def __init__(self,
region_name: str,
system_level: SystemLevel,
ingest_directory_path: Optional[str] = None,
storage_directory_path: Optional[str] = None,
max_delay_sec_between_files: Optional[int] = None):
super().__init__(region_name, system_level)
self.fs = GcsfsFactory.build()
self.max_delay_sec_between_files = max_delay_sec_between_files
if not ingest_directory_path:
ingest_directory_path = \
gcsfs_direct_ingest_directory_path_for_region(region_name,
system_level)
self.ingest_directory_path = \
GcsfsDirectoryPath.from_absolute_path(ingest_directory_path)
if not storage_directory_path:
storage_directory_path = \
gcsfs_direct_ingest_storage_directory_path_for_region(
region_name, system_level)
self.storage_directory_path = \
GcsfsDirectoryPath.from_absolute_path(storage_directory_path)
self.file_prioritizer = \
GcsfsDirectIngestJobPrioritizer(
self.fs,
self.ingest_directory_path,
self._get_file_tag_rank_list())
self.file_split_line_limit = self._FILE_SPLIT_LINE_LIMIT
def __init__(
self,
region_code: str,
dry_run: bool,
):
self.region_code = region_code
self.file_type = GcsfsDirectIngestFileType.UNSPECIFIED
self.dry_run = dry_run
self.project_id = 'recidiviz-123'
self.region_ingest_bucket_dir_path = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_directory_path_for_region(
region_code, SystemLevel.STATE, project_id=self.project_id))
self.region_storage_raw_dir_path = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_storage_directory_path_for_region(
region_code,
SystemLevel.STATE,
GcsfsDirectIngestFileType.RAW_DATA,
project_id=self.project_id))
self.log_output_path = os.path.join(
os.path.dirname(__file__),
f'move_prod_ingest_files_to_raw_start_bound_{self.region_code}_region_dry_run_{dry_run}_'
f'{datetime.datetime.now().isoformat()}.txt')
self.mutex = threading.Lock()
self.move_list: List[Tuple[str, str]] = []
self.move_progress: Optional[Bar] = None
def __init__(self,
region_name: str,
system_level: SystemLevel,
ingest_directory_path: Optional[str] = None,
storage_directory_path: Optional[str] = None,
max_delay_sec_between_files: Optional[int] = None):
super().__init__(region_name, system_level)
self.fs = GcsfsFactory.build()
self.max_delay_sec_between_files = max_delay_sec_between_files
if not ingest_directory_path:
ingest_directory_path = \
gcsfs_direct_ingest_directory_path_for_region(region_name,
system_level)
self.ingest_directory_path = \
GcsfsDirectoryPath.from_absolute_path(ingest_directory_path)
if not storage_directory_path:
storage_directory_path = \
gcsfs_direct_ingest_storage_directory_path_for_region(
region_name, system_level)
self.storage_directory_path = \
GcsfsDirectoryPath.from_absolute_path(storage_directory_path)
self.temp_output_directory_path = \
GcsfsDirectoryPath.from_absolute_path(gcsfs_direct_ingest_temporary_output_directory_path())
ingest_job_file_type_filter = \
GcsfsDirectIngestFileType.INGEST_VIEW \
if self.region.is_raw_vs_ingest_file_name_detection_enabled() else None
self.file_prioritizer = \
GcsfsDirectIngestJobPrioritizer(
self.fs,
self.ingest_directory_path,
self.get_file_tag_rank_list(),
ingest_job_file_type_filter)
self.ingest_file_split_line_limit = self._INGEST_FILE_SPLIT_LINE_LIMIT
self.file_metadata_manager = PostgresDirectIngestFileMetadataManager(
region_code=self.region.region_code)
self.raw_file_import_manager = DirectIngestRawFileImportManager(
region=self.region,
fs=self.fs,
ingest_directory_path=self.ingest_directory_path,
temp_output_directory_path=self.temp_output_directory_path,
big_query_client=BigQueryClientImpl())
self.ingest_view_export_manager = DirectIngestIngestViewExportManager(
region=self.region,
fs=self.fs,
ingest_directory_path=self.ingest_directory_path,
file_metadata_manager=self.file_metadata_manager,
big_query_client=BigQueryClientImpl(),
view_collector=DirectIngestPreProcessedIngestViewCollector(
self.region, self.get_file_tag_rank_list()))
def __init__(self, file_type: GcsfsDirectIngestFileType, region_code: str,
start_date_bound: Optional[str],
end_date_bound: Optional[str], dry_run: bool, project_id: str,
file_filter: Optional[str]):
self.file_type = file_type
self.region_code = region_code
self.start_date_bound = start_date_bound
self.end_date_bound = end_date_bound
self.dry_run = dry_run
self.file_filter = file_filter
self.project_id = project_id
self.region_storage_dir_path = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_storage_directory_path_for_region(
region_code,
SystemLevel.STATE,
self.file_type,
project_id=self.project_id))
self.log_output_path = os.path.join(
os.path.dirname(__file__),
f'move_storage_files_to_deprecated_start_bound_{self.region_code}_region_{self.start_date_bound}'
