def slice_catalog(catalog: ConfiguredAirbyteCatalog,
streams: Set[str]) -> ConfiguredAirbyteCatalog:
sliced_catalog = ConfiguredAirbyteCatalog(streams=[])
for stream in catalog.streams:
if stream.stream.name in streams:
sliced_catalog.streams.append(stream)
return sliced_catalog
def test_streams_outputs_records(self, catalog_path, stream_config):
configured_catalog = ConfiguredAirbyteCatalog.parse_file(catalog_path)
records, states = self._read_records(stream_config, configured_catalog)
assert records, "should have some records returned"
if configured_catalog.streams[0].sync_mode == SyncMode.incremental:
assert states, "should have some states returned"
def configured_catalog() -> ConfiguredAirbyteCatalog:
catalog_filename = HERE.parent / "sample_files" / "configured_catalog.json"
if not catalog_filename.exists():
raise RuntimeError(
f"Please provide configured catalog in {catalog_filename}")
return ConfiguredAirbyteCatalog.parse_file(catalog_filename)
def test_parse_sheet_and_column_names_from_catalog(self):
sheet1 = "soccer_team"
sheet1_columns = frozenset(["arsenal", "chelsea", "manutd", "liverpool"])
sheet1_schema = {"properties": {c: {"type": "string"} for c in sheet1_columns}}
sheet2 = "basketball_teams"
sheet2_columns = frozenset(["gsw", "lakers"])
sheet2_schema = {"properties": {c: {"type": "string"} for c in sheet2_columns}}
catalog = ConfiguredAirbyteCatalog(
streams=[
ConfiguredAirbyteStream(
stream=AirbyteStream(name=sheet1, json_schema=sheet1_schema),
sync_mode=SyncMode.full_refresh,
destination_sync_mode=DestinationSyncMode.overwrite,
),
ConfiguredAirbyteStream(
stream=AirbyteStream(name=sheet2, json_schema=sheet2_schema),
sync_mode=SyncMode.full_refresh,
destination_sync_mode=DestinationSyncMode.overwrite,
),
]
)
actual = Helpers.parse_sheet_and_column_names_from_catalog(catalog)
expected = {sheet1: sheet1_columns, sheet2: sheet2_columns}
self.assertEqual(actual, expected)
def read(self,
logger: AirbyteLogger,
config_container: ConfigContainer,
catalog_path,
state_path: str = None) -> Generator[AirbyteMessage, None, None]:
client = self._client(config_container)
if state_path:
logger.info("Starting sync with provided state file")
state_obj = json.loads(open(state_path, "r").read())
else:
logger.info("No state provided, starting fresh sync")
state_obj = {}
state = defaultdict(dict, state_obj)
catalog = ConfiguredAirbyteCatalog.parse_obj(
self.read_config(catalog_path))
logger.info("Starting syncing mailchimp")
for configured_stream in catalog.streams:
stream = configured_stream.stream
for record in self._read_record(client=client,
stream=stream.name,
state=state):
yield record
logger.info("Finished syncing mailchimp")
def configured_catalog_from_client(
client: BaseClient) -> ConfiguredAirbyteCatalog:
"""Helper to generate configured catalog for testing"""
catalog = ConfiguredAirbyteCatalog(streams=[
ConfiguredAirbyteStream(stream=stream) for stream in client.streams
])
return catalog
def read(self,
logger,
config_container,
catalog_path,
state_path=None) -> Generator[AirbyteMessage, None, None]:
"""
:param logger:
:param config_container:
:param catalog_path:
:param state_path:
:return:
"""
config = config_container.rendered_config
storage = SourceFile.get_storage_scheme(logger,
config["provider"]["storage"],
config["url"])
url = SourceFile.get_simple_url(config["url"])
name = SourceFile.get_stream_name(config)
logger.info(
f"Reading {name} ({storage}{url}, {catalog_path}, {state_path})..."
