Make filters (de-)serialize to the correct thing

saalfeldlab · Oct 19, 2023 · 9dc961c · 9dc961c
1 parent edccdaa
commit 9dc961c
Show file tree

Hide file tree

Showing 4 changed files with 70 additions and 35 deletions.
diff --git a/src/main/java/org/janelia/saalfeldlab/n5/zarr/DType.java b/src/main/java/org/janelia/saalfeldlab/n5/zarr/DType.java
@@ -28,13 +28,11 @@
  */
 package org.janelia.saalfeldlab.n5.zarr;
 
-import java.lang.reflect.Type;
 import java.nio.ByteBuffer;
 import java.nio.ByteOrder;
 import java.util.ArrayList;
 import java.util.Collection;
 import java.util.EnumMap;
-import java.util.List;
 
 import org.janelia.saalfeldlab.n5.ByteArrayDataBlock;
 import org.janelia.saalfeldlab.n5.DataBlock;
@@ -45,13 +43,7 @@
 import org.janelia.saalfeldlab.n5.LongArrayDataBlock;
 import org.janelia.saalfeldlab.n5.ShortArrayDataBlock;
 
-import com.google.gson.JsonDeserializationContext;
-import com.google.gson.JsonDeserializer;
-import com.google.gson.JsonElement;
-import com.google.gson.JsonParseException;
-import com.google.gson.JsonPrimitive;
-import com.google.gson.JsonSerializationContext;
-import com.google.gson.JsonSerializer;
+import static org.janelia.saalfeldlab.n5.zarr.Filter.VLEN_UTF8;
 
 /**
  * Enumerates available zarr data types as defined at
@@ -219,8 +211,15 @@ public DType(final String typestr, final Collection<Filter> filters) {
 		case OBJECT:
 			nBytes = 1;
 			nBits = 0;
-			dataBlockFactory = null;
-			byteBlockFactory = null;
+			if (filters.contains(VLEN_UTF8)) {
+				dataBlockFactory = (blockSize, gridPosition, numElements) ->
+						new ZarrCompatibleStringDataBlock(blockSize, gridPosition, new String[0]);
+				byteBlockFactory = (blockSize, gridPosition, numElements) ->
+						new ByteArrayDataBlock(blockSize, gridPosition, new byte[numElements * nBytes]);
+			} else {
+				dataBlockFactory = null;
+				byteBlockFactory = null;
+			}
 			break;
 //		case BOOLEAN:
 //		case OTHER:     // not sure about this
@@ -237,7 +236,7 @@ public DType(final String typestr, final Collection<Filter> filters) {
 					new ByteArrayDataBlock(blockSize, gridPosition, new byte[numElements * nBytes]);
 		}
 
-		dataType = getDataType(primitive, nBytes);
+		dataType = getDataType(primitive, nBytes, filters);
 	}
 
 	public DType(final DataType dataType, final int nPrimitives) {
@@ -279,6 +278,11 @@ public DType(final DataType dataType, final int nPrimitives) {
 			break;
 //		case INT8:
 //		case UINT8:
+		case STRING:
+			nBytes = 1;
+			dataBlockFactory = (blockSize, gridPosition, numElements) ->
+					new ZarrCompatibleStringDataBlock(blockSize, gridPosition, new String[0]);
+			break;
 		default:
 			nBytes = nPrimitives;
 			dataBlockFactory = (blockSize, gridPosition, numElements) ->
@@ -300,7 +304,8 @@ public DataType getDataType() {
 
 	protected final static DataType getDataType(
 			final Primitive primitive,
-			final int nBytes) {
+			final int nBytes,
+			final Collection<Filter> filters) {
 
 		switch (primitive) {
 		case INT:
@@ -344,7 +349,10 @@ protected final static DataType getDataType(
 				return DataType.UINT8; // fallback
 			}
 		case OBJECT:
-			return DataType.OBJECT;
+			if (filters.contains(VLEN_UTF8))
+				return DataType.STRING;
+			else
+				return DataType.OBJECT;
 		default:
 			return DataType.UINT8; // fallback
 		}
@@ -362,10 +370,10 @@ public String toString() {
 	 *
 	 * @return list of filters
 	 */
-	public List<Filter> getFilters() {
+	public Collection<Filter> getFilters() {
 		if (dataType == DataType.STRING) {
 			ArrayList<Filter> filterSet = new ArrayList<>();
-			filterSet.add(new ZarrCompatibleStringDataBlock.VLenStringFilter());
+			filterSet.add(VLEN_UTF8);
 			return filterSet;
 		}
 		else

