README.md

@@ -48,20 +48,31 @@ The tool reads a token from the environment variable `DUMPTHINGS_TOKEN` if set.
 Move records from inboxes into the curated part of a collection.
 
 ```
-usage: auto-curate [-h] [--exclude [EXCLUDE ...]] [--list-labels] [--list-only] base_url collection
+usage: auto_curate [-h] [--destination-base-url DEST_SERVICE_URL] [--destination-collection DEST_COLLECTION] [--destination-token DEST_TOKEN] [--exclude [EXCLUDE ...]] [--list-labels] [--list-only] [-p PID]
+                   SOURCE_SERVICE_URL SOURCE_COLLECTION
+
+Automatically move records from the incoming areas of a collection to the curated area of the same collection, or to the incoming area of another collection.
 
 positional arguments:
-  base_url
-  collection
+  SOURCE_SERVICE_URL
+  SOURCE_COLLECTION
 
 options:
   -h, --help            show this help message and exit
-  --exclude, -e [EXCLUDE ...]   dont' move records from inbox with label EXCLUDE
-  --list-labels, -l     list all inbox labels, don't move any records
-  --list-only, -o       list all records in all inboxes, don't move any record
+  --destination-base-url DEST_SERVICE_URL
+                        select a different dump-thing-service, i.e. not SOURCE_SERVICE_URL, as destination for auto-curated records
+  --destination-collection DEST_COLLECTION
+                        select a different collection, i.e. not the SOURCE_COLLECTION of SOURCE_SERVICE_URL, as destination for auto-curated records
+  --destination-token DEST_TOKEN
+                        if provided, this token will be used for the destination service, otherwise ${CURATOR_TOKEN} will be used
+  --exclude, -e [EXCLUDE ...]
+                        exclude an inbox on the source collection
+  --list-labels, -l
+  --list-only, -o
+  -p, --pid PID         if provided, process only records that match the given PIDs. NOTE: matching does not involve CURIE-resolution!
 ```
 
-`auto-curate` requires that the environment variable `CURATOR_TOKEN` is set, and contains a valid curator-token
+`auto-curate` requires that the environment variable `CURATOR_TOKEN` is set, and contains a valid curator-token.
 
 
 #### build-local-triple-store

triple_tools/__about__.py

@@ -1 +1 @@
-__version__ = '0.1.0'
+__version__ = '0.2.0'

triple_tools/auto_curate.py

@@ -16,20 +16,54 @@ from triple_tools.communicate import (
 
 
 def main():
-    argument_parser = argparse.ArgumentParser()
-    argument_parser.add_argument('base_url')
-    argument_parser.add_argument('collection')
-    argument_parser.add_argument('--exclude', '-e', nargs='*', default=[])
+    argument_parser = argparse.ArgumentParser(
+        prog='auto_curate',
+        description="""
+            Automatically move records from the incoming areas of a
+            collection to the curated area of the same collection, or to
+            the incoming area of another collection.
+            """
+    )
+    argument_parser.add_argument('base_url', metavar='SOURCE_SERVICE_URL')
+    argument_parser.add_argument('collection', metavar='SOURCE_COLLECTION')
+    argument_parser.add_argument(
+        '--destination-base-url',
+        default=None,
+        metavar='DEST_SERVICE_URL',
+        help='select a different dump-thing-service, i.e. not SOURCE_SERVICE_URL, as destination for auto-curated records',
+    )
+    argument_parser.add_argument(
+        '--destination-collection',
+        default=None,
+        metavar='DEST_COLLECTION',
+        help='select a different collection, i.e. not the SOURCE_COLLECTION of SOURCE_SERVICE_URL, as destination for auto-curated records',
+    ),
+    argument_parser.add_argument(
+        '--destination-token',
+        default=None,
+        metavar='DEST_TOKEN',
+        help='if provided, this token will be used for the destination service, otherwise ${CURATOR_TOKEN} will be used',
+    )
+    argument_parser.add_argument('--exclude', '-e', nargs='*', default=[], help='exclude an inbox on the source collection')
     argument_parser.add_argument('--list-labels', '-l', action='store_true')
     argument_parser.add_argument('--list-only', '-o', action='store_true')
+    argument_parser.add_argument(
+        '-p', '--pid', action='append',
+        help='if provided, process only records that match the given PIDs. NOTE: matching does not involve CURIE-resolution!',
+    )
 
     arguments = argument_parser.parse_args()
+    print(arguments)
 
     curator_token = os.environ.get('CURATOR_TOKEN')
     if curator_token is None:
         print('ERROR: CURATOR_TOKEN not set', file=sys.stderr, flush=True)
         return 1
 
+    destination_url = arguments.destination_base_url or arguments.base_url
+    destination_collection = arguments.destination_collection or arguments.collection
+    destination_token = arguments.destination_token or curator_token
+
     for label in get_labels(
         url_base=arguments.base_url,
         collection=arguments.collection,
@@ -48,6 +82,10 @@ def main():
             label=label,
             token=curator_token
         ):
+            if arguments.pid:
+                if record['pid'] not in arguments.pid:
+                    continue
+
             if arguments.list_only:
                 print(f'{label}:\t{record}')
                 continue
@@ -55,8 +93,8 @@ def main():
             class_name = re.search('([_A-Za-z0-9]*$)', record['schema_type']).group(0)
             # Store record in collection
             post_to_url(
-                f'{arguments.base_url}/{arguments.collection}/curated/record/{class_name}',
-                token=curator_token,
+                f'{destination_url}/{destination_collection}/curated/record/{class_name}',
+                token=destination_token,
                 content=record,
             )