def call(self, files):
"""
This is the generalized function to yield the ``FileInfo`` objects.
``dir_op`` and ``use_src_name`` flags affect which files are used and
ensure the proper destination paths and compare keys are formed.
"""
function_table = {'s3': self.list_objects, 'local': self.list_files}
source = files['src']['path']
src_type = files['src']['type']
dest_type = files['dest']['type']
file_list = function_table[src_type](source, files['dir_op'])
for src_path, size, last_update in file_list:
dest_path, compare_key = find_dest_path_comp_key(files, src_path)
yield FileStat(src=src_path, dest=dest_path,
compare_key=compare_key, size=size,
last_update=last_update, src_type=src_type,
dest_type=dest_type,
operation_name=self.operation_name)
def call(self, files):
"""
This is the generalized function to yield the ``FileInfo`` objects.
``dir_op`` and ``use_src_name`` flags affect which files are used and
ensure the proper destination paths and compare keys are formed.
"""
function_table = {'s3': self.list_objects, 'local': self.list_files}
source = files['src']['path']
src_type = files['src']['type']
dest_type = files['dest']['type']
file_iterator = function_table[src_type](source, files['dir_op'])
for src_path, extra_information in file_iterator:
dest_path, compare_key = find_dest_path_comp_key(files, src_path)
file_stat_kwargs = {
'src': src_path, 'dest': dest_path, 'compare_key': compare_key,
'src_type': src_type, 'dest_type': dest_type,
'operation_name': self.operation_name
}
self._inject_extra_information(file_stat_kwargs, extra_information)
yield FileStat(**file_stat_kwargs)
def call(self, files):
"""
This is the generalized function to yield the ``FileInfo`` objects.
``dir_op`` and ``use_src_name`` flags affect which files are used and
ensure the proper destination paths and compare keys are formed.
"""
function_table = {'s3': self.list_objects, 'local': self.list_files}
source = files['src']['path']
src_type = files['src']['type']
dest_type = files['dest']['type']
file_iterator = function_table[src_type](source, files['dir_op'])
for src_path, extra_information in file_iterator:
dest_path, compare_key = find_dest_path_comp_key(files, src_path)
file_stat_kwargs = {
'src': src_path, 'dest': dest_path, 'compare_key': compare_key,
'src_type': src_type, 'dest_type': dest_type,
'operation_name': self.operation_name
}
self._inject_extra_information(file_stat_kwargs, extra_information)
yield FileStat(**file_stat_kwargs)
def call(self, files):
"""
This is the generalized function to yield the ``FileInfo`` objects.
``dir_op`` and ``use_src_name`` flags affect which files are used and
ensure the proper destination paths and compare keys are formed.
"""
function_table = {'s3': self.list_objects, 'local': self.list_files}
source = files['src']['path']
src_type = files['src']['type']
dest_type = files['dest']['type']
file_list = function_table[src_type](source, files['dir_op'])
for src_path, size, last_update in file_list:
dest_path, compare_key = find_dest_path_comp_key(files, src_path)
yield FileStat(src=src_path,
dest=dest_path,
compare_key=compare_key,
size=size,
last_update=last_update,
src_type=src_type,
dest_type=dest_type,
operation_name=self.operation_name)
def call(self, files):
"""
This is the generalized function to yield the ``FileInfo`` objects.
``dir_op`` and ``use_src_name`` flags affect which files are used and
ensure the proper destination paths and compare keys are formed.
"""
function_table = {"s3": self.list_objects, "local": self.list_files}
source = files["src"]["path"]
src_type = files["src"]["type"]
dest_type = files["dest"]["type"]
file_iterator = function_table[src_type](source, files["dir_op"])
for src_path, extra_information in file_iterator:
dest_path, compare_key = find_dest_path_comp_key(files, src_path)
file_stat_kwargs = {
"src": src_path,
"dest": dest_path,
"compare_key": compare_key,
"src_type": src_type,
"dest_type": dest_type,
"operation_name": self.operation_name,
}
self._inject_extra_information(file_stat_kwargs, extra_information)
yield FileStat(**file_stat_kwargs)
def run(self):
"""
This function wires together all of the generators and completes
the command. First a dictionary is created that is indexed first by
the command name. Then using the instruction, another dictionary
can be indexed to obtain the objects corresponding to the
particular instruction for that command. To begin the wiring,
either a ``FileFormat`` or ``TaskInfo`` object, depending on the
command, is put into a list. Then the function enters a while loop
that pops off an instruction. It then determines the object needed
and calls the call function of the object using the list as the input.