f'_end_bound_{self.end_date_bound}_dry_run_{dry_run}_{datetime.datetime.now().isoformat()}.txt'
)
self.mutex = threading.Lock()
self.move_list: List[Tuple[str, str]] = []
self.move_progress: Optional[Bar] = None
def create_export_manager(self, region):
metadata_manager = PostgresDirectIngestFileMetadataManager(region.region_code)
return DirectIngestIngestViewExportManager(
region=region,
fs=FakeDirectIngestGCSFileSystem(),
ingest_directory_path=GcsfsDirectoryPath.from_absolute_path('ingest_bucket'),
big_query_client=self.mock_client,
file_metadata_manager=metadata_manager,
view_collector=_ViewCollector(region, controller_file_tags=['ingest_view']))
def _copy_files_for_date(self, subdir_path_str: str):
dir_path = GcsfsDirectoryPath.from_absolute_path(subdir_path_str.rstrip('/'))
from_path = f'gs://{self.prod_region_storage_dir_path.bucket_name}/{dir_path.relative_path}*'
to_path = f'gs://{self.staging_region_storage_dir_path.bucket_name}/{dir_path.relative_path}'
if not self.dry_run:
gsutil_cp(from_path=from_path, to_path=to_path)
with self.mutex:
self.copy_list.append((from_path, to_path))
if self.copy_progress:
self.copy_progress.next()
def __init__(self, project_id: str, region: str,
file_type_to_move: GcsfsDirectIngestFileType,
destination_file_type: GcsfsDirectIngestFileType,
start_date_bound: Optional[str],
end_date_bound: Optional[str], dry_run: bool,
file_filter: Optional[str]):
self.project_id = project_id
self.region = region
self.file_type_to_move = file_type_to_move
self.destination_file_type = destination_file_type
if self.file_type_to_move != self.destination_file_type and \
self.file_type_to_move != GcsfsDirectIngestFileType.UNSPECIFIED:
raise ValueError(
'Args file_type_to_move and destination_file_type must match if type to move is UNSPECIFIED'
)
self.start_date_bound = start_date_bound
self.end_date_bound = end_date_bound
self.dry_run = dry_run
self.file_filter = file_filter
self.storage_bucket = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_storage_directory_path_for_region(
region, SystemLevel.STATE, project_id=self.project_id))
self.ingest_bucket = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_directory_path_for_region(
region, SystemLevel.STATE, project_id=self.project_id))
self.mutex = threading.Lock()
self.collect_progress: Optional[Bar] = None
self.move_progress: Optional[Bar] = None
self.moves_list: List[Tuple[str, str]] = []
self.log_output_path = os.path.join(
os.path.dirname(__file__),
f'move_result_{region}_{self.project_id}_start_bound_{self.start_date_bound}_end_bound_'
f'{self.end_date_bound}_dry_run_{self.dry_run}_{datetime.datetime.now().isoformat()}.txt'
)
def __init__(self, region_code: str, file_type: GcsfsDirectIngestFileType,
start_date_bound: Optional[str],
end_date_bound: Optional[str], dry_run: bool):
self.file_type = file_type
self.prod_region_storage_dir_path = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_storage_directory_path_for_region(
region_code, SystemLevel.STATE, project_id='recidiviz-123'))
self.staging_region_storage_dir_path = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_storage_directory_path_for_region(
region_code, SystemLevel.STATE,
project_id='recidiviz-staging'))
self.dry_run = dry_run
self.start_date_bound = start_date_bound
self.end_date_bound = end_date_bound
self.log_output_path = os.path.join(
os.path.dirname(__file__),
f'copy_prod_to_staging_result_{region_code}_start_bound_{self.start_date_bound}_end_bound_'
f'{self.end_date_bound}_dry_run_{dry_run}_{datetime.datetime.now().isoformat()}.txt'
)
self.mutex = threading.Lock()
self.copy_list: List[Tuple[str, str]] = []
self.copy_progress: Optional[Bar] = None
def __init__(self, paths: str, project_id: str, region: str, date: str,
dry_run: bool):
self.paths = paths
self.project_id = project_id
self.region = region.lower()
self.datetime = datetime.datetime.fromisoformat(date)
self.dry_run = dry_run
self.ingest_bucket = GcsfsDirectoryPath.from_absolute_path(
gcsfs_direct_ingest_directory_path_for_region(
region, SystemLevel.STATE, project_id=self.project_id))
self.mutex = threading.Lock()
self.move_progress: Optional[Bar] = None
self.copies_list: List[Tuple[str, str]] = []
self.log_output_path = os.path.join(
os.path.dirname(__file__),
f'upload_to_ingest_result_{region}_{self.project_id}_date_{self.datetime.date().isoformat()}'
f'_dry_run_{self.dry_run}_{datetime.datetime.now().isoformat()}.txt'
)
def _move_files_for_date(self, subdir_path_str: str):
"""Function that loops through each subdirectory and moves files in each subdirectory using the from path
and to path specified."""
