pointblank 0.17.0__py3-none-any.whl → 0.18.0__py3-none-any.whl

pointblank/datascan.py

@@ -3,12 +3,11 @@ from __future__ import annotations
 import contextlib
 import json
 from importlib.metadata import version
-from typing import TYPE_CHECKING, Any
+from typing import TYPE_CHECKING, Any, cast
 
 import narwhals as nw
 from great_tables import GT, google_font, html, loc, style
 from narwhals.dataframe import LazyFrame
-from narwhals.typing import FrameT
 
 from pointblank._utils_html import _create_table_dims_html, _create_table_type_html, _fmt_frac
 from pointblank.scan_profile import ColumnProfile, _as_physical, _DataProfile, _TypeMap
@@ -18,7 +17,7 @@ if TYPE_CHECKING:
     from collections.abc import Mapping, Sequence
 
     from narwhals.dataframe import DataFrame
-    from narwhals.typing import Frame, IntoFrameT
+    from narwhals.typing import Frame
 
     from pointblank.scan_profile_stats import StatGroup
 
@@ -123,7 +122,7 @@ class DataScan:
     """
 
     # TODO: This needs to be generically typed at the class level, ie. DataScan[T]
-    def __init__(self, data: IntoFrameT, tbl_name: str | None = None) -> None:
+    def __init__(self, data: Any, tbl_name: str | None = None) -> None:
         # Import processing functions from validate module
         from pointblank.validate import (
             _process_data,
@@ -172,7 +171,7 @@ class DataScan:
             implementation=self.nw_data.implementation,
         )
         for column in columns:
-            col_data: DataFrame = self.nw_data.select(column)
+            col_data: Frame = self.nw_data.select(column)
 
             ## Handle dtyping:
             native_dtype = schema[column]
@@ -183,7 +182,7 @@ class DataScan:
             except NotImplementedError:
                 continue
 
-            col_profile = ColumnProfile(colname=column, coltype=native_dtype)
+            col_profile = ColumnProfile(colname=column, coltype=str(native_dtype))
 
             ## Collect Sample Data:
             ## This is the most consistent way (i think) to get the samples out of the data.
@@ -205,7 +204,7 @@ class DataScan:
         return profile
 
     @property
-    def summary_data(self) -> IntoFrameT:
+    def summary_data(self) -> Any:
         return self.profile.as_dataframe(strict=False).to_native()
 
     def get_tabular_report(self, *, show_sample_data: bool = False) -> GT:
@@ -318,11 +317,10 @@ class DataScan:
 
         # format fractions:
         # this is an anti-pattern but there's no serious alternative
+        _backend = cast(Any, self.profile.implementation)
         for _fmt_col in ("__frac_n_unique", "__frac_n_missing"):
             _formatted: list[str | None] = _fmt_frac(formatted_data[_fmt_col])
-            formatted: nw.Series = nw.new_series(
-                _fmt_col, values=_formatted, backend=self.profile.implementation
-            )
+            formatted: nw.Series = nw.new_series(_fmt_col, values=_formatted, backend=_backend)
             formatted_data = formatted_data.drop(_fmt_col)
             formatted_data = formatted_data.with_columns(formatted.alias(_fmt_col))
 
@@ -365,10 +363,10 @@ class DataScan:
                         trues.append(None)
                         falses.append(None)
                 true_ser: nw.Series = nw.new_series(
-                    name="__freq_true", values=trues, backend=self.profile.implementation
+                    name="__freq_true", values=trues, backend=_backend
                 )
                 false_ser: nw.Series = nw.new_series(
-                    name="__freq_false", values=falses, backend=self.profile.implementation
+                    name="__freq_false", values=falses, backend=_backend
                 )
                 formatted_data = formatted_data.with_columns(
                     __freq_true=true_ser, __freq_false=false_ser
@@ -382,9 +380,7 @@ class DataScan:
             )
             for _fmt_col in ("__pct_true", "__pct_false"):
                 _formatted: list[str | None] = _fmt_frac(formatted_data[_fmt_col])
-                formatted = nw.new_series(
-                    name=_fmt_col, values=_formatted, backend=self.profile.implementation
-                )
+                formatted = nw.new_series(name=_fmt_col, values=_formatted, backend=_backend)
                 formatted_data = formatted_data.drop(_fmt_col)
                 formatted_data = formatted_data.with_columns(formatted.alias(_fmt_col))
 