Depending on the number of objects in the input list and the number
of components in the list corresponding to the instruction, the call
method of the component can be called two different ways. If the
number of inputs is equal to the number of components a 1:1 mapping of
inputs to components is used when calling the call function. If the
there are more inputs than components, then a 2:1 mapping of inputs to
components is used where the component call method takes two inputs
instead of one. Whatever files are yielded from the call function
is appended to a list and used as the input for the next repetition
of the while loop until there are no more instructions.
"""
src = self.parameters['src']
dest = self.parameters['dest']
paths_type = self.parameters['paths_type']
files = FileFormat().format(src, dest, self.parameters)
rev_files = FileFormat().format(dest, src, self.parameters)
cmd_translation = {}
cmd_translation['locals3'] = {'cp': 'upload', 'sync': 'upload',
'mv': 'move'}
cmd_translation['s3s3'] = {'cp': 'copy', 'sync': 'copy', 'mv': 'move'}
cmd_translation['s3local'] = {'cp': 'download', 'sync': 'download',
'mv': 'move'}
cmd_translation['s3'] = {
'rm': 'delete',
'mb': 'make_bucket',
'rb': 'remove_bucket'
}
result_queue = queue.Queue()
operation_name = cmd_translation[paths_type][self.cmd]
file_generator = FileGenerator(self._source_client,
operation_name,
self.parameters['follow_symlinks'],
self.parameters['page_size'],
result_queue=result_queue)
rev_generator = FileGenerator(self._client, '',
self.parameters['follow_symlinks'],
self.parameters['page_size'],
result_queue=result_queue)
taskinfo = [TaskInfo(src=files['src']['path'],
src_type='s3',
operation_name=operation_name,
client=self._client)]
stream_dest_path, stream_compare_key = find_dest_path_comp_key(files)
stream_file_info = [FileInfo(src=files['src']['path'],
dest=stream_dest_path,
compare_key=stream_compare_key,
src_type=files['src']['type'],
dest_type=files['dest']['type'],
operation_name=operation_name,
client=self._client,
is_stream=True)]
file_info_builder = FileInfoBuilder(
self._client, self._source_client, self.parameters)
s3handler = S3Handler(self.session, self.parameters,
runtime_config=self._runtime_config,
result_queue=result_queue)
s3_stream_handler = S3StreamHandler(self.session, self.parameters,
result_queue=result_queue)
sync_strategies = self.choose_sync_strategies()
command_dict = {}
if self.cmd == 'sync':
command_dict = {'setup': [files, rev_files],
'file_generator': [file_generator,
rev_generator],
'filters': [create_filter(self.parameters),
create_filter(self.parameters)],
'comparator': [Comparator(**sync_strategies)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]}
elif self.cmd == 'cp' and self.parameters['is_stream']:
command_dict = {'setup': [stream_file_info],
's3_handler': [s3_stream_handler]}
elif self.cmd == 'cp':
command_dict = {'setup': [files],
'file_generator': [file_generator],
'filters': [create_filter(self.parameters)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]}
elif self.cmd == 'rm':
command_dict = {'setup': [files],
'file_generator': [file_generator],
'filters': [create_filter(self.parameters)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]}
elif self.cmd == 'mv':
command_dict = {'setup': [files],
'file_generator': [file_generator],
'filters': [create_filter(self.parameters)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]}
elif self.cmd == 'mb':
command_dict = {'setup': [taskinfo],
's3_handler': [s3handler]}
elif self.cmd == 'rb':
command_dict = {'setup': [taskinfo],
's3_handler': [s3handler]}
files = command_dict['setup']
while self.instructions:
instruction = self.instructions.pop(0)
file_list = []
components = command_dict[instruction]
for i in range(len(components)):
if len(files) > len(components):
file_list.append(components[i].call(*files))
else:
file_list.append(components[i].call(files[i]))
