adds unit tests for text word weighting in TextQuery class

Author: justin-cechmanek

redisvl/query/query.py

@@ -1155,7 +1155,9 @@ def _tokenize_and_escape_query(self, user_query: str) -> str:
             )
             for token in user_query.split()
         ]
-        token_list = [token for token in tokens if token and token not in self._stopwords]
+        token_list = [
+            token for token in tokens if token and token not in self._stopwords
+        ]
         for i, token in enumerate(token_list):
             if token in self._text_weights:
                 token_list[i] = f"{token}=>{{weight:{self._text_weights[token]}}}"
@@ -1227,18 +1229,29 @@ def text_field_name(self) -> Union[str, Dict[str, float]]:
                 return field
         return self._field_weights.copy()
 
-    def _parse_text_weights(self, weights: Dict[str, float]) -> Dict[str, float]:
-        parsed_weights = {}
+    def _parse_text_weights(
+        self, weights: Optional[Dict[str, float]]
+    ) -> Dict[str, float]:
+        parsed_weights: Dict[str, float] = {}
+        if not weights:
+            return parsed_weights
         for word, weight in weights.items():
             word = word.strip().lower()
             if not word or " " in word:
-                raise ValueError("Only individual words may be weighted. Got {{ {word}:{weight} }}")
-            if not isinstance(weight, float) or weight <0.0:
-                raise ValueError("Weights must be positive floats. Got {{ {word}:{weight} }}")
+                raise ValueError(
+                    f"Only individual words may be weighted. Got {{ {word}:{weight} }}"
+                )
+            if (
+                not (isinstance(weight, float) or isinstance(weight, int))
+                or weight < 0.0
+            ):
+                raise ValueError(
+                    f"Weights must be positive number. Got {{ {word}:{weight} }}"
+                )
             parsed_weights[word] = weight
         return parsed_weights
 
-    def set_text_weights(self, weights:Dict[str, float]):
+    def set_text_weights(self, weights: Dict[str, float]):
         """Set or update the text weights for the query.
 
         Args:
@@ -1248,7 +1261,7 @@ def set_text_weights(self, weights:Dict[str, float]):
         self._built_query_string = None
 
     @property
-    def text_weights() -> Dict[str, float]:
+    def text_weights(self) -> Dict[str, float]:
         """Get the text weights.
 
         Returns:

tests/unit/test_query_types.py

@@ -333,6 +333,64 @@ def test_text_query_with_string_filter():
     assert "AND" not in query_string_wildcard
 
 
+def test_text_query_word_weights():
+    # verify word weights get added into the raw Redis query syntax
+    query = TextQuery(
+        text="query string alpha bravo delta tango alpha",
+        text_field_name="description",
+        text_weights={"alpha": 2, "delta": 0.555, "gamma": 0.95},
+    )
+
+    assert (
+        str(query)
+        == "@description:(query | string | alpha=>{weight:2} | bravo | delta=>{weight:0.555} | tango | alpha=>{weight:2}) SCORER BM25STD WITHSCORES DIALECT 2 LIMIT 0 10"
+    )
+
+    # raise an error if weights are not positive floats
+    with pytest.raises(ValueError):
+        _ = TextQuery(
+            text="sample text query",
+            text_field_name="description",
+            text_weights={"first": 0.2, "second": -0.1},
+        )
+
+    with pytest.raises(ValueError):
+        _ = TextQuery(
+            text="sample text query",
+            text_field_name="description",
+            text_weights={"first": 0.2, "second": "0.1"},
+        )
+
+    # no error is weights dictiionary is empty or None
+    query = TextQuery(
+        text="sample text query", text_field_name="description", text_weights={}
+    )
+    assert query
+
+    query = TextQuery(
+        text="sample text query", text_field_name="description", text_weights=None
+    )
+    assert query
+
+    # no error if the words in weights dictionary don't appear in query
+    query = TextQuery(
+        text="sample text query",
+        text_field_name="description",
+        text_weights={"alpha": 0.2, "bravo": 0.4},
+    )
+    assert query
+
+    # we can access the word weights on a query object
+    assert query.text_weights == {"alpha": 0.2, "bravo": 0.4}
+
+    # we can change the text weights on a query object
+    query.set_text_weights(weights={"new": 0.3, "words": 0.125, "here": 99})
+    assert query.text_weights == {"new": 0.3, "words": 0.125, "here": 99}
+
+    query.set_text_weights(weights={})
+    assert query.text_weights == {}
+
+
 @pytest.mark.parametrize(
     "query",
     [