Validate.col_vals_expr()

Validate column values using a custom expression.

Usage

Source

Validate.col_vals_expr(
    expr,
    pre=None,
    segments=None,
    thresholds=None,
    actions=None,
    brief=None,
    active=True,
    dimension=None,
)

The col_vals_expr() validation method checks whether column values in a table satisfy a custom expr= expression. This validation will operate over the number of test units that is equal to the number of rows in the table (determined after any pre= mutation has been applied).

Parameters

expr: Any

A column expression that will evaluate each row in the table, returning a boolean value per table row. A Narwhals column expression (e.g., nw.col("a") > 1) works with every table type. Otherwise, use an expression from the table’s own library: a Polars expression for a Polars table, an Ibis expression for an Ibis table, or a PySpark column for a PySpark DataFrame. A function (such as a lambda) can also be given; it receives the table and should return one of those expressions or a boolean Series (e.g., lambda df: df["a"] > 1 for a pandas DataFrame).

pre: Callable | None = None

An optional preprocessing function or lambda to apply to the data table during interrogation. This function should take a table as input and return a modified table. Have a look at the Preprocessing section for more information on how to use this argument.

segments: SegmentSpec | None = None

An optional directive on segmentation, which serves to split a validation step into multiple (one step per segment). Can be a single column name, a tuple that specifies a column name and its corresponding values to segment on, or a combination of both (provided as a list). Read the Segmentation section for usage information.

thresholds: int | float | bool | tuple | dict | Thresholds | None = None

Set threshold failure levels for reporting and reacting to exceedences of the levels. The thresholds are set at the step level and will override any global thresholds set in Validate(thresholds=...). The default is None, which means that no thresholds will be set locally and global thresholds (if any) will take effect. Look at the Thresholds section for information on how to set threshold levels.

actions: Actions | None = None

Optional actions to take when the validation step meets or exceeds any set threshold levels. If provided, the Actions class should be used to define the actions.

brief: str | bool | None = None

An optional brief description of the validation step that will be displayed in the reporting table. You can use the templating elements like "{step}" to insert the step number, or "{auto}" to include an automatically generated brief. If True the entire brief will be automatically generated. If None (the default) then there won’t be a brief.

active: bool | Callable = True

A boolean value or callable that determines whether the validation step should be active. Using False will make the validation step inactive (still reporting its presence and keeping indexes for the steps unchanged). A callable can also be provided; it will receive the data table as its single argument and must return a boolean value. The callable is evaluated before any pre= processing. Inspection functions like has_columns() and has_rows() can be used here to conditionally activate a step based on properties of the target table.

dimension: str | None = None
An optional data quality dimension to categorize this validation step for health scoring. One of "completeness", "validity", "uniqueness", "consistency", "timeliness", or "volume" (or any custom string). If None (the default), the dimension is inferred automatically from the assertion type. This label appears in the validation report and feeds the overall and per-dimension health scores.

Returns

Validate
The Validate object with the added validation step.

Missing Values And Evaluation Errors

A row where the expression evaluates to a missing value (e.g., pl.col("a") > 1 when a is null) counts as a failing test unit. To let such rows pass, allow for them in the expression itself, as in (pl.col("a") > 1) | pl.col("a").is_null().

If the expression can’t be evaluated against the table (e.g., it refers to a column that doesn’t exist, comes from a library that doesn’t match the table, or doesn’t produce boolean values), the step is marked as having an evaluation error and a note explains why. The other validation steps are still interrogated.

Preprocessing

The pre= argument allows for a preprocessing function or lambda to be applied to the data table during interrogation. This function should take a table as input and return a modified table. This is useful for performing any necessary transformations or filtering on the data before the validation step is applied.

The preprocessing function can be any callable that takes a table as input and returns a modified table. For example, you could use a lambda function to filter the table based on certain criteria or to apply a transformation to the data. Regarding the lifetime of the transformed table, it only exists during the validation step and is not stored in the Validate object or used in subsequent validation steps.

