src/main/java/io/github/ilri/cgspace/ctasks/CGSpaceCountriesVocabulary.java

@@ -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;
 

src/main/java/io/github/ilri/cgspace/ctasks/CountriesVocabulary.java

@@ -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;
 
@@ -11,29 +11,26 @@ import javax.annotation.Nullable;
 public class CountriesVocabulary {
 
     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(
+        public Country(String name,
-                String name,
+                       @Nullable String common_name,
-                @Nullable String common_name,
+                       @Nullable String official_name,
-                @Nullable String official_name,
+                       @Nullable String cgspace_name,
-                @Nullable String cgspace_name,
+                       String numeric,
-                String numeric,
+                       String alpha_2,
-                String alpha_2,
+                       String alpha_3) {
-                String alpha_3) {
             this.name = name;
             this.common_name = common_name;
             this.official_name = official_name;
             this.cgspace_name = cgspace_name;
-            this.numeric =
+            this.numeric = numeric; // fuuuuu this is a string and we can't cast to Integer because some values are zeropadded like "004"
-                    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;
         }

src/main/java/io/github/ilri/cgspace/ctasks/CountryCodeTagger.java

@@ -1,13 +1,12 @@
 /*
- * 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;
 
 import com.google.gson.Gson;
-
 import org.apache.log4j.Logger;
 import org.dspace.authorize.AuthorizeException;
 import org.dspace.content.DSpaceObject;
@@ -24,11 +23,11 @@ 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 =
+        private final String cgspaceCountriesJsonPath = "/io/github/ilri/cgspace/ctasks/cgspace-countries.json";
-                "/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);
@@ -58,15 +57,17 @@ 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);
@@ -76,18 +77,16 @@ public class CountryCodeTagger extends AbstractCurationTask {
 
             setResult(alpha2Result.getResult());
             report(alpha2Result.getResult());
-        }
+		}
 
-        return alpha2Result.getStatus();
+		return alpha2Result.getStatus();
     }
 
-    public CountryCodeTaggerResult performAlpha2(Item item, CountryCodeTaggerConfig config)
+    public CountryCodeTaggerResult performAlpha2(Item item, CountryCodeTaggerConfig config) throws IOException, SQLException {
-            throws IOException, SQLException {
         CountryCodeTaggerResult alpha2Result = new CountryCodeTaggerResult();
         String itemHandle = item.getHandle();
 
-        List<MetadataValue> itemCountries =
+        List<MetadataValue> itemCountries = itemService.getMetadataByMetadataString(item, config.iso3166Field);
-                itemService.getMetadataByMetadataString(item, config.iso3166Field);
 
         // skip items that don't have country metadata
         if (itemCountries.size() == 0) {
@@ -96,60 +95,36 @@ public class CountryCodeTagger extends AbstractCurationTask {
         } else {
             Gson gson = new Gson();
 
-            // TODO: convert to try:
+            // TODO: convert to try: https://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html
-            // https://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html
+            BufferedReader reader = new BufferedReader(new InputStreamReader(this.getClass().getResourceAsStream(config.isocodesJsonPath)));
-            BufferedReader reader =
+            ISO3166CountriesVocabulary isocodesCountriesJson = gson.fromJson(reader, ISO3166CountriesVocabulary.class);
-                    new BufferedReader(
-                            new InputStreamReader(
-                                    this.getClass().getResourceAsStream(config.isocodesJsonPath)));
-            ISO3166CountriesVocabulary isocodesCountriesJson =
-                    gson.fromJson(reader, ISO3166CountriesVocabulary.class);
             reader.close();
 
-            reader =
+            reader = new BufferedReader(new InputStreamReader(this.getClass().getResourceAsStream(config.cgspaceCountriesJsonPath)));
-                    new BufferedReader(
+            CGSpaceCountriesVocabulary cgspaceCountriesJson = gson.fromJson(reader, CGSpaceCountriesVocabulary.class);
-                            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
+            // split the alpha2 country code field into schema, element, and qualifier so we can use it with item.addMetadata()
-            // it with item.addMetadata()
             String[] iso3166Alpha2FieldParts = config.iso3166Alpha2Field.split("\\.");
 
             if (config.forceupdate) {
-                itemService.clearMetadata(
+                itemService.clearMetadata(Curator.curationContext(), item, iso3166Alpha2FieldParts[0], iso3166Alpha2FieldParts[1], iso3166Alpha2FieldParts[2], Item.ANY);
-                        Curator.curationContext(),
-                        item,
-                        iso3166Alpha2FieldParts[0],
-                        iso3166Alpha2FieldParts[1],
-                        iso3166Alpha2FieldParts[2],
-                        Item.ANY);
             }
 
