org.dspace.app.util.GoogleMetadata Maven / Gradle / Ivy
/**
* The contents of this file are subject to the license and copyright
* detailed in the LICENSE and NOTICE files at the root of the source
* tree and available online at
*
* http://www.dspace.org/license/
*/
package org.dspace.app.util;
import java.sql.SQLException;
import org.dspace.content.*;
import java.io.IOException;
import java.io.InputStream;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.Enumeration;
import java.util.HashMap;
import java.util.Map;
import java.util.Properties;
import org.apache.log4j.Logger;
import org.dspace.core.ConfigurationManager;
import java.io.File;
import java.io.UnsupportedEncodingException;
import java.util.ArrayList;
import java.util.HashSet;
import java.util.List;
import java.util.Map.Entry;
import java.util.Set;
import org.dspace.core.Constants;
import org.dspace.core.Context;
import org.dspace.handle.HandleManager;
import org.jdom.Element;
/**
*
* @author Sands Fish
*
*/
@SuppressWarnings("deprecation")
public class GoogleMetadata
{
private final static Logger log = Logger.getLogger(GoogleMetadata.class);
private static final String GOOGLE_PREFIX = "google.";
private Item item;
private String itemURL;
// Configuration keys and fields
private static Map configuredFields = new HashMap();
// Google field names (e.g. citation_fieldname) and formatted metadata
// values
private Map metadataMappings = new HashMap();
public static final String TITLE = "citation_title";
public static final String JOURNAL_TITLE = "citation_journal_title";
public static final String PUBLISHER = "citation_publisher";
public static final String AUTHORS = "citation_authors";
public static final String DATE = "citation_date";
public static final String VOLUME = "citation_volume";
public static final String ISSUE = "citation_issue";
public static final String FIRSTPAGE = "citation_firstpage";
public static final String LASTPAGE = "citation_lastpage";
public static final String DOI = "citation_doi";
public static final String PMID = "citation_pmid";
public static final String ABSTRACT = "citation_abstract_html_url";
public static final String FULLTEXT = "citation_fulltext_html_url";
public static final String PDF = "citation_pdf_url";
public static final String ISSN = "citation_issn";
public static final String ISBN = "citation_isbn";
public static final String LANGUAGE = "citation_language";
public static final String KEYWORDS = "citation_keywords";
public static final String CONFERENCE = "citation_conference";
public static final String DISSERTATION_ID = "identifiers.dissertation";
public static final String DISSERTATION_NAME = "citation_dissertation_name";
public static final String DISSERTATION_INSTITUTION = "citation_dissertation_institution";
public static final String PATENT_ID = "identifiers.patent";
public static final String PATENT_NUMBER = "citation_patent_number";
public static final String PATENT_COUNTRY = "citation_patent_country";
public static final String TECH_REPORT_ID = "identifiers.technical_report";
public static final String TECH_REPORT_NUMBER = "citation_technical_report_number";
public static final String TECH_REPORT_INSTITUTION = "citation_technical_report_institution";
private static final int SINGLE = 0;
private static final int MULTI = 1;
private static final int ALL_FIELDS_IN_OPTION = 2;
// Load configured fields from google-metadata.properties
static
{
File loadedFile = null;
URL url = null;
InputStream is = null;
String googleConfigFile = ConfigurationManager
.getProperty("google-metadata.config");
log.info("Using [" + googleConfigFile
+ "] for Google Metadata configuration");
loadedFile = new File(googleConfigFile);
try
{
url = loadedFile.toURL();
}
catch (MalformedURLException mux)
{
log.error("Can't find Google Metadata configuration file: "
+ googleConfigFile, mux);
}
Properties properties = new Properties();
try
{
is = url.openStream();
properties.load(is);
}
catch (IOException iox)
{
log.error("Could not read Google Metadata configuration file: "
+ googleConfigFile, iox);
}
Enumeration propertyNames = properties.propertyNames();
while (propertyNames.hasMoreElements())
{
String key = ((String) propertyNames.nextElement()).trim();
if (key.startsWith(GOOGLE_PREFIX))
{
String name = key.substring(GOOGLE_PREFIX.length());
String field = properties.getProperty(key);
if (null != name && !name.equals("") && null != field
&& !field.equals(""))
{
configuredFields.put(name.trim(), field.trim());
}
}
}
if (log.isDebugEnabled())
{
logConfiguration();
}
}
/**
* Dump Metadata field mapping to log
*
*/
public static void logConfiguration()
{
log.debug("Google Metadata Configuration Mapping:");
for (String name : configuredFields.keySet())
{
log.debug(" " + name + " => " + configuredFields.get(name));
}
}
/**
* Wrap the item, parse all configured fields and generate metadata field
* values.
