diff options
Diffstat (limited to 'bh20seqanalyzer')
-rw-r--r-- | bh20seqanalyzer/main.py | 31 |
1 files changed, 29 insertions, 2 deletions
diff --git a/bh20seqanalyzer/main.py b/bh20seqanalyzer/main.py index dae8eca..78e32c9 100644 --- a/bh20seqanalyzer/main.py +++ b/bh20seqanalyzer/main.py @@ -6,6 +6,8 @@ import subprocess import tempfile import json import logging +import ruamel.yaml +from bh20sequploader.qc_metadata import qc_metadata logging.basicConfig(format="[%(asctime)s] %(levelname)s %(message)s", datefmt="%Y-%m-%d %H:%M:%S", level=logging.INFO) @@ -20,9 +22,12 @@ def validate_upload(api, collection, validated_project): if "sequence.fasta" not in col: valid = False logging.warn("Upload '%s' missing sequence.fasta", collection["name"]) - if "metadata.jsonld" not in col: - logging.warn("Upload '%s' missing metadata.jsonld", collection["name"]) + if "metadata.yaml" not in col: + logging.warn("Upload '%s' missing metadata.yaml", collection["name"]) valid = False + else: + metadata_content = ruamel.yaml.round_trip_load(col.open("metadata.yaml")) + valid = qc_metadata(metadata_content) and valid dup = api.collections().list(filters=[["owner_uuid", "=", validated_project], ["portable_data_hash", "=", col.portable_data_hash()]]).execute() @@ -79,12 +84,31 @@ def start_analysis(api, logging.error(comp.stderr.decode('utf-8')) +def copy_most_recent_result(api, analysis_project, latest_result_uuid): + most_recent_analysis = api.groups().list(filters=[['owner_uuid', '=', analysis_project]], + order="created_at desc", limit=1).execute() + for m in most_recent_analysis["items"]: + cr = api.container_requests().list(filters=[['owner_uuid', '=', m["uuid"]], + ["requesting_container_uuid", "=", None]]).execute() + if cr["items"] and cr["items"][0]["output_uuid"]: + wf = cr["items"][0] + src = api.collections().get(uuid=wf["output_uuid"]).execute() + dst = api.collections().get(uuid=latest_result_uuid).execute() + if src["portable_data_hash"] != dst["portable_data_hash"]: + logging.info("Copying latest result from '%s' to %s", m["name"], latest_result_uuid) + api.collections().update(uuid=latest_result_uuid, + body={"manifest_text": src["manifest_text"], + "description": "latest result from %s %s" % (m["name"], wf["uuid"])}).execute() + break + + def main(): parser = argparse.ArgumentParser(description='Analyze collections uploaded to a project') parser.add_argument('--uploader-project', type=str, default='lugli-j7d0g-n5clictpuvwk8aa', help='') parser.add_argument('--analysis-project', type=str, default='lugli-j7d0g-y4k4uswcqi3ku56', help='') parser.add_argument('--validated-project', type=str, default='lugli-j7d0g-5ct8p1i1wrgyjvp', help='') parser.add_argument('--workflow-uuid', type=str, default='lugli-7fd4e-mqfu9y3ofnpnho1', help='') + parser.add_argument('--latest-result-uuid', type=str, default='lugli-4zz18-z513nlpqm03hpca', help='') args = parser.parse_args() api = arvados.api() @@ -101,4 +125,7 @@ def main(): start_analysis(api, args.analysis_project, args.workflow_uuid, args.validated_project) + + copy_most_recent_result(api, args.analysis_project, args.latest_result_uuid) + time.sleep(10) |