files = file_list
# This is kinda quirky, but each call through the instructions
# will replaces the files attr with the return value of the
# file_list. The very last call is a single list of
# [s3_handler], and the s3_handler returns the number of
# tasks failed and the number of tasks warned.
# This means that files[0] now contains a namedtuple with
# the number of failed tasks and the number of warned tasks.
# In terms of the RC, we're keeping it simple and saying
# that > 0 failed tasks will give a 1 RC and > 0 warned
# tasks will give a 2 RC. Otherwise a RC of zero is returned.
rc = 0
if files[0].num_tasks_failed > 0:
rc = 1
if files[0].num_tasks_warned > 0:
rc = 2
return rc
def run(self):
"""
This function wires together all of the generators and completes
the command. First a dictionary is created that is indexed first by
the command name. Then using the instruction, another dictionary
can be indexed to obtain the objects corresponding to the
particular instruction for that command. To begin the wiring,
either a ``FileFormat`` or ``TaskInfo`` object, depending on the
command, is put into a list. Then the function enters a while loop
that pops off an instruction. It then determines the object needed
and calls the call function of the object using the list as the input.
Depending on the number of objects in the input list and the number
of components in the list corresponding to the instruction, the call
method of the component can be called two different ways. If the
number of inputs is equal to the number of components a 1:1 mapping of
inputs to components is used when calling the call function. If the
there are more inputs than components, then a 2:1 mapping of inputs to
components is used where the component call method takes two inputs
instead of one. Whatever files are yielded from the call function
is appended to a list and used as the input for the next repetition
of the while loop until there are no more instructions.
"""
src = self.parameters['src']
dest = self.parameters['dest']
paths_type = self.parameters['paths_type']
files = FileFormat().format(src, dest, self.parameters)
rev_files = FileFormat().format(dest, src, self.parameters)
cmd_translation = {}
cmd_translation['locals3'] = {
'cp': 'upload',
'sync': 'upload',
'mv': 'move'
}
cmd_translation['s3s3'] = {'cp': 'copy', 'sync': 'copy', 'mv': 'move'}
cmd_translation['s3local'] = {
'cp': 'download',
'sync': 'download',
'mv': 'move'
}
cmd_translation['s3'] = {
'rm': 'delete',
'mb': 'make_bucket',
'rb': 'remove_bucket'
}
result_queue = queue.Queue()
operation_name = cmd_translation[paths_type][self.cmd]
file_generator = FileGenerator(self._service,
self._source_endpoint,
operation_name,
self.parameters['follow_symlinks'],
self.parameters['page_size'],
result_queue=result_queue)
rev_generator = FileGenerator(self._service,
self._endpoint,
'',
self.parameters['follow_symlinks'],
self.parameters['page_size'],
result_queue=result_queue)
taskinfo = [
TaskInfo(src=files['src']['path'],
src_type='s3',
operation_name=operation_name,
service=self._service,
endpoint=self._endpoint)
]
stream_dest_path, stream_compare_key = find_dest_path_comp_key(files)
stream_file_info = [
FileInfo(src=files['src']['path'],
dest=stream_dest_path,
compare_key=stream_compare_key,
src_type=files['src']['type'],
dest_type=files['dest']['type'],
operation_name=operation_name,
service=self._service,
endpoint=self._endpoint,
is_stream=True)
]
file_info_builder = FileInfoBuilder(self._service, self._endpoint,
self._source_endpoint,
self.parameters)
s3handler = S3Handler(self.session,
self.parameters,
runtime_config=self._runtime_config,
result_queue=result_queue)
s3_stream_handler = S3StreamHandler(self.session,
self.parameters,
result_queue=result_queue)
sync_strategies = self.choose_sync_strategies()
command_dict = {}
if self.cmd == 'sync':
command_dict = {
'setup': [files, rev_files],
'file_generator': [file_generator, rev_generator],
'filters': [
create_filter(self.parameters),
