Arvados-DCO-1.1-Signed-off-by: Radhika Chippada <radhika@curoverse.com>

[arvados.git] / crunch_scripts / collection-merge

diff --git a/crunch_scripts/collection-merge b/crunch_scripts/collection-merge
index f16d62466a1e1853b56266ed58ef9935fd3f974b..f3aa5ce9cf5458aa6335ebf42be43c60074fc9fe 100755 (executable)
--- a/crunch_scripts/collection-merge
+++ b/crunch_scripts/collection-merge
@@ -1,8 +1,24 @@
 #!/usr/bin/env python
+# Copyright (C) The Arvados Authors. All rights reserved.
+#
+# SPDX-License-Identifier: Apache-2.0
+
+# collection-merge
+#
+# Merge two or more collections together.  Can also be used to extract specific
+# files from a collection to produce a new collection.
+#
+# input:
+# An array of collections or collection/file paths in script_parameter["input"]
+#
+# output:
+# A manifest with the collections merged.  Duplicate file names will
+# have their contents concatenated in the order that they appear in the input
+# array.
 
 import arvados
 import md5
-import subst
+import crunchutil.subst as subst
 import subprocess
 import os
 import hashlib
@@ -30,28 +46,4 @@ for c in p["input"]:
                 if fn in s.files():
                     merged += s.files()[fn].as_manifest()
 
-crm = arvados.CollectionReader(merged)
-
-combined = crm.manifest_text(strip=True)
-
-m = hashlib.new('md5')
-m.update(combined)
-
-uuid = "{}+{}".format(m.hexdigest(), len(combined))
-
-collection = arvados.api().collections().create(
-    body={
-        'uuid': uuid,
-        'manifest_text': crm.manifest_text(),
-    }).execute()
-
-for s in src:
-    l = arvados.api().links().create(body={
-        "link": {
-            "tail_uuid": s,
-            "head_uuid": uuid,
-            "link_class": "provenance",
-            "name": "provided"
-        }}).execute()
-
-arvados.current_task().set_output(uuid)
+arvados.current_task().set_output(merged)