pixeltable 0.3.14__py3-none-any.whl → 0.5.7__py3-none-any.whl

pixeltable/type_system.py

@@ -5,10 +5,15 @@ import datetime
 import enum
 import io
 import json
+import types
 import typing
 import urllib.parse
 import urllib.request
-from typing import Any, ClassVar, Iterable, Literal, Mapping, Optional, Sequence, Union
+import uuid
+from pathlib import Path
+from typing import Any, ClassVar, Iterable, Literal, Mapping, Sequence, Union
+
+from typing import _GenericAlias  # type: ignore[attr-defined]  # isort: skip
 
 import av
 import jsonschema
@@ -21,10 +26,9 @@ import sqlalchemy as sql
 from typing_extensions import _AnnotatedAlias
 
 import pixeltable.exceptions as excs
+from pixeltable.env import Env
 from pixeltable.utils import parse_local_file_path
 
-from typing import _GenericAlias  # type: ignore[attr-defined]  # isort: skip
-
 
 class ColumnType:
     @enum.unique
@@ -41,6 +45,8 @@ class ColumnType:
         AUDIO = 9
         DOCUMENT = 10
         DATE = 11
+        UUID = 12
+        BINARY = 13
 
         # exprs that don't evaluate to a computable value in Pixeltable, such as an Image member function
         INVALID = 255
@@ -48,11 +54,11 @@ class ColumnType:
         @classmethod
         def supertype(
             cls,
-            type1: Optional['ColumnType.Type'],
-            type2: Optional['ColumnType.Type'],
+            type1: 'ColumnType.Type' | None,
+            type2: 'ColumnType.Type' | None,
             # we need to pass this in because we can't easily append it as a class member
             common_supertypes: dict[tuple['ColumnType.Type', 'ColumnType.Type'], 'ColumnType.Type'],
-        ) -> Optional['ColumnType.Type']:
+        ) -> 'ColumnType.Type' | None:
             if type1 == type2:
                 return type1
             t = common_supertypes.get((type1, type2))
@@ -63,26 +69,8 @@ class ColumnType:
                 return t
             return None
 
-    @enum.unique
-    class DType(enum.Enum):
-        """
-        Base type used in images and arrays
-        """
-
-        BOOL = (0,)
-        INT8 = (1,)
-        INT16 = (2,)
-        INT32 = (3,)
-        INT64 = (4,)
-        UINT8 = (5,)
-        UINT16 = (6,)
-        UINT32 = (7,)
-        UINT64 = (8,)
-        FLOAT16 = (9,)
-        FLOAT32 = (10,)
-        FLOAT64 = 11
-
-    scalar_types: ClassVar[set[Type]] = {Type.STRING, Type.INT, Type.FLOAT, Type.BOOL, Type.TIMESTAMP, Type.DATE}
+    scalar_json_types: ClassVar[set[Type]] = {Type.STRING, Type.INT, Type.FLOAT, Type.BOOL}
+    scalar_types: ClassVar[set[Type]] = scalar_json_types | {Type.TIMESTAMP, Type.DATE, Type.UUID}
     numeric_types: ClassVar[set[Type]] = {Type.INT, Type.FLOAT}
     common_supertypes: ClassVar[dict[tuple[Type, Type], Type]] = {
         (Type.BOOL, Type.INT): Type.INT,
@@ -151,31 +139,37 @@ class ColumnType:
 
     @classmethod
     def make_type(cls, t: Type) -> ColumnType:
-        assert t != cls.Type.INVALID
-        if t == cls.Type.STRING:
-            return StringType()
-        if t == cls.Type.INT:
-            return IntType()
-        if t == cls.Type.FLOAT:
-            return FloatType()
-        if t == cls.Type.BOOL:
-            return BoolType()
-        if t == cls.Type.TIMESTAMP:
-            return TimestampType()
-        if t == cls.Type.ARRAY:
-            return ArrayType()
-        if t == cls.Type.JSON:
-            return JsonType()
-        if t == cls.Type.IMAGE:
-            return ImageType()
-        if t == cls.Type.VIDEO:
-            return VideoType()
-        if t == cls.Type.AUDIO:
-            return AudioType()
-        if t == cls.Type.DOCUMENT:
-            return DocumentType()
-        if t == cls.Type.DATE:
-            return DateType()
+        match t:
+            case cls.Type.STRING:
+                return StringType()
+            case cls.Type.INT:
+                return IntType()
+            case cls.Type.FLOAT:
+                return FloatType()
+            case cls.Type.BOOL:
+                return BoolType()
+            case cls.Type.TIMESTAMP:
+                return TimestampType()
+            case cls.Type.JSON:
+                return JsonType()
+            case cls.Type.ARRAY:
+                return ArrayType()
+            case cls.Type.IMAGE:
+                return ImageType()
+            case cls.Type.VIDEO:
+                return VideoType()
+            case cls.Type.AUDIO:
+                return AudioType()
+            case cls.Type.DOCUMENT:
+                return DocumentType()
+            case cls.Type.DATE:
+                return DateType()
+            case cls.Type.UUID:
+                return UUIDType()
+            case cls.Type.BINARY:
+                return BinaryType()
+            case _:
+                raise AssertionError(t)
 
     def __repr__(self) -> str:
         return self._to_str(as_schema=False)
@@ -185,7 +179,7 @@ class ColumnType:
         if as_schema:
             return base_str if self.nullable else f'Required[{base_str}]'
         else:
-            return f'Optional[{base_str}]' if self.nullable else base_str
+            return f'{base_str} | None' if self.nullable else base_str
 
     def _to_base_str(self) -> str:
         """
@@ -214,7 +208,13 @@ class ColumnType:
         # Default: just compare base types (this works for all types whose only parameter is nullable)
         return self._type == other._type
 