@@ -459,7 +455,11 @@ class DataScan:
             )
             .tab_style(style=style.text(size="12px"), locations=loc.body(columns="colname"))
             .cols_width(
-                icon="35px", colname="200px", **{stat_col: "60px" for stat_col in present_stat_cols}
+                cases={
+                    "icon": "35px",
+                    "colname": "200px",
+                    **{stat_col: "60px" for stat_col in present_stat_cols},
+                }
             )
         )
 
@@ -498,7 +498,7 @@ class DataScan:
             json.dump(json_string, f, indent=4)
 
 
-def col_summary_tbl(data: FrameT | Any, tbl_name: str | None = None) -> GT:
+def col_summary_tbl(data: Any, tbl_name: str | None = None) -> GT:
     """
     Generate a column-level summary table of a dataset.
 

pointblank/draft.py

@@ -4,7 +4,6 @@ from dataclasses import dataclass, field
 from typing import Any
 
 from importlib_resources import files
-from narwhals.typing import FrameT
 
 from pointblank._constants import MODEL_PROVIDERS
 from pointblank.datascan import DataScan
@@ -223,7 +222,7 @@ class DraftValidation:
     be replaced with the actual data variable.
     """
 
-    data: FrameT | Any
+    data: Any
     model: str
     api_key: str | None = None
     verify_ssl: bool = True
@@ -328,7 +327,7 @@ class DraftValidation:
         if provider == "anthropic":  # pragma: no cover
             # Check that the anthropic package is installed
             try:
-                import anthropic  # noqa
+                import anthropic  # noqa  # type: ignore[import-not-found]
             except ImportError:  # pragma: no cover
                 raise ImportError(  # pragma: no cover
                     "The `anthropic` package is required to use the `DraftValidation` class with "

pointblank/scan_profile.py

@@ -5,7 +5,7 @@ from collections import defaultdict
 from collections.abc import Sequence
 from dataclasses import dataclass, field
 from enum import Enum
-from typing import TYPE_CHECKING, Any
+from typing import TYPE_CHECKING, Any, ClassVar
 
 import narwhals as nw
 from narwhals.dataframe import DataFrame
@@ -96,6 +96,7 @@ class ColumnProfile(_ColumnProfileABC):
     colname: str
     coltype: str
     statistics: MutableSequence[Stat] = field(default_factory=lambda: [])
+    _type: ClassVar[_TypeMap]  # Defined by subclasses
 
     @property
     def sample_data(self) -> Sequence[Any]:

pointblank/schema.py

@@ -2,12 +2,16 @@ from __future__ import annotations
 
 import copy
 from dataclasses import dataclass
+from typing import TYPE_CHECKING
 
 import narwhals as nw
 
 from pointblank._constants import IBIS_BACKENDS
 from pointblank._utils import _get_tbl_type, _is_lazy_frame, _is_lib_present, _is_narwhals_table
 
+if TYPE_CHECKING:
+    from typing import Any
+
 __all__ = ["Schema", "_check_schema_match"]
 
 
@@ -269,17 +273,15 @@ class Schema:
     `Schema` object is used in a validation workflow.
     """
 
-    columns: str | list[str] | list[tuple[str, str]] | list[tuple[str]] | dict[str, str] | None = (
-        None
-    )
-    tbl: any | None = None
+    columns: list[tuple[str, ...]] | None = None
+    tbl: Any | None = None
 
     def __init__(
         self,
         columns: (
             str | list[str] | list[tuple[str, str]] | list[tuple[str]] | dict[str, str] | None
         ) = None,
-        tbl: any | None = None,
+        tbl: Any | None = None,
         **kwargs,
     ):
         if tbl is None and columns is None and not kwargs:
@@ -387,6 +389,8 @@ class Schema:
         bool
             True if the columns are the same, False otherwise.
         """
+        if self.columns is None or other.columns is None:
+            return self.columns is None and other.columns is None
 
         if not case_sensitive_colnames:
             this_column_list = [col.lower() for col in self.get_column_list()]
@@ -463,6 +467,8 @@ class Schema:
         bool
             True if the columns are the same, False otherwise.
         """
+        if self.columns is None or other.columns is None:
+            return self.columns is None and other.columns is None
 
         if not case_sensitive_colnames:
             this_column_list = [col.lower() for col in self.get_column_list()]
@@ -547,6 +553,8 @@ class Schema:
         bool
             True if the columns are the same, False otherwise.
         """
+        if self.columns is None or other.columns is None:
+            return self.columns is None and other.columns is None
 
