Parse dashboard operators independently of quoted filter values

great_ai/deploy/routes/dashboard/get_filter_from_datatable.py

@@ -1,23 +1,32 @@
-from typing import Optional, Union
+import re
+from typing import Optional, Union, cast
 
-from ....views import Filter, operators
+from ....views import Filter
+from ....views.operators import Operator
+
+_expression = re.compile(
+    r"\s*\{([^}]+)\}\s*(>=|<=|!=|=|>|<|contains\b|ge\b|le\b|lt\b|gt\b|ne\b|eq\b)\s*(.*?)\s*",
+    re.DOTALL,
+)
+_aliases = {"ge": ">=", "le": "<=", "lt": "<", "gt": ">", "ne": "!=", "eq": "="}
 
 
 def get_filter_from_datatable(description: str) -> Optional[Filter]:
-    for operator in operators:
-        if operator in description:
-            name_part, value_part = description.split(operator, 1)
-            value_part = value_part.strip()
-            name_part = name_part[name_part.find("{") + 1 : name_part.rfind("}")]
-
-            v0 = value_part[0]
-            if v0 == value_part[-1] and v0 in ("'", '"', "`"):
-                value: Union[str, float] = value_part[1:-1].replace("\\" + v0, v0)
-            else:
-                try:
-                    value = float(value_part)
-                except ValueError:
-                    value = value_part
-            return Filter(property=name_part, operator=operator, value=value)
-
-    return None
+    match = _expression.fullmatch(description)
+    if match is None:
+        return None
+    name, token, raw = match.groups()
+    if not raw:
+        return None
+    if raw[0] in ("'", '"', "`"):
+        if len(raw) < 2 or raw[-1] != raw[0]:
+            return None
+        value: Union[str, float] = re.sub(r"\\([\\'\"`])", r"\1", raw[1:-1])
+    else:
+        try:
+            value = float(raw)
+        except ValueError:
+            value = raw
+    return Filter(
+        property=name, operator=cast(Operator, _aliases.get(token, token)), value=value
+    )

tests/test_dashboard_regressions.py

@@ -0,0 +1,32 @@
+from typing import Any
+
+import pytest
+
+from great_ai.deploy.routes.dashboard.get_filter_from_datatable import (
+    get_filter_from_datatable,
+)
+
+
+@pytest.mark.parametrize(
+    "expression, operator, value",
+    [
+        ('{value} contains "a=b"', "contains", "a=b"),
+        ("{value} eq 7", "=", 7),
+        ("{value} ge 2.5", ">=", 2.5),
+        ('{value} contains ""', "contains", ""),
+        (r"{value} contains 'it\'s > 2'", "contains", "it's > 2"),
+    ],
+)
+def test_dashboard_filter_grammar(expression: str, operator: str, value: Any) -> None:
+    result = get_filter_from_datatable(expression)
+    assert result is not None
+    assert result.property == "value"
+    assert result.operator == operator
+    assert result.value == value
+
+
+@pytest.mark.parametrize(
+    "expression", ["{value} > ", '{value} contains "unfinished', "invalid"]
+)
+def test_incomplete_filters_are_ignored(expression: str) -> None:
+    assert get_filter_from_datatable(expression) is None