)
catalog = ConfiguredAirbyteCatalog.parse_obj(
self.read_config(catalog_path))
selection = SourceFile.parse_catalog(catalog)
try:
if "format" in config and config["format"] == "json":
data_list = SourceFile.load_nested_json(config, logger)
for data in data_list:
yield AirbyteMessage(
type=Type.RECORD,
record=AirbyteRecordMessage(
stream=name,
data=data,
emitted_at=int(datetime.now().timestamp()) * 1000),
)
else:
df_list = SourceFile.load_dataframes(config, logger)
for df in df_list:
if len(selection) > 0:
columns = selection.intersection(set(df.columns))
else:
columns = df.columns
df = df.replace(np.nan, "NaN", regex=True)
for data in df[columns].to_dict(orient="records"):
yield AirbyteMessage(
type=Type.RECORD,
record=AirbyteRecordMessage(
stream=name,
data=data,
emitted_at=int(datetime.now().timestamp()) *
1000),
)
except Exception as err:
reason = f"Failed to read data of {name} at {storage}{url}: {repr(err)}\n{traceback.format_exc()}"
logger.error(reason)
raise err
def incremental_only_catalog(configured_catalog: ConfiguredAirbyteCatalog) -> ConfiguredAirbyteCatalog:
"""Transform provided catalog to catalog with all streams configured to use Incremental sync (when possible)"""
streams = []
for stream in configured_catalog.streams:
if SyncMode.incremental in stream.stream.supported_sync_modes:
stream.sync_mode = SyncMode.incremental
streams.append(stream)
configured_catalog.streams = streams
return configured_catalog
def full_refresh_only_catalog(configured_catalog: ConfiguredAirbyteCatalog) -> ConfiguredAirbyteCatalog:
"""Transform provided catalog to catalog with all streams configured to use Full Refresh sync (when possible)"""
streams = []
for stream in configured_catalog.streams:
if SyncMode.full_refresh in stream.stream.supported_sync_modes:
stream.sync_mode = SyncMode.full_refresh
streams.append(stream)
configured_catalog.streams = streams
return configured_catalog
def read(
self, logger: AirbyteLogger, config_container: ConfigContainer, catalog_path: str, state_path: str = None
) -> Generator[AirbyteMessage, None, None]:
"""
Implements the parent class read method.
"""
catalogs = self._discover_internal(logger, config_container.config_path)
masked_airbyte_catalog = ConfiguredAirbyteCatalog.parse_obj(self.read_config(catalog_path))
selected_singer_catalog_path = SingerHelper.create_singer_catalog_with_selection(masked_airbyte_catalog, catalogs.singer_catalog)
read_cmd = self.read_cmd(logger, config_container.config_path, selected_singer_catalog_path, state_path)
return SingerHelper.read(logger, read_cmd)
def read(self,
logger: AirbyteLogger,
config_container: ConfigContainer,
catalog_path,
state=None) -> Generator[AirbyteMessage, None, None]:
client = self._get_client(config_container)
config = self.read_config(catalog_path)
catalog = ConfiguredAirbyteCatalog.parse_obj(config)
logger.info(f"Starting syncing {self.__class__.__name__}")
for configured_stream in catalog.streams:
for record in client.read_stream(configured_stream.stream):
yield AirbyteMessage(type=airbyte_protocol.Type.RECORD,
record=record)
logger.info(f"Finished syncing {self.__class__.__name__}")
def read(self,
logger: AirbyteLogger,
config_container: ConfigContainer,
catalog_path,
state=None) -> Generator[AirbyteMessage, None, None]:
client = self._client(config_container)
catalog = ConfiguredAirbyteCatalog.parse_obj(
self.read_config(catalog_path))
logger.info("Starting syncing mailchimp")
for configured_stream in catalog.streams:
stream = configured_stream.stream
for record in self._read_record(client=client, stream=stream.name):
yield AirbyteMessage(type=Type.RECORD, record=record)
logger.info("Finished syncing mailchimp")
def read(self, logger: AirbyteLogger, config_container, catalog_path, state=None) -> Generator[AirbyteMessage, None, None]:
config = config_container.rendered_config
client = Helpers.get_authenticated_sheets_client(json.loads(config["credentials_json"]))