-    def supertype(self, other: ColumnType) -> Optional[ColumnType]:
+    def supertype(self, other: ColumnType, for_inference: bool = False) -> ColumnType | None:
+        """
+        Returns the most specific type that is a supertype of both `self` and `other`.
+
+        If `for_inference=True`, then we disallow certain type relationships that are technically correct, but may
+        be confusing for schema inference during data imports.
+        """
         if self == other:
             return self
         if self.matches(other):
@@ -229,12 +229,20 @@ class ColumnType:
             t = self.Type.supertype(self._type, other._type, self.common_supertypes)
             if t is not None:
                 return self.make_type(t).copy(nullable=(self.nullable or other.nullable))
-            return None
+
+        # If we see a mix of JSON and/or JSON-compatible scalar types, resolve to JSON.
+        # (For JSON+JSON, we return None to allow JsonType to handle merging the type schemas.)
+        if not for_inference and (
+            (self.is_json_type() and other.is_scalar_json_type())
+            or (self.is_scalar_json_type() and other.is_json_type())
+            or (self.is_scalar_json_type() and other.is_scalar_json_type())
+        ):
+            return JsonType(nullable=(self.nullable or other.nullable))
 
         return None
 
     @classmethod
-    def infer_literal_type(cls, val: Any, nullable: bool = False) -> Optional[ColumnType]:
+    def infer_literal_type(cls, val: Any, nullable: bool = False) -> ColumnType | None:
         if val is None:
             return InvalidType(nullable=True)
         if isinstance(val, str):
@@ -252,6 +260,10 @@ class ColumnType:
             return TimestampType(nullable=nullable)
         if isinstance(val, datetime.date):
             return DateType(nullable=nullable)
+        if isinstance(val, uuid.UUID):
+            return UUIDType(nullable=nullable)
+        if isinstance(val, bytes):
+            return BinaryType(nullable=nullable)
         if isinstance(val, PIL.Image.Image):
             return ImageType(width=val.width, height=val.height, mode=val.mode, nullable=nullable)
         if isinstance(val, np.ndarray):
@@ -268,7 +280,7 @@ class ColumnType:
         return None
 
     @classmethod
-    def infer_common_literal_type(cls, vals: Iterable[Any]) -> Optional[ColumnType]:
+    def infer_common_literal_type(cls, vals: Iterable[Any]) -> ColumnType | None:
         """
         Returns the most specific type that is a supertype of all literals in `vals`. If no such type
         exists, returns None.
@@ -276,13 +288,13 @@ class ColumnType:
         Args:
             vals: A collection of literals.
         """
-        inferred_type: Optional[ColumnType] = None
+        inferred_type: ColumnType | None = None
         for val in vals:
             val_type = cls.infer_literal_type(val)
             if inferred_type is None:
                 inferred_type = val_type
             else:
-                inferred_type = inferred_type.supertype(val_type)
+                inferred_type = inferred_type.supertype(val_type, for_inference=True)
             if inferred_type is None:
                 return None
             if not inferred_type.has_supertype():
@@ -291,8 +303,12 @@ class ColumnType:
 
     @classmethod
     def from_python_type(
-        cls, t: Union[type, _GenericAlias], nullable_default: bool = False, allow_builtin_types: bool = True
-    ) -> Optional[ColumnType]:
+        cls,
+        t: type | _GenericAlias,
+        nullable_default: bool = False,
+        allow_builtin_types: bool = True,
+        infer_pydantic_json: bool = False,
+    ) -> ColumnType | None:
         """
         Convert a Python type into a Pixeltable `ColumnType` instance.
 
