Code changes to make this repo compatible with datalad-catalog>=1.1.0 #46
2 changed files with 51 additions and 83 deletions
|
|
@ -12,8 +12,9 @@ import sys
|
|||
import tempfile
|
||||
from uuid import uuid4
|
||||
|
||||
from datalad_catalog.catalog import Catalog
|
||||
from datalad.api import catalog_add
|
||||
from datalad_catalog.webcatalog import WebCatalog
|
||||
from datalad_catalog.schema_utils import get_metadata_item
|
||||
|
||||
# this points to the top of the ICF data store.
|
||||
# internally it will be amended with the missing components
|
||||
|
|
@ -32,9 +33,6 @@ dicom_metadata_keys = [
|
|||
"PulseSequenceName",
|
||||
]
|
||||
|
||||
# Instantiate interface object for api
|
||||
catalog_api = Catalog()
|
||||
|
||||
|
||||
def main(store_dir: str,
|
||||
study_id: str,
|
||||
|
|
@ -56,9 +54,12 @@ def main(store_dir: str,
|
|||
visit_id,
|
||||
dataset_metadata_path,
|
||||
file_metadata_path)
|
||||
add_to_catalog(visit_entry, str(study_catalog_path))
|
||||
catalog_add(
|
||||
catalog=study_catalog_path,
|
||||
metadata=visit_entry,
|
||||
)
|
||||
# Add visit entry as subdataset to study entry
|
||||
super_dict = read_json_file(ctlg.location / 'metadata' / 'super.json')
|
||||
super_dict = ctlg.get_main_dataset()
|
||||
subdatasets = [
|
||||
{
|
||||
'dataset_id': visit_entry['dataset_id'],
|
||||
|
|
@ -79,22 +80,24 @@ def get_catalog(study_id, catalog_path):
|
|||
""""""
|
||||
package_path = Path(__file__).resolve().parent.parent
|
||||
# Instantiate WebCatalog object
|
||||
ctlg = WebCatalog(
|
||||
location=str(catalog_path),
|
||||
config_file=str(package_path / 'assets' / 'catalog_config.json'),
|
||||
catalog_action='create',
|
||||
)
|
||||
ctlg = WebCatalog(location=str(catalog_path))
|
||||
# If catalog does not exist:
|
||||
if not ctlg.is_created():
|
||||
# 1. create it
|
||||
ctlg.create()
|
||||
ctlg.create(
|
||||
config_file=str(package_path / 'assets' / 'catalog_config.json'),
|
||||
)
|
||||
# 2. generate and add the study-level catalog entry
|
||||
study_entry = generate_study_entry(study_id)
|
||||
add_to_catalog(study_entry, str(catalog_path))
|
||||
catalog_add(
|
||||
catalog=catalog_path,
|
||||
metadata=study_entry,
|
||||
)
|
||||
# 3. set catalog home page
|
||||
ctlg.main_id = study_entry.get('dataset_id')
|
||||
ctlg.main_version = study_entry.get('dataset_version')
|
||||
ctlg.set_main_dataset()
|
||||
ctlg.set_main_dataset(
|
||||
dataset_id=study_entry.get('dataset_id'),
|
||||
dataset_version=study_entry.get('dataset_version'),
|
||||
)
|
||||
return ctlg
|
||||
|
||||
|
||||
|
|
@ -103,23 +106,32 @@ def generate_study_entry(study_id):
|
|||
desc=f"""This data catalog presents the DICOM data collected
|
||||
for all visits of the study: {study_id}. Browse through details
|
||||
of all study visits in the 'Subdatasets' tab below."""
|
||||
return new_dataset_meta_item(
|
||||
ds_id=str(uuid4()),
|
||||
ds_version='latest',
|
||||
ds_name=study_id,
|
||||
ds_description=desc)
|
||||
meta_item = get_metadata_item(
|
||||
item_type='dataset',
|
||||
dataset_id=str(uuid4()),
|
||||
dataset_version='latest',
|
||||
source_name='automated_addition',
|
||||
source_version='0.1.0',
|
||||
)
|
||||
meta_item['name'] = study_id
|
||||
meta_item['description'] = desc
|
||||
return meta_item
|
||||
|
||||
|
||||
def update_entry(ds_id, ds_version, ds_name, key, value, study_catalog_path):
|
||||
meta_item = {
|
||||
'type': 'dataset',
|
||||
'dataset_id': ds_id,
|
||||
'dataset_version': ds_version,
|
||||
'name': ds_name,
|
||||
'metadata_sources': get_metadata_source(),
|
||||
}
|
||||
meta_item = get_metadata_item(
|
||||
item_type='dataset',
|
||||
dataset_id=ds_id,
|
||||
dataset_version=ds_version,
|
||||
source_name='automated_addition',
|
||||
source_version='0.1.0',
|
||||
)
|
||||
meta_item['name'] = ds_name
|
||||
meta_item.update({key: value})
|
||||
add_to_catalog(meta_item, str(study_catalog_path))
|
||||
catalog_add(
|
||||
catalog=study_catalog_path,
|
||||
metadata=meta_item,
|
||||
)
|
||||
return meta_item
|
||||
|
||||
|
||||
|
|
@ -129,18 +141,21 @@ def generate_visit_entry(study_id, visit_id, metapath_dataset, metapath_file):
|
|||
desc=f"""This page presents the DICOM data collected for the visit
|
||||
{visit_id} during the imaging study {study_id}. Browse through details
|
||||
of this particular study visit in the 'DICOM' tab below."""