         if not case_sensitive_colnames:
             this_column_list = [col.lower() for col in self.get_column_list()]
@@ -633,6 +641,8 @@ class Schema:
         bool
             True if the columns are the same, False otherwise.
         """
+        if self.columns is None or other.columns is None:
+            return self.columns is None and other.columns is None
 
         if not case_sensitive_colnames:
             this_column_list = [col.lower() for col in self.get_column_list()]
@@ -702,6 +712,8 @@ class Schema:
         list[str]
             A list of column names.
         """
+        if self.columns is None:
+            return []
         return [col[0] for col in self.columns]
 
     def get_dtype_list(self) -> list[str]:
@@ -713,9 +725,11 @@ class Schema:
         list[str]
             A list of data types.
         """
+        if self.columns is None:
+            return []
         return [col[1] for col in self.columns]
 
-    def get_schema_coerced(self, to: str | None = None) -> dict[str, str]:
+    def get_schema_coerced(self, to: str | None = None) -> Schema:
         # If a table isn't provided, we cannot use this method
         if self.tbl is None:
             raise ValueError(
@@ -755,8 +769,15 @@ class Schema:
                 new_schema = copy.deepcopy(Schema(tbl=(self.tbl.to_pandas())))
                 return new_schema
 
+        raise ValueError(
+            f"Cannot coerce schema from '{self.tbl_type}' to '{to}'. "
+            "Supported conversions: pandas->polars, polars->pandas."
+        )
+
     def __str__(self):
         formatted_columns = []
+        if self.columns is None:
+            return "Pointblank Schema (empty)"
         for col in self.columns:
             if len(col) == 1:  # Only column name provided (no data type)
                 formatted_columns.append(f"  {col[0]}: <ANY>")
@@ -770,8 +791,15 @@ class Schema:
 
 
 def _process_columns(
-    *, columns: str | list[str] | list[tuple[str, str]] | dict[str, str] | None = None, **kwargs
-) -> list[tuple[str, str]]:
+    *,
+    columns: str
+    | list[str]
+    | list[tuple[str, str]]
+    | list[tuple[str]]
+    | dict[str, str]
+    | None = None,
+    **kwargs,
+) -> list[tuple[str, ...]]:
     """
     Process column information provided as individual arguments or as a list of
     tuples/dictionary.
@@ -785,15 +813,18 @@ def _process_columns(
 
     Returns
     -------
-    list[tuple[str, str]]
-        A list of tuples containing column information.
+    list[tuple[str, ...]]
+        A list of tuples containing column information (name only or name and dtype).
     """
     if columns is not None:
         if isinstance(columns, list):
             if all(isinstance(col, str) for col in columns):
-                return [(col,) for col in columns]
+                # Type narrowing: after the all() check, columns contains only strings
+                str_columns: list[str] = columns  # type: ignore[assignment]
+                return [(col,) for col in str_columns]
             else:
-                return columns
+                # Type narrowing: columns contains tuples
+                return columns  # type: ignore[return-value]
 
         if isinstance(columns, str):
             return [(columns,)]
@@ -810,11 +841,11 @@ def _schema_info_generate_colname_dict(
     index_matched: bool,
     matched_to: str | None,
     dtype_present: bool,
-    dtype_input: str | list[str],
+    dtype_input: str | list[str] | None,
     dtype_matched: bool,
     dtype_multiple: bool,
-    dtype_matched_pos: int,
-) -> dict[str, any]:
+    dtype_matched_pos: int | None,
+) -> dict[str, Any]:
     return {
         "colname_matched": colname_matched,
         "index_matched": index_matched,
@@ -829,8 +860,8 @@ def _schema_info_generate_colname_dict(
 
 def _schema_info_generate_columns_dict(
     colnames: list[str] | None,
-    colname_dict: list[dict[str, any]] | None,
-) -> dict[str, dict[str, any]]:
+    colname_dict: list[dict[str, Any]] | None,
+) -> dict[str, dict[str, Any]]:
     """
     Generate the columns dictionary for the schema information dictionary.
 
@@ -847,6 +878,7 @@ def _schema_info_generate_columns_dict(
     dict[str, dict[str, any]]
         The columns dictionary.
     """
+    assert colnames is not None and colname_dict is not None
     return {colnames[i]: colname_dict[i] for i in range(len(colnames))}
 
 
@@ -856,7 +888,7 @@ def _schema_info_generate_params_dict(
     case_sensitive_colnames: bool,
     case_sensitive_dtypes: bool,
     full_match_dtypes: bool,
-) -> dict[str, any]:
+) -> dict[str, Any]:
     """
     Generate the parameters dictionary for the schema information dictionary.
 