@@ -304,16 +320,20 @@ class ColumnType:
                 allowed (as in UDF definitions). If False, then only Pixeltable types such as `pxt.String`,
                 `pxt.Int`, etc., will be allowed (as in schema definitions). `Optional` and `Required`
                 designations will be allowed regardless.
+            infer_pydantic_json: If True, accepts an extended set of built-ins (eg, Enum, Path) and returns the type to
+                which pydantic.BaseModel.model_dump(mode='json') serializes it.
         """
         origin = typing.get_origin(t)
         type_args = typing.get_args(t)
-        if origin is typing.Union:
-            # Check if `t` has the form Optional[T].
+        if origin in (typing.Union, types.UnionType):
+            # Check if `t` has the form T | None.
             if len(type_args) == 2 and type(None) in type_args:
-                # `t` is a type of the form Optional[T] (equivalently, Union[T, None] or Union[None, T]).
+                # `t` is a type of the form T | None (equivalently, T | None or None | T).
                 # We treat it as the underlying type but with nullable=True.
                 underlying_py_type = type_args[0] if type_args[1] is type(None) else type_args[1]
-                underlying = cls.from_python_type(underlying_py_type, allow_builtin_types=allow_builtin_types)
+                underlying = cls.from_python_type(
+                    underlying_py_type, allow_builtin_types=allow_builtin_types, infer_pydantic_json=infer_pydantic_json
+                )
                 if underlying is not None:
                     return underlying.copy(nullable=True)
         elif origin is Required:
@@ -327,7 +347,7 @@ class ColumnType:
             if isinstance(parameters, ColumnType):
                 return parameters.copy(nullable=nullable_default)
         else:
-            # It's something other than Optional[T], Required[T], or an explicitly annotated type.
+            # It's something other than T | None, Required[T], or an explicitly annotated type.
             if origin is not None:
                 # Discard type parameters to ensure that parameterized types such as `list[T]`
                 # are correctly mapped to Pixeltable types.
@@ -340,6 +360,13 @@ class ColumnType:
                     if literal_type is None:
                         return None
                     return literal_type.copy(nullable=(literal_type.nullable or nullable_default))
+                if infer_pydantic_json and isinstance(t, type) and issubclass(t, enum.Enum):
+                    literal_type = cls.infer_common_literal_type(member.value for member in t)
+                    if literal_type is None:
+                        return None
+                    return literal_type.copy(nullable=(literal_type.nullable or nullable_default))
+                if infer_pydantic_json and t is Path:
+                    return StringType(nullable=nullable_default)
                 if t is str:
                     return StringType(nullable=nullable_default)
                 if t is int:
@@ -352,6 +379,10 @@ class ColumnType:
                     return TimestampType(nullable=nullable_default)
                 if t is datetime.date:
                     return DateType(nullable=nullable_default)
+                if t is uuid.UUID:
+                    return UUIDType(nullable=nullable_default)
+                if t is bytes:
+                    return BinaryType(nullable=nullable_default)
                 if t is PIL.Image.Image:
                     return ImageType(nullable=nullable_default)
                 if isinstance(t, type) and issubclass(t, (Sequence, Mapping, pydantic.BaseModel)):
@@ -360,10 +391,7 @@ class ColumnType:
 
     @classmethod
     def normalize_type(
-        cls,
-        t: Union[ColumnType, type, _AnnotatedAlias],
-        nullable_default: bool = False,
-        allow_builtin_types: bool = True,
+        cls, t: ColumnType | type | _AnnotatedAlias, nullable_default: bool = False, allow_builtin_types: bool = True
     ) -> ColumnType:
         """
         Convert any type recognizable by Pixeltable to its corresponding ColumnType.
@@ -382,19 +410,51 @@ class ColumnType:
         (float, 'pxt.Float'),
         (datetime.datetime, 'pxt.Timestamp'),
         (datetime.date, 'pxt.Date'),
+        (uuid.UUID, 'pxt.UUID'),
         (PIL.Image.Image, 'pxt.Image'),
+        (bytes, 'pxt.Binary'),
         (Sequence, 'pxt.Json'),
         (Mapping, 'pxt.Json'),
     ]
 
     @classmethod
-    def __raise_exc_for_invalid_type(cls, t: Union[type, _AnnotatedAlias]) -> None:
+    def __raise_exc_for_invalid_type(cls, t: type | _AnnotatedAlias) -> None:
         for builtin_type, suggestion in cls.__TYPE_SUGGESTIONS:
             if t is builtin_type or (isinstance(t, type) and issubclass(t, builtin_type)):
                 name = t.__name__ if t.__module__ == 'builtins' else f'{t.__module__}.{t.__name__}'
                 raise excs.Error(f'Standard Python type `{name}` cannot be used here; use `{suggestion}` instead')
         raise excs.Error(f'Unknown type: {t}')
 
+    @classmethod
+    def from_json_schema(cls, schema: dict[str, Any]) -> ColumnType | None:
+        # We first express the JSON schema as a Python type, and then convert it to a Pixeltable type.
+        # TODO: Is there a meaningful fallback if one of these operations fails? (Maybe another use case for a pxt Any
+        #     type?)
+        py_type = cls.__json_schema_to_py_type(schema)
+        return cls.from_python_type(py_type) if py_type is not None else None
+
+    @classmethod
+    def __json_schema_to_py_type(cls, schema: dict[str, Any]) -> type | _GenericAlias | None:
+        if 'type' in schema:
+            if schema['type'] == 'null':
+                return type(None)
+            if schema['type'] == 'string':
+                return str
+            if schema['type'] == 'integer':
+                return int
+            if schema['type'] == 'number':
+                return float
+            if schema['type'] == 'boolean':
+                return bool
+            if schema['type'] in ('array', 'object'):
+                return list
+        elif 'anyOf' in schema:
+            subscripts = tuple(cls.__json_schema_to_py_type(subschema) for subschema in schema['anyOf'])
+            if all(subscript is not None for subscript in subscripts):
+                return Union[subscripts]
+
+        return None
+
     def validate_literal(self, val: Any) -> None:
         """Raise TypeError if val is not a valid literal for this type"""
         if val is None:
@@ -443,6 +503,9 @@ class ColumnType:
     def is_scalar_type(self) -> bool:
         return self._type in self.scalar_types
 
+    def is_scalar_json_type(self) -> bool:
+        return self._type in self.scalar_json_types
+
     def is_numeric_type(self) -> bool:
         return self._type in self.numeric_types
 