|
||||
meta_item = new_dataset_meta_item(
|
||||
ds_id=str(uuid4()),
|
||||
ds_version='latest',
|
||||
ds_name=visit_id,
|
||||
ds_description=desc)
|
||||
meta_item = get_metadata_item(
|
||||
item_type='dataset',
|
||||
dataset_id=str(uuid4()),
|
||||
dataset_version='latest',
|
||||
source_name='automated_addition',
|
||||
source_version='0.1.0',
|
||||
)
|
||||
meta_item['name'] = visit_id
|
||||
meta_item['description'] = desc
|
||||
# Load tarball metadata
|
||||
tar_metadata = read_json_file(metapath_dataset)
|
||||
expected_keys = ('size', 'md5', 'dspath', 'storepath')
|
||||
if not all(k in tar_metadata for k in expected_keys):
|
||||
raise ValueError(f'incomplete tarball metadata at {metapath_dataset}')
|
||||
# add dataset url
|
||||
|
||||
access_url_pre = 'datalad-annex::?type=external&externaltype=uncurl&url='
|
||||
access_url_post = '_{{annex_key}}&encryption=none'
|
||||
access_url = f'{access_url_pre}{icfstore_baseurl}/{study_id}/{visit_id}{access_url_post}'
|
||||
|
|
@ -176,14 +191,6 @@ def generate_visit_entry(study_id, visit_id, metapath_dataset, metapath_file):
|
|||
return meta_item
|
||||
|
||||
|
||||
def add_to_catalog(meta_entry: dict, catalog_dir: str ):
|
||||
""""""
|
||||
with tempfile.NamedTemporaryFile(mode="w+t") as f:
|
||||
json.dump(meta_entry, f)
|
||||
f.seek(0)
|
||||
res = catalog_api("add", catalog_dir=catalog_dir, metadata=f.name)
|
||||
|
||||
|
||||
def read_json_file(file_path):
|
||||
"""
|
||||
Load content from catalog metadata file for current node
|
||||
|
|
@ -197,45 +204,6 @@ def read_json_file(file_path):
|
|||
raise("Unexpected error:", sys.exc_info()[0])
|
||||
|
||||
|
||||
def get_gitconfig(conf_name):
|
||||
result = (
|
||||
subprocess.run(['git', 'config', conf_name], capture_output=True)
|
||||
.stdout.decode()
|
||||
.rstrip()
|
||||
)
|
||||
return result
|
||||
|
||||
|
||||
def get_metadata_source():
|
||||
"""Create metadata_sources dict required by catalog schema"""
|
||||
source = {
|
||||
'key_source_map': {},
|
||||
'sources': [
|
||||
{
|
||||
'source_name': 'automated_addition',
|
||||
'source_version': '0.1.0',
|
||||
'source_time': datetime.now().timestamp(),
|
||||
'agent_email': get_gitconfig('user.name'),
|
||||
'agent_name': get_gitconfig('user.email'),
|
||||
}
|
||||
],
|
||||
}
|
||||
return source
|
||||
|
||||
|
||||
def new_dataset_meta_item(ds_id, ds_version, ds_name = '', ds_description = ''):
|
||||
"""Create a minimal valid dataset metadata blob in catalog schema"""
|
||||
meta_item = {
|
||||
'type': 'dataset',
|
||||
'dataset_id': ds_id,
|
||||
'dataset_version': ds_version,
|
||||
'name': ds_name,
|
||||
'description': ds_description,
|
||||
'metadata_sources': get_metadata_source(),
|
||||
}
|
||||
return meta_item
|
||||
|
||||
|
||||
def format_bytes(bytes, decimals=2):
|
||||
if bytes == 0:
|
||||
return "0 Bytes"
|
||||
|
|
|
|||
|
|
@ -3,5 +3,5 @@ datalad-next
|
|||
pydicom
|
||||
pytest
|
||||
pytest-env
|
||||
datalad-catalog==0.2.1b0 --pre
|
||||
datalad-catalog
|
||||
www-authenticate
|
||||
|
I think ideally we wouldn't pin it, but under the circumstances of I think ideally we wouldn't pin it, but under the circumstances of `datalad-catalog` receiving breaking changes sometimes, and the future development being unclear at the moment, I would agree with pinning it here.
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue
I wonder if it would make sense to still pin the version, just to the current one, to avoid having to deal with future changes? Or is that unnecessary?