diff --git a/src/main/java/org/janelia/saalfeldlab/n5/zarr/Filter.java b/src/main/java/org/janelia/saalfeldlab/n5/zarr/Filter.java
@@ -33,39 +33,69 @@
 import com.google.gson.JsonElement;
 import com.google.gson.JsonObject;
 import com.google.gson.JsonParseException;
+import com.google.gson.JsonPrimitive;
+import com.google.gson.JsonSerializationContext;
+import com.google.gson.JsonSerializer;
 
 import java.lang.reflect.Type;
 
 /**
- * Place holder interface for filters
+ * Filter types
  *
  * TODO implement some
  *
  * @author Stephan Saalfeld &lt;[email protected]&gt;
+ * @author Michael Innerberger
  */
-public interface Filter {
+public enum Filter {
+    // Note: the JSON (de-)serializer below is very much tailored to this filter, which serializes to "{"id":"vlen-utf8"}"
+    // If additional filters are implemented, consider also changing the type adapter below
+    VLEN_UTF8("vlen-utf8");
 
-    public static JsonAdapter jsonAdapter = new Filter.JsonAdapter();
+    private final String id;
 
-    class JsonAdapter implements JsonDeserializer<Filter> {
+    Filter(final String id) {
+        this.id = id;
+    }
+
+    public String getId() {
+        return id;
+    }
+
+    public static Filter fromString(final String id) {
+        for (final Filter filter : values())
+            if (filter.getId().equals(id))
+                return filter;
+        return null;
+    }
+
+    public static final JsonAdapter jsonAdapter = new JsonAdapter();
+
+    public static class JsonAdapter implements JsonDeserializer<Filter>, JsonSerializer<Filter> {
 
         @Override
         public Filter deserialize(
                 final JsonElement json,
                 final Type typeOfT,
                 final JsonDeserializationContext context) throws JsonParseException {
 
-            final JsonObject jsonObject = json.getAsJsonObject();
-            final JsonElement jsonId = jsonObject.get("id");
+            final JsonElement jsonId = json.getAsJsonObject().get("id");
             if (jsonId == null)
                 return null;
-            final String id = jsonId.getAsString();
-            switch (id) {
-            case "vlen-utf8":
-                return new ZarrCompatibleStringDataBlock.VLenStringFilter();
-            default:
-                return null;
-            }
+
+            final String stringId = jsonId.getAsString();
+            return Filter.fromString(stringId);
+        }
+
+        @Override
+        public JsonElement serialize(
+                final Filter filter,
+                final Type typeOfSrc,
+                final JsonSerializationContext context) {
+
+            final JsonObject serialization = new JsonObject();
+            serialization.add("id", new JsonPrimitive(filter.getId()));
+            return serialization;
         }
     }
 }
diff --git a/src/main/java/org/janelia/saalfeldlab/n5/zarr/ZarrCompatibleStringDataBlock.java b/src/main/java/org/janelia/saalfeldlab/n5/zarr/ZarrCompatibleStringDataBlock.java
@@ -67,8 +67,4 @@ protected String[] deserialize(byte[] rawBytes) {
 
         return strings;
     }
-
-    public static class VLenStringFilter implements Filter {
-        public final String id = "vlen-utf8";
-    }
 }
diff --git a/src/main/java/org/janelia/saalfeldlab/n5/zarr/ZarrKeyValueWriter.java b/src/main/java/org/janelia/saalfeldlab/n5/zarr/ZarrKeyValueWriter.java
@@ -368,17 +368,18 @@ protected ZArrayAttributes createZArrayAttributes(final DatasetAttributes datase
 		reorder(shape);
 		final int[] chunks = datasetAttributes.getBlockSize().clone();
 		reorder(chunks);
+		final DType dType = new DType(datasetAttributes.getDataType());
 
 		final ZArrayAttributes zArrayAttributes = new ZArrayAttributes(
 				N5ZarrReader.VERSION.getMajor(),
 				shape,
 				chunks,
-				new DType(datasetAttributes.getDataType()),
+				dType,
 				ZarrCompressor.fromCompression(datasetAttributes.getCompression()),
 				"0",
 				'C',
 				dimensionSeparator,
-				null);
+				dType.getFilters());
 
 		return zArrayAttributes;
 	}