from_dir_path = GcsfsDirectoryPath.from_absolute_path(
subdir_path_str.rstrip('/'))
previous_date_format = from_dir_path.relative_path.rstrip('/').split(
'/')[-1]
new_date_format = date.fromisoformat(previous_date_format).strftime(
"%Y/%m/%d/")
from_paths = gsutil_ls(f'{subdir_path_str}*.csv')
for from_path in from_paths:
file_name = GcsfsFilePath(
bucket_name=self.region_storage_dir_path.bucket_name,
blob_name=from_path).file_name
to_file_path = os.path.join(
'gs://', self.region_storage_dir_path.bucket_name,
self.region_code, GcsfsDirectIngestFileType.RAW_DATA.value,
new_date_format, file_name)
normalized_to_file_path = to_normalized_processed_file_path_from_normalized_path(
to_file_path,
file_type_override=GcsfsDirectIngestFileType.RAW_DATA)
to_path = normalized_to_file_path
if not self.dry_run:
gsutil_mv(from_path=from_path, to_path=to_path)
with self.mutex:
self.move_list.append((from_path, to_path))
if self.move_progress:
self.move_progress.next()
class TestGcsfsDirectIngestJobPrioritizer(unittest.TestCase):
"""Tests for the GcsfsDirectIngestJobPrioritizer."""
_DAY_1_TIME_1 = datetime.datetime(year=2019,
month=1,
day=2,
hour=3,
minute=4,
second=5,
microsecond=6789,
tzinfo=datetime.timezone.utc)
_DAY_1_TIME_2 = datetime.datetime(year=2019,
month=1,
day=2,
hour=3,
minute=4,
second=5,
microsecond=7789,
tzinfo=datetime.timezone.utc)
_DAY_1_TIME_3 = datetime.datetime(year=2019,
month=1,
day=2,
hour=10,
minute=4,
second=5,
microsecond=678,
tzinfo=datetime.timezone.utc)
_DAY_2_TIME_1 = datetime.datetime(year=2019,
month=1,
day=3,
hour=3,
minute=4,
second=5,
microsecond=6789,
tzinfo=datetime.timezone.utc)
_DAY_1 = _DAY_1_TIME_1.date()
_DAY_2 = _DAY_2_TIME_1.date()
_INGEST_BUCKET_PATH = \
GcsfsDirectoryPath.from_absolute_path('direct/regions/us_nd/fixtures')
def setUp(self) -> None:
self.fs = FakeDirectIngestGCSFileSystem()
self.prioritizer = GcsfsDirectIngestJobPrioritizer(
self.fs, self._INGEST_BUCKET_PATH, ['tagA', 'tagB'])
FIXTURE_PATH_PREFIX = 'direct/regions/us_nd/fixtures'
def _normalized_path_for_filename(self, filename: str,
dt: datetime.datetime) -> GcsfsFilePath:
normalized_path = \
to_normalized_unprocessed_file_path(
os.path.join(self._INGEST_BUCKET_PATH.abs_path(),
filename), dt)
return GcsfsFilePath.from_absolute_path(normalized_path)
def _process_jobs_for_paths_with_no_gaps_in_expected_order(
self, paths: List[GcsfsFilePath]):
for path in paths:
date_str = filename_parts_from_path(path).date_str
next_job_args = self.prioritizer.get_next_job_args()
self.assertIsNotNone(next_job_args)
if next_job_args is None:
# Make mypy happy
self.fail()
self.assertEqual(next_job_args.file_path, path)
self.assertTrue(
self.prioritizer.are_next_args_expected(next_job_args))
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(date_str))