*
* @param item
* - The item being viewed to extract metadata from
*/
public GoogleMetadata(Context context, Item item) throws SQLException
{
// Hold onto the item in case we need to refresh a stale parse
this.item = item;
itemURL = HandleManager.resolveToURL(context, item.getHandle());
parseItem();
}
/**
* Add a single metadata value to the Google field, defaulting to the
* first-encountered instance of the field for this Item.
*
* @param fieldName
* @param schema
* @param element
* @param qualifier
* @return
*/
private boolean addSingleField(String fieldName)
{
String config = configuredFields.get(fieldName);
if (null == config || config.equals(""))
{
return false;
}
if (log.isDebugEnabled())
{
log.debug("Processing " + fieldName);
}
if (config.equals("$handle"))
{
if (null != itemURL && !itemURL.equals(""))
{
metadataMappings.put(fieldName, itemURL);
return true;
}
else
{
return false;
}
}
if (config.equals("$simple-pdf"))
{
String pdf_url = getPDFSimpleUrl(item);
if(pdf_url.length() > 0)
{
metadataMappings.put(fieldName, pdf_url);
return true;
} else
{
return false;
}
}
DCValue v = resolveMetadataField(config);
if (null != v && (null != v.value) && !v.value.trim().equals(""))
{
metadataMappings.put(fieldName, v.value);
return true;
}
else
{
// No values found
return false;
}
}
/**
* A singular version of resolveMetadata to return only one field value
* instead of an aggregate.
*
* @param configFilter
* @return The first configured match of metadata field for the item.
*/
private DCValue resolveMetadataField(String configFilter)
{
ArrayList fields = resolveMetadata(configFilter, SINGLE);
if (null != fields && fields.size() > 0)
{
return fields.get(0);
}
return null;
}
/**
* A plural version of resolveMetadata for aggregate fields.
*
* @param configFilter
* @return Aggregate of all matching metadata fields configured in the first
* option field-set to return any number of filter matches.
*/
private ArrayList resolveMetadataFields(String configFilter)
{
ArrayList fields = resolveMetadata(configFilter, MULTI);
if (null != fields && fields.size() > 0)
{
return fields;
}
return null;
}
/**
* Aggregate an array of DCValues present on the current item that pass the
* configuration filter.
*
* @param configValue
* @return Array of configuration -> item-field matches
*/
private ArrayList resolveMetadata(String configFilter,
int returnType)
{
if (null == configFilter || configFilter.trim().equals("")
|| !configFilter.contains("."))
{
log.error("The configuration string [" + configFilter
+ "] is invalid.");
return null;
}
else
{
configFilter = configFilter.trim();
}
ArrayList> parsedOptions = new ArrayList>();
parsedOptions = parseOptions(configFilter);
if (log.isDebugEnabled())
{
log
.debug("Resolved Fields For This Item Per Configuration Filter:");
for (int i = 0; i < parsedOptions.size(); i++)
{
ArrayList optionFields = parsedOptions.get(i);
log.debug("Option " + (i + 1) + ":");
for (String f : optionFields)
{
log.debug("{" + f + "}");
}
}
}
// Iterate through each configured option's field-set until
// we have a match.
