>
<classpathentry excluding="**" kind="src" output="target/test-classes" path="src/test/resources">
<attributes>
<attribute name="maven.pomderived" value="true"/>
<attribute name="test" value="true"/>
</attributes>
</classpathentry>
<classpathentry excluding="**" kind="src" output="target/classes" path="src/main/resources">
<attributes>
<attribute name="maven.pomderived" value="true"/>

5
pom.xml
View File

@ -99,6 +99,11 @@
<scope>provided</scope>
</dependency>
<dependency>
<groupId>com.google.guava</groupId>
<artifactId>guava</artifactId>
</dependency>
<!-- JUNIT -->
<dependency>
<groupId>junit</groupId>

6
src/main/java/org/gcube/common/metadataprofilediscovery/MetadataProfileReader.java
View File

@ -28,10 +28,10 @@ import javax.xml.validation.Validator;
import org.gcube.common.metadataprofilediscovery.bean.MetadataProfile;
import org.gcube.common.metadataprofilediscovery.jaxb.MetadataFormat;
import org.gcube.common.metadataprofilediscovery.jaxb.NamespaceCategories;
import org.gcube.common.metadataprofilediscovery.jaxb.NamespaceCategory;
import org.gcube.common.metadataprofilediscovery.reader.MetadataFormatDiscovery;
import org.gcube.common.metadataprofilediscovery.reader.MetadataFormatReader;
import org.gcube.common.metadataprofilediscovery.reader.NamespaceCategoryReader;
import org.gcube.common.scope.api.ScopeProvider;
import org.gcube.common.scope.impl.ScopeBean;
import org.slf4j.Logger;
@ -169,8 +169,8 @@ public class MetadataProfileReader implements MetadataProfileDiscovery {
if (namespaceCategories == null)
namespaceCategories = new ArrayList<NamespaceCategory>();
NamespaceCategoryReader rd = new NamespaceCategoryReader(scope);
namespaceCategories.addAll(rd.getNamespaces().getNamespaceCategories());
NamespaceCategories ncCache = NamespaceCategoriesCache.loadNamespaces(scope.toString());
namespaceCategories.addAll(ncCache.getNamespaceCategories());
}
} catch (Exception e) {
logger.warn("An error occurred during read namespaces for categories: ", e);

145
src/main/java/org/gcube/common/metadataprofilediscovery/NamespaceCategoriesCache.java Normal file
View File

