mirror of
https://github.com/ilri/cgspace-java-helpers.git
synced 2024-11-28 09:38:21 +01:00
Compare commits
4 Commits
f0754ab419
...
b396fba043
Author | SHA1 | Date | |
---|---|---|---|
b396fba043 | |||
38a9cc5188 | |||
16db38967b | |||
2604dc3cce |
@ -1,8 +1,8 @@
|
|||||||
/*
|
/*
|
||||||
* Copyright (C) 2020 Alan Orth
|
* Copyright (C) 2020 Alan Orth
|
||||||
*
|
*
|
||||||
* SPDX-License-Identifier: GPL-3.0-or-later
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
||||||
*/
|
*/
|
||||||
|
|
||||||
package io.github.ilri.cgspace.ctasks;
|
package io.github.ilri.cgspace.ctasks;
|
||||||
|
|
||||||
@ -10,4 +10,4 @@ import java.util.List;
|
|||||||
|
|
||||||
public class CGSpaceCountriesVocabulary extends CountriesVocabulary {
|
public class CGSpaceCountriesVocabulary extends CountriesVocabulary {
|
||||||
List<Country> countries;
|
List<Country> countries;
|
||||||
}
|
}
|
||||||
|
@ -1,8 +1,8 @@
|
|||||||
/*
|
/*
|
||||||
* Copyright (C) 2020 Alan Orth
|
* Copyright (C) 2020 Alan Orth
|
||||||
*
|
*
|
||||||
* SPDX-License-Identifier: GPL-3.0-or-later
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
||||||
*/
|
*/
|
||||||
|
|
||||||
package io.github.ilri.cgspace.ctasks;
|
package io.github.ilri.cgspace.ctasks;
|
||||||
|
|
||||||
@ -11,26 +11,29 @@ import javax.annotation.Nullable;
|
|||||||
public class CountriesVocabulary {
|
public class CountriesVocabulary {
|
||||||
|
|
||||||
class Country {
|
class Country {
|
||||||
private String name; //required
|
private String name; // required
|
||||||
private String common_name; //optional
|
private String common_name; // optional
|
||||||
private String official_name; //optional
|
private String official_name; // optional
|
||||||
private String cgspace_name; //optional
|
private String cgspace_name; // optional
|
||||||
private String numeric; //required Hmmmm need to cast this...
|
private String numeric; // required Hmmmm need to cast this...
|
||||||
private String alpha_2; //required
|
private String alpha_2; // required
|
||||||
private String alpha_3; //required
|
private String alpha_3; // required
|
||||||
|
|
||||||
public Country(String name,
|
public Country(
|
||||||
@Nullable String common_name,
|
String name,
|
||||||
@Nullable String official_name,
|
@Nullable String common_name,
|
||||||
@Nullable String cgspace_name,
|
@Nullable String official_name,
|
||||||
String numeric,
|
@Nullable String cgspace_name,
|
||||||
String alpha_2,
|
String numeric,
|
||||||
String alpha_3) {
|
String alpha_2,
|
||||||
|
String alpha_3) {
|
||||||
this.name = name;
|
this.name = name;
|
||||||
this.common_name = common_name;
|
this.common_name = common_name;
|
||||||
this.official_name = official_name;
|
this.official_name = official_name;
|
||||||
this.cgspace_name = cgspace_name;
|
this.cgspace_name = cgspace_name;
|
||||||
this.numeric = numeric; // fuuuuu this is a string and we can't cast to Integer because some values are zeropadded like "004"
|
this.numeric =
|
||||||
|
numeric; // fuuuuu this is a string and we can't cast to Integer because some
|
||||||
|
// values are zeropadded like "004"
|
||||||
this.alpha_2 = alpha_2;
|
this.alpha_2 = alpha_2;
|
||||||
this.alpha_3 = alpha_3;
|
this.alpha_3 = alpha_3;
|
||||||
}
|
}
|
||||||
@ -63,4 +66,4 @@ public class CountriesVocabulary {
|
|||||||
return cgspace_name;
|
return cgspace_name;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
@ -1,12 +1,13 @@
|
|||||||
/*
|
/*
|
||||||
* Copyright (C) 2020 Alan Orth
|
* Copyright (C) 2020 Alan Orth
|
||||||
*
|
*
|
||||||
* SPDX-License-Identifier: GPL-3.0-or-later
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