for (ArrayList optionFields : parsedOptions)
{
int optionMatches = 0;
String[] components;
DCValue[] values;
ArrayList resolvedFields = new ArrayList();
for (String field : optionFields)
{
components = parseComponents(field);
values = item.getMetadata(components[0], components[1],
components[2], Item.ANY);
if (values.length > 0)
{
for (DCValue v : values)
{
resolvedFields.add(v);
if (returnType == SINGLE)
{
if (!resolvedFields.isEmpty())
{
if (log.isDebugEnabled())
{
log
.debug("Resolved Field Value For This Item:");
for (DCValue r : resolvedFields)
{
log.debug("{" + r.value + "}");
}
}
return resolvedFields;
}
}
}
}
}
// If the item had any of the fields contained in this option,
// return them, otherwise move on to the next option's field-set.
if (!resolvedFields.isEmpty())
{
if (log.isDebugEnabled())
{
log.debug("Resolved Field Values For This Item:");
for (DCValue v : resolvedFields)
{
log.debug("{" + v.value + "}");
}
}
// Check to see if this is a full option match
if (ALL_FIELDS_IN_OPTION == returnType)
{
if (resolvedFields.size() == optionMatches)
{
return resolvedFields;
}
// Otherwise, if there are any matches for the option,
// return them.
}
else if (MULTI == returnType)
{
return resolvedFields;
}
}
}
return null;
}
/**
* Parse first-match path of metadata field-group options for the given
* configuration.
*
* @param configFilter
* @return
*/
private ArrayList> parseOptions(String configFilter)
{
ArrayList options = new ArrayList();
ArrayList> parsedOptions = new ArrayList>();
if (null == configFilter || configFilter.equals(""))
{
return null;
}
if (configFilter.contains("|"))
{
String[] configOptions = configFilter.split("\\|");
for (String option : configOptions)
{
options.add(option.trim());
}
}
else
{
options = new ArrayList();
options.add(configFilter);
}
// Parse first-match path options. The first option (field-set)
// to match fields present in the item is used.
ArrayList parsedFields;
// Parse the fields for each field-set in order.
for (String option : options)
{
ArrayList fields;
parsedFields = new ArrayList();
if (option.contains(","))
{
fields = parseFields(option);
}
else
{
fields = new ArrayList();
fields.add(option);
}
// Parse field list for this field-set, expanding any wildcards.
for (String field : fields)
{
if (field.contains("*"))
{
ArrayList wc = parseWildcard(field);
for (String wcField : wc)
{
if (!parsedFields.contains(wcField))
{
parsedFields.add(wcField);
}
}
}
else
{
if (!parsedFields.contains(field))
{
parsedFields.add(field);
}
}
}
parsedOptions.add(parsedFields);
}
if (null != parsedOptions)
{
return parsedOptions;
}
else
{
return null;
}
}
/**
* Build a Vector of fields that can be added to when expanding wildcards.
*
* @param configString
* - Value of one metadata field configuration
* @return A vector of raw field configurations.
*/
private ArrayList parseFields(String configString)
{
ArrayList fields = new ArrayList();
for (String field : configString.split("\\,"))
{
fields.add(field.trim());
}
return fields;
}
/**
* Pull apart an individual field structure.
*
* @param The
* configured field for one metadata field map
* @return Schema, Element, Qualifier of metadata field
*/
private String[] parseComponents(String field)
{
int index = 0;
String[] components = new String[3];
for (String c : field.split("\\."))
{
components[index] = c.trim();
index++;
}
return components;
}
/**
* Expand any wildcard characters to an array of all matching fields for
* this item. No order consistency is implied.
*
* @param field
* The field identifier containing a wildcard character.
* @return Expanded field list.
*/
private ArrayList parseWildcard(String field)
{
if (!field.contains("*"))
{
return null;
}
else
{
String[] components = parseComponents(field);
for (int i = 0; i < components.length; i++)
{
if (components[i].trim().equals("*"))
{
components[i] = Item.ANY;
}
}
DCValue[] allMD = item.getMetadata(components[0], components[1],
components[2], Item.ANY);
ArrayList expandedDC = new ArrayList();
for (DCValue v : allMD)
{
// De-dup multiple occurances of field names in item
if (!expandedDC.contains(buildFieldName(v)))
{
expandedDC.add(buildFieldName(v));
}
}
if (log.isDebugEnabled())
{
log.debug("Field Names From Expanded Wildcard \"" + field
+ "\"");
for (String v : expandedDC)
{
log.debug(" " + v);
}
}
return expandedDC;
}
}
/**
* Construct metadata field name out of DCValue components
*
* @param v
* The DCValue to construct a name for.