@ -0,0 +1,145 @@
/**
*
*/
package org.gcube.common.metadataprofilediscovery;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import org.gcube.common.metadataprofilediscovery.jaxb.NamespaceCategories;
import org.gcube.common.metadataprofilediscovery.reader.NamespaceCategoryReader;
import org.gcube.common.scope.api.ScopeProvider;
import org.gcube.common.scope.impl.ScopeBean;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.google.common.cache.CacheBuilder;
import com.google.common.cache.CacheLoader;
import com.google.common.cache.LoadingCache;
import com.google.common.cache.RemovalListener;
import com.google.common.cache.RemovalNotification;
/**
* The Class NamespaceCategoriesCache.
*
* @author Francesco Mangiacrapa at ISTI-CNR francesco.mangiacrapa@isti.cnr.it
*
* Mar 3, 2022
*/
public class NamespaceCategoriesCache {
private static Logger LOG = LoggerFactory.getLogger(NamespaceCategoriesCache.class);
private static LoadingCache<String, NamespaceCategories> namespaces;
static {
CacheLoader<String, NamespaceCategories> loader = new CacheLoader<String, NamespaceCategories>() {
@Override
public NamespaceCategories load(String scope) throws Exception {
LOG.info("Loading the NamespaceCategories cache for scope: {}", scope);
NamespaceCategories namespacesCat = loadNamespaces(scope);
if (namespacesCat != null)
LOG.info("Returning {} for the scope name: {}", NamespaceCategories.class.getSimpleName(), scope);
else {
LOG.info("No NamespaceCategories obj for scope {}", scope);
}
return namespacesCat;
}
};
RemovalListener<String, NamespaceCategories> removalListener = new RemovalListener<String, NamespaceCategories>() {
@Override
public void onRemoval(RemovalNotification<String, NamespaceCategories> arg0) {
LOG.debug("cache expired");
}
};
namespaces = CacheBuilder.newBuilder().maximumSize(300).expireAfterWrite(30, TimeUnit.MINUTES)
.removalListener(removalListener).build(loader);
}
/**
* Populate the cache.
*
* @param scope the scope
*/
private static void populateTheCache(ScopeBean scope) {
String origScope = null;
String instScope = scope.toString();
try {
origScope = ScopeProvider.instance.get();
// Populating the cache by using the detachedres-library
ScopeProvider.instance.set(scope.toString());
LOG.info("Trying to populate the Namespaces cache in the scope: "+instScope);
NamespaceCategoryReader rd = new NamespaceCategoryReader(scope);
namespaces.put(instScope, rd.getNamespaces());
} catch (Exception e) {
// SILENT
} finally {
if (origScope != null && !origScope.equals(scope.toString())) {
// Setting original scope
ScopeProvider.instance.set(scope.toString());
}
}
}
/**
* Gets the VRE obj for input VRE name.
*
* @param scope the scope
* @return the vre
* @throws ExecutionException the execution exception
*/
public static NamespaceCategories get(String scope) throws ExecutionException {
try {
return namespaces.get(scope);
} catch (Exception e) {
LOG.info("Error on getting NamespaceCategories obj for scope {}. Is the key {} not found in the cache?",
scope, namespaces);
throw e;
}
}
/**
* Load VRE obj for VRE name.
*
* @param scope the scope
* @return the vre
*/
protected static NamespaceCategories loadNamespaces(String scope) {
NamespaceCategories namespacesCategories = namespaces.getIfPresent(scope);
// THIS CHECK SHOULD NOT BE NEEDED
if (namespacesCategories == null) {
LOG.info("loading Catalogue Namespaces for scope: {}",scope);
ScopeBean scopeB = new ScopeBean(scope);
populateTheCache(scopeB);
namespacesCategories = namespaces.getIfPresent(scope);
LOG.info("NamespaceCategories populated correclty with scope: {}",scope);
}else {
LOG.info("NamespaceCategories cache already populated with scope: {}, returning cached NamespaceCategories",scope);
}
return namespacesCategories;
}
/**
* Gets the cache.
*
* @return the cache
*/
public LoadingCache<String, NamespaceCategories> getCache() {
return namespaces;
}
}

75
src/main/java/org/gcube/common/metadataprofilediscovery/reader/NamespaceCategoryReader.java
View File

