Skip to content

Test Catalogue 🧪

This documentation is intended as an exaustive list of possible tests within Wimsey. Note that examples given intentionally use all possible keywords for demonstrative purposes. This isn't required, and you can give as many or as few keywords as you like with the exception of where column is required.

mean_should

Test that column mean is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: mean_should

{
  "test": "mean_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import mean_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[mean_should(**keywords)])

min_should

Test that column min is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: min_should

{
  "test": "min_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import min_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[min_should(**keywords)])

max_should

Test that column max is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: max_should

{
  "test": "max_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import max_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[max_should(**keywords)])

std_should

Test that column std is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: std_should

{
  "test": "std_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import std_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[std_should(**keywords)])

null_count_should

Test that column null_count is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: null_count_should

{
  "test": "null_count_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import null_count_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[null_count_should(**keywords)])

count_should

Test that column count is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: count_should

{
  "test": "count_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import count_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[count_should(**keywords)])

null_percentage_should

Test that column null_percentage is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
test: null_percentage_should

{
  "test": "null_percentage_should",
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import null_percentage_should

keywords = {
  "column": "column_a",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[null_percentage_should(**keywords)])

columns_should

Test column names match up with expected values

be:
- column_a
- column_b
have:
- column_a
not_have:
- column_c
test: columns_should

{
  "test": "columns_should",
  "have": [
    "column_a"
  ],
  "not_have": [
    "column_c"
  ],
  "be": [
    "column_a",
    "column_b"
  ]
}

from wimsey import test
from wimsey.tests import columns_should

keywords = {
  "have": [
    "column_a"
  ],
  "not_have": [
    "column_c"
  ],
  "be": [
    "column_a",
    "column_b"
  ]
}

result = test(df, contract=[columns_should(**keywords)])

type_should

Test column type matches up with expected value. Note that this will expect polars style types, although does not require that they be case sensitive. For example, if testing a pandas dataframe for integer type, specify "int64" rather than, say, "int64[pyarrow]" or otherwise.

be: int64
be_one_of:
- int64
- float64
column: column_a
not_be: string
test: type_should

{
  "test": "type_should",
  "column": "column_a",
  "be": "int64",
  "not_be": "string",
  "be_one_of": [
    "int64",
    "float64"
  ]
}

from wimsey import test
from wimsey.tests import type_should

keywords = {
  "column": "column_a",
  "be": "int64",
  "not_be": "string",
  "be_one_of": [
    "int64",
    "float64"
  ]
}

result = test(df, contract=[type_should(**keywords)])

row_count_should

Test that dataframe row count is within designated range

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
test: row_count_should

{
  "test": "row_count_should",
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500,
  "be_exactly": 300
}

from wimsey import test
from wimsey.tests import row_count_should

keywords = {
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500,
  "be_exactly": 300
}

result = test(df, contract=[row_count_should(**keywords)])

average_difference_from_other_column_should

Test that the average difference between column and other column are within designated bounds.

be_exactly: 300
be_greater_than: 500
be_greater_than_or_equal_to: 500
be_less_than: 500
be_less_than_or_equal_to: 300
column: column_a
other_column: column_b
test: average_difference_from_other_column_should

{
  "test": "average_difference_from_other_column_should",
  "column": "column_a",
  "other_column": "column_b",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

from wimsey import test
from wimsey.tests import average_difference_from_other_column_should

keywords = {
  "column": "column_a",
  "other_column": "column_b",
  "be_exactly": 300,
  "be_less_than": 500,
  "be_less_than_or_equal_to": 300,
  "be_greater_than": 500,
  "be_greater_than_or_equal_to": 500
}

result = test(df, contract=[average_difference_from_other_column_should(**keywords)])