From e821857e7a9403739f321feb7418d33d6bd8b2c7 Mon Sep 17 00:00:00 2001 From: Peter Amstutz Date: Wed, 8 Jul 2020 17:10:10 -0400 Subject: Give uploader an option to trust local validation (for batch imports) Arvados-DCO-1.1-Signed-off-by: Peter Amstutz --- bh20sequploader/main.py | 17 ++++++++++++++--- bh20sequploader/qc_metadata.py | 10 +++++----- 2 files changed, 19 insertions(+), 8 deletions(-) diff --git a/bh20sequploader/main.py b/bh20sequploader/main.py index 56e0150..f744a8c 100644 --- a/bh20sequploader/main.py +++ b/bh20sequploader/main.py @@ -22,7 +22,7 @@ ARVADOS_API_HOST='lugli.arvadosapi.com' UPLOADER_API_TOKEN='2fbebpmbo3rw3x05ueu2i6nx70zhrsb1p22ycu3ry34m4x4462' ANONYMOUS_API_TOKEN='5o42qdxpxp5cj15jqjf7vnxx5xduhm4ret703suuoa3ivfglfh' UPLOAD_PROJECT='lugli-j7d0g-n5clictpuvwk8aa' -VALIDATED_PROJECT='lugli-j7d0g-n5clictpuvwk8aa' +VALIDATED_PROJECT='lugli-j7d0g-5ct8p1i1wrgyjvp' def qc_stuff(metadata, sequence_p1, sequence_p2, do_qc=True): failed = False @@ -69,9 +69,14 @@ def main(): parser.add_argument('sequence_p2', type=argparse.FileType('rb'), default=None, nargs='?', help='sequence FASTQ pair') parser.add_argument("--validate", action="store_true", help="Dry run, validate only") parser.add_argument("--skip-qc", action="store_true", help="Skip local qc check") + parser.add_argument("--trusted", action="store_true", help="Trust local validation and add directly to validated project") args = parser.parse_args() - api = arvados.api(host=ARVADOS_API_HOST, token=UPLOADER_API_TOKEN, insecure=True) + if args.trusted: + # Use credentials from environment + api = arvados.api() + else: + api = arvados.api(host=ARVADOS_API_HOST, token=UPLOADER_API_TOKEN, insecure=True) target = qc_stuff(args.metadata, args.sequence_p1, args.sequence_p2, not args.skip_qc) seqlabel = target[0][1] @@ -116,7 +121,13 @@ def main(): print("Duplicate of %s" % ([d["uuid"] for d in dup["items"]])) exit(1) - col.save_new(owner_uuid=UPLOAD_PROJECT, name="%s uploaded by %s from %s" % + if args.trusted: + properties["status"] = "validated" + owner_uuid = VALIDATED_PROJECT + else: + owner_uuid = UPLOAD_PROJECT + + col.save_new(owner_uuid=owner_uuid, name="%s uploaded by %s from %s" % (seqlabel, properties['upload_user'], properties['upload_ip']), properties=properties, ensure_unique_name=True) diff --git a/bh20sequploader/qc_metadata.py b/bh20sequploader/qc_metadata.py index 0a0c780..27657b1 100644 --- a/bh20sequploader/qc_metadata.py +++ b/bh20sequploader/qc_metadata.py @@ -16,12 +16,12 @@ def qc_metadata(metadatafile): if metadata_schema is None: schema_resource = pkg_resources.resource_stream(__name__, "bh20seq-schema.yml") cache = {"https://raw.githubusercontent.com/arvados/bh20-seq-resource/master/bh20sequploader/bh20seq-schema.yml": schema_resource.read().decode("utf-8")} - metadata_schema = schema_salad.schema.load_schema("https://raw.githubusercontent.com/arvados/bh20-seq-resource/master/bh20sequploader/bh20seq-schema.yml", cache=cache) + metadata_schema = schema_salad.schema.load_schema("https://raw.githubusercontent.com/arvados/bh20-seq-resource/master/bh20sequploader/bh20seq-schema.yml", cache=cache) - (document_loader, - avsc_names, - schema_metadata, - metaschema_loader) = metadata_schema + (document_loader, + avsc_names, + schema_metadata, + metaschema_loader) = metadata_schema shex = pkg_resources.resource_stream(__name__, "bh20seq-shex.rdf").read().decode("utf-8") -- cgit v1.2.3