* @return The complete metadata field name.
*/
private String buildFieldName(DCValue v)
{
StringBuilder name = new StringBuilder();
name.append(v.schema + "." + v.element);
if (null != v.qualifier)
{
name.append("." + v.qualifier);
}
return name.toString();
}
/**
* Using metadata field mappings contained in the loaded configuration,
* parse through configured metadata fields, building valid Google metadata
* value strings. Field names & values contained in metadataMappings.
*
*/
private void parseItem()
{
// TITLE
addSingleField(TITLE);
// AUTHORS (multi)
addAggregateValues(AUTHORS, ";");
// DATE
addSingleField(DATE);
// ISSN
addSingleField(ISSN);
// ISBN
addSingleField(ISBN);
// VOLUME
addSingleField(VOLUME);
// ISSUE
addSingleField(ISSUE);
// FIRSTPAGE
addSingleField(FIRSTPAGE);
// LASTPAGE
addSingleField(LASTPAGE);
// DOI
addSingleField(DOI);
// PMID
addSingleField(PMID);
// ABSTRACT_HTML_URL ('$handle' variable substitution if present)
addSingleField(ABSTRACT);
// FULLTEXT_HTML_URL ('$handle' variable substitution if present)
addSingleField(FULLTEXT);
// PDF_URL ('$handle' variable substitution if present)
addSingleField(PDF);
// LANGUAGE
addSingleField(LANGUAGE);
// KEYWORDS (multi)
addAggregateValues(KEYWORDS, ";");
// CONFERENCE
addSingleField(CONFERENCE);
// Dissertations
if (itemIsDissertation())
{
if(log.isDebugEnabled()) {
log.debug("ITEM TYPE: DISSERTATION");
}
addSingleField(DISSERTATION_NAME);
addSingleField(DISSERTATION_INSTITUTION);
}
// Patents
if (itemIsPatent())
{
if(log.isDebugEnabled()) {
log.debug("ITEM TYPE: PATENT");
}
addSingleField(PATENT_NUMBER);
// Use config value for patent country. Should be a literal.
String countryConfig = configuredFields.get(PATENT_COUNTRY);
if (null != countryConfig && !countryConfig.trim().equals(""))
{
metadataMappings.put(PATENT_COUNTRY, countryConfig.trim());
}
addSingleField(PUBLISHER);
}
// Tech Reports
if (itemIsTechReport())
{
if(log.isDebugEnabled()) {
log.debug("ITEM TYPE: TECH REPORT");
}
addSingleField(TECH_REPORT_NUMBER);
addSingleField(TECH_REPORT_INSTITUTION);
}
if(!itemIsDissertation() && !itemIsTechReport()) {
// PUBLISHER
addSingleField(PUBLISHER);
}
}
/**
* Fetch all metadata mappings
*
* Usage: GoogleMetadata gmd = new GoogleMetadata(item); for(Entry mapping : googlemd.getMappings()) { ... }
*
* @return Iterable of metadata fields mapped to Google-formatted values
*/
public Set> getMappings()
{
return new HashSet>(metadataMappings.entrySet());
}
/**
* Produce meta elements that can easily be put into the head.
* @return
*/
public List disseminateList()
{
List metas = new ArrayList();
for (Entry m : getMappings())
{
Element e = new Element("meta");
e.setNamespace(null);
e.setAttribute("name", m.getKey());
e.setAttribute("content", m.getValue());
metas.add(e);
}
return metas;
}
// Getters for individual metadata fields...