@@ -467,12 +530,18 @@ class ColumnType:
     def is_date_type(self) -> bool:
         return self._type == self.Type.DATE
 
+    def is_uuid_type(self) -> bool:
+        return self._type == self.Type.UUID
+
     def is_json_type(self) -> bool:
         return self._type == self.Type.JSON
 
     def is_array_type(self) -> bool:
         return self._type == self.Type.ARRAY
 
+    def is_binary_type(self) -> bool:
+        return self._type == self.Type.BINARY
+
     def is_image_type(self) -> bool:
         return self._type == self.Type.IMAGE
 
@@ -489,6 +558,10 @@ class ColumnType:
         # types that refer to external media files
         return self.is_image_type() or self.is_video_type() or self.is_audio_type() or self.is_document_type()
 
+    def supports_file_offloading(self) -> bool:
+        # types that can be offloaded to file-based storage via a CellMaterializationNode
+        return self.is_array_type() or self.is_json_type() or self.is_binary_type()
+
     @classmethod
     @abc.abstractmethod
     def to_sa_type(cls) -> sql.types.TypeEngine:
@@ -505,6 +578,35 @@ class ColumnType:
     def _to_json_schema(self) -> dict[str, Any]:
         raise excs.Error(f'Pixeltable type {self} is not a valid JSON type')
 
+    @classmethod
+    def from_np_dtype(cls, dtype: np.dtype, nullable: bool) -> ColumnType | None:
+        """
+        Return pixeltable type corresponding to a given simple numpy dtype
+        """
+        if np.issubdtype(dtype, np.integer):
+            return IntType(nullable=nullable)
+
+        if np.issubdtype(dtype, np.floating):
+            return FloatType(nullable=nullable)
+
+        if dtype == np.bool_:
+            return BoolType(nullable=nullable)
+
+        if np.issubdtype(dtype, np.str_):
+            return StringType(nullable=nullable)
+
+        if np.issubdtype(dtype, np.character):
+            return StringType(nullable=nullable)
+
+        if np.issubdtype(dtype, np.datetime64):
+            unit, _ = np.datetime_data(dtype)
+            if unit in ('D', 'M', 'Y'):
+                return DateType(nullable=nullable)
+            else:
+                return TimestampType(nullable=nullable)
+
+        return None
+
 
 class InvalidType(ColumnType):
     def __init__(self, nullable: bool = False):
@@ -629,8 +731,9 @@ class TimestampType(ColumnType):
     def _create_literal(self, val: Any) -> Any:
         if isinstance(val, str):
             return datetime.datetime.fromisoformat(val)
-        if isinstance(val, datetime.datetime):
-            return val
+        # Place naive timestamps in the default time zone
+        if isinstance(val, datetime.datetime) and val.tzinfo is None:
+            return val.replace(tzinfo=Env.get().default_time_zone)
         return val
 
 
@@ -657,11 +760,57 @@ class DateType(ColumnType):
         return val
 
 
+class UUIDType(ColumnType):
+    def __init__(self, nullable: bool = False):
+        super().__init__(self.Type.UUID, nullable=nullable)
+
+    def has_supertype(self) -> bool:
+        return not self.nullable
+
+    @classmethod
+    def to_sa_type(cls) -> sql.types.TypeEngine:
+        return sql.UUID(as_uuid=True)
+
+    def _to_json_schema(self) -> dict[str, Any]:
+        return {'type': 'string', 'format': 'uuid'}
+
+    def print_value(self, val: Any) -> str:
+        return f"'{val}'"
+
+    def _to_base_str(self) -> str:
+        return 'UUID'
+
+    def _validate_literal(self, val: Any) -> None:
+        if not isinstance(val, uuid.UUID):
+            raise TypeError(f'Expected uuid.UUID, got {val.__class__.__name__}')
+
+    def _create_literal(self, val: Any) -> Any:
+        if isinstance(val, str):
+            return uuid.UUID(val)
+        return val
+
+
+class BinaryType(ColumnType):
+    def __init__(self, nullable: bool = False):
+        super().__init__(self.Type.BINARY, nullable=nullable)
+
+    @classmethod
+    def to_sa_type(cls) -> sql.types.TypeEngine:
+        return sql.LargeBinary()
+
+    def _to_base_str(self) -> str:
+        return 'Binary'
+
+    def _validate_literal(self, val: Any) -> None:
+        if not isinstance(val, bytes):
+            raise TypeError(f'Expected `bytes`, got `{val.__class__.__name__}`')
+
+
 class JsonType(ColumnType):
-    json_schema: Optional[dict[str, Any]]
-    __validator: Optional[jsonschema.protocols.Validator]
+    json_schema: dict[str, Any] | None
+    __validator: jsonschema.protocols.Validator | None
 
-    def __init__(self, json_schema: Optional[dict[str, Any]] = None, nullable: bool = False):
+    def __init__(self, json_schema: dict[str, Any] | None = None, nullable: bool = False):
         super().__init__(self.Type.JSON, nullable=nullable)
         self.json_schema = json_schema
         if json_schema is None:
@@ -716,7 +865,7 @@ class JsonType(ColumnType):
 
