def copy_records_to_db(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, DatabaseStorageApi)
mdr = from_storage_api.get(from_name)
to_storage_api.bulk_insert_records(to_name, mdr.records_object, schema)
def copy_file_object_iterator_to_records_iterator(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
itr = (read_csv(chunk) for chunk in with_header(mdr.records_object))
to_mdr = as_records(itr, data_format=RecordsIteratorFormat, schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)
def copy_df_to_records(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
df = dataframe_to_records(mdr.records_object, schema)
to_mdr = as_records(df, data_format=RecordsFormat, schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)
def copy_file_object_to_records(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
obj = read_csv(mdr.records_object)
to_mdr = as_records(obj, data_format=RecordsFormat, schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)
def copy_df_iterator_to_records_iterator(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
itr = (dataframe_to_records(df, schema) for df in mdr.records_object)
to_mdr = as_records(itr, data_format=RecordsIteratorFormat, schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)
def copy_file_object_to_delim_file(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, FileSystemStorageApi)
mdr = from_storage_api.get(from_name)
file_obj_iterator = mdr.records_object
if isinstance(mdr.records_object, IOBase):
file_obj_iterator = [file_obj_iterator]
with to_storage_api.open(to_name, "w") as to_file:
for file_obj in file_obj_iterator:
to_file.write(file_obj)
def copy_records_iterator_to_records(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
all_records = []
for records in mdr.records_object:
all_records.extend(records)
to_mdr = as_records(all_records, data_format=RecordsFormat, schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)
def copy_file_object_to_records_iterator(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
# Note: must keep header on each chunk when iterating delimited file object!
# TODO: ugly hard-coded 1000 here, but how could we ever make it configurable? Not a big deal I guess
itr = (read_csv(chunk)
for chunk in with_header(iterate_chunks(mdr.records_object, 1000)))
to_mdr = as_records(itr, data_format=RecordsIteratorFormat, schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)
def copy_records_to_delim_file(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, FileSystemStorageApi)
mdr = from_storage_api.get(from_name)
records_iterator = mdr.records_object
if not isinstance(mdr.records_object, Iterator):
records_iterator = [records_iterator]
with to_storage_api.open(to_name, "w") as f:
append = False
for records in records_iterator:
write_csv(records, f, append=append)
append = True
def copy_dataframe_iterator_to_dataframe(
from_name: str,
to_name: str,
conversion: Conversion,
from_storage_api: StorageApi,
to_storage_api: StorageApi,
schema: Schema,
):
assert isinstance(from_storage_api, PythonStorageApi)
assert isinstance(to_storage_api, PythonStorageApi)
mdr = from_storage_api.get(from_name)
all_dfs = []
for df in mdr.records_object:
all_dfs.append(df)
to_mdr = as_records(pd.concat(all_dfs),
data_format=DataFrameFormat,
schema=schema)
to_mdr = to_mdr.conform_to_schema()
to_storage_api.put(to_name, to_mdr)