pixeltable 0.4.0rc3__py3-none-any.whl → 0.4.20__py3-none-any.whl

pixeltable/catalog/column.py

@@ -3,18 +3,21 @@ from __future__ import annotations
 import logging
 import warnings
 from textwrap import dedent
-from typing import TYPE_CHECKING, Any, Optional
+from typing import TYPE_CHECKING, Any
 
 import sqlalchemy as sql
 
 import pixeltable.exceptions as excs
+import pixeltable.exprs as exprs
 import pixeltable.type_system as ts
-from pixeltable import exprs
+from pixeltable.env import Env
+from pixeltable.metadata import schema
 
-from .globals import MediaValidation, is_valid_identifier
+from .globals import MediaValidation, QColumnId, is_valid_identifier
 
 if TYPE_CHECKING:
     from .table_version import TableVersion
+    from .table_version_handle import ColumnHandle, TableVersionHandle
     from .table_version_path import TableVersionPath
 
 _logger = logging.getLogger('pixeltable')
@@ -25,122 +28,161 @@ class Column:
 
     A Column contains all the metadata necessary for executing queries and updates against a particular version of a
     table/view.
+
+    Args:
+        name: column name; None for system columns (eg, index columns)
+        col_type: column type; can be None if the type can be derived from ``computed_with``
+        computed_with: an Expr that computes the column value
+        is_pk: if True, this column is part of the primary key
+        stored: determines whether a computed column is present in the stored table or recomputed on demand
+        destination: An object store reference for persisting computed files
+        col_id: column ID (only used internally)
+
+    Computed columns: those have a non-None ``computed_with`` argument
+    - when constructed by the user: ``computed_with`` was constructed explicitly and is passed in;
+        col_type is None
+    - when loaded from md store: ``computed_with`` is set and col_type is set
+
+    ``stored`` (only valid for computed columns):
+    - if True: the column is present in the stored table
+    - if False: the column is not present in the stored table and recomputed during a query
+    - if None: the system chooses for you (at present, this is always False, but this may change in the future)
     """
 
-    name: str
-    id: Optional[int]
+    name: str | None
+    id: int | None
     col_type: ts.ColumnType
     stored: bool
     is_pk: bool
-    _media_validation: Optional[MediaValidation]  # if not set, TableVersion.media_validation applies
-    schema_version_add: Optional[int]
-    schema_version_drop: Optional[int]
-    _records_errors: Optional[bool]
-    sa_col: Optional[sql.schema.Column]
-    sa_col_type: Optional[sql.sqltypes.TypeEngine]
-    sa_errormsg_col: Optional[sql.schema.Column]
-    sa_errortype_col: Optional[sql.schema.Column]
-    _value_expr: Optional[exprs.Expr]
-    value_expr_dict: Optional[dict[str, Any]]
-    dependent_cols: set[Column]
-    # we store a TableVersion here, not a TableVersionHandle, because this column is owned by that TableVersion instance
-    # (re-resolving it later to a different instance doesn't make sense)
-    tbl: Optional[TableVersion]
-    # tbl: Optional[TableVersionHandle]
+    is_iterator_col: bool
+    _explicit_destination: str | None  # An object store reference for computed files
+    _media_validation: MediaValidation | None  # if not set, TableVersion.media_validation applies
+    schema_version_add: int | None
+    schema_version_drop: int | None
+    stores_cellmd: bool
+    sa_col: sql.schema.Column | None
+    sa_col_type: sql.types.TypeEngine
+    sa_cellmd_col: sql.schema.Column | None  # JSON metadata for the cell, e.g. errortype, errormsg for media columns
+    _value_expr: exprs.Expr | None
+    value_expr_dict: dict[str, Any] | None
+    # we store a handle here in order to allow Column construction before there is a corresponding TableVersion
+    tbl_handle: 'TableVersionHandle' | None
 