||||||
*/
|
*/
|
||||||
|
|
||||||
package io.github.ilri.cgspace.ctasks;
|
package io.github.ilri.cgspace.ctasks;
|
||||||
|
|
||||||
import com.google.gson.Gson;
|
import com.google.gson.Gson;
|
||||||
|
|
||||||
import org.apache.log4j.Logger;
|
import org.apache.log4j.Logger;
|
||||||
import org.dspace.authorize.AuthorizeException;
|
import org.dspace.authorize.AuthorizeException;
|
||||||
import org.dspace.content.DSpaceObject;
|
import org.dspace.content.DSpaceObject;
|
||||||
@ -23,11 +24,11 @@ import java.sql.SQLException;
|
|||||||
import java.util.ArrayList;
|
import java.util.ArrayList;
|
||||||
import java.util.List;
|
import java.util.List;
|
||||||
|
|
||||||
public class CountryCodeTagger extends AbstractCurationTask
|
public class CountryCodeTagger extends AbstractCurationTask {
|
||||||
{
|
|
||||||
public class CountryCodeTaggerConfig {
|
public class CountryCodeTaggerConfig {
|
||||||
private final String isocodesJsonPath = "/io/github/ilri/cgspace/ctasks/iso_3166-1.json";
|
private final String isocodesJsonPath = "/io/github/ilri/cgspace/ctasks/iso_3166-1.json";
|
||||||
private final String cgspaceCountriesJsonPath = "/io/github/ilri/cgspace/ctasks/cgspace-countries.json";
|
private final String cgspaceCountriesJsonPath =
|
||||||
|
"/io/github/ilri/cgspace/ctasks/cgspace-countries.json";
|
||||||
private final String iso3166Field = taskProperty("iso3166.field");
|
private final String iso3166Field = taskProperty("iso3166.field");
|
||||||
private final String iso3166Alpha2Field = taskProperty("iso3166-alpha2.field");
|
private final String iso3166Alpha2Field = taskProperty("iso3166-alpha2.field");
|
||||||
private final boolean forceupdate = taskBooleanProperty("forceupdate", false);
|
private final boolean forceupdate = taskBooleanProperty("forceupdate", false);
|
||||||
@ -57,17 +58,15 @@ public class CountryCodeTagger extends AbstractCurationTask
|
|||||||
}
|
}
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public int perform(DSpaceObject dso) throws IOException
|
public int perform(DSpaceObject dso) throws IOException {
|
||||||
{
|
|
||||||
// gotta define this here so we can access it after the if context...
|
// gotta define this here so we can access it after the if context...
|
||||||
CountryCodeTaggerResult alpha2Result = new CountryCodeTaggerResult();
|
CountryCodeTaggerResult alpha2Result = new CountryCodeTaggerResult();
|
||||||
|
|
||||||
if (dso.getType() == Constants.ITEM)
|
if (dso.getType() == Constants.ITEM) {
|
||||||
{
|
|
||||||
// Load configuration
|
// Load configuration
|
||||||
CountryCodeTaggerConfig config = new CountryCodeTaggerConfig();
|
CountryCodeTaggerConfig config = new CountryCodeTaggerConfig();
|
||||||
|
|
||||||
Item item = (Item)dso;
|
Item item = (Item) dso;
|
||||||
|
|
||||||
try {
|
try {
|
||||||
alpha2Result = performAlpha2(item, config);
|
alpha2Result = performAlpha2(item, config);
|
||||||
@ -77,16 +76,18 @@ public class CountryCodeTagger extends AbstractCurationTask
|
|||||||
|
|
||||||
setResult(alpha2Result.getResult());
|
setResult(alpha2Result.getResult());
|
||||||
report(alpha2Result.getResult());
|
report(alpha2Result.getResult());
|
||||||
}
|
}
|
||||||
|
|
||||||
return alpha2Result.getStatus();
|
return alpha2Result.getStatus();
|
||||||
}
|
}
|
||||||
|
|
||||||
public CountryCodeTaggerResult performAlpha2(Item item, CountryCodeTaggerConfig config) throws IOException, SQLException {
|
public CountryCodeTaggerResult performAlpha2(Item item, CountryCodeTaggerConfig config)
|
||||||
|
throws IOException, SQLException {
|
||||||
CountryCodeTaggerResult alpha2Result = new CountryCodeTaggerResult();
|
