Add support for boolean column filtering with is: and has: syntax

Co-authored-by: williambdean <57733339+williambdean@users.noreply.github.com>

Author: Copilot

src/frame_search/search.py

@@ -8,7 +8,7 @@
 
 Operator = Literal[":", ">", "<", ">=", "<="]
 
-Value = Union[str, float, int, datetime]
+Value = Union[str, float, int, datetime, bool]
 
 
 @dataclass
@@ -39,20 +39,28 @@ def is_standalone(self) -> bool:
 def parse_query(query: str):
     # Regex to capture:
     # 1. Optional negation operator (NOT or -)
-    # 2. key:operator:value (e.g., name:alice, age:">30", city:"New York")
-    #    - Group 2: key (\w+)
-    #    - Group 3: operator (:|>|<)
-    #    - Group 4: value (either "[^"]*" for quoted strings or \S+ for non-whitespace)
-    # 3. standalone value (\S+)
-    #    - Group 5: standalone value
-    pattern = r'(?:(NOT|-)\s*)?(\w+):("[^"]*"|\S+)|(?:(NOT|-)\s*)?(\S+)'
+    # 2. is:column or has:column for boolean columns (must come before key:value)
+    #    - Group 2: is or has
+    #    - Group 3: column name
+    # 3. key:operator:value (e.g., name:alice, age:">30", city:"New York")
+    #    - Group 4: key (\w+)
+    #    - Group 5: value (either "[^"]*" for quoted strings or \S+ for non-whitespace)
+    # 4. standalone value (\S+)
+    #    - Group 7: standalone value
+    pattern = r'(?:(NOT|-)\s*)?(is|has):(\w+)|(?:(NOT|-)\s*)?(\w+):("[^"]*"|\S+)|(?:(NOT|-)\s*)?(\S+)'
     return list(re.findall(pattern, query))
 
 
 def _parse_value(value: str) -> Value:
     if value.startswith('"') and value.endswith('"'):
         value = value[1:-1]
 
+    # Handle boolean values
+    if value.lower() == "true":
+        return True
+    elif value.lower() == "false":
+        return False
+
     try:
         return datetime.fromisoformat(value)
     except ValueError:
@@ -84,10 +92,24 @@ def get_search_parts(query: str) -> list[SearchPart]:
 
     search_parts = []
     for match in matches:
-        negation, key, value, standalone_negation, standalone_value = match
-        negated = bool(negation or standalone_negation)
-
-        if standalone_value:  # This is the standalone value group
+        (
+            is_has_negation,
+            is_has_keyword,
+            is_has_column,
+            negation,
+            key,
+            value,
+            standalone_negation,
+            standalone_value,
+        ) = match
+        negated = bool(negation or standalone_negation or is_has_negation)
+
+        if is_has_keyword:  # This is the is:column or has:column group
+            # For is: and has:, we treat them as column:True
+            search_parts.append(
+                SearchPart(key=is_has_column, operator=":", value=True, negated=negated)
+            )
+        elif standalone_value:  # This is the standalone value group
             search_parts.append(
                 SearchPart(
                     key=None, operator=None, value=standalone_value, negated=negated

tests/conftest.py

@@ -59,6 +59,8 @@ def search():
                 "Hobby": nw.String,
                 "City of Interest": nw.String,
                 "First Visit": nw.Datetime,
+                "seen_movie": nw.Boolean,
+                "older_than_30": nw.Boolean,
             },
         ),
     )

tests/test_search.py

@@ -39,29 +39,36 @@ def test_range_different_type_raises(lower, upper) -> None:
     "query, expected",
     [
         pytest.param("", [], id="empty-query"),
-        pytest.param("bob", [("", "", "", "", "bob")], id="default"),
+        pytest.param("bob", [("", "", "", "", "", "", "", "bob")], id="default"),
         pytest.param(
-            "name:alice", [("", "name", "alice", "", "")], id="contains-match"
+            "name:alice",
+            [("", "", "", "", "name", "alice", "", "")],
+            id="contains-match",
+        ),
+        pytest.param("age:>=30", [("", "", "", "", "age", ">=30", "", "")], id="ge"),
+        pytest.param("age:>30", [("", "", "", "", "age", ">30", "", "")], id="gt"),
+        pytest.param("age:<30", [("", "", "", "", "age", "<30", "", "")], id="lt"),
+        pytest.param(
+            "age:35.5", [("", "", "", "", "age", "35.5", "", "")], id="exact-numeric"
         ),
-        pytest.param("age:>=30", [("", "age", ">=30", "", "")], id="ge"),
-        pytest.param("age:>30", [("", "age", ">30", "", "")], id="gt"),
-        pytest.param("age:<30", [("", "age", "<30", "", "")], id="lt"),
-        pytest.param("age:35.5", [("", "age", "35.5", "", "")], id="exact-numeric"),
         pytest.param(
             "opening_date:<2023-01-01",
-            [("", "opening_date", "<2023-01-01", "", "")],
+            [("", "", "", "", "opening_date", "<2023-01-01", "", "")],
             id="date-lt",
         ),
         pytest.param(
             "bob age:>30",
-            [("", "", "", "", "bob"), ("", "age", ">30", "", "")],
+            [
+                ("", "", "", "", "", "", "", "bob"),
+                ("", "", "", "", "age", ">30", "", ""),
+            ],
             id="mixed",
         ),
         pytest.param(
             'hobby:reading city:"New York"',
             [
-                ("", "hobby", "reading", "", ""),
-                ("", "city", '"New York"', "", ""),
+                ("", "", "", "", "hobby", "reading", "", ""),
+                ("", "", "", "", "city", '"New York"', "", ""),
             ],
             id="multiple-conditions",
         ),
@@ -162,25 +169,21 @@ def test_get_search_parts(query, expected) -> None:
             "is:older_than_30",
             [2, 3],
             id="boolean-is",
-            marks=pytest.mark.xfail,
         ),
         pytest.param(
             "older_than_30:True",
             [2, 3],
             id="boolean-True",
-            marks=pytest.mark.xfail,
         ),
         pytest.param(
             "older_than_30:False",
             [0, 1],
             id="boolean-False",
-            marks=pytest.mark.xfail,
         ),
         pytest.param(
             "has:seen_movie",
             [0, 2],
             id="boolean-has",
-            marks=pytest.mark.xfail,
         ),
     ],
 )