Merge branch 'posit-dev:main' into add-get_dataframe

Author: Meghansaha

.github/workflows/code-checks.yaml

@@ -17,4 +17,11 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: "3.10"
+      - name: Setup uv
+        uses: astral-sh/setup-uv@v2
+      - name: Check validate.pyi is up to date
+        run: |
+          uv run make pyi
+          git diff --exit-code pointblank/validate.pyi || \
+            (echo "validate.pyi is out of date — run 'make pyi' and commit the result" && exit 1)
       - uses: pre-commit/action@v3.0.1

.pre-commit-config.yaml

@@ -1,15 +1,22 @@
 exclude: "(.*\\.svg)|(.*\\.qmd)|(.*\\.ambr)|(.*\\.csv)|(.*\\.txt)|(.*\\.json)|(.*\\.ipynb)|(.*\\.html)"
 repos:
+  - repo: local
+    hooks:
+      - id: check-pyi-sync
+        name: validate.pyi must be up to date
+        language: system
+        entry: bash -c 'make pyi && git diff --exit-code pointblank/validate.pyi'
+        pass_filenames: false
+        stages: [commit]
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v4.6.0
     hooks:
       - id: trailing-whitespace
       - id: end-of-file-fixer
+  # NOTE: ruff version must match the pin in pyproject.toml [dependency-groups] dev
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.9.9
+    rev: v0.14.10
     hooks:
-      # Run the linter.
       - id: ruff
         args: [--fix]
-      # Run the formatter.
       - id: ruff-format

.vscode/settings.json

@@ -13,11 +13,13 @@
     "editor.codeActionsOnSave": {
       "source.fixAll": "explicit",
       "source.organizeImports": "explicit"
-    },
+    }
   },
   "[markdown]": {
     "editor.formatOnSave": false
   },
+  "ruff.format.args": ["--exclude", "*.pyi", "--exclude", "*.md"],
+  "ruff.lint.args": ["--exclude", "*.pyi", "--exclude", "*.md"],
   "python.testing.pytestArgs": ["tests"],
   "python.testing.unittestEnabled": false,
   "python.testing.pytestEnabled": true,

CITATION.cff

@@ -3,7 +3,7 @@ message: 'If you wish to cite the "Pointblank" package use:'
 type: software
 license: MIT
 title: "Pointblank: data validation toolkit for assessing and monitoring data quality."
-version: 0.21.0
+version: 0.22.0
 abstract: Validate data in Polars and Pandas DataFrames and database tables.
   Validation pipelines can be made using easily-readable, consecutive validation
   steps. Upon execution of the validation plan, several reporting options are available.

Makefile

@@ -6,6 +6,8 @@ pyi: ## Generate .pyi stub files
 		--include-private \
 		-o  .
 	@uv run scripts/generate_agg_validate_pyi.py
+	@uv run ruff check --fix pointblank/validate.pyi
+	@uv run ruff format pointblank/validate.pyi
 
 .PHONY: test
 test:

pointblank/_agg.py

@@ -86,6 +86,20 @@ def _generic_between(real: float, lower: float, upper: float) -> bool:
     return bool(lower <= real <= upper)
 
 
+def split_agg_name(name: str) -> tuple[str, str]:
+    """Split an aggregation method name into aggregator and comparator names.
+
+    Args:
+        name (str): The aggregation method name (e.g., "col_sum_eq" or "sum_eq").
+
+    Returns:
+        tuple[str, str]: A tuple of (agg_name, comp_name) e.g., ("sum", "eq").
+    """
+    name = name.removeprefix("col_")
+    agg_name, comp_name = name.rsplit("_", 1)
+    return agg_name, comp_name
+
+
 def resolve_agg_registries(name: str) -> tuple[Aggregator, Comparator]:
     """Resolve the assertion name to a valid aggregator
 
@@ -95,8 +109,7 @@ def resolve_agg_registries(name: str) -> tuple[Aggregator, Comparator]:
     Returns:
         tuple[Aggregator, Comparator]: The aggregator and comparator functions.
     """
-    name = name.removeprefix("col_")
-    agg_name, comp_name = name.split("_")[-2:]
+    agg_name, comp_name = split_agg_name(name)
 
     aggregator = AGGREGATOR_REGISTRY.get(agg_name)
     comparator = COMPARATOR_REGISTRY.get(comp_name)

pointblank/validate.py

@@ -28,7 +28,12 @@
 from great_tables.vals import fmt_integer, fmt_number
 from importlib_resources import files
 
-from pointblank._agg import is_valid_agg, load_validation_method_grid, resolve_agg_registries
+from pointblank._agg import (
+    is_valid_agg,
+    load_validation_method_grid,
+    resolve_agg_registries,
+    split_agg_name,
+)
 from pointblank._constants import (
     ASSERTION_TYPE_METHOD_MAP,
     CHECK_MARK_SPAN,
@@ -19174,6 +19179,15 @@ def _create_autobrief_or_failure_text(
             for_failure=for_failure,
         )
 
+    if is_valid_agg(assertion_type):
+        return _create_text_agg(
+            lang=lang,
+            assertion_type=assertion_type,
+            column=column,
+            values=values,
+            for_failure=for_failure,
+        )
+
     return None
 