             // check the item's country codes, if any
-            List<MetadataValue> itemAlpha2CountryCodes =
+            List<MetadataValue> itemAlpha2CountryCodes = itemService.getMetadataByMetadataString(item, config.iso3166Alpha2Field);
-                    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())
+                        if (itemCountry.getValue().equalsIgnoreCase(country.getName()) || itemCountry.getValue().equalsIgnoreCase(country.get_official_name()) || itemCountry.getValue().equalsIgnoreCase(country.get_common_name())) {
-                                || 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());
@@ -159,14 +134,7 @@ public class CountryCodeTagger extends AbstractCurationTask {
 
                 if (newAlpha2Codes.size() > 0) {
                     try {
-                        itemService.addMetadata(
+                        itemService.addMetadata(Curator.curationContext(), item, iso3166Alpha2FieldParts[0], iso3166Alpha2FieldParts[1], iso3166Alpha2FieldParts[2], "en_US", newAlpha2Codes);
-                                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());
@@ -174,11 +142,7 @@ public class CountryCodeTagger extends AbstractCurationTask {
                         alpha2Result.setStatus(Curator.CURATE_ERROR);
                     }
 
-                    alpha2Result.setResult(
+                    alpha2Result.setResult(itemHandle + ": added " + newAlpha2Codes.size() + " alpha2 country code(s)");
-                            itemHandle
-                                    + ": added "
-                                    + newAlpha2Codes.size()
-                                    + " alpha2 country code(s)");
                 } else {
                     alpha2Result.setResult(itemHandle + ": no matching countries found");
                 }

src/main/java/io/github/ilri/cgspace/ctasks/ISO3166CountriesVocabulary.java

@@ -1,18 +1,15 @@
 /*
- * 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;
 
 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
+    // support reading iso_3166-1.json from Debian's iso-codes package using SerializedName since our class needs to match the JSON exactly
-    // our class needs to match the JSON exactly
+    @SerializedName("3166-1") List<Country> countries;
-    @SerializedName("3166-1")
-    List<Country> countries;
 }

src/main/java/io/github/ilri/cgspace/scripts/FixJpgJpgThumbnails.java

@@ -1,27 +1,21 @@
 /*
- * 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;
 
 import org.apache.commons.lang.StringUtils;
 import org.dspace.authorize.AuthorizeException;
-import org.dspace.content.Bitstream;
+import org.dspace.content.*;
-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.Context;
+import org.dspace.content.factory.ContentServiceFactory;
+import org.dspace.content.service.ItemService;
 import org.dspace.handle.factory.HandleServiceFactory;
 import org.dspace.handle.service.HandleService;
+import org.dspace.content.service.BundleService;
 
 import java.io.IOException;
 import java.sql.SQLException;
@@ -35,139 +29,118 @@ import java.util.List;
  * @since 5.1
  */
 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 =
+	public static HandleService handleService = HandleServiceFactory.getInstance().getHandleService();
-            HandleServiceFactory.getInstance().getHandleService();
+	public static BundleService bundleService = ContentServiceFactory.getInstance().getBundleService();
-    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(
+							process(context, itemService.findByCollection(context, (Collection) parent));
-                                    context,
+							break;
-                                    itemService.findByCollection(context, (Collection) parent));
+						case Constants.COMMUNITY:
-                            break;
+							List<Collection> collections = ((Community) parent).getCollections();
-                        case Constants.COMMUNITY:
+							for (Collection collection : collections) {
-                            List<Collection> collections = ((Community) parent).getCollections();
+								process(context, itemService.findAllByCollection(context, collection));
-                            for (Collection collection : collections) {
+							}
-                                process(
+							break;
-                                        context,
+						case Constants.SITE:
-                                        itemService.findAllByCollection(context, collection));
+							process(context, itemService.findAll(context));
-                            }
+							break;
-                            break;
+						case Constants.ITEM:
-                        case Constants.SITE:
+							processItem(context, (Item) parent);
-                            process(context, itemService.findAll(context));
+							context.commit();
-                            break;
+							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()) {
-        } catch (SQLException | AuthorizeException | IOException e) {
+				context.abort();
-            e.printStackTrace(System.err);
+			}
-        } finally {
+		}
-            if (context != null && context.isValid()) {
+	}
-                context.abort();
-            }
-        }
-    }
 
