Skip to content

Commit f4c1d35

Browse files
mwojtyczkamwojtyczka
authored
Merge branch 'main' into add_context_info_to_output
2 parents ec7e2f8 + bce1d41 commit f4c1d35

File tree

4 files changed

+136
-6
lines changed

4 files changed

+136
-6
lines changed

‎CHANGELOG.md

+6
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,11 @@
11
# Version changelog
22

3+
## 0.2.0
4+
5+
* Added uniqueness check([#200](https://github.com/databrickslabs/dqx/issues/200)). A uniqueness check has been added, which reports an issue for each row containing a duplicate value in a specified column. This resolves issue [154](https://github.com/databrickslabs/dqx/issues/154).
6+
* Added column expression support for limits in not less and not greater than checks, and updated docs ([#200](https://github.com/databrickslabs/dqx/issues/200)). This commit introduces several changes to simplify and enhance data quality checking in PySpark workloads for both streaming and batch data. The naming conventions of rule functions have been unified, and the `is_not_less_than` and `is_not_greater_than` functions now accept column names or expressions as limits. The input parameters for range checks have been unified, and the logic of `is_not_in_range` has been updated to be inclusive of the boundaries. The project's documentation has been improved, with the addition of comprehensive examples, and the contribution guidelines have been clarified. This change includes a breaking change for some of the checks. Users are advised to review and test the changes before implementation to ensure compatibility and avoid any disruptions. Reslves issues: [131](https://github.com/databrickslabs/dqx/issues/131), [197](https://github.com/databrickslabs/dqx/pull/200), [175](https://github.com/databrickslabs/dqx/issues/175), [205](https://github.com/databrickslabs/dqx/issues/205)
7+
* Include predefined check functions by default when applying custom checks by metadata ([#203](https://github.com/databrickslabs/dqx/issues/203)). The data quality engine has been updated to include predefined check functions by default when applying custom checks using metadata in the form of YAML or JSON. This change simplifies the process of defining custom checks, as users no longer need to manually import predefined functions, which were previously required and could be cumbersome. The default behavior now is to import all predefined checks. The `validate_checks` method has been updated to accept a dictionary of custom check functions instead of global variables. This improvement resolves issue [#48](https://github.com/databrickslabs/dqx/issues/48).
8+
39
## 0.1.13
410

511
* Fixed cli installation and demo ([#177](https://github.com/databrickslabs/dqx/issues/177)). In this release, changes have been made to adjust the dashboard name, ensuring compliance with new API naming rules. The dashboard name now only contains alphanumeric characters, hyphens, or underscores, and the reference section has been split for clarity. In addition, demo for the tool has been updated to work regardless if a path or UC table is provided in the config. Furthermore, documentation has been refactored and udpated to improve clarity. The following issue have been closed: [#171](https://github.com/databrickslabs/dqx/issues/171) and [#198](https://github.com/databrickslabs/dqx/issues/198).

‎docs/dqx/docs/reference/quality_rules.mdx

+4-3
Original file line numberDiff line numberDiff line change
@@ -282,7 +282,7 @@ For brevity, the `name` field in the examples is omitted, meaning it will be aut
282282
col_name: col1
283283