 
@@ -19208,6 +19222,52 @@ def _create_text_comparison(
     )
 
 
+def _create_text_agg(
+    lang: str,
+    assertion_type: str,
+    column: str | list[str],
+    values: dict[str, Any],
+    for_failure: bool = False,
+) -> str:
+    """Create autobrief text for aggregation methods like col_sum_eq, col_avg_gt, etc."""
+    type_ = _expect_failure_type(for_failure=for_failure)
+
+    agg_type, comp_type = split_agg_name(assertion_type)
+
+    # this is covered by the test `test_brief_auto_all_agg_methods` to make sure we don't
+    # create any weird secret agg constants.
+    agg_display_names: dict[str, str] = {
+        "sum": "sum",
+        "avg": "average",
+        "sd": "standard deviation",
+    }
+    try:
+        agg_display: str = agg_display_names[agg_type]
+    except KeyError as ke:  # pragma: no cover
+        raise AssertionError from ke  # This should never happen in prod, it's caught in CI.
+
+    # Get the operator
+    comparison_assertion = f"col_vals_{comp_type}"
+    if lang == "ar":  # pragma: no cover
+        operator = COMPARISON_OPERATORS_AR.get(comparison_assertion, comp_type)
+    else:
+        operator = COMPARISON_OPERATORS.get(comparison_assertion, comp_type)
+
+    column_text = _prep_column_text(column=column)
+
+    value = values.get("value", values) if isinstance(values, dict) else values
+    values_text = _prep_values_text(values=str(value), lang=lang, limit=3)
+
+    # "Expect that the {agg} of {column} should be {operator} {value}."
+    agg_expectation_text = EXPECT_FAIL_TEXT[f"compare_{type_}_text"][lang]
+
+    return agg_expectation_text.format(
+        column_text=f"the {agg_display} of {column_text}",
+        operator=operator,
+        values_text=values_text,
+    )
+
+
 def _create_text_between(
     lang: str,
     column: str,

pointblank/validate.pyi

@@ -1,8 +1,3 @@
-from pointblank import Actions, Thresholds
-from pointblank._utils import _PBUnresolvedColumn
-from pointblank.column import Column, ReferenceColumn
-from pointblank._typing import Tolerance
-
 import datetime
 from collections.abc import Collection
 from dataclasses import dataclass
@@ -17,20 +12,20 @@ from pointblank.thresholds import Actions, FinalActions, Thresholds
 from typing import Any, Callable, Literal, ParamSpec, TypeVar
 
 __all__ = [
-    "Validate",
+    "get_action_metadata",
+    "get_validation_summary",
+    "config",
     "load_dataset",
     "read_file",
     "write_file",
-    "config",
-    "connect_to_table",
-    "print_database_tables",
+    "get_data_path",
     "preview",
     "missing_vals_tbl",
-    "get_action_metadata",
     "get_column_count",
-    "get_data_path",
     "get_row_count",
-    "get_validation_summary",
+    "connect_to_table",
+    "print_database_tables",
+    "Validate",
 ]
 
 P = ParamSpec("P")
@@ -586,7 +581,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sum to a value eq some `value`.
 
@@ -622,7 +617,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sum to a value gt some `value`.
 
@@ -658,7 +653,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sum to a value ge some `value`.
 
@@ -694,7 +689,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sum to a value lt some `value`.
 
@@ -730,7 +725,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sum to a value le some `value`.
 
@@ -766,7 +761,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column avg to a value eq some `value`.
 
@@ -802,7 +797,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column avg to a value gt some `value`.
 
@@ -838,7 +833,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column avg to a value ge some `value`.
 
@@ -874,7 +869,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column avg to a value lt some `value`.
 
@@ -910,7 +905,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column avg to a value le some `value`.
 
@@ -946,7 +941,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sd to a value eq some `value`.
 
@@ -982,7 +977,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sd to a value gt some `value`.
 
@@ -1018,7 +1013,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sd to a value ge some `value`.
 
@@ -1054,7 +1049,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sd to a value lt some `value`.
 
@@ -1090,7 +1085,7 @@ class Validate:
         thresholds: float | bool | tuple | dict | Thresholds | None = None,
         brief: str | bool = False,
         actions: Actions | None = None,
-        active: bool = True,
+        active: bool | Callable = True,
     ) -> Validate:
         """Assert the values in a column sd to a value le some `value`.
 

pyproject.toml

@@ -107,7 +107,7 @@ dev = [
     "pytest-xdist>=3.6.1",
     "pytz>=2025.2",
     "quartodoc>=0.8.1; python_version >= '3.9'",
-    "ruff>=0.9.9",
+    "ruff==0.14.10",  # NOTE: must match rev in .pre-commit-config.yaml
     "shiny>=1.4.0",
     "openpyxl>=3.0.0",
     "mcp[cli]>=1.10.1",
@@ -136,7 +136,7 @@ testpaths = ["tests"]
 line-length = 100
 
 [tool.ruff.lint]
-exclude = ["docs", ".venv", "tests/*"]
+exclude = ["docs", ".venv", "tests/*", "pointblank/validate.pyi"]
 
 ignore = [
     "E402", # module level import not at top of file