def _create_body(self) -> dict:
body = {
DESCRIPTION: self.description,
STATUS: GcpTransferJobsStatus.ENABLED,
TRANSFER_SPEC: {
AWS_S3_DATA_SOURCE: {
BUCKET_NAME: self.s3_bucket,
PATH: normalize_directory_path(self.s3_path),
},
GCS_DATA_SINK: {
BUCKET_NAME: self.gcs_bucket,
PATH: normalize_directory_path(self.gcs_path),
},
},
}
if self.project_id is not None:
body[PROJECT_ID] = self.project_id
if self.schedule is not None:
body[SCHEDULE] = self.schedule
if self.object_conditions is not None:
body[TRANSFER_SPEC][OBJECT_CONDITIONS] = self.object_conditions # type: ignore[index]
if self.transfer_options is not None:
body[TRANSFER_SPEC][TRANSFER_OPTIONS] = self.transfer_options # type: ignore[index]
return body
def sync(
self,
source_bucket: str,
destination_bucket: str,
source_object: Optional[str] = None,
destination_object: Optional[str] = None,
recursive: bool = True,
allow_overwrite: bool = False,
delete_extra_files: bool = False,
) -> None:
"""
Synchronizes the contents of the buckets.
Parameters ``source_object`` and ``destination_object`` describe the root sync directories. If they
are not passed, the entire bucket will be synchronized. If they are passed, they should point
to directories.
.. note::
The synchronization of individual files is not supported. Only entire directories can be
synchronized.
:param source_bucket: The name of the bucket containing the source objects.
:type source_bucket: str
:param destination_bucket: The name of the bucket containing the destination objects.
:type destination_bucket: str
:param source_object: The root sync directory in the source bucket.
:type source_object: Optional[str]
:param destination_object: The root sync directory in the destination bucket.
:type destination_object: Optional[str]
:param recursive: If True, subdirectories will be considered
:type recursive: bool
:param recursive: If True, subdirectories will be considered
:type recursive: bool
:param allow_overwrite: if True, the files will be overwritten if a mismatched file is found.
By default, overwriting files is not allowed
:type allow_overwrite: bool
:param delete_extra_files: if True, deletes additional files from the source that not found in the
destination. By default extra files are not deleted.
.. note::
This option can delete data quickly if you specify the wrong source/destination combination.
:type delete_extra_files: bool
:return: none
"""
client = self.get_conn()
# Create bucket object
source_bucket_obj = client.bucket(source_bucket)
destination_bucket_obj = client.bucket(destination_bucket)
# Normalize parameters when they are passed
source_object = normalize_directory_path(source_object)
destination_object = normalize_directory_path(destination_object)
# Calculate the number of characters that remove from the name, because they contain information
# about the parent's path
source_object_prefix_len = len(source_object) if source_object else 0
# Prepare synchronization plan
to_copy_blobs, to_delete_blobs, to_rewrite_blobs = self._prepare_sync_plan(
source_bucket=source_bucket_obj,
destination_bucket=destination_bucket_obj,
source_object=source_object,
destination_object=destination_object,
recursive=recursive,
)
self.log.info(
"Planned synchronization. To delete blobs count: %s, to upload blobs count: %s, "
"to rewrite blobs count: %s",
len(to_delete_blobs),
len(to_copy_blobs),
len(to_rewrite_blobs),
)
# Copy missing object to new bucket
if not to_copy_blobs:
self.log.info("Skipped blobs copying.")
else:
for blob in to_copy_blobs:
dst_object = self._calculate_sync_destination_path(
blob, destination_object, source_object_prefix_len)
self.copy(
source_bucket=source_bucket_obj.name,
source_object=blob.name,
destination_bucket=destination_bucket_obj.name,
destination_object=dst_object,
)
self.log.info("Blobs copied.")
# Delete redundant files
if not to_delete_blobs:
self.log.info("Skipped blobs deleting.")
elif delete_extra_files:
# TODO: Add batch. I tried to do it, but the Google library is not stable at the moment.
for blob in to_delete_blobs:
self.delete(blob.bucket.name, blob.name)
self.log.info("Blobs deleted.")
# Overwrite files that are different
if not to_rewrite_blobs:
self.log.info("Skipped blobs overwriting.")
elif allow_overwrite:
for blob in to_rewrite_blobs:
dst_object = self._calculate_sync_destination_path(
blob, destination_object, source_object_prefix_len)
self.rewrite(
source_bucket=source_bucket_obj.name,
source_object=blob.name,
destination_bucket=destination_bucket_obj.name,
destination_object=dst_object,
)
self.log.info("Blobs rewritten.")
self.log.info("Synchronization finished.")
