Allow negative constraints that match an existing positive constraint. Actually, there may not even be a reason to disallow two positive constraints if they are both regex

Author: AngledLuffa

src/edu/stanford/nlp/semgraph/semgrex/NodeAttributes.java

@@ -12,7 +12,7 @@
 /**
  * Stores attributes for a Semgrex NodePattern.
  *<br>
- * Refactored out of the parser itself for a couple reason:
+ * Refactored out of the parser itself for a couple reasons:
  *<ul>
  *<li> Allows combining isRoot ($) with node restrictions (word:foo)
  *<li> Can pass this object around, allowing for more refactoring in the semgrex parser
@@ -60,10 +60,10 @@ public boolean empty() {
   }
 
   public void setAttribute(String key, String value, boolean negated) {
-    if (positiveAttributes.contains(key)) {
-      throw new SemgrexParseException("Duplicate attribute " + key + " found in semgrex expression");
-    }
     if (!negated) {
+      if (positiveAttributes.contains(key)) {
+        throw new SemgrexParseException("Duplicate attribute " + key + " found in semgrex expression");
+      }
       positiveAttributes.add(key);
     }
     attributes.add(new Triple(key, value, negated));

test/src/edu/stanford/nlp/semgraph/semgrex/SemgrexTest.java

@@ -1207,6 +1207,34 @@ public void testIllegal() {
     }
   }
 
+  public void testDuplicateConstraints() {
+    // There should be an exception if the same attribute shows up
+    // twice as a positive attribute
+    // Although it isn't clear that's necessary,
+    // since both portions could be regex which match different things
+    try {
+      String pattern = "{word:foo;word:bar}";
+      SemgrexPattern semgrex = SemgrexPattern.compile(pattern);
+      throw new RuntimeException("This expression is now illegal");
+    } catch (SemgrexParseException e) {
+      // yay
+    }
+
+    // this should parse since negative constraints which
+    // match positive constraints are allowed
+    String pattern = "{word:/.*i.*/;word!:/.*m.*/}";
+    SemgrexPattern semgrex = SemgrexPattern.compile(pattern);
+    runTest(pattern,
+            "[ate/NN subj>Bill/NN obj>[muffins compound>blueberry]]",
+            "Bill/NN");
+
+    pattern = "{word:/.*i.*/;word!:/.*z.*/}";
+    semgrex = SemgrexPattern.compile(pattern);
+    runTest(pattern,
+            "[ate/NN subj>Bill/NN obj>[muffins compound>blueberry]]",
+            "Bill/NN", "muffins");
+  }
+
   public void testAdjacent() {
     // test using a colon expression so that the targeted nodes
     // are the nodes which show up