-    private static void process(Context context, Iterator<Item> items)
+	private static void process(Context context, Iterator<Item> items) throws SQLException, IOException, AuthorizeException {
-            throws SQLException, IOException, AuthorizeException {
+		while (items.hasNext()) {
-        while (items.hasNext()) {
+			Item item = items.next();
-            Item item = items.next();
+			processItem(context, item);
-            processItem(context, item);
+			itemService.update(context, item);
-            itemService.update(context, item);
+		}
-        }
+	}
-    }
 
-    private static void processItem(Context context, Item item)
+	private static void processItem(Context context, Item item) throws SQLException, AuthorizeException, IOException {
-            throws SQLException, AuthorizeException, IOException {
+		// Some bitstreams like Infographics are large JPGs and put in the ORIGINAL bundle on purpose so we shouldn't
-        // Some bitstreams like Infographics and Maps are large JPEGs and put in the ORIGINAL bundle
+		// swap them.
-        // on purpose so we shouldn't
+		List<MetadataValue> itemTypes = itemService.getMetadataByMetadataString(item, "dcterms.type");
-        // swap them.
+		boolean itemHasInfographic = false;
-        List<MetadataValue> itemTypes =
+		for (MetadataValue itemType: itemTypes) {
-                itemService.getMetadataByMetadataString(item, "dcterms.type");
+			if (itemType.getValue().equals("Infographic")) {
-        for (MetadataValue itemType : itemTypes) {
+				itemHasInfographic = true;
-            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();
 
-                // There is no point continuing if the thumbnail's description is empty or null
+				if (thumbnailName.toLowerCase().contains(".jpg.jpg")) {
-                if (StringUtils.isEmpty(thumbnailDescription)) {
+					List<Bundle> originalBundles = item.getBundles("ORIGINAL");
-                    continue;
+					for (Bundle originalBundle : originalBundles) {
-                }
+						List<Bitstream> originalBundleBitstreams = originalBundle.getBitstreams();
 
-                if (thumbnailName.toLowerCase().contains(".jpg.jpg")) {
+						for (Bitstream originalBitstream : originalBundleBitstreams) {
-                    List<Bundle> originalBundles = item.getBundles("ORIGINAL");
+							String originalName = originalBitstream.getName();
-                    for (Bundle originalBundle : originalBundles) {
-                        List<Bitstream> originalBundleBitstreams = originalBundle.getBitstreams();
 
-                        for (Bitstream originalBitstream : originalBundleBitstreams) {
+							long originalBitstreamBytes = originalBitstream.getSize();
-                            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
-                            - check if the original file name is the same as the thumbnail name minus the extra ".jpg"
+								bundleService.addBitstream(context, thumbnailBundle, originalBitstream);
-                            - check if the thumbnail description indicates it was automatically generated
+								//remove the original bitstream from the ORIGINAL bundle
-                            - check if the original bitstream is less than ~100KiB
+								originalBundle.removeBitstream(originalBitstream);
-                                - Note: in my tests there were 4022 items with ".jpg.jpg" thumbnails totaling 394549249
+								//remove the JpgJpg bitstream from the THUMBNAIL bundle
-                                  bytes for an average of about 98KiB so ~100KiB seems like a good cut off
+								thumbnailBundle.removeBitstream(thumbnailBitstream);
-                            */
+							}
-                            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);
-                            }
-                        }
-                    }
-                }
-            }
-        }
-    }
 }

src/main/java/io/github/ilri/cgspace/scripts/FixLowQualityThumbnails.java

@@ -1,19 +1,14 @@
 /*
- * 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;
 
 import org.apache.commons.lang.StringUtils;
 import org.dspace.authorize.AuthorizeException;
-import org.dspace.content.Bitstream;
+import org.dspace.content.*;
-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.service.BundleService;
 import org.dspace.content.service.ItemService;