284284
# is_unique check with custom window
285-
# default value for NULL in the time column of the window spec must be provided using coalesce() to prevent rows exclusion!
285+
# provide default value for NULL in the time column of the window spec using coalesce() to prevent rows exclusion!
286286
- criticality: error
287287
name: col1_is_not_unique2
288288
check:
@@ -315,6 +315,7 @@ For brevity, the `name` field in the examples is omitted, meaning it will be aut
315315
<details style={{ backgroundColor: 'transparent', color: 'neutral' }}>
316316
<summary>**Checks defined using DQX classes**</summary>
317317
```python
318+
from databricks.labs.dqx.rule import DQRule
318319
from databricks.labs.dqx.col_functions import *
319320
from datetime import datetime
320321

@@ -448,8 +449,8 @@ checks = [
448449
DQRule(
449450
criticality="error",
450451
name="col1_is_not_unique2",
451-
# default value for NULL in the time column of the window spec must be provided using coalesce() to prevent rows exclusion!
452-
check_func=is_unique, col_name="col1", check_func_kwargs={"window_spec": F.window(F.coalesce(F.col("col6"), F.lit(datetime(1970, 1, 1))), "10 minutes")}
452+
# provide default value for NULL in the time column of the window spec using coalesce() to prevent rows exclusion!
453+
check=is_unique("col1", window_spec=F.window(F.coalesce(F.col("col6"), F.lit(datetime(1970, 1, 1))), "10 minutes"))
453454
),
454455
DQRule(
455456
criticality="error",

‎src/databricks/labs/dqx/__about__.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
__version__ = "0.1.13"
1+
__version__ = "0.2.0"

‎tests/integration/test_apply_checks.py

+125-2
Original file line numberDiff line numberDiff line change
@@ -5,15 +5,35 @@
55
import pytest
66
from pyspark.sql import Column
77
from chispa.dataframe_comparer import assert_df_equality # type: ignore
8-
from databricks.labs.dqx.col_functions import is_not_null_and_not_empty, make_condition
8+
from databricks.labs.dqx.col_functions import (
9+
is_not_null_and_not_empty,
10+
make_condition,
11+
sql_expression,
12+
regex_match,
13+
is_unique,
14+
is_older_than_col2_for_n_days,
15+
is_older_than_n_days,
16+
is_not_in_near_future,
17+
is_not_in_future,
18+
is_valid_timestamp,
19+
is_valid_date,
20+
is_not_greater_than,
21+
is_not_less_than,
22+
is_not_in_range,
23+
is_in_range,
24+
is_not_null_and_not_empty_array,
25+
is_not_null_and_is_in_list,
26+
is_in_list,
27+
is_not_empty,
28+
is_not_null,
29+
)
930
from databricks.labs.dqx.engine import (
1031
DQEngine,
1132
ExtraParams,
1233
)
1334
from databricks.labs.dqx.rule import DQRule, DQRuleColSet, ColumnArguments
1435
from databricks.labs.dqx.schema import validation_result_schema
1536

16-
1737
SCHEMA = "a: int, b: int, c: int"
1838
REPORTING_COLUMNS = (
1939
f", _errors: {validation_result_schema.simpleString()}, _warnings: {validation_result_schema.simpleString()}"
@@ -1866,3 +1886,106 @@ def test_apply_checks_all_checks_as_yaml(ws, spark):
18661886
expected_schema,
18671887
)
18681888
assert_df_equality(checked, expected, ignore_nullable=True)
1889+
1890+
1891+
def test_apply_checks_all_checks_using_classes(ws, spark):
1892+
checks = [
1893+
DQRule(criticality="error", check=is_not_null("col1")),
1894+
DQRule(criticality="error", check=is_not_empty("col1")),
1895+
DQRule(criticality="error", check=is_not_null_and_not_empty("col1", trim_strings=True)),
1896+
DQRule(criticality="error", check=is_in_list("col2", [1, 2, 3])),
1897+
DQRule(criticality="error", check=is_not_null_and_is_in_list("col2", [1, 2, 3])),
1898+
DQRule(criticality="error", check=is_not_null_and_not_empty_array("col4")),
1899+
DQRule(criticality="error", check=is_in_range("col2", min_limit=1, max_limit=10)),
1900+
DQRule(
1901+
criticality="error",
1902+
check=is_in_range("col5", min_limit=datetime(2025, 1, 1).date(), max_limit=datetime(2025, 2, 24).date()),
1903+
),
1904+
DQRule(
1905+
criticality="error",
1906+
check=is_in_range(
1907+
"col6", min_limit=datetime(2025, 1, 1, 0, 0, 0), max_limit=datetime(2025, 2, 24, 1, 0, 0)
1908+
),