CountryCodeTaggerResult alpha2Result = new CountryCodeTaggerResult();
|
||||||
String itemHandle = item.getHandle();
|
String itemHandle = item.getHandle();
|
||||||
|
|
||||||
List<MetadataValue> itemCountries = itemService.getMetadataByMetadataString(item, config.iso3166Field);
|
List<MetadataValue> itemCountries =
|
||||||
|
itemService.getMetadataByMetadataString(item, config.iso3166Field);
|
||||||
|
|
||||||
// skip items that don't have country metadata
|
// skip items that don't have country metadata
|
||||||
if (itemCountries.size() == 0) {
|
if (itemCountries.size() == 0) {
|
||||||
@ -95,36 +96,60 @@ public class CountryCodeTagger extends AbstractCurationTask
|
|||||||
} else {
|
} else {
|
||||||
Gson gson = new Gson();
|
Gson gson = new Gson();
|
||||||
|
|
||||||
// TODO: convert to try: https://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html
|
// TODO: convert to try:
|
||||||
BufferedReader reader = new BufferedReader(new InputStreamReader(this.getClass().getResourceAsStream(config.isocodesJsonPath)));
|
// https://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html
|
||||||
ISO3166CountriesVocabulary isocodesCountriesJson = gson.fromJson(reader, ISO3166CountriesVocabulary.class);
|
BufferedReader reader =
|
||||||
|
new BufferedReader(
|
||||||
|
new InputStreamReader(
|
||||||
|
this.getClass().getResourceAsStream(config.isocodesJsonPath)));
|
||||||
|
ISO3166CountriesVocabulary isocodesCountriesJson =
|
||||||
|
gson.fromJson(reader, ISO3166CountriesVocabulary.class);
|
||||||
reader.close();
|
reader.close();
|
||||||
|
|
||||||
reader = new BufferedReader(new InputStreamReader(this.getClass().getResourceAsStream(config.cgspaceCountriesJsonPath)));
|
reader =
|
||||||
CGSpaceCountriesVocabulary cgspaceCountriesJson = gson.fromJson(reader, CGSpaceCountriesVocabulary.class);
|
new BufferedReader(
|
||||||
|
new InputStreamReader(
|
||||||
|
this.getClass()
|
||||||
|
.getResourceAsStream(config.cgspaceCountriesJsonPath)));
|
||||||
|
CGSpaceCountriesVocabulary cgspaceCountriesJson =
|
||||||
|
gson.fromJson(reader, CGSpaceCountriesVocabulary.class);
|
||||||
reader.close();
|
reader.close();
|
||||||
|
|
||||||
// split the alpha2 country code field into schema, element, and qualifier so we can use it with item.addMetadata()
|
// split the alpha2 country code field into schema, element, and qualifier so we can use
|
||||||
|
// it with item.addMetadata()
|
||||||
String[] iso3166Alpha2FieldParts = config.iso3166Alpha2Field.split("\\.");
|
String[] iso3166Alpha2FieldParts = config.iso3166Alpha2Field.split("\\.");
|
||||||
|
|
||||||
if (config.forceupdate) {
|
if (config.forceupdate) {
|
||||||
itemService.clearMetadata(Curator.curationContext(), item, iso3166Alpha2FieldParts[0], iso3166Alpha2FieldParts[1], iso3166Alpha2FieldParts[2], Item.ANY);
|
itemService.clearMetadata(
|
||||||
|
Curator.curationContext(),
|
||||||
|
item,
|
||||||
|
iso3166Alpha2FieldParts[0],
|
||||||
|
iso3166Alpha2FieldParts[1],
|
||||||
|
iso3166Alpha2FieldParts[2],
|
||||||
|
Item.ANY);
|
||||||
}
|
}
|
||||||
|
|
||||||
// check the item's country codes, if any
|
// check the item's country codes, if any
|
||||||
List<MetadataValue> itemAlpha2CountryCodes = itemService.getMetadataByMetadataString(item, config.iso3166Alpha2Field);
|
List<MetadataValue> itemAlpha2CountryCodes =
|
||||||
|
itemService.getMetadataByMetadataString(item, config.iso3166Alpha2Field);
|
||||||
|
|
||||||
if (itemAlpha2CountryCodes.size() == 0) {
|
if (itemAlpha2CountryCodes.size() == 0) {
|
||||||
List<String> newAlpha2Codes = new ArrayList<String>();
|
List<String> newAlpha2Codes = new ArrayList<String>();
