def create_project_batch( # pylint: disable=too-many-arguments
project_id,
batch_type,
batch_name,
sample_ids,
host,
email,
password,
api_key,
):
"""Create a batch in a project."""
if sample_ids:
sample_ids = [s_id.strip() for s_id in sample_ids.split(",")]
echo_debug("Sample ids translation: {}".format(sample_ids))
else:
sample_ids = []
CreateBatch(
project_id,
batch_type,
batch_name,
sample_ids,
Credentials(email=email, password=password, api_key=api_key),
Optionals(host=host),
).run()
def download_file(destination, download_url, no_progress=False):
"""Download a file to file system.
Args:
destination (File): file object
download_url (str): url of the file to download
no_progress (bool): don't show progress bar
Returns:
str : file path
location of the downloaded file
"""
stream_params = dict(stream=True,
allow_redirects=False,
headers=dict(),
timeout=30)
with requests.get(download_url, **stream_params) as req:
req.raise_for_status()
echo_debug("Starting download")
if not no_progress:
pbar = get_progress_bar(int(req.headers["content-length"]),
"Downloading: ")
pbar.start()
for chunk in req.iter_content(chunk_size=CHUNK_SIZE):
destination.write(chunk)
if not no_progress:
pbar.update(pbar.value + len(chunk))
if not no_progress:
pbar.finish()
echo_debug("Finished downloading")
def validate_credentials(credentials):
"""Validate user credentials."""
if credentials.email and credentials.password and credentials.api_key:
echo_debug("User provided 2 sets of credentials.")
echo_warning("Multiple sets of credentials provided."
"Please provide either username/password or API key.")
return False
return True
def seek_files_to_upload(path, path_root=""):
"""Generate a list of valid fastq files."""
for root, dirs, files in os.walk(path):
files.sort()
for file in files:
file_path = os.path.join(path_root, root, file)
if file_path.lower().endswith(FASTQ_EXTENSIONS):
echo_debug("Found file to upload: {}".format(file_path))
yield file_path
dirs.sort()
for folder in dirs:
seek_files_to_upload(folder, root)
def build_file_path(deliverable, file_with_prefix, download_to, filename=None):
"""Create and return file system path where the file will be downloaded to.
Args:
deliverable (dict): used to get download url and file type
file_with_prefix (str): used as a template for download path
download_to (str): general location where to download the file to
Returns:
str : file path on current file system
"""
prefix = _get_prefix_parts(file_with_prefix)
download_url, file_type = "", ""
if isinstance(deliverable, SampleFile):
download_url, file_type = (
deliverable.download_url,
deliverable.file_type,
)
else:
download_url, file_type = (
deliverable.get("download_url"),
deliverable.get("file_type"),
)
source_filename = (filename if filename else
get_filename_from_download_url(download_url))
destination_filename = prefix.filename
if prefix.file_extension:
destination_filename = "{}.{}".format(prefix.filename,
prefix.file_extension)
# turning off formatting for improved code readability
# fmt: off
destination_filename = destination_filename.format(
**{
DownloadTemplateParts.FILE_TYPE.value:
FILE_TYPES_MAPPER.get(file_type) or file_type, # noqa: E501 # pylint: disable=line-too-long
DownloadTemplateParts.FILE_EXTENSION.value:
deliverable_type_from_filename(source_filename), # noqa: E501 # pylint: disable=line-too-long
DownloadTemplateParts.DEFAULT_FILENAME.value:
source_filename,
})
# fmt: on
echo_debug(
"Calculated destination filename: {}".format(destination_filename))
return _create_filepath(download_to, prefix.dirs, destination_filename)
def delete_project_samples( # pylint: disable=too-many-arguments
project_id,
sample_ids,
host,
email,
password,
api_key,
):
"""Delete samples in a project."""
sample_ids = ([s_id.strip()
for s_id in sample_ids.split(",")] if sample_ids else [])
echo_debug("Sample ids translation: {}".format(sample_ids))
DeleteSamples(
project_id,
sample_ids,
Credentials(email=email, password=password, api_key=api_key),
Optionals(host=host),
).run()
def get_filename_from_download_url(url):
"""Deduce filename from url.