@ -8,12 +8,21 @@ import static org.gcube.resources.discovery.icclient.ICFactory.client;
import java.io.ByteArrayInputStream;
import java.io.InputStream;
import java.io.StringReader;
import java.io.StringWriter;
import java.util.List;
import javax.xml.bind.JAXBContext;
import javax.xml.bind.Unmarshaller;
import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerException;
import javax.xml.transform.TransformerFactory;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.stream.StreamResult;
import javax.xml.xpath.XPath;
import javax.xml.xpath.XPathConstants;
import javax.xml.xpath.XPathFactory;
import org.gcube.common.metadataprofilediscovery.jaxb.NamespaceCategories;
import org.gcube.common.resources.gcore.utils.XPathHelper;
@ -23,7 +32,9 @@ import org.gcube.resources.discovery.client.queries.api.Query;
import org.gcube.resources.discovery.client.queries.impl.QueryBox;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.w3c.dom.Document;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.InputSource;
@ -35,6 +46,8 @@ import org.xml.sax.InputSource;
*/
public class NamespaceCategoryReader {
private static final String PATH_RESOURCE_PROFILE_BODY_NAMESPACES = "/Resource/Profile/Body/namespaces";
private static Logger logger = LoggerFactory.getLogger(NamespaceCategoryReader.class);
private static final String GENERIC_RESOURCE_NAMESPACES_NAME = "Namespaces Catalogue Categories";
@ -89,12 +102,11 @@ public class NamespaceCategoryReader {
String theResource = null;
try{
theResource = appProfile.get(0);
logger.debug("Resource (Namespaces Catalogue Categories) found");
logger.debug("Resource "+GENERIC_RESOURCE_NAMESPACES_NAME+" found");
logger.trace("Resource "+GENERIC_RESOURCE_NAMESPACES_NAME+" is: "+theResource);
DocumentBuilder docBuilder = DocumentBuilderFactory.newInstance().newDocumentBuilder();
Node node = docBuilder.parse(new InputSource(new StringReader(theResource))).getDocumentElement();
XPathHelper helper = new XPathHelper(node);
logger.debug("Unmarshalling it..");
readNamespaceCategoryFromResource(helper);
Document xmlDocument = docBuilder.parse(new InputSource(new StringReader(theResource)));
readNamespaceCategoryFromResource(xmlDocument);
}catch(Exception e){
logger.error("Error while parsing Resource "+theResource+" from the infrastructure, the scope is "+scopeString,e);
}
@ -109,6 +121,52 @@ public class NamespaceCategoryReader {
}
}
/**
* Read namespace category from resource.
*
* @param helper the helper
* @throws Exception the exception
*/
private void readNamespaceCategoryFromResource(Document xmlDocument) throws Exception{
try {
XPath xPath = XPathFactory.newInstance().newXPath();
String expression = PATH_RESOURCE_PROFILE_BODY_NAMESPACES;
NodeList nodeList = (NodeList) xPath.compile(expression).evaluate(xmlDocument, XPathConstants.NODESET);
String namespaceSources = getNodeString(nodeList.item(0));
JAXBContext jaxbContext = JAXBContext.newInstance(NamespaceCategories.class);
Unmarshaller jaxbUnmarshaller = jaxbContext.createUnmarshaller();
if(namespaceSources==null || namespaceSources.length()==0){
throw new Exception("Resource does not contain <namespaces> <namespace>...</namespace> </namespaces> in the body");
}
logger.debug("Unmarshalling document path "+PATH_RESOURCE_PROFILE_BODY_NAMESPACES +" to "+NamespaceCategories.class.getSimpleName());
InputStream stream = new ByteArrayInputStream(namespaceSources.getBytes());
namespaces = (NamespaceCategories) jaxbUnmarshaller.unmarshal(stream);
}catch(Exception e){
String error = "An error occurred in readNamespaceCategoryFromResource " + e.getMessage();
logger.error("An error occurred in readNamespaceCategoryFromResource ", e);
throw new Exception(error);
}
}
private String getNodeString(Node node) {
try {
StringWriter writer = new StringWriter();
Transformer transformer = TransformerFactory.newInstance().newTransformer();
transformer.transform(new DOMSource(node), new StreamResult(writer));
String output = writer.toString();
return output;
} catch (TransformerException e) {
e.printStackTrace();
}
return node.getTextContent();
}
/**
* Read namespace category from resource.
@ -120,7 +178,12 @@ public class NamespaceCategoryReader {
try {
List<String> namespaceSources = helper.evaluate("/Resource/Profile/Body/namespaces");
if(helper==null) {
logger.warn("**************************\n\nXpathHelper is null!!!!\n\n");
}
List<String> namespaceSources = helper.evaluate(PATH_RESOURCE_PROFILE_BODY_NAMESPACES);
JAXBContext jaxbContext = JAXBContext.newInstance(NamespaceCategories.class);
Unmarshaller jaxbUnmarshaller = jaxbContext.createUnmarshaller();

47
src/test/java/org/gcube/common/metadataprofilediscovery/TestMetadataFormatReader.java
View File