create_filter(self.parameters)
],
'comparator': [Comparator(**sync_strategies)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]
}
elif self.cmd == 'cp' and self.parameters['is_stream']:
command_dict = {
'setup': [stream_file_info],
's3_handler': [s3_stream_handler]
}
elif self.cmd == 'cp':
command_dict = {
'setup': [files],
'file_generator': [file_generator],
'filters': [create_filter(self.parameters)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]
}
elif self.cmd == 'rm':
command_dict = {
'setup': [files],
'file_generator': [file_generator],
'filters': [create_filter(self.parameters)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]
}
elif self.cmd == 'mv':
command_dict = {
'setup': [files],
'file_generator': [file_generator],
'filters': [create_filter(self.parameters)],
'file_info_builder': [file_info_builder],
's3_handler': [s3handler]
}
elif self.cmd == 'mb':
command_dict = {'setup': [taskinfo], 's3_handler': [s3handler]}
elif self.cmd == 'rb':
command_dict = {'setup': [taskinfo], 's3_handler': [s3handler]}
files = command_dict['setup']
while self.instructions:
instruction = self.instructions.pop(0)
file_list = []
components = command_dict[instruction]
for i in range(len(components)):
if len(files) > len(components):
file_list.append(components[i].call(*files))
else:
file_list.append(components[i].call(files[i]))
files = file_list
# This is kinda quirky, but each call through the instructions
# will replaces the files attr with the return value of the
# file_list. The very last call is a single list of
# [s3_handler], and the s3_handler returns the number of
# tasks failed and the number of tasks warned.
# This means that files[0] now contains a namedtuple with
# the number of failed tasks and the number of warned tasks.
# In terms of the RC, we're keeping it simple and saying
# that > 0 failed tasks will give a 1 RC and > 0 warned
# tasks will give a 2 RC. Otherwise a RC of zero is returned.
rc = 0
if files[0].num_tasks_failed > 0:
rc = 1
if files[0].num_tasks_warned > 0:
rc = 2
return rc
def run(self):
"""
This function wires together all of the generators and completes
the command. First a dictionary is created that is indexed first by
the command name. Then using the instruction, another dictionary
can be indexed to obtain the objects corresponding to the
particular instruction for that command. To begin the wiring,
either a ``FileFormat`` or ``TaskInfo`` object, depending on the
command, is put into a list. Then the function enters a while loop
that pops off an instruction. It then determines the object needed
and calls the call function of the object using the list as the input.
Depending on the number of objects in the input list and the number
of components in the list corresponding to the instruction, the call
method of the component can be called two different ways. If the
number of inputs is equal to the number of components a 1:1 mapping of
inputs to components is used when calling the call function. If the
there are more inputs than components, then a 2:1 mapping of inputs to
components is used where the component call method takes two inputs
instead of one. Whatever files are yielded from the call function
is appended to a list and used as the input for the next repetition
of the while loop until there are no more instructions.
"""
src = self.parameters["src"]
dest = self.parameters["dest"]
paths_type = self.parameters["paths_type"]
files = FileFormat().format(src, dest, self.parameters)
rev_files = FileFormat().format(dest, src, self.parameters)
cmd_translation = {}
cmd_translation["locals3"] = {"cp": "upload", "sync": "upload", "mv": "move"}
cmd_translation["s3s3"] = {"cp": "copy", "sync": "copy", "mv": "move"}
cmd_translation["s3local"] = {"cp": "download", "sync": "download", "mv": "move"}
cmd_translation["s3"] = {"rm": "delete", "mb": "make_bucket", "rb": "remove_bucket"}
result_queue = queue.Queue()
operation_name = cmd_translation[paths_type][self.cmd]