# ... job runs ...
self.fs.mv_path_to_processed_path(path)
def test_empty_fs(self):
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1_TIME_1.date().isoformat()))
self.assertIsNone(self.prioritizer.get_next_job_args())
def test_single_expected_file(self):
path = self._normalized_path_for_filename('tagA.csv',
self._DAY_1_TIME_1)
self.fs.test_add_path(path)
self._process_jobs_for_paths_with_no_gaps_in_expected_order([path])
self.assertIsNone(self.prioritizer.get_next_job_args())
# We still expect a file for tagB
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
def test_multiple_files(self):
paths = [
self._normalized_path_for_filename('tagA.csv', self._DAY_1_TIME_1),
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_2)
]
for path in paths:
self.fs.test_add_path(path)
self._process_jobs_for_paths_with_no_gaps_in_expected_order(paths)
self.assertIsNone(self.prioritizer.get_next_job_args())
self.assertFalse(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
def test_unexpected_file(self):
# Only file is out of order
path = self._normalized_path_for_filename('tagB.csv',
self._DAY_1_TIME_1)
self.fs.test_add_path(path)
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
next_job_args = self.prioritizer.get_next_job_args()
self.assertIsNotNone(next_job_args)
self.assertEqual(next_job_args.file_path, path)
self.assertFalse(
self.prioritizer.are_next_args_expected(next_job_args))
# ... job runs eventually even though unexpected...
self.fs.mv_path_to_processed_path(path)
self.assertIsNone(self.prioritizer.get_next_job_args())
# We still expect a file for tagA
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
def test_files_on_multiple_days(self):
paths = [
self._normalized_path_for_filename('tagA.csv', self._DAY_1_TIME_1),
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_2),
self._normalized_path_for_filename('tagA.csv', self._DAY_2_TIME_1),
]
for path in paths:
self.fs.test_add_path(path)
self._process_jobs_for_paths_with_no_gaps_in_expected_order(paths)
self.assertIsNone(self.prioritizer.get_next_job_args())
self.assertFalse(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_2.isoformat()))
def test_files_on_multiple_days_with_gap(self):
"""Runs a test where there are files on multiple days and there is a gap
in the expected files for the first day.
"""
paths = [
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_2),
self._normalized_path_for_filename('tagA.csv', self._DAY_2_TIME_1),
]
for path in paths:
self.fs.test_add_path(path)
for i, path in enumerate(paths):
date_str = filename_parts_from_path(path).date_str
next_job_args = self.prioritizer.get_next_job_args()
self.assertIsNotNone(next_job_args)
self.assertEqual(next_job_args.file_path, path)
are_args_expected = \
self.prioritizer.are_next_args_expected(next_job_args)
if i == 0:
self.assertFalse(are_args_expected)
else:
self.assertTrue(are_args_expected)
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(date_str))
# ... job runs ...
self.fs.mv_path_to_processed_path(path)
self.assertIsNone(self.prioritizer.get_next_job_args())
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
self.assertTrue(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_2.isoformat()))
def test_multiple_files_same_tag(self):
paths = [
self._normalized_path_for_filename('tagA.csv', self._DAY_1_TIME_1),
self._normalized_path_for_filename('tagA.csv', self._DAY_1_TIME_2),
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_3),
]
for path in paths:
self.fs.test_add_path(path)
self._process_jobs_for_paths_with_no_gaps_in_expected_order(paths)
self.assertIsNone(self.prioritizer.get_next_job_args())
self.assertFalse(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
def test_multiple_files_times_out_of_order(self):
"""Runs a test where there are no gaps but the files have been added
(i.e. have creation times) out of order.
"""
paths = [
self._normalized_path_for_filename('tagA.csv', self._DAY_1_TIME_2),
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_1),
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_3),
]
for path in paths:
self.fs.test_add_path(path)
for i, path in enumerate(paths):
date_str = filename_parts_from_path(path).date_str
next_job_args = self.prioritizer.get_next_job_args()
self.assertIsNotNone(next_job_args)
self.assertEqual(next_job_args.file_path, path)
self.assertTrue(
self.prioritizer.are_next_args_expected(next_job_args))
are_more_jobs_expected = \
self.prioritizer.are_more_jobs_expected_for_day(date_str)
if i == 2:
self.assertFalse(are_more_jobs_expected)
else:
self.assertTrue(are_more_jobs_expected)
# ... job runs ...
self.fs.mv_path_to_processed_path(path)
self.assertIsNone(self.prioritizer.get_next_job_args())
self.assertFalse(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
def test_run_multiple_copies_of_same_tag(self):
paths = [
self._normalized_path_for_filename('tagA.csv', self._DAY_1_TIME_2),
self._normalized_path_for_filename('tagA_2.csv',
self._DAY_1_TIME_1),
self._normalized_path_for_filename('tagB.csv', self._DAY_1_TIME_3),
]
for path in paths:
self.fs.test_add_path(path)
self._process_jobs_for_paths_with_no_gaps_in_expected_order(paths)
self.assertIsNone(self.prioritizer.get_next_job_args())
self.assertFalse(
self.prioritizer.are_more_jobs_expected_for_day(
self._DAY_1.isoformat()))