     def __init__(
         self,
-        name: Optional[str],
-        col_type: Optional[ts.ColumnType] = None,
-        computed_with: Optional[exprs.Expr] = None,
+        name: str | None,
+        col_type: ts.ColumnType | None = None,
+        computed_with: exprs.Expr | None = None,
         is_pk: bool = False,
+        is_iterator_col: bool = False,
         stored: bool = True,
-        media_validation: Optional[MediaValidation] = None,
-        col_id: Optional[int] = None,
-        schema_version_add: Optional[int] = None,
-        schema_version_drop: Optional[int] = None,
-        sa_col_type: Optional[sql.sqltypes.TypeEngine] = None,
-        records_errors: Optional[bool] = None,
-        value_expr_dict: Optional[dict[str, Any]] = None,
+        media_validation: MediaValidation | None = None,
+        col_id: int | None = None,
+        schema_version_add: int | None = None,
+        schema_version_drop: int | None = None,
+        sa_col_type: sql.types.TypeEngine | None = None,
+        stores_cellmd: bool | None = None,
+        value_expr_dict: dict[str, Any] | None = None,
+        tbl_handle: 'TableVersionHandle' | None = None,
+        destination: str | None = None,
     ):
-        """Column constructor.
-
-        Args:
-            name: column name; None for system columns (eg, index columns)
-            col_type: column type; can be None if the type can be derived from ``computed_with``
-            computed_with: an Expr that computes the column value
-            is_pk: if True, this column is part of the primary key
-            stored: determines whether a computed column is present in the stored table or recomputed on demand
-            col_id: column ID (only used internally)
-
-        Computed columns: those have a non-None ``computed_with`` argument
-        - when constructed by the user: ``computed_with`` was constructed explicitly and is passed in;
-          col_type is None
-        - when loaded from md store: ``computed_with`` is set and col_type is set
-
-        ``stored`` (only valid for computed columns):
-        - if True: the column is present in the stored table
-        - if False: the column is not present in the stored table and recomputed during a query
-        - if None: the system chooses for you (at present, this is always False, but this may change in the future)
-        """
         if name is not None and not is_valid_identifier(name):
-            raise excs.Error(f"Invalid column name: '{name}'")
+            raise excs.Error(f'Invalid column name: {name}')
         self.name = name
+        self.tbl_handle = tbl_handle
         if col_type is None and computed_with is None:
-            raise excs.Error(f'Column `{name}`: col_type is required if computed_with is not specified')
+            raise excs.Error(f'Column {name!r}: `col_type` is required if `computed_with` is not specified')
 
-        self._value_expr: Optional[exprs.Expr] = None
+        self._value_expr = None
         self.value_expr_dict = value_expr_dict
         if computed_with is not None:
             value_expr = exprs.Expr.from_object(computed_with)
             if value_expr is None:
+                # TODO: this shouldn't be a user-facing error
                 raise excs.Error(
-                    f'Column {name}: computed_with needs to be a valid Pixeltable expression, '
+                    f'Column {name!r}: `computed_with` needs to be a valid Pixeltable expression, '
                     f'but it is a {type(computed_with)}'
                 )
             else:
                 self._value_expr = value_expr.copy()
                 self.col_type = self._value_expr.col_type
+        if self._value_expr is not None and self.value_expr_dict is None:
+            self.value_expr_dict = self._value_expr.as_dict()
 
         if col_type is not None:
             self.col_type = col_type
         assert self.col_type is not None
 
         self.stored = stored
-        self.dependent_cols = set()  # cols with value_exprs that reference us; set by TableVersion
+        # self.dependent_cols = set()  # cols with value_exprs that reference us; set by TableVersion
         self.id = col_id
         self.is_pk = is_pk
+        self.is_iterator_col = is_iterator_col
         self._media_validation = media_validation
         self.schema_version_add = schema_version_add
         self.schema_version_drop = schema_version_drop
 
-        self._records_errors = records_errors
+        if stores_cellmd is not None:
+            self.stores_cellmd = stores_cellmd
+        else:
+            self.stores_cellmd = stored and (
+                self.is_computed
+                or self.col_type.is_media_type()
+                or self.col_type.is_json_type()
+                or self.col_type.is_array_type()
+            )
 
         # column in the stored table for the values of this Column
         self.sa_col = None
-        self.sa_col_type = sa_col_type
+        self.sa_col_type = self.col_type.to_sa_type() if sa_col_type is None else sa_col_type
 