/**
* @return the citation_title
*/
public String getTitle()
{
return metadataMappings.get(TITLE);
}
/**
* @return the citation_journal_title
*/
public String getJournalTitle()
{
return metadataMappings.get(JOURNAL_TITLE);
}
/**
* @return the citation_publisher
*/
public String getPublisher()
{
return metadataMappings.get(PUBLISHER);
}
/**
* @return the citation_authors
*/
public String getAuthors()
{
return metadataMappings.get(AUTHORS);
}
/**
* @return the citation_date
*/
public String getDate()
{
return metadataMappings.get(DATE);
}
/**
* @return the citation_volume
*/
public String getVolume()
{
return metadataMappings.get(VOLUME);
}
/**
* @return the citation_issue
*/
public String getIssue()
{
return metadataMappings.get(ISSUE);
}
/**
* @return the citation_firstpage
*/
public String getFirstpage()
{
return metadataMappings.get(FIRSTPAGE);
}
/**
* @return the citation_lastpage
*/
public String getLastpage()
{
return metadataMappings.get(LASTPAGE);
}
/**
* @return the citation_doi
*/
public String getDOI()
{
return metadataMappings.get(DOI);
}
/**
* @return the citation_pmid
*/
public String getPmid()
{
return metadataMappings.get(PMID);
}
/**
* @return the citation_abstract_html_url
*/
public String getAbstractHTMLURL()
{
return metadataMappings.get(ABSTRACT);
}
/**
* @return the citation_fulltext_html_url
*/
public String getFulltextHTMLURL()
{
return metadataMappings.get(FULLTEXT);
}
/**
* @return the citation_pdf_url
*/
public String getPDFURL()
{
return metadataMappings.get(PDF);
}
/**
* @return the citation_issn
*/
public String getISSN()
{
return metadataMappings.get(ISSN);
}
/**
* @return the citation_isbn
*/
public String getISBN()
{
return metadataMappings.get(ISBN);
}
/**
* @return the citation_language
*/
public String getLanguage()
{
return metadataMappings.get(LANGUAGE);
}
/**
* @return the citation_keywords
*/
public String getKeywords()
{
return metadataMappings.get(KEYWORDS);
}
/**
* @return the citation_conference
*/
public String getConference()
{
return metadataMappings.get(CONFERENCE);
}
/**
* @return the citation_dissertation_name
*/
public String getDissertationName()
{
return metadataMappings.get(DISSERTATION_NAME);
}
/**
* @return the citation_dissertation_institution
*/
public String getDissertationInstitution()
{
return metadataMappings.get(DISSERTATION_INSTITUTION);
}
/**
* @return the citation_patent_number
*/
public String getPatentNumber()
{
return metadataMappings.get(PATENT_NUMBER);
}
/**
* @return the citation_patent_country
*/
public String getPatentCountry()
{
return metadataMappings.get(PATENT_COUNTRY);
}
/**
* @return the citation_technical_report_number
*/
public String getTechnicalReportNumber()
{
return metadataMappings.get(TECH_REPORT_NUMBER);
}
/**
* @return the citation_technical_report_institution
*/
public String getTechnicalReportInstitution()
{
return metadataMappings.get(TECH_REPORT_INSTITUTION);
}
/**
* Gets the URL to a PDF using a very basic strategy by assuming that the PDF
* is in the default content bundle, and that the item only has one public bitstream
* and it is a PDF.
