Fix optional handling and misc small fixes (#16)

Properly handle optional files Use enhanced switch where possible Check if element is null before getting as float Optimize imports Add description method (slightly pointless)
qupath · Feb 5, 2025 · d2e0175 · d2e0175
1 parent 7578c59
commit d2e0175
Show file tree

Hide file tree

Showing 6 changed files with 53 additions and 41 deletions.
diff --git a/src/main/java/qupath/bioimageio/spec/Model.java b/src/main/java/qupath/bioimageio/spec/Model.java
@@ -31,7 +31,6 @@
 import java.util.Collections;
 import java.util.List;
 import java.util.Map;
-import java.util.Optional;
 import java.util.Set;
 import java.util.stream.Collectors;
 import java.util.stream.Stream;
@@ -285,7 +284,7 @@ public List<String> getTestInputs() {
         if (ti.isEmpty() && isFormatNewerThan("0.5")) {
             ti = inputs.stream()
                     .map(BaseTensor::getTestTensor)
-                    .map(fd -> fd.get().source())
+                    .map(ofd -> ofd.orElse(NULL_FILE).source())
                     .collect(Collectors.toList());
         }
         return toUnmodifiableList(ti);
@@ -304,8 +303,7 @@ public List<String> getTestOutputs() {
         if (to.isEmpty() && isFormatNewerThan("0.5")) {
             to = outputs.stream()
                     .map(BaseTensor::getTestTensor)
-                    .map(ofd -> ofd.flatMap(fd -> Optional.of(fd.source())))
-                    .map(Object::toString)
+                    .map(ofd -> ofd.orElse(NULL_FILE).source())
                     .collect(Collectors.toList());
         }
         return toUnmodifiableList(to);
@@ -412,7 +410,7 @@ private static Path findRdf(Path path, Collection<String> names) throws IOExcept
         if (Files.isDirectory(path)) {
             // Check directory
             try (Stream<Path> pathStream = Files.list(path)) {
-                List<Path> yamlFiles = pathStream.filter(Model::isYamlPath).collect(Collectors.toList());
+                List<Path> yamlFiles = pathStream.filter(Model::isYamlPath).toList();
                 if (yamlFiles.isEmpty())
                     return null;
                 if (yamlFiles.size() == 1)

diff --git a/src/main/java/qupath/bioimageio/spec/Resource.java b/src/main/java/qupath/bioimageio/spec/Resource.java
@@ -135,7 +135,7 @@ public String getFormatVersion() {
     }
 
     /**
-     * A [SPDX license identifier](https://spdx.org/licenses/).
+     * A <a href="https://spdx.org/licenses/">SPDX license identifier</a>.
      * We do not support custom license beyond the SPDX license list.
      * @return the license
      */

