def test_downloadable_files_additional_metadata_default(clean_db):
TrialMetadata.create(TRIAL_ID, METADATA)
df = DownloadableFiles(
trial_id=TRIAL_ID,
upload_type="wes_bam",
object_url="10021/Patient 1/sample 1/aliquot 1/wes_forward.fastq",
file_size_bytes=1,
md5_hash="hash1234",
uploaded_timestamp=datetime.now(),
)
# Check no value passed
df.insert()
assert df.additional_metadata == {}
for nullish_value in [None, "null", {}]:
df.additional_metadata = nullish_value
df.update()
assert df.additional_metadata == {}
# Non-nullish value doesn't get overridden
non_nullish_value = {"foo": "bar"}
df.additional_metadata = non_nullish_value
df.update()
assert df.additional_metadata == non_nullish_value
def create_df(facet_group, additional_metadata={}) -> DownloadableFiles:
df = DownloadableFiles(
facet_group=facet_group,
additional_metadata=additional_metadata,
trial_id=TRIAL_ID,
uploaded_timestamp=datetime.now(),
file_size_bytes=0,
object_url=facet_group, # just filler, not relevant to the test
upload_type="",
)
df.insert()
clean_db.refresh(df)
return df
def test_get_filelist(cidc_api, clean_db, monkeypatch):
"""Check that getting a filelist.tsv works as expected"""
user_id = setup_user(cidc_api, monkeypatch)
file_id_1, file_id_2 = setup_downloadable_files(cidc_api)
client = cidc_api.test_client()
url = "/downloadable_files/filelist"
# A JSON body containing a file ID list must be provided
res = client.post(url)
assert res.status_code == 422
# User has no permissions, so no files should be found
short_file_list = {"file_ids": [file_id_1, file_id_2]}
res = client.post(url, json=short_file_list)
assert res.status_code == 404
# Give the user one permission
with cidc_api.app_context():
perm = Permissions(
granted_to_user=user_id,
trial_id=trial_id_1,
upload_type=upload_types[0],
granted_by_user=user_id,
)
perm.insert()
# User has one permission, so the filelist should contain a single file
res = client.post(url, json=short_file_list)
assert res.status_code == 200
assert "text/tsv" in res.headers["Content-Type"]
assert "filename=filelist.tsv" in res.headers["Content-Disposition"]
assert res.data.decode("utf-8") == (
f"gs://{GOOGLE_DATA_BUCKET}/{trial_id_1}/wes/.../reads_123.bam\t{trial_id_1}_wes_..._reads_123.bam\n"
)
# Admins don't need permissions to get files
make_admin(user_id, cidc_api)
res = client.post(url, json=short_file_list)
assert res.status_code == 200
assert res.data.decode("utf-8") == (
f"gs://{GOOGLE_DATA_BUCKET}/{trial_id_1}/wes/.../reads_123.bam\t{trial_id_1}_wes_..._reads_123.bam\n"
f"gs://{GOOGLE_DATA_BUCKET}/{trial_id_2}/cytof/.../analysis.zip\t{trial_id_2}_cytof_..._analysis.zip\n"
)
# Clear inserted file records
with cidc_api.app_context():
clean_db.query(DownloadableFiles).delete()
# Filelists don't get paginated
ids = []
with cidc_api.app_context():
for id in range(1000):
df = DownloadableFiles(
trial_id=trial_id_1,
object_url=str(id),
upload_type="",
file_size_bytes=0,
uploaded_timestamp=datetime.now(),
)
df.insert()
ids.append(df.id)
res = client.post(url, json={"file_ids": ids})
assert res.status_code == 200
# newly inserted files + EOF newline
assert len(res.data.decode("utf-8").split("\n")) == len(ids) + 1