|
||||||
for (MetadataValue itemCountry : itemCountries) {
|
for (MetadataValue itemCountry : itemCountries) {
|
||||||
//check ISO 3166-1 countries
|
// check ISO 3166-1 countries
|
||||||
for (CountriesVocabulary.Country country : isocodesCountriesJson.countries) {
|
for (CountriesVocabulary.Country country : isocodesCountriesJson.countries) {
|
||||||
if (itemCountry.getValue().equalsIgnoreCase(country.getName()) || itemCountry.getValue().equalsIgnoreCase(country.get_official_name()) || itemCountry.getValue().equalsIgnoreCase(country.get_common_name())) {
|
if (itemCountry.getValue().equalsIgnoreCase(country.getName())
|
||||||
|
|| itemCountry
|
||||||
|
.getValue()
|
||||||
|
.equalsIgnoreCase(country.get_official_name())
|
||||||
|
|| itemCountry
|
||||||
|
.getValue()
|
||||||
|
.equalsIgnoreCase(country.get_common_name())) {
|
||||||
newAlpha2Codes.add(country.getAlpha_2());
|
newAlpha2Codes.add(country.getAlpha_2());
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
//check CGSpace countries
|
// check CGSpace countries
|
||||||
for (CountriesVocabulary.Country country : cgspaceCountriesJson.countries) {
|
for (CountriesVocabulary.Country country : cgspaceCountriesJson.countries) {
|
||||||
if (itemCountry.getValue().equalsIgnoreCase(country.getCgspace_name())) {
|
if (itemCountry.getValue().equalsIgnoreCase(country.getCgspace_name())) {
|
||||||
newAlpha2Codes.add(country.getAlpha_2());
|
newAlpha2Codes.add(country.getAlpha_2());
|
||||||
@ -134,7 +159,14 @@ public class CountryCodeTagger extends AbstractCurationTask
|
|||||||
|
|
||||||
if (newAlpha2Codes.size() > 0) {
|
if (newAlpha2Codes.size() > 0) {
|
||||||
try {
|
try {
|
||||||
itemService.addMetadata(Curator.curationContext(), item, iso3166Alpha2FieldParts[0], iso3166Alpha2FieldParts[1], iso3166Alpha2FieldParts[2], "en_US", newAlpha2Codes);
|
itemService.addMetadata(
|
||||||
|
Curator.curationContext(),
|
||||||
|
item,
|
||||||
|
iso3166Alpha2FieldParts[0],
|
||||||
|
iso3166Alpha2FieldParts[1],
|
||||||
|
iso3166Alpha2FieldParts[2],
|
||||||
|
"en_US",
|
||||||
|
newAlpha2Codes);
|
||||||
itemService.update(Curator.curationContext(), item);
|
itemService.update(Curator.curationContext(), item);
|
||||||
} catch (SQLException | AuthorizeException sqle) {
|
} catch (SQLException | AuthorizeException sqle) {
|
||||||
config.log.debug(sqle.getMessage());
|
config.log.debug(sqle.getMessage());
|
||||||
@ -142,7 +174,11 @@ public class CountryCodeTagger extends AbstractCurationTask
|
|||||||
alpha2Result.setStatus(Curator.CURATE_ERROR);
|
alpha2Result.setStatus(Curator.CURATE_ERROR);
|
||||||
}
|
}
|
||||||
|
|
||||||
alpha2Result.setResult(itemHandle + ": added " + newAlpha2Codes.size() + " alpha2 country code(s)");
|
alpha2Result.setResult(
|
||||||
|
itemHandle
|
||||||
|
+ ": added "
|
||||||
|
+ newAlpha2Codes.size()
|
||||||
|
+ " alpha2 country code(s)");
|
||||||
} else {
|
} else {
|
||||||
alpha2Result.setResult(itemHandle + ": no matching countries found");
|
alpha2Result.setResult(itemHandle + ": no matching countries found");
|
||||||
}
|
}
|
||||||
@ -155,4 +191,4 @@ public class CountryCodeTagger extends AbstractCurationTask
|
|||||||
|
|
||||||
return alpha2Result;
|
return alpha2Result;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
@ -1,15 +1,18 @@
|
|||||||
/*
|
/*
|
||||||
* Copyright (C) 2020 Alan Orth
|
* Copyright (C) 2020 Alan Orth
|
||||||
*
|
*
|
||||||
* SPDX-License-Identifier: GPL-3.0-or-later
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
||||||
*/
|
*/
|
||||||
|
|
||||||
package io.github.ilri.cgspace.ctasks;
|
package io.github.ilri.cgspace.ctasks;
|
||||||
|
|
||||||
import com.google.gson.annotations.SerializedName;
|
import com.google.gson.annotations.SerializedName;
|
||||||
|
|
||||||
import java.util.List;
|
import java.util.List;
|
||||||
|
|
||||||
public class ISO3166CountriesVocabulary extends CountriesVocabulary {
|
public class ISO3166CountriesVocabulary extends CountriesVocabulary {
|
||||||
// support reading iso_3166-1.json from Debian's iso-codes package using SerializedName since our class needs to match the JSON exactly
|
// support reading iso_3166-1.json from Debian's iso-codes package using SerializedName since
|
||||||
@SerializedName("3166-1") List<Country> countries;
|
// our class needs to match the JSON exactly
|
||||||
}
|
@SerializedName("3166-1")
|
||||||
|
List<Country> countries;
|
||||||
|
}
|
||||||
|
@ -1,21 +1,27 @@
|
|||||||
/*
|
/*
|
||||||
* Copyright (C) 2020 Alan Orth
|
* Copyright (C) 2020 Alan Orth
|
||||||
*
|
*
|
||||||
* SPDX-License-Identifier: GPL-3.0-or-later
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
||||||
*/
|
*/
|
||||||
|
|
||||||
package io.github.ilri.cgspace.scripts;
|
package io.github.ilri.cgspace.scripts;
|
||||||
|
|
||||||
import org.apache.commons.lang.StringUtils;
|
import org.apache.commons.lang.StringUtils;
|
||||||
import org.dspace.authorize.AuthorizeException;
|
import org.dspace.authorize.AuthorizeException;
|
||||||
import org.dspace.content.*;
|
import org.dspace.content.Bitstream;
|
||||||
|
import org.dspace.content.Bundle;
|
||||||
|
import org.dspace.content.Collection;
|
||||||
|
import org.dspace.content.Community;
|
||||||
|
import org.dspace.content.DSpaceObject;
|
||||||
|
import org.dspace.content.Item;
|
||||||
|
import org.dspace.content.MetadataValue;
|
||||||
|
import org.dspace.content.factory.ContentServiceFactory;
|
||||||
|
import org.dspace.content.service.BundleService;
|
||||||
|
import org.dspace.content.service.ItemService;
|
||||||
import org.dspace.core.Constants;
|
import org.dspace.core.Constants;
|
||||||
import org.dspace.core.Context;
|
import org.dspace.core.Context;
|
||||||
import org.dspace.content.factory.ContentServiceFactory;
|
|
||||||
import org.dspace.content.service.ItemService;
|
|
||||||
import org.dspace.handle.factory.HandleServiceFactory;
|
import org.dspace.handle.factory.HandleServiceFactory;
|
||||||
import org.dspace.handle.service.HandleService;
|
import org.dspace.handle.service.HandleService;
|
||||||
import org.dspace.content.service.BundleService;
|
|
||||||
|
|
||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
import java.sql.SQLException;
|
import java.sql.SQLException;
|
||||||
@ -29,118 +35,139 @@ import java.util.List;
|
|||||||
* @since 5.1
|
* @since 5.1
|
||||||
*/
|
*/
|
||||||
public class FixJpgJpgThumbnails {
|
public class FixJpgJpgThumbnails {
|
||||||
//note: static members belong to the class itself, not any one instance
|
// note: static members belong to the class itself, not any one instance
|
||||||
public static ItemService itemService = ContentServiceFactory.getInstance().getItemService();
|
public static ItemService itemService = ContentServiceFactory.getInstance().getItemService();
|
||||||
public static HandleService handleService = HandleServiceFactory.getInstance().getHandleService();
|
public static HandleService handleService =
|
||||||
public static BundleService bundleService = ContentServiceFactory.getInstance().getBundleService();
|
HandleServiceFactory.getInstance().getHandleService();
|
||||||
|
public static BundleService bundleService =
|
||||||
|
ContentServiceFactory.getInstance().getBundleService();
|
||||||
|
|
||||||
public static void main(String[] args) {
|