Args:
url (str): URL string
Returns:
str: filename
"""
try:
filename = re.findall(
FILENAME_RE,
parse_qs(urlparse(url).query)["response-content-disposition"][0],
)[0]
except (KeyError, IndexError):
echo_debug("URL didn't contain filename query argument. "
"Assume filename from url")
filename = urlparse(url).path.split("/")[-1]
return filename
def login(api_client, credentials):
"""Login user into Gencove's system."""
if credentials.api_key:
echo_debug("User authorized via api key")
api_client.set_api_key(credentials.api_key)
return True
if not credentials.email or not credentials.password:
echo_info("Login required")
if not credentials.email:
credentials.email = click.prompt("Email", type=str, err=True)
if not credentials.password:
credentials.password = click.prompt("Password",
type=str,
hide_input=True,
err=True)
try:
api_client.login(credentials.email, credentials.password,
credentials.otp_token)
echo_debug("User logged in successfully")
return True
except APIClientError as err:
if "otp_token" in err.message:
echo_info("One time password required")
credentials.otp_token = click.prompt("One time password",
type=str,
err=True)
return login(api_client, credentials)
echo_debug("Failed to login: {}".format(err))
echo_error(
"Failed to login. Please verify your credentials and try again")
return False
def basespace_import( # pylint: disable=too-many-arguments
basespace_project_ids,
project_id,
metadata_json,
host,
email,
password,
api_key,
): # pylint: disable=line-too-long
"""Import all Biosamples from BaseSpace projects to a project. Optionally add
metadata to the samples.
Examples:
Import Biosamples to a project:
gencove basespace projects import 12345678 06a5d04b-526a-4471-83ba-fb54e0941758
Import Biosamples from multiple BaseSpace projects to a project:
gencove basespace projects import 12345678,87654321 06a5d04b-526a-4471-83ba-fb54e0941758
Import Biosamples to a project with metadata:
gencove basespace projects import 12345678 06a5d04b-526a-4471-83ba-fb54e0941758 --metadata-json='{"batch": "batch1"}'
""" # noqa: E501
basespace_project_ids = [
basespace_project_id.strip()
for basespace_project_id in basespace_project_ids.split(",")
]
echo_debug(
"BaseSpace project ids translation: {}".format(basespace_project_ids))
BaseSpaceImport(
basespace_project_ids,
project_id,
Credentials(email=email, password=password, api_key=api_key),
BaseSpaceImportOptionals(host=host, metadata_json=metadata_json),
).run()
def _create_filepath(download_to, prefix_dirs, filename):
"""Build full file path and ensure that directory structure exists.
Args:
download_to (str): top level directory path
prefix_dirs (str): subdirectories structure to create under
download_to.
filename (str): name of the file inside download_to/file_prefix
structure.
"""
echo_debug("_create_filepath Downloading to: {}".format(download_to))
echo_debug("_create_filepath file prefix is: {}".format(prefix_dirs))
path = os.path.join(download_to, prefix_dirs)
# Cross-platform cross-python-version directory creation
if not os.path.exists(path):
echo_debug("creating path: {}".format(path))
os.makedirs(path)
file_path = os.path.join(path, filename)
echo_debug("Deduced full file path is {}".format(file_path))
return file_path
def echo_debug(msg, **kwargs):
"""Output debug message."""
echo_debug(msg, **kwargs)
def deliverable_type_from_filename(filename):
"""Deduce deliverable type based on dot notation."""
filetype = ".".join(filename.split(".")[1:])
echo_debug("Deduced filetype to be: {} "
"from filename: {}".format(filetype, filename))
return filetype
def download_file(file_path,
download_url,
skip_existing=True,
no_progress=False):
"""Download a file to file system.
Args:
file_path (str): full file path, according to destination
and download template
download_url (str): url of the file to download
skip_existing (bool): skip already downloaded files
no_progress (bool): don't show progress bar
Returns:
str : file path
location of the downloaded file
"""
file_path_tmp = "{}.tmp".format(file_path)
if os.path.exists(file_path_tmp):
file_mode = "ab"
headers = dict(
Range="bytes={}-".format(os.path.getsize(file_path_tmp)))
echo_info("Resuming previous download: {}".format(file_path))
else:
file_mode = "wb"
headers = dict()
echo_info("Downloading file to {}".format(file_path))
stream_params = dict(stream=True,
allow_redirects=False,
headers=headers,
timeout=30)
with requests.get(download_url, **stream_params) as req:
req.raise_for_status()
total = int(req.headers["content-length"])
# pylint: disable=E0012,C0330
if (skip_existing and os.path.isfile(file_path)
and os.path.getsize(file_path) == total):
echo_info("Skipping existing file: {}".format(file_path))
return file_path
echo_debug("Starting to download file to: {}".format(file_path))
with open(file_path_tmp, file_mode) as downloaded_file:
if not no_progress:
pbar = get_progress_bar(int(req.headers["content-length"]),
"Downloading: ")
pbar.start()
for chunk in req.iter_content(chunk_size=CHUNK_SIZE):
downloaded_file.write(chunk)
if not no_progress:
pbar.update(pbar.value + len(chunk))
if not no_progress:
pbar.finish()
# Cross-platform cross-python-version file overwriting
if os.path.exists(file_path):
echo_debug("Found old file under same name: {}. "
"Removing it.".format(file_path))
os.remove(file_path)
os.rename(file_path_tmp, file_path)
echo_info("Finished downloading a file: {}".format(file_path))
return file_path
def _refresh_authentication(self):
echo_debug("Refreshing authentication")
jwt = self.refresh_token(self._jwt_refresh_token)
self._set_jwt(jwt.access)
def _request(
self,
endpoint="",
params=None,
method="get",
custom_headers=None,
timeout=60,
sensitive=False,
):
url = urljoin(text(self.host), text(endpoint))
headers = {
"content-type": "application/json",
"date": None,
"Gencove-cli-version": cli_version(),
}
if custom_headers:
headers.update(custom_headers)
if not params:
params = {}
echo_debug(
"Contacting url: {} with payload: {}".format(
url, "[SENSITIVE CONTENT]" if sensitive else params
)
)
start = time.time()
try:
if method == "get":
response = get(
url=url, params=params, headers=headers, timeout=timeout
)
elif method == "delete":
post_payload = APIClient._serialize_post_payload(params)
response = delete(
url=url,
data=post_payload,
headers=headers,
timeout=timeout,
)
else:
post_payload = APIClient._serialize_post_payload(params)
response = post(
url=url,
data=post_payload,
headers=headers,
timeout=timeout,
)
if response.status_code == 429:
raise APIClientTooManyRequestsError("Too Many Requests")
except (ConnectTimeout, ConnectionError):
