From f77a9b414cb28c9429a7f100b76272355bf78bb6 Mon Sep 17 00:00:00 2001
From: John Bauer <horatio@gmail.com>
Date: Fri, 27 Oct 2023 21:46:47 -0700
Subject: [PATCH] Sort morphological features in lowercase alphabetical order. 
 Better fits with what is expected in UD datasets

---
 .../stanford/nlp/trees/ud/CoNLLUFeatures.java | 24 ++++++++++++++++---
 1 file changed, 21 insertions(+), 3 deletions(-)

diff --git a/src/edu/stanford/nlp/trees/ud/CoNLLUFeatures.java b/src/edu/stanford/nlp/trees/ud/CoNLLUFeatures.java
index 8e769b5dea..7363ce6a6b 100644
--- a/src/edu/stanford/nlp/trees/ud/CoNLLUFeatures.java
+++ b/src/edu/stanford/nlp/trees/ud/CoNLLUFeatures.java
@@ -15,6 +15,23 @@
  * which is necessary for the CoNLLU format
  */
 public class CoNLLUFeatures extends TreeMap<String, String> {
+  public static class LowercaseComparator implements Comparator<String> {
+    public int compare(String x, String y) {
+      if (x == null && y == null) {
+        return 0;
+      }
+      if (x == null) {
+        return -1;
+      }
+      if (y == null) {
+        return 1;
+      }
+      return x.compareToIgnoreCase(y);
+    }
+  }
+
+  static final LowercaseComparator comparator = new LowercaseComparator();
+
   /**
    * Parses the value of the feature column in a CoNLL-U file
    * and returns them in a HashMap with the feature names as keys
@@ -24,7 +41,7 @@ public class CoNLLUFeatures extends TreeMap<String, String> {
    * @return A {@code HashMap<String,String>} with the feature values.
    */
   public CoNLLUFeatures(String featureString) {
-    super();
+    super(comparator);
 
     if (!featureString.equals("_")) {
       String[] featValPairs = featureString.split("\\|");
@@ -36,11 +53,12 @@ public CoNLLUFeatures(String featureString) {
   }
 
   public CoNLLUFeatures(Map<String, String> features) {
-    super(features);
+    super(comparator);
+    putAll(features);
   }
 
   public CoNLLUFeatures() {
-    super();
+    super(comparator);
   }