def parsed_pdf(request):
request.cls.parsed_pdf = tn_aggregate_ingest.parse(
'', fixtures.as_filepath('_jailjanuary2019.pdf'))
request.cls.parsed_female_pdf = tn_aggregate_ingest.parse(
'', fixtures.as_filepath('_jailfemalejanuary2019.pdf'))
request.cls.parsed_newer_pdf = tn_aggregate_ingest.parse(
'', fixtures.as_filepath('_jailmarch2020.pdf'))
def setUpClass(cls) -> None:
# Cache the parsed pdf between tests since it's expensive to compute
cls.parsed_pdf = fl_aggregate_ingest.parse(
fixtures.as_filepath("jails-2018-01.pdf"))
cls.parsed_pdf_2 = fl_aggregate_ingest.parse(
fixtures.as_filepath(
"florida__pub_jails_2019_2019_06 june fcdf.pdf"))
def parsed_pdf(request):
request.cls.parsed_pdf = fl_aggregate_ingest.parse(
"", fixtures.as_filepath("jails-2018-01.pdf")
)
request.cls.parsed_pdf_2 = fl_aggregate_ingest.parse(
"", fixtures.as_filepath("florida__pub_jails_2019_2019_06 june fcdf.pdf")
)
def parsed_pdf(request):
request.cls.parsed_pdf = ky_aggregate_ingest.parse(
"", fixtures.as_filepath("12-20-18.pdf"))
request.cls.parsed_pdf_2 = ky_aggregate_ingest.parse(
"", fixtures.as_filepath("08-23-18.pdf"))
request.cls.parsed_pdf_3 = ky_aggregate_ingest.parse(
"", fixtures.as_filepath("08-22-19.pdf"))
def setUpClass(cls) -> None:
# Cache the parsed pdf between tests since it's expensive to compute
cls.parsed_pdf = ny_aggregate_ingest.parse(
fixtures.as_filepath("jail_population.pdf")
)
cls.parsed_pdf_3_pages = ny_aggregate_ingest.parse(
fixtures.as_filepath("jail_population_2019.pdf")
)
def setUpClass(cls) -> None:
# Cache the parsed pdfs between tests since it's expensive to compute
cls.parsed_pdf = tn_aggregate_ingest.parse(
fixtures.as_filepath("_jailjanuary2019.pdf"))
cls.parsed_female_pdf = tn_aggregate_ingest.parse(
fixtures.as_filepath("_jailfemalejanuary2019.pdf"))
cls.parsed_newer_pdf = tn_aggregate_ingest.parse(
fixtures.as_filepath("_jailmarch2020.pdf"))
def setUpClass(cls) -> None:
# Cache the parsed pdf between tests since it's expensive to compute
cls.parsed_pdf = ky_aggregate_ingest.parse(
fixtures.as_filepath("12-20-18.pdf"))
cls.parsed_pdf_2 = ky_aggregate_ingest.parse(
fixtures.as_filepath("08-23-18.pdf"))
cls.parsed_pdf_3 = ky_aggregate_ingest.parse(
fixtures.as_filepath("08-22-19.pdf"))
def setUpClass(cls) -> None:
# Cache the parsed pdf between tests since it's expensive to compute
cls.parsed_pdf_before_1996 = tx_aggregate_ingest.parse(
fixtures.as_filepath("abbreviated pop rpt march 1994.pdf"))
cls.parsed_pdf_1996 = tx_aggregate_ingest.parse(
fixtures.as_filepath(
"texas_url_abbreviated pop rpt June 1996.pdf"))
cls.parsed_pdf_after_1996 = tx_aggregate_ingest.parse(
fixtures.as_filepath("Abbreviated Pop Rpt Dec 2017.pdf"))
cls.parsed_pdf_concat = tx_aggregate_ingest.parse(
fixtures.as_filepath(
"docs_abbreviatedpopreports_abbreviated pop rpt oct 2003.pdf"))
def test_get_export_config_valid(self) -> None:
product_configs = ProductConfigs.from_file(
path=fixtures.as_filepath("fixture_products.yaml"))
_export_config = product_configs.get_export_config(
export_job_name="EXPORT",
state_code="US_XX",
)
def test_read_with_exception(self) -> None:
class _TestException(ValueError):
pass
class _ExceptionDelegate(_TestGcsfsCsvReaderDelegate):
def on_dataframe(self, encoding: str, chunk_num: int,
df: pd.DataFrame) -> bool:
should_continue = super().on_dataframe(encoding, chunk_num, df)
if chunk_num > 0:
raise _TestException("We crashed processing!")