# If request timed out,
# let upper level handle it the way it sees fit.
# one place might want to retry another might not.
raise APIClientTimeout( # pylint: disable=W0707
"Could not connect to the api server"
)
except ReadTimeout:
raise APIClientTimeout( # pylint: disable=W0707
"API server did not respond in timely manner"
)
echo_debug(
"API response is {} status is {} in {}ms".format(
"[SENSITIVE CONTENT]" if sensitive else response.content,
response.status_code,
(time.time() - start) * 1000,
)
)
# pylint: disable=no-member
if response.status_code >= 200 and response.status_code < 300:
return response.json() if response.text else {}
http_error_msg = ""
if 400 <= response.status_code < 500:
http_error_msg = "API Client Error: {}".format(response.reason)
if response.text:
response_json = response.json()
if "detail" in response_json:
http_error_msg += ": {}".format(response_json["detail"])
else:
try:
error_msg = "\n".join(
[
# create-batch can return error details that
# is a dict, not a list
" {}: {}".format(
key,
value[0]
if isinstance(value, list)
else str(value),
)
for key, value in response_json.items()
]
)
except AttributeError:
error_msg = "\n".join(response_json)
http_error_msg += ":\n{}".format(error_msg)
elif 500 <= response.status_code < 600:
http_error_msg = "Server Error: {}".format(response.reason)
raise APIClientError(http_error_msg, response.status_code)
def download( # pylint: disable=E0012,C0330,R0913
destination,
project_id,
sample_ids,
file_types,
skip_existing,
download_urls,
download_template,
host,
email,
password,
api_key,
no_progress,
): # noqa: D413,D301,D412 # pylint: disable=C0301
"""Download deliverables of a project.
Must specify either project id or sample ids.
Examples:
Download all samples results:
gencove download ./results --project-id d9eaa54b-aaac-4b85-92b0-0b564be6d7db
Download some samples:
gencove download ./results --sample-ids 59f5c1fd-cce0-4c4c-90e2-0b6c6c525d71,7edee497-12b5-4a1d-951f-34dc8dce1c1d
Download specific deliverables:
gencove download ./results --project-id d9eaa54b-aaac-4b85-92b0-0b564be6d7db --file-types alignment-bam,impute-vcf,fastq-r1,fastq-r2
Skip download entirely and print out the deliverables as a JSON:
gencove download - --project-id d9eaa54b-aaac-4b85-92b0-0b564be6d7db --download-urls
\f
Args:
destination (str): path/to/save/deliverables/to.
project_id (str): project id in Gencove's system.
sample_ids (list(str), optional): specific samples for which
to download the results. if not specified, download deliverables
for all samples.
file_types (list(str), optional): specific deliverables to download
results for. if not specified, all file types will be downloaded.
skip_existing (bool, optional, default True): skip downloading existing
files.
download_urls (bool, optional): output the files available for a
download. if the destination parameter is "-", it goes to the
stdout.
no_progress (bool, optional, default False): do not show progress
bar.
""" # noqa: E501
s_ids = tuple()
if sample_ids:
s_ids = tuple(s_id.strip() for s_id in sample_ids.split(","))
echo_debug("Sample ids translation: {}".format(s_ids))
f_types = tuple()
if file_types:
f_types = tuple(f_type.strip() for f_type in file_types.split(","))
echo_debug("File types translation: {}".format(f_types))
Download(
destination,
DownloadFilters(project_id=project_id,
sample_ids=s_ids,
file_types=f_types),
Credentials(email=email, password=password, api_key=api_key),
DownloadOptions(
host=host,
skip_existing=skip_existing,
download_template=download_template,
),
download_urls,
no_progress,
).run()