         # computed cols also have storage columns for the exception string and type
-        self.sa_errormsg_col = None
-        self.sa_errortype_col = None
+        self.sa_cellmd_col = None
+        self._explicit_destination = destination
+
+    def to_md(self, pos: int | None = None) -> tuple[schema.ColumnMd, schema.SchemaColumn | None]:
+        """Returns the Column and optional SchemaColumn metadata for this Column."""
+        assert self.is_pk is not None
+        col_md = schema.ColumnMd(
+            id=self.id,
+            col_type=self.col_type.as_dict(),
+            is_pk=self.is_pk,
+            schema_version_add=self.schema_version_add,
+            schema_version_drop=self.schema_version_drop,
+            value_expr=self.value_expr.as_dict() if self.value_expr is not None else None,
+            stored=self.stored,
+            destination=self._explicit_destination,
+        )
+        if pos is None:
+            return col_md, None
+        assert self.name is not None, 'Column name must be set for user-facing columns'
+        sch_md = schema.SchemaColumn(
+            name=self.name,
+            pos=pos,
+            media_validation=self._media_validation.name.lower() if self._media_validation is not None else None,
+        )
+        return col_md, sch_md
 
-        self.tbl = None  # set by owning TableVersion
+    def init_value_expr(self, tvp: 'TableVersionPath' | None) -> None:
+        """
+        Initialize the value_expr from its dict representation, if necessary.
 
-    @property
-    def value_expr(self) -> Optional[exprs.Expr]:
-        """Instantiate value_expr on-demand"""
-        # TODO: instantiate expr in the c'tor and add an Expr.prepare() that can create additional state after the
-        # catalog has been fully loaded; that way, we encounter bugs in the serialization/deserialization logic earlier
-        if self.value_expr_dict is not None and self._value_expr is None:
-            from pixeltable import exprs
+        If `tvp` is not None, retarget the value_expr to the given TableVersionPath.
+        """
+        from pixeltable import exprs
 
+        if self._value_expr is None and self.value_expr_dict is None:
+            return
+
+        if self._value_expr is None:
+            # Instantiate the Expr from its dict
             self._value_expr = exprs.Expr.from_dict(self.value_expr_dict)
             self._value_expr.bind_rel_paths()
             if not self._value_expr.is_valid:
                 message = (
                     dedent(
                         f"""
-                        The computed column {self.name!r} in table {self.tbl.name!r} is no longer valid.
+                        The computed column {self.name!r} in table {self.get_tbl().name!r} is no longer valid.
                         {{validation_error}}
                         You can continue to query existing data from this column, but evaluating it on new data will raise an error.
                         """  # noqa: E501
@@ -149,37 +191,76 @@ class Column:
                     .format(validation_error=self._value_expr.validation_error)
                 )
                 warnings.warn(message, category=excs.PixeltableWarning, stacklevel=2)
+
+        if tvp is not None:
+            # Retarget the Expr
+            self._value_expr = self._value_expr.retarget(tvp)
+
+    def get_tbl(self) -> TableVersion:
+        tv = self.tbl_handle.get()
+        return tv
+
+    @property
+    def destination(self) -> str | None:
+        if self._explicit_destination is not None:
+            # An expilicit destination was set as part of the column definition
+            return self._explicit_destination
+
+        # Otherwise, if this is a stored media column, use the default destination if one is configured (input
+        #     destination or output destination, depending on whether this is a computed column)
+        # TODO: The `self.name is not None` clause is necessary because index columns currently follow the type of
+        #     the underlying media column. We should move to using pxt.String as the col_type of index columns; this
+        #     would be a more robust solution, and then `self.name is not None` could be removed.
+        if self.is_stored and self.col_type.is_media_type() and self.name is not None:
+            if self.is_computed:
+                return Env.get().default_output_media_dest
+            else:
+                return Env.get().default_input_media_dest
+
+        return None
+
+    @property
+    def handle(self) -> 'ColumnHandle':
+        """Returns a ColumnHandle for this Column."""
+        from .table_version_handle import ColumnHandle
+
+        assert self.tbl_handle is not None
+        assert self.id is not None
+        return ColumnHandle(self.tbl_handle, self.id)
+
+    @property
+    def qid(self) -> QColumnId:
+        assert self.tbl_handle is not None
+        assert self.id is not None
+        return QColumnId(self.tbl_handle.id, self.id)
+
+    @property
+    def value_expr(self) -> exprs.Expr | None:
+        assert self.value_expr_dict is None or self._value_expr is not None
         return self._value_expr
 