return should_continue
file_path = fixtures.as_filepath("encoded_utf_8.csv")
delegate = _ExceptionDelegate()
with self.assertRaises(_TestException):
self.reader.streaming_read(
GcsfsFilePath.from_absolute_path(file_path),
delegate=delegate,
chunk_size=1,
)
self.assertEqual(1, len(delegate.encodings_attempted))
self.assertEqual("UTF-8", delegate.encodings_attempted[0])
self.assertIsNone(delegate.successful_encoding)
self.assertEqual(2, len(delegate.dataframes))
self.assertEqual({"UTF-8"},
{encoding
for encoding, df in delegate.dataframes})
self.assertEqual(0, delegate.decode_errors)
self.assertEqual(1, delegate.exceptions)
def test_direct_ingest_preprocessed_view_detect_row_deletion_unknown_pk_table_specified(
self, ) -> None:
region_config = DirectIngestRegionRawFileConfig(
region_code="us_xx",
yaml_config_file_dir=fixtures.as_filepath("us_xx"),
)
view_query_template = """SELECT * FROM {file_tag_first}
LEFT OUTER JOIN {tagFullHistoricalExport}
USING (col1);"""
with self.assertRaises(ValueError) as e:
DirectIngestPreProcessedIngestView(
ingest_view_name="ingest_view_tag",
view_query_template=view_query_template,
region_raw_table_config=region_config,
order_by_cols="col1, col2",
is_detect_row_deletion_view=True,
primary_key_tables_for_entity_deletion=[
"tagFullHistoricalExport",
"unknown",
],
)
self.assertTrue(
str(e.exception).startswith(
"Ingest view ingest_view_tag has specified unknown in "
"`primary_key_tables_for_entity_deletion`, but that "
"raw file tag was not found as a dependency."))
def test_parse_empty_yaml_throws(self):
with self.assertRaises(ValueError):
_ = DirectIngestRegionRawFileConfig(
region_code='us_xx',
yaml_config_file_path=fixtures.as_filepath(
'empty_raw_data_files.yaml'),
)
def test_happy_path(self) -> None:
yaml_path = fixtures.as_filepath("schema_config.yaml")
validation_schema_config = DatasetSchemaInfo.from_yaml(yaml_path)
expected = DatasetSchemaInfo(
dataset="fixture_schema",
tables=[
TableSchemaInfo(
table_name="incarceration_population_by_facility",
columns=[
"date_of_stay",
"facility",
"month",
"population_count",
"region_code",
"year",
],
),
TableSchemaInfo(
table_name="incarceration_population_person_level",
columns=[
"date_of_stay",
"facility",
"person_external_id",
"region_code",
],
),
],
)
self.assertEqual(expected, validation_schema_config)
def _build(
self,
*,
dataset_overrides: Optional[Dict[str, str]] = None
) -> DirectIngestPreProcessedIngestView:
region_config = DirectIngestRegionRawFileConfig(
region_code="us_xx",
yaml_config_file_dir=fixtures.as_filepath("us_xx"),
)
query = "select * from {file_tag_first} JOIN {tagFullHistoricalExport} USING (COL_1)"
primary_key_tables_for_entity_deletion = (
[] if not self.is_detect_row_deletion_view else
["tagFullHistoricalExport"])
return DirectIngestPreProcessedIngestView(
ingest_view_name=self.tag,
view_query_template=query,
region_raw_table_config=region_config,
order_by_cols="colA, colC",
is_detect_row_deletion_view=self.is_detect_row_deletion_view,
primary_key_tables_for_entity_deletion=
primary_key_tables_for_entity_deletion,
materialize_raw_data_table_views=self.