*
* @param item
* @return URL that the PDF can be directly downloaded from
*/
private String getPDFSimpleUrl(Item item)
{
try {
Bundle[] contentBundles = item.getBundles("ORIGINAL");
if (contentBundles.length > 0) {
Bitstream[] bitstreams = contentBundles[0].getBitstreams();
if (bitstreams.length == 1) {
if (bitstreams[0].getFormat().getMIMEType().equals("application/pdf")) {
StringBuilder path = new StringBuilder();
path.append(ConfigurationManager.getProperty("dspace.url"));
if (item.getHandle() != null) {
path.append("/bitstream/");
path.append(item.getHandle());
path.append("/");
path.append(bitstreams[0].getSequenceID());
} else {
path.append("/retrieve/");
path.append(bitstreams[0].getID());
}
path.append("/");
path.append(Util.encodeBitstreamName(bitstreams[0].getName(), Constants.DEFAULT_ENCODING));
return path.toString();
}
}
}
} catch (UnsupportedEncodingException ex) {
log.debug(ex.getMessage());
} catch (SQLException ex) {
log.debug(ex.getMessage());
}
return "";
}
/**
*
*
* @param Field
* to aggregate all values of in a matching option
* @param delimiter
* to delimit field values with
*/
private void addAggregateValues(String FIELD, String delim)
{
String authorConfig = configuredFields.get(FIELD);
ArrayList fields = resolveMetadataFields(authorConfig);
if (null != fields && !fields.isEmpty())
{
StringBuilder fieldMetadata = new StringBuilder();
int count = 0;
for (DCValue field : fields)
{
fieldMetadata.append(field.value);
if (count < fields.size() - 1)
{
fieldMetadata.append(delim + " ");
count++;
}
}
metadataMappings.put(FIELD, fieldMetadata.toString());
}
}
/**
* Determine, based on config values, if this item is a dissertation.
*
* @return boolean
*/
private boolean itemIsDissertation()
{
String dConfig = configuredFields.get(DISSERTATION_ID);
if (null == dConfig || dConfig.trim().equals(""))
{
return false;
}
else
{
return identifyItemType(dConfig);
}
}
/**
* Determine, based on config values, if this item is a patent.
*
* @return boolean
*/
private boolean itemIsPatent()
{
String dConfig = configuredFields.get(PATENT_ID);
if (null == dConfig || dConfig.trim().equals(""))
{
return false;
}
else
{
return identifyItemType(dConfig);
}
}
/**
* Determine, based on config values, if this item is a tech report.
*
* @return boolean
*/
private boolean itemIsTechReport()
{
String dConfig = configuredFields.get(TECH_REPORT_ID);
if (null == dConfig || dConfig.trim().equals(""))
{
return false;
}
else
{
return identifyItemType(dConfig);
}
}
/**
* Identifies if this item matches a particular configuration of fields and
* values for those fields to identify the type based on a type- cataloging
* metadata practice.
*
* @param dConfig
* @return
*/
private boolean identifyItemType(String dConfig)
{
// FIXME: Shouldn't have to parse identifiers for every identification.
ArrayList> options = parseOptions(dConfig);
HashMap> mdPairs = new HashMap>();
// Parse field/value pairs from field identifier string
for (ArrayList option : options)
{
String pair = option.get(0);
String[] parsedPair = pair.split("\\:");
if (2 == parsedPair.length)
{
// If we've encountered this field before, add the value to the
// list
if (mdPairs.containsKey(parsedPair[0].trim()))
{
mdPairs.get(parsedPair[0].trim()).add(parsedPair[1]);
if(log.isDebugEnabled()) {
log.debug("Registering Type Identifier: " + parsedPair[0] + " => " + parsedPair[1]);
}
}
else
{
// Otherwise, add it as the first occurance of this field
ArrayList newField = new ArrayList();
newField.add(parsedPair[1].trim());
mdPairs.put(parsedPair[0].trim(), newField);
if (log.isDebugEnabled()) {
log.debug("Registering Type Identifier: " + parsedPair[0] + " => " + parsedPair[1]);
}
}
}
else
{
log.error("Malformed field identifier name/value pair");
}
}
// Build config string without values, only field names
StringBuilder sb = new StringBuilder();
for (String value : mdPairs.keySet())
{
sb.append(value + " | ");
}
// Check resolved/present metadata fields against configured values
ArrayList presentMD = resolveMetadataFields(sb.toString());
if (null != presentMD && presentMD.size() != 0)
{
for (DCValue v : presentMD)
{
String fieldName = buildFieldName(v);
if (mdPairs.containsKey(fieldName))
{
for (String configValue : mdPairs.get(fieldName))
{
if (configValue.equals(v.value))
{
return true;
}
}
}
}
}
return false;
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy