alanorth
/
cgspace-java-helpers
mirror of https://github.com/ilri/cgspace-java-helpers.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

src: format Java files with google-java-format 

Using AOSP format so we get four spaces instead of two.
This commit is contained in:
Alan Orth 2022-10-06 14:27:51 +03:00
parent 38a9cc5188
commit b396fba043
Signed by: alanorth
GPG Key ID: 0FB860CC9C45B1B9
6 changed files with 252 additions and 195 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
src/main/java/io/github/ilri/cgspace/ctasks/CGSpaceCountriesVocabulary.java
View File

@ -1,8 +1,8 @@
/*
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
package io.github.ilri.cgspace.ctasks;
@ -10,4 +10,4 @@ import java.util.List;
public class CGSpaceCountriesVocabulary extends CountriesVocabulary {
List<Country> countries;
}
}

43
src/main/java/io/github/ilri/cgspace/ctasks/CountriesVocabulary.java
View File

@ -1,8 +1,8 @@
/*
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
package io.github.ilri.cgspace.ctasks;
@ -11,26 +11,29 @@ import javax.annotation.Nullable;
public class CountriesVocabulary {
class Country {
private String name; //required
private String common_name; //optional
private String official_name; //optional
private String cgspace_name; //optional
private String numeric; //required Hmmmm need to cast this...
private String alpha_2; //required
private String alpha_3; //required
private String name; // required
private String common_name; // optional
private String official_name; // optional
private String cgspace_name; // optional
private String numeric; // required Hmmmm need to cast this...
private String alpha_2; // required
private String alpha_3; // required
public Country(String name,
@Nullable String common_name,
@Nullable String official_name,
@Nullable String cgspace_name,
String numeric,
String alpha_2,
String alpha_3) {
public Country(
String name,
@Nullable String common_name,
@Nullable String official_name,
@Nullable String cgspace_name,
String numeric,
String alpha_2,
String alpha_3) {
this.name = name;
this.common_name = common_name;
this.official_name = official_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_3 = alpha_3;
}
@ -63,4 +66,4 @@ public class CountriesVocabulary {
return cgspace_name;
}
}
}
}

109
src/main/java/io/github/ilri/cgspace/ctasks/CountryCodeTagger.java
View File

@ -1,17 +1,12 @@
/*
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
package io.github.ilri.cgspace.ctasks;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.sql.SQLException;
import java.util.ArrayList;
import java.util.List;
import com.google.gson.Gson;
import org.apache.log4j.Logger;
import org.dspace.authorize.AuthorizeException;
@ -22,13 +17,18 @@ import org.dspace.core.Constants;
import org.dspace.curate.AbstractCurationTask;
import org.dspace.curate.Curator;
import com.google.gson.Gson;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.sql.SQLException;
import java.util.ArrayList;
import java.util.List;
public class CountryCodeTagger extends AbstractCurationTask
{
public class CountryCodeTagger extends AbstractCurationTask {
public class CountryCodeTaggerConfig {
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 iso3166Alpha2Field = taskProperty("iso3166-alpha2.field");
private final boolean forceupdate = taskBooleanProperty("forceupdate", false);
@ -58,17 +58,15 @@ public class CountryCodeTagger extends AbstractCurationTask
}
@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...
CountryCodeTaggerResult alpha2Result = new CountryCodeTaggerResult();
if (dso.getType() == Constants.ITEM)
{
if (dso.getType() == Constants.ITEM) {
// Load configuration
CountryCodeTaggerConfig config = new CountryCodeTaggerConfig();
Item item = (Item)dso;
Item item = (Item) dso;
try {
alpha2Result = performAlpha2(item, config);
@ -78,16 +76,18 @@ public class CountryCodeTagger extends AbstractCurationTask
setResult(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();
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
if (itemCountries.size() == 0) {
@ -96,36 +96,60 @@ public class CountryCodeTagger extends AbstractCurationTask
} else {
Gson gson = new Gson();
// TODO: convert to try: https://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html
BufferedReader reader = new BufferedReader(new InputStreamReader(this.getClass().getResourceAsStream(config.isocodesJsonPath)));
ISO3166CountriesVocabulary isocodesCountriesJson = gson.fromJson(reader, ISO3166CountriesVocabulary.class);
// TODO: convert to try:
// https://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html
BufferedReader reader =
new BufferedReader(
new InputStreamReader(
this.getClass().getResourceAsStream(config.isocodesJsonPath)));
ISO3166CountriesVocabulary isocodesCountriesJson =
gson.fromJson(reader, ISO3166CountriesVocabulary.class);
reader.close();
reader = new BufferedReader(new InputStreamReader(this.getClass().getResourceAsStream(config.cgspaceCountriesJsonPath)));
CGSpaceCountriesVocabulary cgspaceCountriesJson = gson.fromJson(reader, CGSpaceCountriesVocabulary.class);
reader =
new BufferedReader(
new InputStreamReader(
this.getClass()
.getResourceAsStream(config.cgspaceCountriesJsonPath)));
CGSpaceCountriesVocabulary cgspaceCountriesJson =
gson.fromJson(reader, CGSpaceCountriesVocabulary.class);
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("\\.");
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
List<MetadataValue> itemAlpha2CountryCodes = itemService.getMetadataByMetadataString(item, config.iso3166Alpha2Field);
List<MetadataValue> itemAlpha2CountryCodes =
itemService.getMetadataByMetadataString(item, config.iso3166Alpha2Field);
if (itemAlpha2CountryCodes.size() == 0) {
List<String> newAlpha2Codes = new ArrayList<String>();
for (MetadataValue itemCountry : itemCountries) {
//check ISO 3166-1 countries
// check ISO 3166-1 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());
}
}
//check CGSpace countries
// check CGSpace countries
for (CountriesVocabulary.Country country : cgspaceCountriesJson.countries) {
if (itemCountry.getValue().equalsIgnoreCase(country.getCgspace_name())) {
newAlpha2Codes.add(country.getAlpha_2());
@ -135,7 +159,14 @@ public class CountryCodeTagger extends AbstractCurationTask
if (newAlpha2Codes.size() > 0) {
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);
} catch (SQLException | AuthorizeException sqle) {
config.log.debug(sqle.getMessage());
@ -143,7 +174,11 @@ public class CountryCodeTagger extends AbstractCurationTask
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 {
alpha2Result.setResult(itemHandle + ": no matching countries found");
}
@ -156,4 +191,4 @@ public class CountryCodeTagger extends AbstractCurationTask
return alpha2Result;
}
}
}

20
src/main/java/io/github/ilri/cgspace/ctasks/ISO3166CountriesVocabulary.java
View File

@ -1,16 +1,18 @@
/*
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
package io.github.ilri.cgspace.ctasks;
import java.util.List;
import com.google.gson.annotations.SerializedName;
import java.util.List;
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
@SerializedName("3166-1") List<Country> countries;
}
// support reading iso_3166-1.json from Debian's iso-codes package using SerializedName since
// our class needs to match the JSON exactly
@SerializedName("3166-1")
List<Country> countries;
}

247
src/main/java/io/github/ilri/cgspace/scripts/FixJpgJpgThumbnails.java
View File

@ -1,16 +1,11 @@
/*
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
* Copyright (C) 2020 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
package io.github.ilri.cgspace.scripts;
import java.io.IOException;
import java.sql.SQLException;
import java.util.Iterator;
import java.util.List;
import org.apache.commons.lang.StringUtils;
import org.dspace.authorize.AuthorizeException;
import org.dspace.content.Bitstream;
@ -28,6 +23,11 @@ import org.dspace.core.Context;
import org.dspace.handle.factory.HandleServiceFactory;
import org.dspace.handle.service.HandleService;
import java.io.IOException;
import java.sql.SQLException;
import java.util.Iterator;
import java.util.List;
/**
* @author Andrea Schweer schweer@waikato.ac.nz for the LCoNZ Institutional Research Repositories
* @author Alan Orth for the International Livestock Research Institute
@ -35,122 +35,139 @@ import org.dspace.handle.service.HandleService;
* @since 5.1
*/
public class FixJpgJpgThumbnails {
//note: static members belong to the class itself, not any one instance
public static ItemService itemService = ContentServiceFactory.getInstance().getItemService();
public static HandleService handleService = HandleServiceFactory.getInstance().getHandleService();
public static BundleService bundleService = ContentServiceFactory.getInstance().getBundleService();
// note: static members belong to the class itself, not any one instance
public static ItemService itemService = ContentServiceFactory.getInstance().getItemService();
public static HandleService handleService =
HandleServiceFactory.getInstance().getHandleService();
public static BundleService bundleService =
ContentServiceFactory.getInstance().getBundleService();
public static void main(String[] args) {
String parentHandle = null;
if (args.length >= 1) {
parentHandle = args[0];
}
public static void main(String[] args) {
String parentHandle = null;
if (args.length >= 1) {
parentHandle = args[0];
}
Context context = null;
try {
context = new Context();
context.turnOffAuthorisationSystem();
Context context = null;
try {
context = new Context();
context.turnOffAuthorisationSystem();
if (StringUtils.isBlank(parentHandle)) {
process(context, itemService.findAll(context));
} else {
DSpaceObject parent = handleService.resolveToObject(context, parentHandle);
if (parent != null) {
switch (parent.getType()) {
case Constants.COLLECTION:
process(context, itemService.findByCollection(context, (Collection) parent));
break;
case Constants.COMMUNITY:
List<Collection> collections = ((Community) parent).getCollections();
for (Collection collection : collections) {
process(context, itemService.findAllByCollection(context, collection));
}
break;
case Constants.SITE:
process(context, itemService.findAll(context));
break;
case Constants.ITEM:
processItem(context, (Item) parent);
context.commit();
break;
}
}
}
} catch (SQLException | AuthorizeException | IOException e) {
e.printStackTrace(System.err);
} finally {
if (context != null && context.isValid()) {
context.abort();
}
}
}
if (StringUtils.isBlank(parentHandle)) {
process(context, itemService.findAll(context));
} else {
DSpaceObject parent = handleService.resolveToObject(context, parentHandle);
if (parent != null) {
switch (parent.getType()) {
case Constants.COLLECTION:
process(
context,
itemService.findByCollection(context, (Collection) parent));
break;
case Constants.COMMUNITY:
List<Collection> collections = ((Community) parent).getCollections();
for (Collection collection : collections) {
process(
context,
itemService.findAllByCollection(context, collection));
}
break;
case Constants.SITE:
process(context, itemService.findAll(context));
break;
case Constants.ITEM:
processItem(context, (Item) parent);
context.commit();
break;
}
}
}
} 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 {
while (items.hasNext()) {
Item item = items.next();
processItem(context, item);
itemService.update(context, item);
}
}
private static void process(Context context, Iterator<Item> items)
throws SQLException, IOException, AuthorizeException {
while (items.hasNext()) {
Item item = items.next();
processItem(context, item);
itemService.update(context, item);
}
}
private static void processItem(Context context, Item item) throws SQLException, AuthorizeException, IOException {
// Some bitstreams like Infographics and Maps are large JPEGs and put in the ORIGINAL bundle on purpose so we shouldn't
// swap them.
List<MetadataValue> itemTypes = itemService.getMetadataByMetadataString(item, "dcterms.type");
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;
}
}
private static void processItem(Context context, Item item)
throws SQLException, AuthorizeException, IOException {
// Some bitstreams like Infographics and Maps are large JPEGs and put in the ORIGINAL bundle
// on purpose so we shouldn't
// swap them.
List<MetadataValue> itemTypes =
itemService.getMetadataByMetadataString(item, "dcterms.type");
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");
for (Bundle thumbnailBundle : thumbnailBundles) {
List<Bitstream> thumbnailBundleBitstreams = thumbnailBundle.getBitstreams();
for (Bitstream thumbnailBitstream : thumbnailBundleBitstreams) {
String thumbnailName = thumbnailBitstream.getName();
String thumbnailDescription = thumbnailBitstream.getDescription();
List<Bundle> thumbnailBundles = item.getBundles("THUMBNAIL");
for (Bundle thumbnailBundle : thumbnailBundles) {
List<Bitstream> thumbnailBundleBitstreams = thumbnailBundle.getBitstreams();
for (Bitstream thumbnailBitstream : thumbnailBundleBitstreams) {
String thumbnailName = thumbnailBitstream.getName();
String thumbnailDescription = thumbnailBitstream.getDescription();
// There is no point continuing if the thumbnail's description is empty or null
if (StringUtils.isEmpty(thumbnailDescription)) {
continue;
}
// There is no point continuing if the thumbnail's description is empty or null
if (StringUtils.isEmpty(thumbnailDescription)) {
continue;
}
if (thumbnailName.toLowerCase().contains(".jpg.jpg")) {
List<Bundle> originalBundles = item.getBundles("ORIGINAL");
for (Bundle originalBundle : originalBundles) {
List<Bitstream> originalBundleBitstreams = originalBundle.getBitstreams();
if (thumbnailName.toLowerCase().contains(".jpg.jpg")) {
List<Bundle> originalBundles = item.getBundles("ORIGINAL");
for (Bundle originalBundle : originalBundles) {
List<Bitstream> originalBundleBitstreams = originalBundle.getBitstreams();
for (Bitstream originalBitstream : originalBundleBitstreams) {
String originalName = originalBitstream.getName();
for (Bitstream originalBitstream : originalBundleBitstreams) {
String originalName = originalBitstream.getName();
long originalBitstreamBytes = originalBitstream.getSize();
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 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(thumbnailDescription) || "IM Thumbnail".equals(thumbnailDescription))
&& originalBitstreamBytes < 100000
) {
System.out.println(item.getHandle() + ": replacing " + thumbnailName + " with " + originalName);
/*
- 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 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(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);
}
}
}
}
}
}
}
// 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);
}
}
}
}
}
}
}
}

18
src/main/java/io/github/ilri/cgspace/scripts/FixLowQualityThumbnails.java
View File

@ -1,16 +1,11 @@
/*
* Copyright (C) 2022 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
* Copyright (C) 2022 Alan Orth
*
* SPDX-License-Identifier: GPL-3.0-or-later
*/
package io.github.ilri.cgspace.scripts;
import java.io.IOException;
import java.sql.SQLException;
import java.util.Iterator;
import java.util.List;
import org.apache.commons.lang.StringUtils;
import org.dspace.authorize.AuthorizeException;
import org.dspace.content.Bitstream;
@ -27,6 +22,11 @@ import org.dspace.core.Context;
import org.dspace.handle.factory.HandleServiceFactory;
import org.dspace.handle.service.HandleService;
import java.io.IOException;
import java.sql.SQLException;
import java.util.Iterator;
import java.util.List;
/**
* Fix low-quality thumbnails in a DSpace repository.
*