def download_file(self, context, target_file):
check.str_param(target_file, 'target_file')
target_path = os.path.join(self.target_folder, target_file)
if self.skip_if_present and safe_isfile(target_path):
context.log.info(
'Skipping download, file already present at {target_path}'.
format(target_path=target_path))
else:
full_key = self.key + '/' + target_file
if os.path.dirname(target_path):
mkdir_p(os.path.dirname(target_path))
context.log.info(
'Starting download of {bucket}/{key} to {target_path}'.format(
bucket=self.bucket, key=full_key, target_path=target_path))
headers = context.resources.s3.head_object(Bucket=self.bucket,
Key=full_key)
logger = S3Logger(context.log.debug, self.bucket, full_key,
target_path, int(headers['ContentLength']))
context.resources.s3.download_file(Bucket=self.bucket,
Key=full_key,
Filename=target_path,
Callback=logger)
return target_path
def _download_from_s3_to_file(session, context, bucket, key, target_folder, skip_if_present):
# TODO: remove context argument once we support resource logging
# file name is S3 key path suffix after last /
target_file = os.path.join(target_folder, key.split('/')[-1])
if skip_if_present and safe_isfile(target_file):
context.log.info(
'Skipping download, file already present at {target_file}'.format(
target_file=target_file
)
)
else:
if not os.path.exists(target_folder):
mkdir_p(target_folder)
context.log.info(
'Starting download of {bucket}/{key} to {target_file}'.format(
bucket=bucket, key=key, target_file=target_file
)
)
headers = session.head_object(Bucket=bucket, Key=key)
logger = S3Logger(
context.log.debug, bucket, key, target_file, int(headers['ContentLength'])
)
session.download_file(Bucket=bucket, Key=key, Filename=target_file, Callback=logger)
return target_file
def execute_create_notebook(notebook, force_overwrite, **kwargs):
if not re.match(r'^[a-zA-Z0-9\-_\\/]+$', notebook):
raise click.BadOptionUsage(
notebook,
('Notebook name {name} is not valid, '
'cannot contain anything except alphanumeric characters, '
'-, _, \\ and / for path manipulation').format(name=notebook),
)
notebook_path = os.path.join(
os.getcwd(),
notebook if notebook.endswith('.ipynb') else notebook + ".ipynb")
notebook_dir = os.path.dirname(notebook_path)
if not os.path.exists(notebook_dir):
os.makedirs(notebook_dir)
if not force_overwrite and safe_isfile(notebook_path):
click.confirm(
('Warning, {notebook_path} already exists and continuing '
'will overwrite the existing notebook. '
'Are you sure you want to continue?').format(
notebook_path=notebook_path),
abort=True,
)
register_repo_info = get_register_repo_info(kwargs)
with open(notebook_path, 'w') as f:
f.write(get_notebook_scaffolding(register_repo_info))
click.echo("Created new dagstermill notebook at {path}".format(
path=notebook_path))
def download_from_s3(context):
'''Download an object from s3.
Args:
info (ExpectationExecutionInfo): Must expose a boto3 S3 client as its `s3` resource.
Returns:
str:
The path to the downloaded object.
'''
results = []
for file_ in context.solid_config:
bucket = file_['bucket']
key = file_['key']
target_path = file_.get('target_path') or key
if target_path is None:
target_path = context.resources.tempfile.tempfile().name
if file_['skip_if_present'] and safe_isfile(target_path):
context.log.info(
'Skipping download, file already present at {target_path}'.
format(target_path=target_path))
else:
if os.path.dirname(target_path):
mkdir_p(os.path.dirname(target_path))
context.resources.s3.download_file(bucket, key, target_path)
results.append(target_path)
return results
def download_from_s3(context):
(bucket, key, target_folder,
skip_if_present) = (context.solid_config.get(k)
for k in ('bucket', 'key', 'target_folder',
'skip_if_present'))
# file name is S3 key path suffix after last /
target_file = os.path.join(target_folder, key.split('/')[-1])
if skip_if_present and safe_isfile(target_file):
context.log.info(
'Skipping download, file already present at {target_file}'.format(
target_file=target_file))
else:
if not os.path.exists(target_folder):
mkdir_p(target_folder)
context.log.info(
'Starting download of {bucket}/{key} to {target_file}'.format(
bucket=bucket, key=key, target_file=target_file))
s3 = boto3.client('s3')
headers = s3.head_object(Bucket=bucket, Key=key)
logger = S3Logger(context.log.debug, bucket, key, target_file,
int(headers['ContentLength']))
s3.download_file(Bucket=bucket,
Key=key,
Filename=target_file,
Callback=logger)
return target_file
def file_exists_at_path_type_check(value):
if not isinstance(value, six.string_types):
raise Failure(
'FileExistsAtPath must be a string in memory. Got {value}'.format(
value=repr(value)))
if not safe_isfile(value):
raise Failure(
('FileExistsAtPath must be a path that points to a file that '
'exists. "{value}" does not exist on disk').format(value=value))
def unzip_file(
context,
archive_paths,
archive_members,
# destination_dir=None
):
# FIXME
# archive_path = info.config['archive_path']
# archive_member = info.config['archive_member']
results = []
for (i, archive_path) in enumerate(archive_paths):
destination_dir = (
# info.config['destination_dir'] or
os.path.dirname(archive_path))
if archive_members:
archive_member = archive_members[i]
else:
archive_member = None
with zipfile.ZipFile(archive_path, 'r') as zip_ref:
if archive_member is not None:
target_path = os.path.join(destination_dir, archive_member)
is_file = safe_isfile(target_path)
is_dir = os.path.isdir(target_path)
if not (context.solid_config['skip_if_present'] and
(is_file or is_dir)):
zip_ref.extract(archive_member, destination_dir)
else:
if is_file:
context.log.info(
'Skipping unarchive of {archive_member} from {archive_path}, '
'file already present at {target_path}'.format(
archive_member=archive_member,
archive_path=archive_path,
target_path=target_path,
))
if is_dir:
context.log.info(
'Skipping unarchive of {archive_member} from {archive_path}, '
'directory already present at {target_path}'.