     @classmethod
     def __is_valid_json(cls, val: Any) -> bool:
-        if val is None or isinstance(val, (str, int, float, bool)):
+        if val is None or isinstance(val, (str, int, float, bool, np.ndarray, PIL.Image.Image, bytes)):
             return True
         if isinstance(val, (list, tuple)):
             return all(cls.__is_valid_json(v) for v in val)
@@ -731,7 +880,7 @@ class JsonType(ColumnType):
             return val.model_dump()
         return val
 
-    def supertype(self, other: ColumnType) -> Optional[JsonType]:
+    def supertype(self, other: ColumnType, for_inference: bool = False) -> JsonType | None:
         # Try using the (much faster) supertype logic in ColumnType first. That will work if, for example, the types
         # are identical except for nullability. If that doesn't work and both types are JsonType, then we will need to
         # merge their schemas.
@@ -753,7 +902,7 @@ class JsonType(ColumnType):
         )
 
     @classmethod
-    def __superschema(cls, a: dict[str, Any], b: dict[str, Any]) -> Optional[dict[str, Any]]:
+    def __superschema(cls, a: dict[str, Any], b: dict[str, Any]) -> dict[str, Any] | None:
         # Defining a general type hierarchy over all JSON schemas would be a challenging problem. In order to keep
         # things manageable, we only define a hierarchy among "conforming" schemas, which provides enough generality
         # for the most important use cases (unions for type inference, validation of inline exprs). A schema is
@@ -813,7 +962,7 @@ class JsonType(ColumnType):
         return {}  # Unresolvable type conflict; the supertype is an unrestricted JsonType.
 
     @classmethod
-    def __superschema_with_nulls(cls, a: dict[str, Any], b: dict[str, Any]) -> Optional[dict[str, Any]]:
+    def __superschema_with_nulls(cls, a: dict[str, Any], b: dict[str, Any]) -> dict[str, Any] | None:
         a, a_nullable = cls.__unpack_null_from_schema(a)
         b, b_nullable = cls.__unpack_null_from_schema(b)
 
@@ -841,33 +990,64 @@ class JsonType(ColumnType):
             return f'Json[{self.json_schema}]'
 
 
+ARRAY_SUPPORTED_NUMPY_DTYPES = [
+    np.bool_,
+    np.uint8,
+    np.uint16,
+    np.uint32,
+    np.uint64,
+    np.int8,
+    np.int16,
+    np.int32,
+    np.int64,
+    np.float16,
+    np.float32,
+    np.float64,
+    np.str_,
+]
+
+
 class ArrayType(ColumnType):
-    shape: Optional[tuple[Optional[int], ...]]
-    pxt_dtype: Optional[ColumnType]
-    dtype: Optional[ColumnType.Type]
+    pxt_dtype_to_numpy_dtype: ClassVar[dict[ColumnType.Type, np.dtype]] = {
+        ColumnType.Type.INT: np.dtype(np.int64),
+        ColumnType.Type.FLOAT: np.dtype(np.float32),
+        ColumnType.Type.BOOL: np.dtype(np.bool_),
+        ColumnType.Type.STRING: np.dtype(np.str_),
+    }
+
+    shape: tuple[int | None, ...] | None
+    dtype: np.dtype | None
 
     def __init__(
         self,
-        shape: Optional[tuple[Optional[int], ...]] = None,
-        dtype: Optional[ColumnType] = None,
+        shape: tuple[int | None, ...] | None = None,
+        dtype: ColumnType | np.dtype | None = None,
         nullable: bool = False,
     ):
         super().__init__(self.Type.ARRAY, nullable=nullable)
         assert shape is None or dtype is not None, (shape, dtype)  # cannot specify a shape without a dtype
-        assert (
-            dtype is None
-            or dtype.is_int_type()
-            or dtype.is_float_type()
-            or dtype.is_bool_type()
-            or dtype.is_string_type()
-        )
-
         self.shape = shape
-        self.pxt_dtype = dtype  # we need this for copy() and __str__()
-        self.dtype = None if dtype is None else dtype._type
+        if dtype is None:
+            self.dtype = None
+        elif isinstance(dtype, np.dtype):
+            # Numpy string has some specifications (endianness, max length, encoding) that we don't support, so we just
+            # strip them out.
+            if dtype.type == np.str_:
+                self.dtype = np.dtype(np.str_)
+            else:
+                if dtype not in ARRAY_SUPPORTED_NUMPY_DTYPES:
+                    raise ValueError(f'Unsupported dtype: {dtype}')
+                self.dtype = dtype
+        elif isinstance(dtype, ColumnType):
+            self.dtype = self.pxt_dtype_to_numpy_dtype.get(dtype._type, None)
+            if self.dtype is None:
+                raise ValueError(f'Unsupported dtype: {dtype}')
+            assert self.dtype in ARRAY_SUPPORTED_NUMPY_DTYPES
+        else:
+            raise ValueError(f'Unsupported dtype: {dtype}')
 
     def copy(self, nullable: bool) -> ColumnType:
-        return ArrayType(self.shape, self.pxt_dtype, nullable=nullable)
+        return ArrayType(self.shape, self.dtype, nullable=nullable)
 
     def matches(self, other: ColumnType) -> bool:
         return isinstance(other, ArrayType) and self.shape == other.shape and self.dtype == other.dtype
@@ -875,7 +1055,7 @@ class ArrayType(ColumnType):
     def __hash__(self) -> int:
         return hash((self._type, self.nullable, self.shape, self.dtype))
 