diff --git a/src/main/java/qupath/bioimageio/spec/tensor/Processing.java b/src/main/java/qupath/bioimageio/spec/tensor/Processing.java
@@ -354,53 +354,61 @@ public Processing deserialize(JsonElement json, Type typeOfT, JsonDeserializatio
             var obj = json.getAsJsonObject();
             var name = obj.has("name") ? obj.get("name").getAsString() : obj.get("id").getAsString();
             JsonObject kwargs = deserializeField(context, obj, "kwargs", JsonObject.class, null);
-            switch (name) {
-                case "binarize":
-                    var binarize = new Processing.Binarize();
+            return switch (name) {
+                case "binarize" -> {
+                    var binarize = new Binarize();
                     binarize.threshold = deserializeField(context, kwargs, "threshold", Double.class, true);
-                    return binarize;
-                case "clip":
-                    var clip = new Processing.Clip();
+                    yield binarize;
+                }
+                case "clip" -> {
+                    var clip = new Clip();
                     clip.min = deserializeField(context, kwargs, "min", Double.class, Double.NEGATIVE_INFINITY);
                     clip.max = deserializeField(context, kwargs, "max", Double.class, Double.POSITIVE_INFINITY);
-                    return clip;
-                case "scale_linear":
-                    var scaleLinear = new Processing.ScaleLinear();
+                    yield clip;
+                }
+                case "scale_linear" -> {
+                    var scaleLinear = new ScaleLinear();
                     scaleLinear.axes = deserializeField(context, kwargs, "axes", Axis[].class, false);
                     scaleLinear.gain = deserializeField(context, kwargs, "gain", double[].class, false);
                     scaleLinear.offset = deserializeField(context, kwargs, "offset", double[].class, false);
-                    return scaleLinear;
-                case "scale_mean_variance":
-                    var scaleMeanVariance = new Processing.ScaleMeanVariance();
-                    scaleMeanVariance.mode = deserializeField(context, kwargs, "mode", Processing.ProcessingMode.class, false);
+                    yield scaleLinear;
+                }
+                case "scale_mean_variance" -> {
+                    var scaleMeanVariance = new ScaleMeanVariance();
+                    scaleMeanVariance.mode = deserializeField(context, kwargs, "mode", ProcessingMode.class, false);
                     scaleMeanVariance.axes = deserializeField(context, kwargs, "axes", Axis[].class, false);
                     scaleMeanVariance.eps = deserializeField(context, kwargs, "eps", Double.class, 1e-6);
                     scaleMeanVariance.referenceTensor = deserializeField(context, kwargs, "reference_tensor", String.class, null);
-                    return scaleMeanVariance;
-                case "scale_range":
-                    var scaleRange = new Processing.ScaleRange();
-                    scaleRange.mode = deserializeField(context, kwargs, "mode", Processing.ProcessingMode.class, false);
+                    yield scaleMeanVariance;
+                }
+                case "scale_range" -> {
+                    var scaleRange = new ScaleRange();
+                    scaleRange.mode = deserializeField(context, kwargs, "mode", ProcessingMode.class, false);
                     scaleRange.axes = deserializeField(context, kwargs, "axes", Axis[].class, false);
                     scaleRange.eps = deserializeField(context, kwargs, "eps", Double.class, 1e-6);
                     scaleRange.referenceTensor = deserializeField(context, obj, "reference_tensor", String.class, null);
                     scaleRange.maxPercentile = deserializeField(context, kwargs, "max_percentile", Double.class, 0.0);
                     scaleRange.minPercentile = deserializeField(context, kwargs, "min_percentile", Double.class, 100.0);
-                    return scaleRange;
-                case "sigmoid":
-                    return new Processing.Sigmoid();
-                case "zero_mean_unit_variance":
-                    var zeroMeanUnitVariance = new Processing.ZeroMeanUnitVariance();
-                    zeroMeanUnitVariance.mode = deserializeField(context, kwargs, "mode", Processing.ProcessingMode.class, false);
+                    yield scaleRange;
+                }
+                case "sigmoid" -> {
+                    yield new Sigmoid();
+                }
+                case "zero_mean_unit_variance" -> {
+                    var zeroMeanUnitVariance = new ZeroMeanUnitVariance();
+                    zeroMeanUnitVariance.mode = deserializeField(context, kwargs, "mode", ProcessingMode.class, false);
                     zeroMeanUnitVariance.axes = deserializeField(context, kwargs, "axes", Axis[].class, false);
                     zeroMeanUnitVariance.eps = deserializeField(context, kwargs, "eps", Double.class, 1e-6);
                     zeroMeanUnitVariance.mean = deserializeField(context, kwargs, "mean", double[].class, false);
                     zeroMeanUnitVariance.std = deserializeField(context, kwargs, "std", double[].class, false);
-                    return zeroMeanUnitVariance;
-                default:
+                    yield zeroMeanUnitVariance;
+                }
+                default -> {
                     var processing = new Processing(name);
                     processing.kwargs = kwargs == null ? Collections.emptyMap() : context.deserialize(kwargs, Map.class);
-                    return processing;
-            }
+                    yield processing;
+                }
+            };
         }
 
     }

diff --git a/src/main/java/qupath/bioimageio/spec/tensor/Tensors.java b/src/main/java/qupath/bioimageio/spec/tensor/Tensors.java
@@ -214,8 +214,11 @@ public qupath.bioimageio.spec.tensor.TensorDataDescription deserialize(JsonEleme
                     range = List.of(Optional.empty(), Optional.empty());
                 } else {
                     range = r.getAsJsonArray().asList().stream()
-                            .map(JsonElement::getAsFloat)
-                            .map(Optional::of)
+                            .map(je -> {
+                                Optional<Float> o ;
+                                o = je.isJsonNull() ? Optional.empty() : Optional.of(je.getAsFloat());
+                                return o;
+                            })
                             .collect(Collectors.toList());
                 }
                 JsonElement unit = jsonObject.get("unit");

diff --git a/src/main/java/qupath/bioimageio/spec/tensor/axes/Axis.java b/src/main/java/qupath/bioimageio/spec/tensor/axes/Axis.java
@@ -16,15 +16,10 @@
 
 package qupath.bioimageio.spec.tensor.axes;
 
-import com.google.gson.JsonDeserializer;
-import qupath.bioimageio.spec.tensor.sizes.Size;
 import qupath.bioimageio.spec.tensor.BaseTensor;
+import qupath.bioimageio.spec.tensor.sizes.Size;
 
-import java.util.Arrays;
-import java.util.HashMap;
 import java.util.List;
-import java.util.Map;
-import java.util.stream.Collectors;
 
 /**
  * Base axis class for 0.4 and 0.5 axes.

diff --git a/src/main/java/qupath/bioimageio/spec/tensor/axes/AxisBase.java b/src/main/java/qupath/bioimageio/spec/tensor/axes/AxisBase.java
@@ -31,4 +31,12 @@ abstract class AxisBase implements Axis {
     public String getID() {
         return this.id;
     }
+
+    /**
+     * Gets the description of the axis, if present
+     * @return the free text description (usually null)
+     */
+    public String getDescription() {
+        return description;
+    }
 }