format(
archive_member=archive_member,
archive_path=archive_path,
target_path=target_path,
))
else:
if not (context.solid_config['skip_if_present'] and is_dir):
zip_ref.extractall(destination_dir)
else:
context.log.info(
'Skipping unarchive of {archive_path}, directory already present '
'at {target_path}'.format(archive_path=archive_path,
target_path=target_path))
results.append(target_path)
return results
def execute_create_notebook(notebook, solid_name, force_overwrite, **kwargs):
if not re.match(r'^[a-zA-Z0-9\-_\\/]+$', notebook):
raise click.BadOptionUsage(
notebook,
('Notebook name {name} is not valid, '
'cannot contain anything except alphanumeric characters, '
'-, _, \\ and / for path manipulation').format(name=notebook),
)
notebook_path = os.path.join(
os.getcwd(),
notebook if notebook.endswith('.ipynb') else notebook + ".ipynb")
notebook_dir = os.path.dirname(notebook_path)
if not os.path.exists(notebook_dir):
os.makedirs(notebook_dir)
if not force_overwrite and safe_isfile(notebook_path):
click.confirm(
('Warning, {notebook_path} already exists and continuing '
'will overwrite the existing notebook. '
'Are you sure you want to continue?').format(
notebook_path=notebook_path),
abort=True,
)
if not solid_name:
solid_name = os.path.basename(notebook_path).split(".")[0]
repository_target_info = load_target_info_from_cli_args(kwargs)
module_target_info = get_module_target_function(repository_target_info)
if module_target_info:
module = module_target_info.module_name
fn_name = module_target_info.fn_name
RegisterRepoInfo = namedtuple(
'RegisterRepoInfo', 'import_statement declaration_statement')
register_repo_info = RegisterRepoInfo(
"from {module} import {fn_name}".format(module=module,
fn_name=fn_name),
"dm.declare_as_solid({fn_name}(), '{solid_name}')".format(
fn_name=fn_name, solid_name=solid_name),
)
else:
raise click.UsageError(
"Cannot instantiate notebook with repository definition given by a function from a file"
)
with open(notebook_path, 'w') as f:
f.write(get_notebook_scaffolding(register_repo_info))
click.echo("Created new dagstermill notebook at {path}".format(
path=notebook_path))
def gunzipper(_, gzip_file):
"""gunzips /path/to/foo.gz to /path/to/raw/2019/01/01/data.json"""
# TODO: take date as an input
path_prefix = os.path.dirname(gzip_file)
output_folder = os.path.join(path_prefix, "raw/2019/01/01")
outfile = os.path.join(output_folder, "data.json")
if not safe_isfile(outfile):
mkdir_p(output_folder)
with gzip.open(gzip_file, "rb") as f_in, open(outfile, "wb") as f_out:
shutil.copyfileobj(f_in, f_out)
return [path_prefix]
def gunzipper(_, gzip_file):
'''gunzips /path/to/foo.gz to /path/to/raw/2019/01/01/data.json
'''
# TODO: take date as an input
path_prefix = os.path.dirname(gzip_file)
output_folder = os.path.join(path_prefix, 'raw/2019/01/01')
outfile = os.path.join(output_folder, 'data.json')
if not safe_isfile(outfile):
mkdir_p(output_folder)
with gzip.open(gzip_file, 'rb') as f_in, open(outfile, 'wb') as f_out:
shutil.copyfileobj(f_in, f_out)
return [path_prefix]
def file_exists_at_path_type_check(_, value):
if not isinstance(value, six.string_types):
return TypeCheck(
success=False,
description='FileExistsAtPath must be a string in memory. Got {value}'.format(
value=repr(value)
),
)
if not safe_isfile(value):
return TypeCheck(
success=False,
description=(
'FileExistsAtPath must be a path that points to a file that '
'exists. "{value}" does not exist on disk'
).format(value=value),
)
return True
def execute_create_notebook(notebook, force_overwrite, kernel):
notebook_path = os.path.join(
os.getcwd(),
notebook if notebook.endswith('.ipynb') else notebook + ".ipynb")
notebook_dir = os.path.dirname(notebook_path)
mkdir_p(notebook_dir)
if not force_overwrite and safe_isfile(notebook_path):
click.confirm(
('Warning, {notebook_path} already exists and continuing '
'will overwrite the existing notebook. '
'Are you sure you want to continue?').format(
notebook_path=notebook_path),
abort=True,
)
with open(notebook_path, 'w') as f:
f.write(get_notebook_scaffolding(get_kernelspec(kernel)))
click.echo("Created new dagstermill notebook at {path}".format(
path=notebook_path))
def test_safe_isfile():
assert safe_isfile(file_relative_path(__file__, 'test_file_utils.py'))
assert not safe_isfile(file_relative_path(__file__, 'not_a_file.py'))
def test_safe_isfile():
assert safe_isfile(script_relative_path('test_safe_isfile.py'))
assert not safe_isfile(script_relative_path('test_safe_isfile_foobar.py'))
def test_safe_isfile():
assert safe_isfile(script_relative_path('test_file_utils.py'))
assert not safe_isfile(script_relative_path('not_a_file.py'))