-    def supertype(self, other: ColumnType) -> Optional[ArrayType]:
+    def supertype(self, other: ColumnType, for_inference: bool = False) -> ArrayType | None:
         basic_supertype = super().supertype(other)
         if basic_supertype is not None:
             assert isinstance(basic_supertype, ArrayType)
@@ -884,162 +1064,121 @@ class ArrayType(ColumnType):
         if not isinstance(other, ArrayType):
             return None
 
-        super_dtype = self.Type.supertype(self.dtype, other.dtype, self.common_supertypes)
-        if super_dtype is None:
-            # if the dtypes are incompatible, then the supertype is a fully general array
+        # Supertype has dtype only if dtypes are identical. We can change this behavior to consider casting rules or
+        # something else if there's demand for it.
+        if self.dtype != other.dtype:
             return ArrayType(nullable=(self.nullable or other.nullable))
-        super_shape: Optional[tuple[Optional[int], ...]]
+        super_dtype = self.dtype
+
+        # Determine the shape of the supertype
+        super_shape: tuple[int | None, ...] | None
         if self.shape is None or other.shape is None or len(self.shape) != len(other.shape):
             super_shape = None
         else:
             super_shape = tuple(n1 if n1 == n2 else None for n1, n2 in zip(self.shape, other.shape))
-        return ArrayType(super_shape, self.make_type(super_dtype), nullable=(self.nullable or other.nullable))
+        return ArrayType(super_shape, super_dtype, nullable=(self.nullable or other.nullable))
 
     def _as_dict(self) -> dict:
         result = super()._as_dict()
         shape_as_list = None if self.shape is None else list(self.shape)
-        dtype_value = None if self.dtype is None else self.dtype.value
-        result.update(shape=shape_as_list, dtype=dtype_value)
+        result.update(shape=shape_as_list)
+
+        if self.dtype is None:
+            result.update(numpy_dtype=None)
+        elif self.dtype == np.str_:
+            # str(np.str_) would be something like '<U', but since we don't support the string specifications, just use
+            # 'str' instead to avoid confusion.
+            result.update(numpy_dtype='str')
+        else:
+            result.update(numpy_dtype=str(self.dtype))
         return result
 
     def _to_base_str(self) -> str:
         if self.shape is None and self.dtype is None:
             return 'Array'
         if self.shape is None:
-            return f'Array[{self.pxt_dtype}]'
+            return f'Array[{self.dtype.name}]'
         assert self.dtype is not None
-        return f'Array[{self.shape}, {self.pxt_dtype}]'
+        return f'Array[{self.shape}, {self.dtype.name}]'
 
     @classmethod
     def _from_dict(cls, d: dict) -> ColumnType:
+        assert 'numpy_dtype' in d
+        dtype = None if d['numpy_dtype'] is None else np.dtype(d['numpy_dtype'])
         assert 'shape' in d
-        assert 'dtype' in d
         shape = None if d['shape'] is None else tuple(d['shape'])
-        dtype = None if d['dtype'] is None else cls.make_type(cls.Type(d['dtype']))
         return cls(shape, dtype, nullable=d['nullable'])
 
     @classmethod
-    def from_np_dtype(cls, dtype: np.dtype, nullable: bool) -> Optional[ColumnType]:
-        """
-        Return pixeltable type corresponding to a given simple numpy dtype
-        """
-        if np.issubdtype(dtype, np.integer):
-            return IntType(nullable=nullable)
-
-        if np.issubdtype(dtype, np.floating):
-            return FloatType(nullable=nullable)
-
-        if dtype == np.bool_:
-            return BoolType(nullable=nullable)
-
-        if np.issubdtype(dtype, np.str_):
-            return StringType(nullable=nullable)
-
-        if np.issubdtype(dtype, np.character):
-            return StringType(nullable=nullable)
-
-        if np.issubdtype(dtype, np.datetime64):
-            unit, _ = np.datetime_data(dtype)
-            if unit in ['D', 'M', 'Y']:
-                return DateType(nullable=nullable)
-            else:
-                return TimestampType(nullable=nullable)
-
-        return None
-
-    @classmethod
-    def from_literal(cls, val: np.ndarray, nullable: bool = False) -> Optional[ArrayType]:
-        # determine our dtype
+    def from_literal(cls, val: np.ndarray, nullable: bool = False) -> ArrayType | None:
         assert isinstance(val, np.ndarray)
-        pxttype: Optional[ColumnType] = cls.from_np_dtype(val.dtype, nullable)
-        if pxttype is None:
+        if val.dtype.type not in ARRAY_SUPPORTED_NUMPY_DTYPES:
             return None
-        return cls(val.shape, dtype=pxttype, nullable=nullable)
+        return cls(val.shape, dtype=val.dtype, nullable=nullable)
 
-    def is_valid_literal(self, val: np.ndarray) -> bool:
+    def _to_json_schema(self) -> dict[str, Any]:
+        schema: dict[str, Any] = {'type': 'array'}
+        if self.dtype == np.str_:
+            schema.update({'items': {'type': 'str'}})
+        elif self.dtype is not None:
+            schema.update({'items': {'type': str(self.dtype)}})
+        return schema
+
+    def _validate_literal(self, val: Any) -> None:
         if not isinstance(val, np.ndarray):
-            return False
+            raise TypeError(f'Expected numpy.ndarray, got {val.__class__.__name__}')
 