public static void main(String[] args) {
|
||||||
String parentHandle = null;
|
String parentHandle = null;
|
||||||
if (args.length >= 1) {
|
if (args.length >= 1) {
|
||||||
parentHandle = args[0];
|
parentHandle = args[0];
|
||||||
}
|
}
|
||||||
|
|
||||||
Context context = null;
|
Context context = null;
|
||||||
try {
|
try {
|
||||||
context = new Context();
|
context = new Context();
|
||||||
context.turnOffAuthorisationSystem();
|
context.turnOffAuthorisationSystem();
|
||||||
|
|
||||||
if (StringUtils.isBlank(parentHandle)) {
|
if (StringUtils.isBlank(parentHandle)) {
|
||||||
process(context, itemService.findAll(context));
|
process(context, itemService.findAll(context));
|
||||||
} else {
|
} else {
|
||||||
DSpaceObject parent = handleService.resolveToObject(context, parentHandle);
|
DSpaceObject parent = handleService.resolveToObject(context, parentHandle);
|
||||||
if (parent != null) {
|
if (parent != null) {
|
||||||
switch (parent.getType()) {
|
switch (parent.getType()) {
|
||||||
case Constants.COLLECTION:
|
case Constants.COLLECTION:
|
||||||
process(context, itemService.findByCollection(context, (Collection) parent));
|
process(
|
||||||
break;
|
context,
|
||||||
case Constants.COMMUNITY:
|
itemService.findByCollection(context, (Collection) parent));
|
||||||
List<Collection> collections = ((Community) parent).getCollections();
|
break;
|
||||||
for (Collection collection : collections) {
|
case Constants.COMMUNITY:
|
||||||
process(context, itemService.findAllByCollection(context, collection));
|
List<Collection> collections = ((Community) parent).getCollections();
|
||||||
}
|
for (Collection collection : collections) {
|
||||||
break;
|
process(
|
||||||
case Constants.SITE:
|
context,
|
||||||
process(context, itemService.findAll(context));
|
itemService.findAllByCollection(context, collection));
|
||||||
break;
|
}
|
||||||
case Constants.ITEM:
|
break;
|
||||||
processItem(context, (Item) parent);
|
case Constants.SITE:
|
||||||
context.commit();
|
process(context, itemService.findAll(context));
|
||||||
break;
|
break;
|
||||||
}
|
case Constants.ITEM:
|
||||||
}
|
processItem(context, (Item) parent);
|
||||||
}
|
context.commit();
|
||||||
} catch (SQLException | AuthorizeException | IOException e) {
|
break;
|
||||||
e.printStackTrace(System.err);
|
}
|
||||||
} finally {
|
}
|
||||||
if (context != null && context.isValid()) {
|
}
|
||||||
context.abort();
|
} catch (SQLException | AuthorizeException | IOException e) {
|
||||||
}
|
e.printStackTrace(System.err);
|
||||||
}
|
} finally {
|
||||||
}
|
if (context != null && context.isValid()) {
|
||||||
|
context.abort();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
private static void process(Context context, Iterator<Item> items) throws SQLException, IOException, AuthorizeException {
|
private static void process(Context context, Iterator<Item> items)
|
||||||
while (items.hasNext()) {
|
throws SQLException, IOException, AuthorizeException {
|
||||||
Item item = items.next();
|
while (items.hasNext()) {
|
||||||
processItem(context, item);
|
Item item = items.next();
|
||||||
itemService.update(context, item);
|
processItem(context, item);
|
||||||
}
|
itemService.update(context, item);
|
||||||
}
|
}
|
||||||
|
}
|
||||||
|
|
||||||
private static void processItem(Context context, Item item) throws SQLException, AuthorizeException, IOException {
|
private static void processItem(Context context, Item item)
|
||||||
// Some bitstreams like Infographics are large JPGs and put in the ORIGINAL bundle on purpose so we shouldn't
|
throws SQLException, AuthorizeException, IOException {
|
||||||
// swap them.