materialize_raw_data_table_views,
)
def test_read_completely_empty_file(self) -> None:
empty_file_path = fixtures.as_filepath("tagA.csv")
delegate = _TestGcsfsCsvReaderDelegate()
self.reader.streaming_read(
GcsfsFilePath.from_absolute_path(empty_file_path),
delegate=delegate,
chunk_size=1,
)
self.assertEqual(1, len(delegate.encodings_attempted))
self.assertEqual(delegate.encodings_attempted[0],
delegate.successful_encoding)
self.assertEqual(0, len(delegate.dataframes))
self.assertEqual(0, delegate.decode_errors)
self.assertEqual(0, delegate.exceptions)
delegate = _TestGcsfsCsvReaderDelegate()
self.reader.streaming_read(
GcsfsFilePath.from_absolute_path(empty_file_path),
delegate=delegate,
chunk_size=10,
)
self.assertEqual(1, len(delegate.encodings_attempted))
self.assertEqual(delegate.encodings_attempted[0],
delegate.successful_encoding)
self.assertEqual(0, len(delegate.dataframes))
self.assertEqual(0, delegate.decode_errors)
self.assertEqual(0, delegate.exceptions)
def test_direct_ingest_preprocessed_view_same_table_multiple_places(self):
region_config = DirectIngestRegionRawFileConfig(
region_code='us_xx',
yaml_config_file_path=fixtures.as_filepath(
'us_xx_raw_data_files.yaml'),
)
view_query_template = """SELECT * FROM {file_tag_first}
LEFT OUTER JOIN {file_tag_first}
USING (col1);"""
view = DirectIngestPreProcessedIngestView(
ingest_view_name='ingest_view_tag',
view_query_template=view_query_template,
region_raw_table_config=region_config)
self.assertEqual(
['file_tag_first'],
[c.file_tag for c in view.raw_table_dependency_configs])
expected_view_query = """WITH
file_tag_first_generated_view AS (
SELECT * FROM `recidiviz-456.us_xx_raw_data_up_to_date_views.file_tag_first_latest`
)
SELECT * FROM file_tag_first_generated_view
LEFT OUTER JOIN file_tag_first_generated_view
USING (col1);"""
self.assertEqual(expected_view_query, view.view_query)
def test_direct_ingest_preprocessed_view_detect_row_deletion_no_historical_table(
self, ) -> None:
region_config = DirectIngestRegionRawFileConfig(
region_code="us_xx",
yaml_config_file_dir=fixtures.as_filepath("us_xx"),
)
view_query_template = """SELECT * FROM {file_tag_first}
LEFT OUTER JOIN {file_tag_second}
USING (col1);"""
with self.assertRaises(ValueError) as e:
DirectIngestPreProcessedIngestView(
ingest_view_name="ingest_view_tag",
view_query_template=view_query_template,
region_raw_table_config=region_config,
order_by_cols="col1, col2",
is_detect_row_deletion_view=True,
primary_key_tables_for_entity_deletion=["file_tag_second"],
)
self.assertTrue(
str(e.exception).startswith(
"Ingest view ingest_view_tag is marked as `is_detect_row_deletion_view` and has table file_tag_second "
"specified in `primary_key_tables_for_entity_deletion`; however the raw data file is not marked as always "
"being exported as historically."))