fgen_kwargs = {
"client": self._source_client,
"operation_name": operation_name,
"follow_symlinks": self.parameters["follow_symlinks"],
"page_size": self.parameters["page_size"],
"result_queue": result_queue,
}
rgen_kwargs = {
"client": self._client,
"operation_name": "",
"follow_symlinks": self.parameters["follow_symlinks"],
"page_size": self.parameters["page_size"],
"result_queue": result_queue,
}
fgen_request_parameters = {}
fgen_head_object_params = {}
fgen_request_parameters["HeadObject"] = fgen_head_object_params
fgen_kwargs["request_parameters"] = fgen_request_parameters
# SSE-C may be neaded for HeadObject for copies/downloads/deletes
# If the operation is s3 to s3, the FileGenerator should use the
# copy source key and algorithm. Otherwise, use the regular
# SSE-C key and algorithm. Note the reverse FileGenerator does
# not need any of these because it is used only for sync operations
# which only use ListObjects which does not require HeadObject.
RequestParamsMapper.map_head_object_params(fgen_head_object_params, self.parameters)
if paths_type == "s3s3":
RequestParamsMapper.map_head_object_params(
fgen_head_object_params,
{
"sse_c": self.parameters.get("sse_c_copy_source"),
"sse_c_key": self.parameters.get("sse_c_copy_source_key"),
},
)
file_generator = FileGenerator(**fgen_kwargs)
rev_generator = FileGenerator(**rgen_kwargs)
taskinfo = [
TaskInfo(src=files["src"]["path"], src_type="s3", operation_name=operation_name, client=self._client)
]
stream_dest_path, stream_compare_key = find_dest_path_comp_key(files)
stream_file_info = [
FileInfo(
src=files["src"]["path"],
dest=stream_dest_path,
compare_key=stream_compare_key,
src_type=files["src"]["type"],
dest_type=files["dest"]["type"],
operation_name=operation_name,
client=self._client,
is_stream=True,
)
]
file_info_builder = FileInfoBuilder(self._client, self._source_client, self.parameters)
s3handler = S3Handler(
self.session, self.parameters, runtime_config=self._runtime_config, result_queue=result_queue
)
s3_stream_handler = S3StreamHandler(self.session, self.parameters, result_queue=result_queue)
sync_strategies = self.choose_sync_strategies()
command_dict = {}
if self.cmd == "sync":
command_dict = {
"setup": [files, rev_files],
"file_generator": [file_generator, rev_generator],
"filters": [create_filter(self.parameters), create_filter(self.parameters)],
"comparator": [Comparator(**sync_strategies)],
"file_info_builder": [file_info_builder],
"s3_handler": [s3handler],
}
elif self.cmd == "cp" and self.parameters["is_stream"]:
command_dict = {"setup": [stream_file_info], "s3_handler": [s3_stream_handler]}
elif self.cmd == "cp":
command_dict = {
"setup": [files],
"file_generator": [file_generator],
"filters": [create_filter(self.parameters)],
"file_info_builder": [file_info_builder],
"s3_handler": [s3handler],
}
elif self.cmd == "rm":
command_dict = {
"setup": [files],
"file_generator": [file_generator],
"filters": [create_filter(self.parameters)],
"file_info_builder": [file_info_builder],
"s3_handler": [s3handler],
}
elif self.cmd == "mv":
command_dict = {
"setup": [files],
"file_generator": [file_generator],
"filters": [create_filter(self.parameters)],
"file_info_builder": [file_info_builder],
"s3_handler": [s3handler],
}
elif self.cmd == "mb":
command_dict = {"setup": [taskinfo], "s3_handler": [s3handler]}
elif self.cmd == "rb":
command_dict = {"setup": [taskinfo], "s3_handler": [s3handler]}
files = command_dict["setup"]
while self.instructions:
instruction = self.instructions.pop(0)
file_list = []
components = command_dict[instruction]
for i in range(len(components)):
if len(files) > len(components):
file_list.append(components[i].call(*files))
else:
file_list.append(components[i].call(files[i]))
files = file_list
# This is kinda quirky, but each call through the instructions
# will replaces the files attr with the return value of the
# file_list. The very last call is a single list of
# [s3_handler], and the s3_handler returns the number of
# tasks failed and the number of tasks warned.
# This means that files[0] now contains a namedtuple with
# the number of failed tasks and the number of warned tasks.
# In terms of the RC, we're keeping it simple and saying
# that > 0 failed tasks will give a 1 RC and > 0 warned
# tasks will give a 2 RC. Otherwise a RC of zero is returned.
rc = 0
if files[0].num_tasks_failed > 0:
rc = 1
if files[0].num_tasks_warned > 0:
rc = 2
return rc