Create command to requeue outstanding ScannableURIs #539

Signed-off-by: Jono Yang <[email protected]>

minecode/management/commands/requeue_outstanding_scannableuris.py

@@ -0,0 +1,40 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# purldb is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/aboutcode-org/purldb for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+
+import logging
+import sys
+
+from minecode.management.commands import VerboseCommand
+from minecode.models import ScannableURI
+
+
+logger = logging.getLogger(__name__)
+logging.basicConfig(stream=sys.stdout)
+logger.setLevel(logging.INFO)
+
+TRACE = False
+if TRACE:
+    logger.setLevel(logging.DEBUG)
+
+
+class Command(VerboseCommand):
+    help = "Check for outstanding ScannableURIs and requeue them."
+
+    def handle(self, *args, **options):
+        """
+        Check for outstanding ScannableURIs and requeue them.
+        """
+
+        outstanding_scannable_uris = ScannableURI.objects.get_outstanding_scannable_resource_uris()
+        if outstanding_scannable_uris.exists():
+            logger.info(f"Resetting {outstanding_scannable_uris.count()} ScannableURIs")
+            for scannable_uri in outstanding_scannable_uris:
+                if TRACE:
+                    logger.debug(f"Resetting ScannableURI for: {scannable_uri}")
+                scannable_uri.reset()

minecode/models.py

@@ -7,6 +7,7 @@
 # See https://aboutcode.org for more information about nexB OSS projects.
 #
 
+import datetime
 import logging
 import sys
 import uuid
@@ -653,6 +654,14 @@ def _recent(self, scan_status, extra_value=None, most_recent=10):
                 recent[extra_value] = getattr(scauri, extra_value)
             yield recent
 
+    def get_outstanding_scannable_uris(self):
+        """
+        Return ScannableURIs that are older than a day from when this
+        queryset is called and that are being worked on.
+        """
+        time_delta = datetime.datetime.now() - datetime.timedelta(days=1)
+        return self.filter(scan_date__lte=time_delta, wip_date__isnull=False)
+
 
 class ScannableURI(BaseURI):
     """
@@ -818,6 +827,17 @@ def process_scan_results(
         )
         return job
 
+    def reset(self):
+        """
+        Reset fields such that a ScannableURI can be sent off for scanning again
+        """
+        self.scan_status = ScannableURI.SCAN_NEW
+        self.scan_error = None
+        self.index_error = None
+        self.scan_date = None
+        self.wip_date = None
+        self.save()
+
 
 # TODO: Use the QuerySet.as_manager() for more flexibility and chaining.
 class PriorityResourceURIManager(models.Manager):