def test_read_file_with_columns_no_contents(self):
empty_file_path = fixtures.as_filepath('tagB.csv')
delegate = TestGcsfsCsvReaderDelegate()
self.reader.streaming_read(
GcsfsFilePath.from_absolute_path(empty_file_path),
delegate=delegate,
chunk_size=1)
self.assertEqual(1, len(delegate.encodings_attempted))
self.assertEqual(delegate.encodings_attempted[0],
delegate.successful_encoding)
self.assertEqual(1, len(delegate.dataframes))
encoding, df = delegate.dataframes[0]
self.assertEqual(encoding, delegate.successful_encoding)
self.assertEqual(0, df.shape[0]) # No rows
self.assertEqual(7, df.shape[1]) # 7 columns
self.assertEqual(0, delegate.decode_errors)
self.assertEqual(0, delegate.exceptions)
delegate = TestGcsfsCsvReaderDelegate()
self.reader.streaming_read(
GcsfsFilePath.from_absolute_path(empty_file_path),
delegate=delegate,
chunk_size=10)
self.assertEqual(1, len(delegate.encodings_attempted))
self.assertEqual(delegate.encodings_attempted[0],
delegate.successful_encoding)
self.assertEqual(1, len(delegate.dataframes))
encoding, df = delegate.dataframes[0]
self.assertEqual(encoding, delegate.successful_encoding)
self.assertEqual(0, df.shape[0]) # No rows
self.assertEqual(7, df.shape[1]) # 7 columns
self.assertEqual(0, delegate.decode_errors)
self.assertEqual(0, delegate.exceptions)
def test_direct_ingest_preprocessed_view_detect_row_deletion_no_pk_tables_specified(
self, ) -> None:
region_config = DirectIngestRegionRawFileConfig(
region_code="us_xx",
yaml_config_file_dir=fixtures.as_filepath("us_xx"),
)
view_query_template = """SELECT * FROM {file_tag_first}
LEFT OUTER JOIN {tagFullHistoricalExport}
USING (col1);"""
with self.assertRaises(ValueError) as e:
DirectIngestPreProcessedIngestView(
ingest_view_name="ingest_view_tag",
view_query_template=view_query_template,
region_raw_table_config=region_config,
order_by_cols="col1, col2",
is_detect_row_deletion_view=True,
primary_key_tables_for_entity_deletion=[],
)
self.assertTrue(
str(e.exception).startswith(
"Ingest view ingest_view_tag was marked as `is_detect_row_deletion_view`; however no "
"`primary_key_tables_for_entity_deletion` were defined."))
def test_direct_ingest_preprocessed_view_other_materialized_subquery_fails(
self, ) -> None:
region_config = DirectIngestRegionRawFileConfig(
region_code="us_xx",
yaml_config_file_dir=fixtures.as_filepath("us_xx"),
)
view_query_template = """
CREATE TEMP TABLE my_subquery AS (SELECT * FROM {file_tag_first});
SELECT * FROM my_subquery;"""
with self.assertRaises(ValueError) as e:
_ = DirectIngestPreProcessedIngestView(
ingest_view_name="ingest_view_tag",
view_query_template=view_query_template,
region_raw_table_config=region_config,
order_by_cols="col1, col2",
is_detect_row_deletion_view=False,
primary_key_tables_for_entity_deletion=[],
)
self.assertEqual(
str(e.exception),
"Found CREATE TEMP TABLE clause in this query - ingest views cannot contain CREATE clauses.",
)
def _parsed_result() -> Dict[DeclarativeMeta, pd.DataFrame]:
global _PARSED_RESULT
if not _PARSED_RESULT:
_PARSED_RESULT = pa_aggregate_ingest.parse(
fixtures.as_filepath(
"2018 County Statistics _ General Information - 2017 Data.xlsx"
), )
return _PARSED_RESULT
def test_get_export_config_missing_state_code(self) -> None:
product_configs = ProductConfigs.from_file(
path=fixtures.as_filepath("fixture_products.yaml"))
with self.assertRaisesRegex(
BadProductExportSpecificationError,
"Missing required state_code parameter for export_job_name EXPORT",
