@@ -45,57 +45,8 @@ validation = (
4545print(validation.get_json_report())
4646```
4747
48- Preview a table (any type) and get a consistent view of the data no matter what type of table it is.
49- This is good for getting your bearings on a new dataset.
50-
51- ``` {python}
52- import pointblank as pb
53-
54- data = pb.load_dataset(dataset="game_revenue", tbl_type="duckdb")
55-
56- pb.preview(data=data)
57- ```
58-
59- ### Perform a more comprehensive data validation with sample data
60-
61- The validation plan can be quite large if you want it to be. Here is validation of the ` penguins `
62- dataset as a DuckDB table.
63-
64- ``` {python}
65- import pointblank as pb
66- import ibis
67-
68- con = ibis.connect("duckdb://penguins.ddb")
69- con.create_table(
70- "penguins", ibis.examples.penguins.fetch().to_pyarrow(), overwrite=True
71- )
72-
73- penguins = con.table("penguins")
74-
75- validation = (
76- pb.Validate(
77- data=penguins,
78- tbl_name="penguins",
79- label="The penguins dataset"
80- )
81- .col_vals_gt(columns="body_mass_g", value=1000000)
82- .col_vals_lt(columns="year", value=2024)
83- .col_vals_eq(columns="year", value=2007)
84- .col_vals_ne(columns="year", value=2007)
85- .col_vals_ge(columns="year", value=2007)
86- .col_vals_regex(columns="sex", pattern="fe")
87- .col_exists(columns="body_mass_g")
88- .col_vals_in_set(columns="year", set=[2020, 2018])
89- .col_vals_not_in_set(columns="year", set=[2010, 2011])
90- .col_vals_between(columns="year", left=2007, right=2010)
91- .col_vals_outside(columns="year", left=2008, right=2013)
92- .interrogate()
93- )
94-
95- validation
96- ```
97-
98- We can use the ` pre= ` argument to mutate table for a specific validation step.
48+ We can use the ` pre= ` argument (available in every validation method) to mutate table for a specific
49+ validation step.
9950
10051``` {python}
10152import polars as pl
@@ -134,8 +85,21 @@ validation = (
13485validation
13586```
13687
137- The following validation plan (collection of validation steps) is even larger. Here, in two
138- different validation step, we use the ` pre= ` argument to mutate the target table with a lambda.
88+ Preview a table (any type) and get a consistent view of the data no matter what type of table it is.
89+ This is good for getting your bearings on a new dataset.
90+
91+ ``` {python}
92+ import pointblank as pb
93+
94+ data = pb.load_dataset(dataset="game_revenue", tbl_type="duckdb")
95+
96+ pb.preview(data=data)
97+ ```
98+
99+ ### Perform a more comprehensive data validation with sample data
100+
101+ The validation plan can be quite large if you want it to be. Here, in two different validation
102+ steps, we use the ` pre= ` argument to mutate the target table with a lambda.
139103
140104``` {python}
141105import pointblank as pb
0 commit comments