1909+
),
1910+
DQRule(criticality="error", check=is_in_range("col3", min_limit="col2", max_limit="col2 * 2")),
1911+
DQRule(criticality="error", check=is_not_in_range("col2", min_limit=11, max_limit=20)),
1912+
DQRule(
1913+
criticality="error",
1914+
check=is_not_in_range(
1915+
"col5", min_limit=datetime(2025, 2, 25).date(), max_limit=datetime(2025, 2, 26).date()
1916+
),
1917+
),
1918+
DQRule(
1919+
criticality="error",
1920+
check=is_not_in_range(
1921+
"col6", min_limit=datetime(2025, 2, 25, 0, 0, 0), max_limit=datetime(2025, 2, 26, 1, 0, 0)
1922+
),
1923+
),
1924+
DQRule(criticality="error", check=is_not_in_range("col3", min_limit="col2 + 10", max_limit="col2 * 10")),
1925+
DQRule(criticality="error", check=is_not_less_than("col2", limit=0)),
1926+
DQRule(criticality="error", check=is_not_less_than("col5", limit=datetime(2025, 1, 1).date())),
1927+
DQRule(criticality="error", check=is_not_less_than("col6", limit=datetime(2025, 1, 1, 1, 0, 0))),
1928+
DQRule(criticality="error", check=is_not_less_than("col3", limit="col2 - 10")),
1929+
DQRule(criticality="error", check=is_not_greater_than("col2", limit=10)),
1930+
DQRule(criticality="error", check=is_not_greater_than("col5", limit=datetime(2025, 3, 1).date())),
1931+
DQRule(criticality="error", check=is_not_greater_than("col6", limit=datetime(2025, 3, 24, 1, 0, 0))),
1932+
DQRule(criticality="error", check=is_not_greater_than("col3", limit="col2 + 10")),
1933+
DQRule(criticality="error", check=is_valid_date("col5")),
1934+
DQRule(
1935+
criticality="error", check=is_valid_date("col5", date_format="yyyy-MM-dd"), name="col5_is_not_valid_date2"
1936+
),
1937+
DQRule(criticality="error", check=is_valid_timestamp("col6")),
1938+
DQRule(
1939+
criticality="error",
1940+
check=is_valid_timestamp("col6", timestamp_format="yyyy-MM-dd HH:mm:ss"),
1941+
name="col6_is_not_valid_timestamp2",
1942+
),
1943+
DQRule(criticality="error", check=is_not_in_future("col6", offset=86400)),
1944+
DQRule(criticality="error", check=is_not_in_near_future("col6", offset=36400)),
1945+
DQRule(criticality="error", check=is_older_than_n_days("col5", days=10000)),
1946+
DQRule(criticality="error", check=is_older_than_col2_for_n_days("col5", "col6", days=2)),
1947+
DQRule(criticality="error", check=is_unique("col1")),
1948+
DQRule(
1949+
criticality="error",
1950+
name="col1_is_not_unique2",
1951+
# provide default value for NULL in the time column of the window spec using coalesce()
1952+
# to prevent rows exclusion!
1953+
check=is_unique(
1954+
"col1", window_spec=F.window(F.coalesce(F.col("col6"), F.lit(datetime(1970, 1, 1))), "10 minutes")
1955+
),
1956+
),
1957+
DQRule(criticality="error", check=regex_match("col2", regex="[0-9]+", negate=False)),
1958+
DQRule(
1959+
criticality="error",
1960+
check=sql_expression(
1961+
expression="col3 > col2 and col3 < 10",
1962+
msg="col3 is greater than col2 and col3 less than 10",
1963+
name="custom_output_name",
1964+
negate=False,
1965+
),
1966+
),
1967+
]
1968+
dq_engine = DQEngine(ws)
1969+
1970+
schema = "col1: string, col2: int, col3: int, col4 array<int>, col5: date, col6: timestamp"
1971+
test_df = spark.createDataFrame(
1972+
[
1973+
["val1", 1, 1, [1], datetime(2025, 1, 2).date(), datetime(2025, 1, 2, 1, 0, 0)],
1974+
["val2", 2, 2, [2], datetime(2025, 1, 2).date(), datetime(2025, 1, 2, 2, 0, 0)],
1975+
["val3", 3, 3, [3], datetime(2025, 1, 2).date(), datetime(2025, 1, 2, 3, 0, 0)],
1976+
],
1977+
schema,
1978+
)
1979+
1980+
checked = dq_engine.apply_checks(test_df, checks)
1981+
1982+
expected_schema = schema + REPORTING_COLUMNS
1983+
expected = spark.createDataFrame(
1984+
[
1985+
["val1", 1, 1, [1], datetime(2025, 1, 2).date(), datetime(2025, 1, 2, 1, 0, 0), None, None],
1986+
["val2", 2, 2, [2], datetime(2025, 1, 2).date(), datetime(2025, 1, 2, 2, 0, 0), None, None],
1987+
["val3", 3, 3, [3], datetime(2025, 1, 2).date(), datetime(2025, 1, 2, 3, 0, 0), None, None],
1988+
],
1989+
expected_schema,
1990+
)
1991+
assert_df_equality(checked, expected, ignore_nullable=True)

0 commit comments

Comments
 (0)