):
product_configs.get_export_config(export_job_name="EXPORT", )
def _get_JID() -> pd.DataFrame:
global _JID
if _JID is None:
_JID = pd.read_csv(as_filepath("jid.csv", subdir="data_sets"),
dtype={"fips": str})
return _JID
def _get_FIPS() -> pd.DataFrame:
global _FIPS
if _FIPS is None:
_FIPS = pd.read_csv(as_filepath("fips.csv", subdir="data_sets"),
dtype={"fips": str})
return _FIPS
def _get_FID() -> pd.DataFrame:
global _FID
if _FID is None:
_FID = pd.read_csv(as_filepath("fid.csv", subdir="data_sets"),
dtype={"vera_jid": str})
return _FID
def test_parse_no_defaults_throws(self) -> None:
with self.assertRaises(ValueError) as e:
_ = DirectIngestRegionRawFileConfig(
region_code="us_yy",
yaml_config_file_dir=fixtures.as_filepath("us_yy"),
)
self.assertEqual(str(e.exception),
"Missing default raw data configs for region: us_yy")
def _parsed_result() -> Dict[DeclarativeMeta, pd.DataFrame]:
global _PARSED_RESULT
if not _PARSED_RESULT:
_PARSED_RESULT = ca_aggregate_ingest.parse(
fixtures.as_filepath("QueryResult.xls")
)
return _PARSED_RESULT
def test_direct_ingest_preprocessed_view_with_reference_table(self):
region_config = DirectIngestRegionRawFileConfig(
region_code='us_xx',
yaml_config_file_path=fixtures.as_filepath(
'us_xx_raw_data_files.yaml'),
)
view_query_template = """SELECT * FROM {file_tag_first}
LEFT OUTER JOIN `{{project_id}}.reference_tables.my_table`
USING (col1);"""
view = DirectIngestPreProcessedIngestView(
ingest_view_name='ingest_view_tag',
view_query_template=view_query_template,
region_raw_table_config=region_config,
order_by_cols='col1, col2')
self.assertEqual(
['file_tag_first'],
[c.file_tag for c in view.raw_table_dependency_configs])
expected_view_query = """WITH
file_tag_first_generated_view AS (
SELECT * FROM `recidiviz-456.us_xx_raw_data_up_to_date_views.file_tag_first_latest`
)
SELECT * FROM file_tag_first_generated_view
LEFT OUTER JOIN `recidiviz-456.reference_tables.my_table`
USING (col1)
ORDER BY col1, col2;"""
self.assertEqual(expected_view_query, view.view_query)
expected_date_parametrized_view_query = """WITH
file_tag_first_generated_view AS (
WITH rows_with_recency_rank AS (
SELECT
* EXCEPT (file_id, update_datetime),
ROW_NUMBER() OVER (PARTITION BY col_name_1a, col_name_1b
ORDER BY update_datetime DESC) AS recency_rank
FROM
`recidiviz-456.us_xx_raw_data.file_tag_first`
WHERE
update_datetime <= @my_param
)
SELECT *
EXCEPT (recency_rank)
FROM rows_with_recency_rank
WHERE recency_rank = 1
)
SELECT * FROM file_tag_first_generated_view
LEFT OUTER JOIN `recidiviz-456.reference_tables.my_table`
USING (col1)
ORDER BY col1, col2;"""
self.assertEqual(expected_date_parametrized_view_query,
view.date_parametrized_view_query('my_param'))
def test_missing_configs_for_region(self) -> None:
with self.assertRaises(ValueError) as e:
region_config = DirectIngestRegionRawFileConfig(
region_code="us_xy",
yaml_config_file_dir=fixtures.as_filepath("us_xy"),
)
_configs = region_config.raw_file_configs
self.assertEqual(str(e.exception),
"Missing raw data configs for region: us_xy")
def test_get_export_config_too_many_exports(self) -> None:
product_configs = ProductConfigs.from_file(
path=fixtures.as_filepath("fixture_products.yaml"))
product_configs.products.append(product_configs.products[0])
with self.assertRaisesRegex(
BadProductExportSpecificationError,
"Wrong number of products returned for export for export_job_name EXPORT",
):
product_configs.get_export_config(export_job_name="EXPORT", )