Segmentation

The segments= argument allows for the segmentation of a validation step into multiple segments. This is useful for applying the same validation step to different subsets of the data. The segmentation can be done based on a single column or specific fields within a column.

Providing a single column name will result in a separate validation step for each unique value in that column. For example, if you have a column called "region" with values "North", "South", and "East", the validation step will be applied separately to each region.

Alternatively, you can provide a tuple that specifies a column name and its corresponding values to segment on. For example, if you have a column called "date" and you want to segment on only specific dates, you can provide a tuple like ("date", ["2023-01-01", "2023-01-02"]). Any other values in the column will be disregarded (i.e., no validation steps will be created for them).

A list with a combination of column names and tuples can be provided as well. This allows for more complex segmentation scenarios. The following inputs are both valid:

# Segments from all unique values in the `region` column
# and specific dates in the `date` column
segments=["region", ("date", ["2023-01-01", "2023-01-02"])]

# Segments from all unique values in the `region` and `date` columns
segments=["region", "date"]

The segmentation is performed during interrogation, and the resulting validation steps will be numbered sequentially. Each segment will have its own validation step, and the results will be reported separately. This allows for a more granular analysis of the data and helps identify issues within specific segments.

Importantly, the segmentation process will be performed after any preprocessing of the data table. Because of this, one can conceivably use the pre= argument to generate a column that can be used for segmentation. For example, you could create a new column called "segment" through use of pre= and then use that column for segmentation.

Thresholds

The thresholds= parameter is used to set the failure-condition levels for the validation step. If they are set here at the step level, these thresholds will override any thresholds set at the global level in Validate(thresholds=...).

There are three threshold levels: ‘warning’, ‘error’, and ‘critical’. The threshold values can either be set as a proportion failing of all test units (a value between 0 to 1), or, the absolute number of failing test units (as integer that’s 1 or greater).

Thresholds can be defined using one of these input schemes:

  1. use the Thresholds class (the most direct way to create thresholds)
  2. provide a tuple of 1-3 values, where position 0 is the ‘warning’ level, position 1 is the ‘error’ level, and position 2 is the ‘critical’ level
  3. create a dictionary of 1-3 value entries; the valid keys: are ‘warning’, ‘error’, and ‘critical’
  4. a single integer/float value denoting absolute number or fraction of failing test units for the ‘warning’ level only

If the number of failing test units exceeds set thresholds, the validation step will be marked as ‘warning’, ‘error’, or ‘critical’. All of the threshold levels don’t need to be set, you’re free to set any combination of them.

Aside from reporting failure conditions, thresholds can be used to determine the actions to take for each level of failure (using the actions= parameter).

Examples

For the examples here, we’ll use a simple Polars DataFrame with three columns (a, b, and c). The table is shown below:

import pointblank as pb
import polars as pl

tbl = pl.DataFrame(
    {
        "a": [1, 2, 1, 7, 8, 6],
        "b": [0, 0, 0, 1, 1, 1],
        "c": [0.5, 0.3, 0.8, 1.4, 1.9, 1.2],
    }
)

pb.preview(tbl)
a
Int64
b
Int64
c
Float64
1 1 0 0.5
2 2 0 0.3
3 1 0 0.8
4 7 1 1.4
5 8 1 1.9
6 6 1 1.2

Let’s validate that the values in column a are all integers. We’ll determine if this validation had any failing test units (there are six test units, one for each row).

validation = (
    pb.Validate(data=tbl)
    .col_vals_expr(expr=pl.col("a") % 1 == 0)
    .interrogate()
)

validation
STEP COLUMNS VALUES TBL EVAL UNITS PASS FAIL W E C EXT
#4CA64C 1
col_vals_expr
col_vals_expr()
 — COLUMN EXPR ✓ 6 6
1.00		 0
0.00		 — — — —

Printing the validation object shows the validation table in an HTML viewing environment. The validation table shows the single entry that corresponds to the validation step created by using col_vals_expr(). All test units passed, with no failing test units.