|
// Some bitstreams like Infographics and Maps are large JPEGs and put in the ORIGINAL bundle
|
||||||
List<MetadataValue> itemTypes = itemService.getMetadataByMetadataString(item, "dcterms.type");
|
// on purpose so we shouldn't
|
||||||
boolean itemHasInfographic = false;
|
// swap them.
|
||||||
for (MetadataValue itemType: itemTypes) {
|
List<MetadataValue> itemTypes =
|
||||||
if (itemType.getValue().equals("Infographic")) {
|
itemService.getMetadataByMetadataString(item, "dcterms.type");
|
||||||
itemHasInfographic = true;
|
for (MetadataValue itemType : itemTypes) {
|
||||||
}
|
if (itemType.getValue().equals("Infographic") || itemType.getValue().equals("Map")) {
|
||||||
}
|
System.out.println(
|
||||||
|
item.getHandle() + ": item has an Infographic or Map, skipping.");
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
List<Bundle> thumbnailBundles = item.getBundles("THUMBNAIL");
|
List<Bundle> thumbnailBundles = item.getBundles("THUMBNAIL");
|
||||||
for (Bundle thumbnailBundle : thumbnailBundles) {
|
for (Bundle thumbnailBundle : thumbnailBundles) {
|
||||||
List<Bitstream> thumbnailBundleBitstreams = thumbnailBundle.getBitstreams();
|
List<Bitstream> thumbnailBundleBitstreams = thumbnailBundle.getBitstreams();
|
||||||
for (Bitstream thumbnailBitstream : thumbnailBundleBitstreams) {
|
for (Bitstream thumbnailBitstream : thumbnailBundleBitstreams) {
|
||||||
String thumbnailName = thumbnailBitstream.getName();
|
String thumbnailName = thumbnailBitstream.getName();
|
||||||
|
String thumbnailDescription = thumbnailBitstream.getDescription();
|
||||||
|
|
||||||
if (thumbnailName.toLowerCase().contains(".jpg.jpg")) {
|
// There is no point continuing if the thumbnail's description is empty or null
|
||||||
List<Bundle> originalBundles = item.getBundles("ORIGINAL");
|
if (StringUtils.isEmpty(thumbnailDescription)) {
|
||||||
for (Bundle originalBundle : originalBundles) {
|
continue;
|
||||||
List<Bitstream> originalBundleBitstreams = originalBundle.getBitstreams();
|
}
|
||||||
|
|
||||||
for (Bitstream originalBitstream : originalBundleBitstreams) {
|
if (thumbnailName.toLowerCase().contains(".jpg.jpg")) {
|
||||||
String originalName = originalBitstream.getName();
|
List<Bundle> originalBundles = item.getBundles("ORIGINAL");
|
||||||
|
for (Bundle originalBundle : originalBundles) {
|
||||||
|
List<Bitstream> originalBundleBitstreams = originalBundle.getBitstreams();
|
||||||
|
|
||||||
long originalBitstreamBytes = originalBitstream.getSize();
|
for (Bitstream originalBitstream : originalBundleBitstreams) {
|
||||||
|
String originalName = originalBitstream.getName();
|
||||||
|
|
||||||
/*
|
long originalBitstreamBytes = originalBitstream.getSize();
|
||||||
- check if the original file name is the same as the thumbnail name minus the extra ".jpg"
|
|
||||||
- check if the thumbnail description indicates it was automatically generated
|
|
||||||
- check if the item has dc.type Infographic (JPG could be the "real" item!)