     def set_value_expr(self, value_expr: exprs.Expr) -> None:
         self._value_expr = value_expr
-        self.value_expr_dict = None
+        self.value_expr_dict = self._value_expr.as_dict()
 
     def check_value_expr(self) -> None:
         assert self._value_expr is not None
-        if self.stored == False and self.is_computed and self.has_window_fn_call():
+        if not self.stored and self.is_computed and self.has_window_fn_call():
             raise excs.Error(
-                f'Column {self.name}: stored={self.stored} not supported for columns computed with window functions:'
-                f'\n{self.value_expr}'
+                f'Column {self.name!r}: `stored={self.stored}` not supported for columns '
+                f'computed with window functions:\n{self.value_expr}'
             )
 
     def has_window_fn_call(self) -> bool:
-        if self.value_expr is None:
-            return False
         from pixeltable import exprs
 
+        if self.value_expr is None:
+            return False
         window_fn_calls = list(
             self.value_expr.subexprs(filter=lambda e: isinstance(e, exprs.FunctionCall) and e.is_window_fn_call)
         )
         return len(window_fn_calls) > 0
 
-    # TODO: This should be moved out of `Column` (its presence in `Column` doesn't anticipate indices being defined on
-    #     multiple dependents)
-    def get_idx_info(self, reference_tbl: Optional['TableVersionPath'] = None) -> dict[str, 'TableVersion.IndexInfo']:
-        assert self.tbl is not None
-        tbl = reference_tbl.tbl_version.get() if reference_tbl is not None else self.tbl
-        return {name: info for name, info in tbl.idxs_by_name.items() if info.col == self}
-
     @property
     def is_computed(self) -> bool:
         return self._value_expr is not None or self.value_expr_dict is not None
@@ -190,25 +271,17 @@ class Column:
         assert self.stored is not None
         return self.stored
 
-    @property
-    def records_errors(self) -> bool:
-        """True if this column also stores error information."""
-        # default: record errors for computed and media columns
-        if self._records_errors is not None:
-            return self._records_errors
-        return self.is_stored and (self.is_computed or self.col_type.is_media_type())
-
     @property
     def qualified_name(self) -> str:
-        assert self.tbl is not None
-        return f'{self.tbl.name}.{self.name}'
+        assert self.get_tbl() is not None
+        return f'{self.get_tbl().name}.{self.name}'
 
     @property
     def media_validation(self) -> MediaValidation:
         if self._media_validation is not None:
             return self._media_validation
-        assert self.tbl is not None
-        return self.tbl.media_validation
+        assert self.get_tbl() is not None
+        return self.get_tbl().media_validation
 
     @property
     def is_required_for_insert(self) -> bool:
@@ -227,48 +300,46 @@ class Column:
 
     def create_sa_cols(self) -> None:
         """
-        These need to be recreated for every new table schema version.
+        These need to be recreated for every sql.Table instance
         """
         assert self.is_stored
+        assert self.stores_cellmd is not None
         # all storage columns are nullable (we deal with null errors in Pixeltable directly)
-        self.sa_col = sql.Column(
-            self.store_name(),
-            self.col_type.to_sa_type() if self.sa_col_type is None else self.sa_col_type,
-            nullable=True,
-        )
-        if self.is_computed or self.col_type.is_media_type():
-            self.sa_errormsg_col = sql.Column(self.errormsg_store_name(), ts.StringType().to_sa_type(), nullable=True)
-            self.sa_errortype_col = sql.Column(self.errortype_store_name(), ts.StringType().to_sa_type(), nullable=True)
+        self.sa_col = sql.Column(self.store_name(), self.sa_col_type, nullable=True)
+        if self.stores_cellmd:
+            self.sa_cellmd_col = sql.Column(self.cellmd_store_name(), self.sa_cellmd_type(), nullable=True)
 
-    def get_sa_col_type(self) -> sql.sqltypes.TypeEngine:
-        return self.col_type.to_sa_type() if self.sa_col_type is None else self.sa_col_type
+    @classmethod
+    def cellmd_type(cls) -> ts.ColumnType:
+        return ts.JsonType(nullable=True)
+
+    @classmethod
+    def sa_cellmd_type(cls) -> sql.types.TypeEngine:
+        return cls.cellmd_type().to_sa_type()
 
     def store_name(self) -> str:
         assert self.id is not None
         assert self.is_stored
         return f'col_{self.id}'
 