@@ -889,7 +921,7 @@ def _schema_info_generate_params_dict(
 
 
 def _get_schema_validation_info(
-    data_tbl: any,
+    data_tbl: Any,
     schema: Schema,
     passed: bool,
     complete: bool,
@@ -897,7 +929,7 @@ def _get_schema_validation_info(
     case_sensitive_colnames: bool,
     case_sensitive_dtypes: bool,
     full_match_dtypes: bool,
-) -> dict[str, any]:
+) -> dict[str, Any]:
     """
     Get the schema validation information dictionary.
 
@@ -949,6 +981,10 @@ def _get_schema_validation_info(
     schema_exp = schema
     schema_tgt = Schema(tbl=data_tbl)
 
+    # Both schemas must have columns for validation
+    assert schema_exp.columns is not None, "Expected schema must have columns"
+    assert schema_tgt.columns is not None, "Target schema must have columns"
+
     # Initialize the schema information dictionary
     schema_info = {
         "passed": passed,
@@ -1122,6 +1158,11 @@ def _get_schema_validation_info(
         #
 
         if colname_matched and dtype_present:
+            # Type narrowing: matched_to is not None when colname_matched is True
+            # and dtype_input is not None when dtype_present is True
+            assert matched_to is not None
+            assert dtype_input is not None
+
             # Get the dtype of the column in the target table
             dtype_tgt = schema_tgt.columns[tgt_colnames.index(matched_to)][1]
 

pointblank/thresholds.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from dataclasses import dataclass, field
-from typing import Callable
+from typing import Callable, cast
 
 __all__ = ["Thresholds", "Actions", "FinalActions"]
 
@@ -180,14 +180,15 @@ class Thresholds:
         # The threshold value might be an absolute count, but we need to convert
         # it to a fractional value
         if isinstance(threshold_value, int):
-            threshold_value = _convert_abs_count_to_fraction(
-                value=threshold_value, test_units=test_units
-            )
+            converted = _convert_abs_count_to_fraction(value=threshold_value, test_units=test_units)
+            if converted is None:
+                return None
+            threshold_value = converted
 
         return fraction_failing >= threshold_value
 
 
-def _convert_abs_count_to_fraction(value: int | None, test_units: int) -> float:
+def _convert_abs_count_to_fraction(value: int | None, test_units: int) -> float | None:
     # Using a integer value signifying the total number of 'test units' (in the
     # context of a validation), we convert an integer count (absolute) threshold
     # value to a fractional threshold value
@@ -251,12 +252,12 @@ def _normalize_thresholds_creation(
         # any of these keys
 
         # Check keys for invalid entries and raise a ValueError if any are found
-        invalid_keys = set(thresholds.keys()) - {"warning", "error", "critical"}
+        invalid_keys: set = set(thresholds.keys()) - {"warning", "error", "critical"}
 
         if invalid_keys:
             raise ValueError(f"Invalid keys in the thresholds dictionary: {invalid_keys}")
 
-        thresholds = Thresholds(**thresholds)
+        thresholds = Thresholds(**cast(dict[str, int | float | None], thresholds))
 
     elif isinstance(thresholds, Thresholds):
         pass
@@ -483,12 +484,12 @@ class Actions:
 
     def _ensure_list(
         self, value: str | Callable | list[str | Callable] | None
-    ) -> list[str | Callable]:
+    ) -> list[str | Callable] | None:
         if value is None:
             return None
-        if not isinstance(value, list):
-            return [value]
-        return value
+        if isinstance(value, list):
+            return cast(list[str | Callable], value)
+        return [value]
 
     def __repr__(self) -> str:
         return f"Actions(warning={self.warning}, error={self.error}, critical={self.critical})"
@@ -627,13 +628,14 @@ class FinalActions:
     def __repr__(self) -> str:
         if isinstance(self.actions, list):
             action_reprs = ", ".join(
-                f"'{a}'" if isinstance(a, str) else a.__name__ for a in self.actions
+                f"'{a}'" if isinstance(a, str) else getattr(a, "__name__", repr(a))
+                for a in self.actions
             )
             return f"FinalActions([{action_reprs}])"
         elif isinstance(self.actions, str):
             return f"FinalActions('{self.actions}')"
         elif callable(self.actions):
-            return f"FinalActions({self.actions.__name__})"
+            return f"FinalActions({getattr(self.actions, '__name__', repr(self.actions))})"
         else:
             return f"FinalActions({self.actions})"  # pragma: no cover