warn when an unsupported/unknown key is given to the dependency matcher

svlandeg · Aug 21, 2023 · 1d0248d · 1d0248d
1 parent 198488e
commit 1d0248d
Show file tree

Hide file tree

Showing 3 changed files with 14 additions and 0 deletions.
diff --git a/spacy/errors.py b/spacy/errors.py
@@ -219,6 +219,7 @@ class Warnings(metaclass=ErrorsWithCodes):
     W125 = ("The StaticVectors key_attr is no longer used. To set a custom "
             "key attribute for vectors, configure it through Vectors(attr=) or "
             "'spacy init vectors --attr'")
+    W126 = ("These keys are unsupported: {unsupported}")
 
 
 class Errors(metaclass=ErrorsWithCodes):

diff --git a/spacy/matcher/dependencymatcher.pyx b/spacy/matcher/dependencymatcher.pyx
@@ -129,13 +129,21 @@ cdef class DependencyMatcher:
             else:
                 required_keys = {"RIGHT_ID", "RIGHT_ATTRS", "REL_OP", "LEFT_ID"}
                 relation_keys = set(relation.keys())
+                # Identify required keys that have not been specified
                 missing = required_keys - relation_keys
                 if missing:
                     missing_txt = ", ".join(list(missing))
                     raise ValueError(Errors.E100.format(
                         required=required_keys,
                         missing=missing_txt
                     ))
+                # Identify additional, unsupported keys
+                unsupported = relation_keys - required_keys
+                if unsupported:
+                    unsupported_txt = ", ".join(list(unsupported))
+                    warnings.warn(Warnings.W126.format(
+                        unsupported=unsupported_txt
+                    ))
                 if (
                     relation["RIGHT_ID"] in visited_nodes
                     or relation["LEFT_ID"] not in visited_nodes

diff --git a/spacy/tests/matcher/test_dependency_matcher.py b/spacy/tests/matcher/test_dependency_matcher.py
@@ -216,6 +216,11 @@ def test_dependency_matcher_pattern_validation(en_vocab):
         pattern2 = copy.deepcopy(pattern)
         pattern2[1]["RIGHT_ID"] = "fox"
         matcher.add("FOUNDED", [pattern2])
+    # invalid key
+    with pytest.warns(UserWarning):
+        pattern2 = copy.deepcopy(pattern)
+        pattern2[1]["FOO"] = "BAR"
+        matcher.add("FOUNDED", [pattern2])
 
 
 def test_dependency_matcher_callback(en_vocab, doc):