-    def errormsg_store_name(self) -> str:
-        return f'{self.store_name()}_errormsg'
-
-    def errortype_store_name(self) -> str:
-        return f'{self.store_name()}_errortype'
+    def cellmd_store_name(self) -> str:
+        return f'{self.store_name()}_cellmd'
 
     def __str__(self) -> str:
         return f'{self.name}: {self.col_type}'
 
     def __repr__(self) -> str:
-        return f'Column({self.id!r}, {self.name!r}, tbl={self.tbl.name!r})'
+        return f'Column({self.id!r}, {self.name!r}, tbl={self.get_tbl().name!r})'
 
     def __hash__(self) -> int:
         # TODO(aaron-siegel): This and __eq__ do not capture the table version. We need to rethink the Column
         #     abstraction (perhaps separating out the version-dependent properties into a different abstraction).
-        assert self.tbl is not None
-        return hash((self.tbl.id, self.id))
+        assert self.tbl_handle is not None
+        return hash((self.tbl_handle.id, self.id))
 
     def __eq__(self, other: object) -> bool:
         if not isinstance(other, Column):
             return False
-        assert self.tbl is not None
-        assert other.tbl is not None
-        return self.tbl.id == other.tbl.id and self.id == other.id
+        assert self.tbl_handle is not None
+        assert other.tbl_handle is not None
+        return self.tbl_handle.id == other.tbl_handle.id and self.id == other.id

pixeltable/catalog/dir.py

@@ -34,8 +34,7 @@ class Dir(SchemaObject):
         dir = cls(dir_record.id, parent_id, name)
         return dir
 
-    @classmethod
-    def _display_name(cls) -> str:
+    def _display_name(self) -> str:
         return 'directory'
 
     def _path(self) -> str:

pixeltable/catalog/globals.py

@@ -1,12 +1,10 @@
 from __future__ import annotations
 
-import dataclasses
 import enum
 import itertools
 import logging
-from typing import Optional
-
-from typing_extensions import Self
+from dataclasses import dataclass
+from uuid import UUID
 
 import pixeltable.exceptions as excs
 
@@ -18,45 +16,15 @@ _ROWID_COLUMN_NAME = '_rowid'
 
 # Set of symbols that are predefined in the `InsertableTable` class (and are therefore not allowed as column names).
 # This will be populated lazily to avoid circular imports.
-_PREDEF_SYMBOLS: Optional[set[str]] = None
-
-
-@dataclasses.dataclass
-class UpdateStatus:
-    """
-    Information about updates that resulted from a table operation.
-    """
-
-    num_rows: int = 0
-    # TODO: disambiguate what this means: # of slots computed or # of columns computed?
-    num_computed_values: int = 0
-    num_excs: int = 0
-    updated_cols: list[str] = dataclasses.field(default_factory=list)
-    cols_with_excs: list[str] = dataclasses.field(default_factory=list)
-
-    def __iadd__(self, other: 'UpdateStatus') -> Self:
-        self.num_rows += other.num_rows
-        self.num_computed_values += other.num_computed_values
-        self.num_excs += other.num_excs
-        self.updated_cols = list(dict.fromkeys(self.updated_cols + other.updated_cols))
-        self.cols_with_excs = list(dict.fromkeys(self.cols_with_excs + other.cols_with_excs))
-        return self
-
-    @property
-    def insert_msg(self) -> str:
-        """Return a message describing the results of an insert operation."""
-        if self.num_excs == 0:
-            cols_with_excs_str = ''
-        else:
-            cols_with_excs_str = (
-                f' across {len(self.cols_with_excs)} column{"" if len(self.cols_with_excs) == 1 else "s"}'
-            )
-            cols_with_excs_str += f' ({", ".join(self.cols_with_excs)})'
-        msg = (
-            f'Inserted {self.num_rows} row{"" if self.num_rows == 1 else "s"} '
-            f'with {self.num_excs} error{"" if self.num_excs == 1 else "s"}{cols_with_excs_str}.'
-        )
-        return msg
+_PREDEF_SYMBOLS: set[str] | None = None
+
+
+@dataclass(frozen=True)
+class QColumnId:
+    """Qualified column id"""
+
+    tbl_id: UUID
+    col_id: int
 
 
 class MediaValidation(enum.Enum):