-        # If a dtype is specified, check that there's a match
-        if self.dtype is not None and not np.issubdtype(val.dtype, self.numpy_dtype()):
-            return False
+        # If column type has a dtype, check if it matches
+        if self.dtype == np.str_:
+            if val.dtype.type != np.str_:
+                raise TypeError(f'Expected numpy.ndarray of dtype {self.dtype}, got numpy.ndarray of dtype {val.dtype}')
+        elif self.dtype is not None and self.dtype != val.dtype:
+            raise TypeError(f'Expected numpy.ndarray of dtype {self.dtype}, got numpy.ndarray of dtype {val.dtype}')
 
-        # If no dtype is specified, we still need to check that the dtype is one of the supported types
-        if self.dtype is None and not any(
-            np.issubdtype(val.dtype, ndtype) for ndtype in [np.int64, np.float32, np.bool_, np.str_]
-        ):
-            return False
+        # Check that the dtype is one of the supported types
+        if val.dtype.type != np.str_ and val.dtype not in ARRAY_SUPPORTED_NUMPY_DTYPES:
+            raise TypeError(f'Unsupported dtype {val.dtype}')
 
         # If a shape is specified, check that there's a match
         if self.shape is not None:
             if len(val.shape) != len(self.shape):
-                return False
+                raise TypeError(
+                    f'Expected numpy.ndarray({self.shape}, dtype={self.dtype}), '
+                    f'got numpy.ndarray({val.shape}, dtype={val.dtype})'
+                )
             # check that the shapes are compatible
             for n1, n2 in zip(val.shape, self.shape):
                 assert n1 is not None  # `val` must have a concrete shape
                 if n2 is None:
                     continue  # wildcard
                 if n1 != n2:
-                    return False
-
-        return True
-
-    def _to_json_schema(self) -> dict[str, Any]:
-        return {'type': 'array', 'items': self.pxt_dtype._to_json_schema()}
-
-    def _validate_literal(self, val: Any) -> None:
-        if not isinstance(val, np.ndarray):
-            raise TypeError(f'Expected numpy.ndarray, got {val.__class__.__name__}')
-        if not self.is_valid_literal(val):
-            if self.shape is not None:
-                raise TypeError(
-                    f'Expected numpy.ndarray({self.shape}, dtype={self.numpy_dtype()}), '
-                    f'got numpy.ndarray({val.shape}, dtype={val.dtype})'
-                )
-            elif self.dtype is not None:
-                raise TypeError(
-                    f'Expected numpy.ndarray of dtype {self.numpy_dtype()}, got numpy.ndarray of dtype {val.dtype}'
-                )
-            else:
-                raise TypeError(f'Unsupported dtype for numpy.ndarray: {val.dtype}')
+                    raise TypeError(
+                        f'Expected numpy.ndarray({self.shape}, dtype={self.dtype}), '
+                        f'got numpy.ndarray({val.shape}, dtype={val.dtype})'
+                    )
 
     def _create_literal(self, val: Any) -> Any:
         if isinstance(val, (list, tuple)):
             # map python float to whichever numpy float is
             # declared for this type, rather than assume float64
-            return np.array(val, dtype=self.numpy_dtype())
+            return np.array(val, dtype=self.dtype)
         return val
 
     @classmethod
     def to_sa_type(cls) -> sql.types.TypeEngine:
         return sql.LargeBinary()
 
-    def numpy_dtype(self) -> Optional[np.dtype]:
-        if self.dtype is None:
-            return None
-        if self.dtype == self.Type.INT:
-            return np.dtype(np.int64)
-        if self.dtype == self.Type.FLOAT:
-            return np.dtype(np.float32)
-        if self.dtype == self.Type.BOOL:
-            return np.dtype(np.bool_)
-        if self.dtype == self.Type.STRING:
-            return np.dtype(np.str_)
-        raise AssertionError(self.dtype)
-
 
 class ImageType(ColumnType):
     def __init__(
         self,
-        width: Optional[int] = None,
-        height: Optional[int] = None,
-        size: Optional[tuple[int, int]] = None,
-        mode: Optional[str] = None,
+        width: int | None = None,
+        height: int | None = None,
+        size: tuple[int, int] | None = None,
+        mode: str | None = None,
         nullable: bool = False,
     ):
-        """
-        TODO: does it make sense to specify only width or height?
-        """
+        # TODO: does it make sense to specify only width or height?
         super().__init__(self.Type.IMAGE, nullable=nullable)
         assert not (width is not None and size is not None)
         assert not (height is not None and size is not None)
@@ -1077,7 +1216,7 @@ class ImageType(ColumnType):
     def __hash__(self) -> int:
         return hash((self._type, self.nullable, self.size, self.mode))
 
-    def supertype(self, other: ColumnType) -> Optional[ImageType]:
+    def supertype(self, other: ColumnType, for_inference: bool = False) -> ImageType | None:
         basic_supertype = super().supertype(other)
         if basic_supertype is not None:
             assert isinstance(basic_supertype, ImageType)
@@ -1092,7 +1231,7 @@ class ImageType(ColumnType):
         return ImageType(width=width, height=height, mode=mode, nullable=(self.nullable or other.nullable))
 