catalog = ConfiguredAirbyteCatalog.parse_obj(self.read_config(catalog_path))
sheet_to_column_name = Helpers.parse_sheet_and_column_names_from_catalog(catalog)
spreadsheet_id = config["spreadsheet_id"]
logger.info(f"Starting syncing spreadsheet {spreadsheet_id}")
# For each sheet in the spreadsheet, get a batch of rows, and as long as there hasn't been
# a blank row, emit the row batch
sheet_to_column_index_to_name = Helpers.get_available_sheets_to_column_index_to_name(client, spreadsheet_id, sheet_to_column_name)
for sheet in sheet_to_column_index_to_name.keys():
logger.info(f"Syncing sheet {sheet}")
column_index_to_name = sheet_to_column_index_to_name[sheet]
row_cursor = 2 # we start syncing past the header row
encountered_blank_row = False
while not encountered_blank_row:
range = f"{sheet}!{row_cursor}:{row_cursor + ROW_BATCH_SIZE}"
logger.info(f"Fetching range {range}")
row_batch = SpreadsheetValues.parse_obj(
client.values().batchGet(spreadsheetId=spreadsheet_id, ranges=range, majorDimension="ROWS").execute()
)
row_cursor += ROW_BATCH_SIZE + 1
# there should always be one range since we requested only one
value_ranges = row_batch.valueRanges[0]
if not value_ranges.values:
break
row_values = value_ranges.values
if len(row_values) == 0:
break
for row in row_values:
if Helpers.is_row_empty(row):
encountered_blank_row = True
break
elif Helpers.row_contains_relevant_data(row, column_index_to_name.keys()):
yield AirbyteMessage(type=Type.RECORD, record=Helpers.row_data_to_record_message(sheet, row, column_index_to_name))
logger.info(f"Finished syncing spreadsheet {spreadsheet_id}")
def read(self,
logger: AirbyteLogger,
config_container: ConfigContainer,
catalog_path,
state=None) -> Generator[AirbyteMessage, None, None]:
client = self._client(config_container)
config = self.read_config(catalog_path)
catalog = ConfiguredAirbyteCatalog.parse_obj(config)
logger.info("Starting syncing recurly")
for configured_stream in catalog.streams:
# TODO handle incremental syncs
stream = configured_stream.stream
if stream.name not in client.ENTITIES:
logger.warn(
f"Stream '{stream}' not found in the recognized entities")
continue
for record in self._read_record(client=client, stream=stream.name):
yield AirbyteMessage(type=Type.RECORD, record=record)
logger.info("Finished syncing recurly")
def read_catalog(self, catalog_path: str) -> ConfiguredAirbyteCatalog:
catalog = ConfiguredAirbyteCatalog.parse_obj(
self.read_config(catalog_path))
if not self.reports_to_read:
self.reports_to_read = [i.stream.name for i in catalog.streams]
return catalog_path
def get_catalog(self) -> ConfiguredAirbyteCatalog:
raw_catalog = pkgutil.get_data(
self.__class__.__module__.split(".")[0],
self.CONFIGURED_CATALOG_FILENAME)
return ConfiguredAirbyteCatalog.parse_obj(json.loads(raw_catalog))
def get_catalog(self) -> ConfiguredAirbyteCatalog:
return ConfiguredAirbyteCatalog.parse_raw(pkgutil.get_data(self.__class__.__module__.split(".")[0], "catalog_subset.json"))
def get_catalog(self) -> ConfiguredAirbyteCatalog:
raw_catalog = pkgutil.get_data(
self.__class__.__module__.split(".")[0], "configured_catalog.json")
return ConfiguredAirbyteCatalog.parse_obj(json.loads(raw_catalog))
def configured_catalog_fixture(
configured_catalog_path) -> Optional[ConfiguredAirbyteCatalog]:
if configured_catalog_path:
return ConfiguredAirbyteCatalog.parse_file(configured_catalog_path)
return None
def read_catalog(self, catalog_path: str) -> ConfiguredAirbyteCatalog:
return ConfiguredAirbyteCatalog.parse_obj(
self.read_config(catalog_path))
def configured_catalog_fixture():
return ConfiguredAirbyteCatalog.parse_file(
"sample_files/configured_catalog.json")
def configured_catalog_fixture():
return ConfiguredAirbyteCatalog.parse_file(
BASE_DIRECTORY /
"sample_files/configured_catalog_activities_overview.json")
def _read_catalog(path):
return ConfiguredAirbyteCatalog.parse_raw(open(path, "r").read())