@ -3,41 +3,50 @@
*/
package org.gcube.common.metadataprofilediscovery;
import org.gcube.common.metadataprofilediscovery.bean.MetadataProfile;
import org.gcube.common.metadataprofilediscovery.jaxb.NamespaceCategory;
import java.io.InputStream;
import org.gcube.common.metadataprofilediscovery.jaxb.MetadataField;
import org.gcube.common.metadataprofilediscovery.jaxb.MetadataFormat;
import org.gcube.common.metadataprofilediscovery.reader.MetadataFormatReader;
import org.gcube.common.scope.api.ScopeProvider;
import org.gcube.common.scope.impl.ScopeBean;
/**
*
* @author Francesco Mangiacrapa francesco.mangiacrapa@isti.cnr.it
* Jun 8, 2016
* @author Francesco Mangiacrapa francesco.mangiacrapa@isti.cnr.it Jun 8, 2016
*/
public class TestMetadataFormatReader {
//@Test
// @Test
public void test() {
String scopeString = "/gcube/devsec/devVRE";
final ScopeBean scope = new ScopeBean(scopeString);
MetadataProfileReader reader;
final ScopeBean scope = new ScopeBean(scopeString);
MetadataFormatReader reader;
try {
ScopeProvider.instance.set(scopeString);
reader = new MetadataProfileReader("GeoNaMetadata");
int i = 0;
for (MetadataProfile metadataProfile : reader.getListOfMetadataProfiles()) {
System.out.println(i++ +")"+metadataProfile);
}
i = 0;
for (NamespaceCategory namespaceCategory : reader.getListOfNamespaceCategories()) {
System.out.println(i++ +")"+namespaceCategory);
}
reader = new MetadataFormatReader(scope, "0d29d7a9-d779-478c-a13d-d70708dc66c4");
System.out.println(reader.getMetadataFormat());
} catch (Exception e) {
e.printStackTrace();
}
}
// @Test
public void testReadInputStream() {
String fileNameMeatadataProfile = "HarvestedObject.xml";
try {
InputStream in = ClassLoader.getSystemResourceAsStream(fileNameMeatadataProfile);
// InputStream in =
// TestMetadataFormatReader.class.getResourceAsStream(fileNameMeatadataProfile);
MetadataFormat mf = MetadataProfileReader.toMetadataFormat(in);
System.out.println("Source: " + mf.getMetadataSource());
for (MetadataField field : mf.getMetadataFields()) {
System.out.println(field);
}
} catch (Exception e) {
e.printStackTrace();
}
}
}

57
src/test/java/org/gcube/common/metadataprofilediscovery/TestMetadataProfileReader.java
View File

@ -3,51 +3,46 @@
*/
package org.gcube.common.metadataprofilediscovery;
import java.io.InputStream;
import org.gcube.common.metadataprofilediscovery.jaxb.MetadataField;
import org.gcube.common.metadataprofilediscovery.jaxb.MetadataFormat;
import org.gcube.common.metadataprofilediscovery.reader.MetadataFormatReader;
import org.gcube.common.metadataprofilediscovery.bean.MetadataProfile;
import org.gcube.common.metadataprofilediscovery.jaxb.NamespaceCategory;
import org.gcube.common.scope.api.ScopeProvider;
import org.gcube.common.scope.impl.ScopeBean;
import org.junit.Test;
/**
*
* @author Francesco Mangiacrapa francesco.mangiacrapa@isti.cnr.it
* Jun 8, 2016
* @author Francesco Mangiacrapa francesco.mangiacrapa@isti.cnr.it Jun 8, 2016
*/
public class TestMetadataProfileReader {
// @Test
//@Test
public void test() {
String scopeString = "/gcube/devsec/devVRE";
final ScopeBean scope = new ScopeBean(scopeString);
MetadataFormatReader reader;
try {
ScopeProvider.instance.set(scopeString);
reader = new MetadataFormatReader(scope, "0d29d7a9-d779-478c-a13d-d70708dc66c4");
System.out.println(reader.getMetadataFormat());
} catch (Exception e) {
e.printStackTrace();
}
}
final ScopeBean scope = new ScopeBean(scopeString);
//@Test
public void testReadInputStream() {
String fileNameMeatadataProfile = "HarvestedObject.xml";
try {
InputStream in = ClassLoader.getSystemResourceAsStream(fileNameMeatadataProfile);
//InputStream in = TestMetadataFormatReader.class.getResourceAsStream(fileNameMeatadataProfile);
MetadataFormat mf = MetadataProfileReader.toMetadataFormat(in);
System.out.println("Source: "+mf.getMetadataSource());
String[] genericResourceNames = new String[] {"Informazioni_di_progetto","Relazione_di_Scavo"};
for (MetadataField field : mf.getMetadataFields()) {
System.out.println(field);
for (int i = 0; i < 2; i++) {
MetadataProfileReader reader;
try {
ScopeProvider.instance.set(scope.toString());
reader = new MetadataProfileReader("GeoNaMetadata",genericResourceNames[i]);
int j = 0;
for (MetadataProfile metadataProfile : reader.getListOfMetadataProfiles()) {
System.out.println(j++ + ")" + metadataProfile);
}
j = 0;
for (NamespaceCategory namespaceCategory : reader.getListOfNamespaceCategories()) {
System.out.println(j++ + ")" + namespaceCategory);
}
} catch (Exception e) {
e.printStackTrace();
}
} catch (Exception e) {
e.printStackTrace();
}
}
}