     @property
-    def size(self) -> Optional[tuple[int, int]]:
+    def size(self) -> tuple[int, int] | None:
         if self.width is None or self.height is None:
             return None
         return (self.width, self.height)
@@ -1123,8 +1262,8 @@ class ImageType(ColumnType):
                 img.load()
                 return img
             except Exception as exc:
-                errormsg_val = val if len(val) < 50 else val[:50] + '...'
-                raise excs.Error(f'data URL could not be decoded into a valid image: {errormsg_val}') from exc
+                error_msg_val = val if len(val) < 50 else val[:50] + '...'
+                raise excs.Error(f'data URL could not be decoded into a valid image: {error_msg_val}') from exc
         return val
 
     def _validate_literal(self, val: Any) -> None:
@@ -1211,7 +1350,7 @@ class DocumentType(ColumnType):
         TXT = 4
 
         @classmethod
-        def from_extension(cls, ext: str) -> Optional['DocumentType.DocumentFormat']:
+        def from_extension(cls, ext: str) -> 'DocumentType.DocumentFormat' | None:
             if ext in ('.htm', '.html'):
                 return cls.HTML
             if ext == '.md':
@@ -1224,7 +1363,7 @@ class DocumentType(ColumnType):
                 return cls.TXT
             return None
 
-    def __init__(self, nullable: bool = False, doc_formats: Optional[str] = None):
+    def __init__(self, nullable: bool = False, doc_formats: str | None = None):
         super().__init__(self.Type.DOCUMENT, nullable=nullable)
         self.doc_formats = doc_formats
         if doc_formats is not None:
@@ -1278,6 +1417,8 @@ Float = typing.Annotated[float, FloatType(nullable=False)]
 Bool = typing.Annotated[bool, BoolType(nullable=False)]
 Timestamp = typing.Annotated[datetime.datetime, TimestampType(nullable=False)]
 Date = typing.Annotated[datetime.date, DateType(nullable=False)]
+UUID = typing.Annotated[uuid.UUID, UUIDType(nullable=False)]
+Binary = typing.Annotated[bytes, BinaryType(nullable=False)]
 
 
 class _PxtType:
@@ -1320,14 +1461,17 @@ class Json(_PxtType):
 class Array(np.ndarray, _PxtType):
     def __class_getitem__(cls, item: Any) -> _AnnotatedAlias:
         """
-        `item` (the type subscript) must be a tuple with exactly two elements (in any order):
-        - A tuple of `Optional[int]`s, specifying the shape of the array
-        - A type, specifying the dtype of the array
-        Example: Array[(3, None, 2), pxt.Float]
+        `item` (the type subscript) must be a tuple with at most two elements (in any order):
+        - An optional tuple of `int | None`s, specifying the shape of the array
+        - A type (`ColumnType | np.dtype`), specifying the dtype of the array
+        Examples:
+        * Array[(3, None, 2), pxt.Float]
+        * Array[(4, 4), np.uint8]
+        * Array[np.bool]
         """
         params = item if isinstance(item, tuple) else (item,)
-        shape: Optional[tuple] = None
-        dtype: Optional[ColumnType] = None
+        shape: tuple | None = None
+        dtype: ColumnType | np.dtype | None = None
         if not any(isinstance(param, (type, _AnnotatedAlias)) for param in params):
             raise TypeError('Array type parameter must include a dtype.')
         for param in params:
@@ -1340,7 +1484,10 @@ class Array(np.ndarray, _PxtType):
             elif isinstance(param, (type, _AnnotatedAlias)):
                 if dtype is not None:
                     raise TypeError(f'Duplicate Array type parameter: {param}')
-                dtype = ColumnType.normalize_type(param, allow_builtin_types=False)
+                if isinstance(param, type) and param in ARRAY_SUPPORTED_NUMPY_DTYPES:
+                    dtype = np.dtype(param)
+                else:
+                    dtype = ColumnType.normalize_type(param, allow_builtin_types=False)
             else:
                 raise TypeError(f'Invalid Array type parameter: {param}')
         return typing.Annotated[np.ndarray, ArrayType(shape=shape, dtype=dtype, nullable=False)]
@@ -1367,8 +1514,8 @@ class Image(PIL.Image.Image, _PxtType):
         else:
             # Not a tuple (single arg)
             params = (item,)
-        size: Optional[tuple] = None
-        mode: Optional[str] = None
+        size: tuple | None = None
+        mode: str | None = None
         for param in params:
             if isinstance(param, tuple):
                 if (
@@ -1413,4 +1560,19 @@ class Document(str, _PxtType):
         return DocumentType(nullable=nullable)
 
 
-ALL_PIXELTABLE_TYPES = (String, Bool, Int, Float, Timestamp, Json, Array, Image, Video, Audio, Document)
+ALL_PIXELTABLE_TYPES = (
+    String,
+    Bool,
+    Int,
+    Float,
+    Timestamp,
+    Json,
+    Array,
+    Image,
+    Video,
+    Audio,
+    Document,
+    Date,
+    UUID,
+    Binary,
+)