|
|
||||||
- check if the original bitstream is less than ~100KiB
|
|
||||||
- Note: in my tests there were 4022 items with ".jpg.jpg" thumbnails totaling 394549249
|
|
||||||
bytes for an average of about 98KiB so ~100KiB seems like a good cut off
|
|
||||||
*/
|
|
||||||
if (
|
|
||||||
originalName.equalsIgnoreCase(StringUtils.removeEndIgnoreCase(thumbnailName, ".jpg"))
|
|
||||||
&& ("Generated Thumbnail".equals(thumbnailBitstream.getDescription()) || "IM Thumbnail".equals(thumbnailBitstream.getDescription()))
|
|
||||||
&& !itemHasInfographic
|
|
||||||
&& originalBitstreamBytes < 100000
|
|
||||||
) {
|
|
||||||
System.out.println(item.getHandle() + ": replacing " + thumbnailName + " with " + originalName);
|
|
||||||
|
|
||||||
//add the original bitstream to the THUMBNAIL bundle
|
/*
|
||||||
bundleService.addBitstream(context, thumbnailBundle, originalBitstream);
|
- check if the original file name is the same as the thumbnail name minus the extra ".jpg"
|
||||||
//remove the original bitstream from the ORIGINAL bundle
|
- check if the thumbnail description indicates it was automatically generated
|
||||||
originalBundle.removeBitstream(originalBitstream);
|
- check if the original bitstream is less than ~100KiB
|
||||||
//remove the JpgJpg bitstream from the THUMBNAIL bundle
|
- Note: in my tests there were 4022 items with ".jpg.jpg" thumbnails totaling 394549249
|
||||||
thumbnailBundle.removeBitstream(thumbnailBitstream);
|
bytes for an average of about 98KiB so ~100KiB seems like a good cut off
|
||||||
}
|
*/
|
||||||
}
|
if (originalName.equalsIgnoreCase(
|
||||||
}
|
StringUtils.removeEndIgnoreCase(thumbnailName, ".jpg"))
|
||||||
}
|
&& ("Generated Thumbnail".equals(thumbnailDescription)
|
||||||
}
|
|| "IM Thumbnail".equals(thumbnailDescription))
|
||||||
}
|
&& originalBitstreamBytes < 100000) {
|
||||||
}
|
System.out.println(
|
||||||
|
item.getHandle()
|
||||||
|
+ ": replacing "
|
||||||
|
+ thumbnailName
|
||||||
|
+ " with "
|
||||||
|
+ originalName);
|
||||||
|
|
||||||
|
// add the original bitstream to the THUMBNAIL bundle
|
||||||
|
bundleService.addBitstream(
|
||||||
|
context, thumbnailBundle, originalBitstream);
|
||||||
|
// remove the original bitstream from the ORIGINAL bundle
|
||||||
|
originalBundle.removeBitstream(originalBitstream);
|
||||||
|
// remove the JpgJpg bitstream from the THUMBNAIL bundle
|
||||||
|
thumbnailBundle.removeBitstream(thumbnailBitstream);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
@ -1,14 +1,19 @@
|
|||||||
/*
|
/*
|
||||||
* Copyright (C) 2022 Alan Orth
|
* Copyright (C) 2022 Alan Orth
|
||||||
*
|
*
|
||||||
* SPDX-License-Identifier: GPL-3.0-or-later
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
||||||
*/
|
*/
|
||||||
|
|
||||||
package io.github.ilri.cgspace.scripts;
|
package io.github.ilri.cgspace.scripts;
|
||||||
|
|
||||||
import org.apache.commons.lang.StringUtils;
|
import org.apache.commons.lang.StringUtils;
|
||||||
import org.dspace.authorize.AuthorizeException;
|
import org.dspace.authorize.AuthorizeException;
|
||||||
import org.dspace.content.*;
|
import org.dspace.content.Bitstream;
|
||||||
|
import org.dspace.content.Bundle;
|
||||||
|
import org.dspace.content.Collection;
|
||||||
|
import org.dspace.content.Community;
|
||||||
|
import org.dspace.content.DSpaceObject;
|
||||||
|
import org.dspace.content.Item;
|
||||||
import org.dspace.content.factory.ContentServiceFactory;
|
import org.dspace.content.factory.ContentServiceFactory;
|
||||||
import org.dspace.content.service.BundleService;
|
import org.dspace.content.service.BundleService;
|
||||||
import org.dspace.content.service.ItemService;
|
import org.dspace.content.service.ItemService;
|
||||||
|
Loading…
Reference in New Issue
Block a user